This is the sixth iteration of Head and Neck Cancer: United Kingdom National Multidisciplinary Guidelines, produced by a broad range of UK professionals involved in the care of patients with head and neck cancer (see Author List).
The Guidelines are divided into 30 chapters that aim to cover all aspects of care for patients with head and neck cancer, ranging from the provision of services, management in general, to the management of specific tumours. Rather than a simple update to the last iteration, most chapters have been rewritten. The major change is the increased emphasis on multidisciplinary care. With that in mind, the Guidelines are split into three sections. These are: (Section 1) general issues underpinning the basis of care of patients with head and neck cancer, which apply across the spectrum of management; (Section 2) patient support, describing mainly the support required throughout the patient journey, and the pre-, peri- and post-treatment provided by various health professionals; and (Section 3) site-specific chapters, as in the previous iteration. An addition to these Guidelines is a chapter dealing with head and neck paraganglioma. Although not usually a malignant disease, it is an area that involves input from a variety of disciplines and is probably subject to significant variation in management.
The Guidelines have been authored by over 170 UK-based authors covering 22 specialties (anaesthetics, clinical nurse specialists, dermatology, dietetics, endocrinology, endocrine surgery, otolaryngology – head and neck surgery (ENT), genetics, oral maxillofacial surgery, oncology, oral surgery, palliative care, pathology, physiotherapy, pharmacology, plastic surgery, psychology, public health, radiology, restorative dentistry, speech and language therapy, and vascular surgery).
The Guidelines have been endorsed by 12 national bodies:
• Association of Palliative Medicine (‘APM’)
• British Association of Endocrine and Thyroid Surgeons (‘BAETS’)
• British Association of Head and Neck Oncologists (‘BAHNO’)
• British Association of Head and Neck Oncology Nurses (‘BAHNON’)
• British Association of Oral and Maxillofacial Surgeons (‘BAOMS’)
• British Association of Plastic Reconstructive and Aesthetic Surgeons (‘BAPRAS’)
• British Dietetic Association
• ENT-UK
• Restorative Dentistry – UK
• Royal College of Speech and Language Therapists
• The Royal College of Pathologists (‘RCPath’)
• The Royal College of Radiologists (Faculty of Clinical Oncology)
The aim has been to set standards of care and best practice. The Guidelines can be also used as an educational resource for clinicians, and for insight into head and neck cancer care for non-clinicians involved in commissioning and managing services for patients with head and neck cancer.
As with the last Guidelines, summary recommendations appear throughout the document. These are qualified by ‘(R)’ or ‘(G)’. ‘(R)’ can be taken as an accepted standard of care and/or with a strong evidence base (an evidence-based recommendation). It can be interpreted as equivalent to ‘offer’ in National Institute for Health and Care Excellence (NICE) guideline parlance. ‘(G)’ can be taken as something that should be at least considered, with some supporting evidence (a good practice point). It can be interpreted as equivalent to ‘consider’ in NICE guideline parlance. When recommendations concern the provision of services, these terms may be replaced with ‘(E)’ (essential) and ‘(D)’ (desirable).
The Guidelines are guidelines, and not black and white diktats of what constitutes acceptable and non-acceptable care. Every patient has different individual circumstances (tumour, patient, environment), and care is therefore variably nuanced.
The variety of treatment options can be difficult for a patient to navigate through. It is thus critical that the clinical team help patients to make very difficult decisions, with the provision of clear and understandable information on the risks and benefits associated with treatments offered.
There have been some significant developments in practice since the last Guidelines. These mainly have involved increasing recognition and understanding of the nature of human papillomavirus related squamous cell carcinomas, increasing experience of systemic therapies, in particular immunotherapy, and the updated staging classification (TNM Classification of Malignant Tumours, eighth edition) by the American Joint Committee on Cancer (‘AJCC’) and Union for International Cancer Control (‘UICC’). However, perhaps the most powerful area of change is the continued development of pre-habilitation, which features prominently in Section 2, and involves input from various health professionals with large areas of overlap.
Chapters on non-melanoma skin cancer and cancer of an unknown primary follow previous consensus-gathering initiatives, with more detail regarding evidence sources and wider input by those taking part in these consultations. This may be a model to follow in future.
The authors of each chapter were invited to suggest areas in which research or evidence is required, and to indicate important on-going clinical trials that are due to report in the next few years.
UK clinical trials can be found via the National Cancer Research Institute, which supports a wide variety of clinical trials in head and neck cancer (https://www.ncri.org.uk (search portfolio maps)). Major USA trials undertaken by the NRG co-operative group can be viewed here: https://www.nrgoncology.org/Clinical-Trials/Protocol-Search. Other global trials can be found at: https://clinicaltrials.gov.
The guidelines have been generously peer reviewed (Appendix 1).
Association for Palliative Medicine (APM)
British Association of Endocrine & Thyroid Surgeons (BAETS)
British Association of Head & Neck Oncologists (BAHNO)
It gives me great pleasure to write a foreword for the 6th edition of the Head and Neck Cancer: United Kingdom National Multidisciplinary Guidelines.
In the five years since the last edition was published, a significant evidence base has accrued in the management of head and neck cancer, especially for oropharyngeal cancer and recurrent metastatic cancer. Additionally, the impact of the coronavirus disease pandemic, which has been felt across all domains of clinical practice, has certainly changed practice in the realms of cancer diagnoses.
Professors Jarrod Homer and Stuart Winter are to be congratulated on compiling this edition of the guidelines, with comprehensive multidisciplinary stakeholder engagement, which significantly builds on the previous edition. Having very carefully sieved through the 30 chapters that comprise the guidelines, I believe this truly captures the best of British expertise in head and neck cancer care. The multidisciplinary authorship reflects the clinicians, researchers and thought leaders in the specialty, who make recommendations based on existing evidence and current practice. I am only too well aware of the long and arduous hours it takes to put together a work of this kind, and the quality of the contributions indicates that the guidelines will be well received by the head and neck oncology community.
In addition to the clinical recommendations, these guidelines have also used a new approach to generate consensus across areas that are not supported by high-level evidence. Working collaboratively with six stakeholder organisations, novel comprehensive methodology was used to achieve consensus recommendations; this is demonstrated to good effect for the unknown primary.
On behalf of the British Association of Head & Neck Oncologists, I whole heartedly endorse these guidelines for practice within these Isles and beyond.
British Association of Head & Neck Oncology Nurses (BAHNON)
On behalf of the British Association of Head & Neck Oncology Nurses Committee, we are delighted to endorse this document. This multidisciplinary document will have a positive impact on the delivery of the highest standard of research-based patient care for many years to come. It will be a valuable resource to all members of the multidisciplinary team. I would like to take the opportunity to thank the authors and editors for all their hard work.
British Association of Oral and Maxillofacial Surgeons (BAOMS)
On behalf of the British Association of Oral and Maxillofacial Surgeons, as the Oncology Subspecialty Interest Group Leads, it is a privilege to write a foreword for this essential scientific document.
Best practice is driven by robust guidelines, which in turn must derive from a high-quality evidence base. The complexity and relatively low incidence of head and neck cancer renders this task extremely challenging.
The 6th edition of Head and Neck Cancer: United Kingdom National Multidisciplinary Guidelines embrace the current best available evidence, with comprehensive multidisciplinary specialist team consensus, for practices in caring for head and neck cancer patients.
The publication addresses the challenges met in the complex field of head and neck cancer, but also offers succinct insight into new scientific approaches, such as non-surgical systemic treatments and minimally invasive procedures, either for diagnosis or definitive management.
Whilst cure is the primary goal in the management of patients with head and neck cancer, the document also guides us well regarding the palliative patient, with a focus on quality of life, and updates on psychosocial support and targeted novel treatments.
With a high calibre authorship of renowned experts, this guidance document undoubtedly offers a state-of-the-art overview in head and neck cancer and its management.
We congratulate the cohesive collegiate teamwork in producing this superb set of guidelines.
British Association of Plastic, Reconstructive and Aesthetic Surgeons (BAPRAS)
On behalf of the British Association of Plastic, Reconstructive and Aesthetic Surgeons, I am delighted to endorse the publication of these guidelines, and for British Association of Plastic, Reconstructive and Aesthetic Surgeons to be involved in such a collaborative effort, bringing together the knowledge of a truly multidisciplinary group of specialists involved in the management of patients suffering from cancer of the head and neck. The latest guidelines will continue to support multidisciplinary teams and to help provide the best possible outcome for patients.
ENT-UK
I am delighted to write a short introduction to the 6th edition of the Head and Neck Cancer: United Kingdom National Multidisciplinary Guidelines. I have been privileged to contribute to all of the previous excellent editions, and without doubt this edition of the guidelines is the most comprehensive to date. The multidisciplinary team (MDT) model is the core of head and neck services in the UK, and the list of authors clearly demonstrates the truly multidisciplinary nature of these guidelines. There was no shortage of excellent authors from all specialties who were keen to contribute to these guidelines, which indicates how important this established publication has become amongst the head and neck specialty. Stuart Winter and Jarrod Homer should be congratulated on the tremendous amount of work they have put in, and it is reflected in a comprehensive, concise, contemporary and eminently readable textbook. These guidelines will be essential reading for all those involved in the management of head and neck cancer, and will form an excellent reference to help inform MDT decisions.
Restorative Dentistry – UK (RD-UK)
We are fortunate to have so many talented and dedicated colleagues involved in delivering healthcare across the UK. Our patients benefit so much more when we work together, within well-structured multidisciplinary teams. We therefore have the opportunity, and, indeed, the responsibility, to collaborate with others, sharing what we currently believe is best practice, caring deeply for our patients, and being open to both teaching and learning, so that we are well prepared for those who will need us in the future.
On behalf of Restorative Dentistry – UK and our specialty of restorative dentistry, I am delighted to endorse the 6th edition of these essential guidelines, and welcome the recognition of our specialty as a core member of the multidisciplinary head and neck cancer team. Restorative dentistry is for those patients who have complex dental problems, requiring multidisciplinary, specialist care for their oral rehabilitation. These guidelines confirm the need for and benefit of restorative dentistry at all stages of the head and neck cancer pathway, ensuring better clinical and personal outcomes for our patients, throughout their treatment and recovery.
These guidelines are, once again, a considerable step towards ensuring that high standards of care are delivered to our patients. We congratulate and thank all those who have contributed to writing, collating and publishing this edition.
Royal College of Speech and Language Therapists (RCSLT)
The Royal College of Speech and Language Therapists is pleased to endorse this high-quality document, in which a great deal of thought and effort has gone into its creation. It will be a very useful resource for speech and language therapists working in head and neck cancer, providing key point summaries and detailed information on speech and language therapy interventions at different stages of the cancer pathway. Clear guidance will be invaluable to speech and language therapists, enabling them to perform their best, which will benefit patients and their families, and foster interprofessional working.
The Royal College of Pathologists (RCPath)
On behalf of The Royal College of Pathologists, I am delighted to endorse the latest update of these guidelines, and I thank the editors and authors for their hard work, particularly Drs Betts, Damato, Robinson and Thavaraj and Prof. Hunter, who contributed the pathology content. The guidelines are testament to the importance of collaborative working between specialties in the treatment of head and neck cancer, and rightly acknowledge cellular pathology as a core component of this. An exciting addition to the guidelines is the molecular testing of head and neck tumours, which has developed immeasurably since the last edition, and places cellular pathology at the centre of diagnosis, prognostication and therapy. This quality-assured pathology guidance provides reassurance to clinical teams that pathology information is based on good evidence, robustly accredited, and has the confidence of members of the College. Congratulations on an excellent document.
The Royal College of Radiologists (RCR)
On behalf of The Royal College of Radiologists, I am delighted to welcome this 6th edition of the Head and Neck Cancer: United Kingdom National Multidisciplinary Guidelines. These guidelines are a key clinical resource for all those involved in the treatment of patients with head and neck cancer, and I hope this new edition will continue to support essential multidisciplinary working in this field and promote the highest possible standards of patient care. I would like to congratulate the editors and authors – who include a number of Fellows of the Royal College of Radiologists – for their hard work and dedication in producing these excellent guidelines.
Elizabeth C Abbey, Department of Palliative Medicine, Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, UK (Chapter 15)
Hiba Aga, Head and Neck Department, Nottingham University Hospitals, Nottingham, UK (Chapter 26)
Reshma Agrawal, Histopathology, University College London Hospitals NHS Trust, London, UK (Chapter 25)
Derfel ap Dafydd, Radiology, Royal Marsden NHS Foundation Trust, London, UK (Chapter 29)
Takhar Arunjit, Department of Otolaryngology – Head and Neck Surgery, St George's Hospital NHS Trust, London, UK (Chapter 30)
Patrick Axon, Department of ENT, Addenbrookes Hospital, Cambridge, UK (Chapter 22)
Eleanor Aynsley, Department of Oncology, James Cook University Hospital, Middlesbrough, UK (Chapter 18)
Izhar N Bagwan, Histopathology, Royal Surrey County NHS Foundation Trust, Guildford, UK (Chapter 29)
Arun Batra, Department of Radiology, North Tees University Hospital, Stockton-on-Tees, UK (Chapter 18)
Donna Begg, Clinical Nursing, University Hospitals Birmingham, Birmingham, UK (Chapter 29)
Jonathan M Bernstein, Department of Otolaryngology – Head and Neck, Surgery, Charing Cross Hospital, Imperial College Healthcare NHS Trust, London, UK (Chapters 17, 24 and 30*)
Guy Betts, Adult Histopathology Department, Manchester Royal Infirmary, Manchester, UK (Chapters 3 and 21)
Colin Bicknell, Department of Vascular Surgery, St Mary's Hospital, Imperial College Healthcare NHS Trust, London, UK (Chapter 30)
Brian Bisase, Department of Oral and Maxillofacial Surgery, Queen Victoria Hospital, East Grinstead, UK (Chapter 7*)
Grainne C Brady, Department of Speech, Voice and Swallowing, Royal Marsden NHS Foundation Trust, London, UK (Chapter 10)
Peter Brennan, Oral and Maxillofacial, Head and Neck Surgery, Portsmouth Hospitals NHS Trust, Portsmouth, UK (Chapter 16*)
Aina Brunet, Department of Otorhinolaryngology, Head and Neck Surgery, and Head, Neck Thyroid Cancer Unit, Guy's and St Thomas’ NHS Foundation Trust, London, UK (Chapter 20)
Val Bryant, Patient representative, Sunderland, UK (Chapter 8)
Linda Cantwell, University of Liverpool School of Health Sciences, Liverpool, UK, and Department of Dietietics, Merseycare NHS Foundation Trust, Liverpool, UK (Chapter 8)
Ashish Chandra, Department of Cellular Pathology, St Thomas' Hospital, Guy's and St Thomas' NHS Foundation Trust, London, UK (Chapter 25)
Preetha Chengot, Department of Pathology, Leeds Teaching Hospitals, Leeds, UK (Chapter 24)
Melvin L K Chua, Division of Radiation Oncology, National Cancer Centre, Singapore (Chapter 19)
Peter Clarke, Department of ENT, Imperial College Healthcare NHS Trust, London, UK (Chapter 23)
Gemma Clunie, Complex Laryngology/Airways Service, National Centre for Airway Reconstruction, Imperial College Healthcare NHS Trust, London, UK (Chapter 10)
Margaret Coffey, Speech and Language Therapy, Imperial College Healthcare NHS Trust, London, UK (Chapter 10)
Clare Conlon, Department of Anaesthesia, Manchester University Hospitals NHS Foundation Trust, Manchester, UK (Chapter 8)
David I Conway, School of Medicine, Dentistry, and Nursing, University of Glasgow, Glasgow, Scotland, UK (Chapter 6)
Florence Cook, Head and Neck Centre, University College London Hospital NHS Foundation Trust, London, UK (Chapter 9*)
Matthew R Cooper, Librarian, Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, UK (Chapter 15)
Declan Costello, Otolaryngology, Head and Neck Surgery, Wexham Park Hospital, Slough, UK (Chapter 16*)
Ben Cosway, Department of Otolaryngology, James Cook University Hospital, Middlesbrough, UK (Chapter 18)
Neil J A Cozens, Department of Radiology, Royal Derby Hospital, Derby, UK (Chapter 25)
Grant Creaney, School of Medicine, Dentistry, and Nursing, University of Glasgow, Glasgow, Scotland, UK (Chapter 6)
Daljit K Gahir, Clinical Oncology, University Hospital of North Midlands NHS Trust, Stoke-on-Trent, UK (Chapter 26)
Stephen Damato, Department of Cellular Pathology, John Radcliffe Hospital, Oxford, UK (Chapter 3)
Joe Davies, Department of Oncology, Poole Hospital, Poole, UK (Chapter 16)
Katharine S Davies, Department of Otolaryngology, Head and Neck Surgery, Aintree Hospital, Liverpool University Hospitals Trust, Liverpool, UK (Chapter 2)
Alina D Dragan, Radiology Department, Royal Marsden Hospital, London, UK (Chapter 26)
Yong Du, Department of Nuclear Medicine and PETCT, Royal Marsden NHS Foundation Trust, Sutton, UK (Chapter 30)
Mark R D Edmond, Department of Otolaryngology, Head and Neck Surgery, Poole Hospital, Poole, UK (Chapters 16 and 29)
Stefano Fedele, Department of Oral Medicine, University College London Eastman Dental Institute, London, UK (Chapter 16*)
Harriet Finze, Physiotherapy, Churchill Hospital, Oxford, UK (Chapter 11)
Jason C Fleming, Liverpool Head and Neck Centre, Liverpool University Hospitals NHS Foundation Trust/University of Liverpool, Liverpool, UK (Chapter 23)
Bernadette H Foran, Weston Park Cancer Centre, Sheffield, UK (Chapter 21)
Beth Fordham, Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of Oxford, UK (Chapter 11)
Mohammed M A S Foridi, Clinical Pharmacy, Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, UK (Chapter 15)
Lesley Freeman, Department of Nutrition and Dietetics, Sunderland Royal Hospital, Tyneside and Sunderland NHS Foundation Trust, Sunderland, UK (Chapter 9)
Katherine E Frew, Palliative Medicine, Wansbeck General Hospital, Northumbria Healthcare NHS Foundation Trust, Ashington, UK (Chapter 15)
Pallavi Gaitonde, Restorative Dentistry, Castle Hill Hospital, Hull and East Ridings NHS Trust, Cottingham, UK (Chapter 13)
Victoria Gallyer, Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of Oxford, Oxford, UK (Chapter 11)
Fraser W Gibb, Department of Endocrinology, NHS Lothian, Edinburgh, Scotland, UK (Chapter 25)
Sinclair M Gore, Blenheim Head and Neck Unit, Churchill Hospital, Oxford, UK (Chapters 1 and 17)
Mark Gormley, MRC Integrative Epidemiology Unit, Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, UK (Chapter 6)
Roganie Govender, Head and Neck Academic Centre and Research, University College London, London, UK (Chapter 10)
J Greedy, Head and Neck Cancer, Beacon Cancer Centre, Musgrove Park Hospital, Taunton, UK (Chapter 12)
Teresa Guerrero Urbano, Department of Clinical Oncology, Guy's and St Thomas’ NHS Foundation Trust, London, UK (Chapter 20)
Dorothy Gujral, Department of Clinical Oncology, Charing Cross Hospital, Imperial College Healthcare NHS Trust, London, UK (Chapter 30)
David W Hamilton, Department of ENT, Freeman Hospital, Newcastle upon Tyne, UK (Chapter 21)
John C Hardman, Department of Head and Neck Surgery, Royal Marsden Hospital, London, UK (Chapter 27*)
Kevin Harrington, Department of Oncology, Royal Marsden NHS Foundation Trust, London, UK (Chapters 27 and 29)
Samantha Holmes, Speech and Language Therapy, Oxford University Hospitals NHS Trust, Oxford, UK (Chapter 8)
Jarrod J Homer, Manchester Head and Neck Centre, Manchester Royal Infirmary, Manchester University NHS Foundation Trust, Manchester, UK (Introduction, and Chapters 1, 25 and 30)
Deborah Howland, Department of Nutrition and Dietetics, Torbay Hospital, Torbay and South Devon NHS Foundation Trust, Torquay, UK (Chapter 9)
Gerald Humphris, Medical School, University of St Andrews, St Andrews, Scotland, UK (Chapter 14*)
Keith D Hunter, Academic Unit of Oral Medicine and Pathology, School of Clinical Dentistry, Sheffield, UK (Chapter 3)
Kate Ingarfield, Centre for Trials Research, Neuadd Meirionnydd, Cardiff, Wales, UK (Chapter 6)
Richard Irving, Department of Otolaryngology, Queen Elizabeth Hospital NHS Foundation Trust, Birmingham, UK, and Department of Otolaryngology, Diana, Princess of Wales Hospital for Children, Birmingham, UK (Chapter 30)
Kristina Isand, Department of Endocrinology, Oxford University Hospital, Oxford, UK (Chapter 16)
Yatin Jain, Department of Radiology, Christie Hospital NHS Foundation Trust, Manchester, UK (Chapters 2 and 21)
Sachin Jauhar, Restorative Dentistry, Glasgow Dental Hospital, Glasgow, Scotland, UK (Chapter 13)
Sarra Jawad, Restorative Dentistry, Guy's Hospital, London, UK (Chapter 13)
Glyndwr W Jenkins, Department of Maxillofacial Surgery, Freeman Hospital, Newcastle upon Tyne Hospitals Trust, Newcastle upon Tyne, UK (Chapter 2)
Anastasios Kanatas, Department of Head and Neck Surgery, St James Institute of Oncology, Leeds Dental Institute, Leeds General Infirmary and University of Leeds, Leeds, UK (Chapter 16*)
Stephen Keohane, Department of Dermatology, Portsmouth Hospitals NHS Trust, Portsmouth, UK (Chapter 28)
Cyrus J Kerawala, Head and Neck Unit, Royal Marsden Hospital, London, UK (Chapters 17 and 29)
William Keys, Restorative Dentistry, Aberdeen Dental Institute, Aberdeen, Scotland, UK (Chapter 13)
Emma V King, Department of Otolaryngology, Head and Neck Surgery, Poole Hospital, Poole, UK (Chapters 16* and 29*)
Anthony Kong, Comprehensive Cancer Centre, Guy's Campus, King's College London, London, UK, and Department of Clinical Oncology, Guy's and St Thomas’ NHS Foundation Trust, London, UK (Chapters 5 and 20)
Fiona Lalloo, Manchester Centre for Genomic Medicine, St Mary's Hospital, Manchester University Foundation Trust, Manchester, UK (Chapter 30)
Kirsten Laws, Department of Oncology, Aberdeen Royal Infirmary, Aberdeen, Scotland, UK (Chapter 4)
Samuel C Leong, Liverpool Head and Neck Centre, Liverpool University Hospitals NHS Foundation Trust/University of Liverpool, Liverpool, UK (Chapter 23)
Shane Lester, Department of Otolaryngology, James Cook University Hospital, Middlesbrough, UK (Chapter 18*)
Miles Levy, Department of Endocrinology, University Hospitals of Leicester, Leicester Royal Infirmary, Leicester, UK (Chapter 30)
Ken Lingley, Carer representative, UK (Chapter 29)
Gitta Madani, Department of Radiology, Charing Cross Hospital, Imperial College Healthcare NHS Trust, London, UK (Chapter 30)
Navin Mani, Manchester Head and Neck Centre, Manchester Royal Infirmary, Manchester, UK (Chapter 22*)
Paolo L Matteucci, Department of Plastic and Reconstructive Surgery, Castle Hill Hospital, Cottingham, UK (Chapter 28*)
Catriona R Mayland, Department of Palliative Medicine, Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, UK, and Yorkshire Cancer Research, Harrogate, UK (Chapter 15*)
James McCaul, Department of Maxillofacial/Head and Neck Surgery, Queen Elizabeth University Hospital, Glasgow, Scotland, UK (Chapter 24)
Lorna K McCaul, Restorative Dentistry, Glasgow Dental Hospital, Glasgow, Scotland, UK (Chapter 13*)
Pádraig McDonnell, Clinical Psychology, Greater Manchester Mental Health NHS Foundation Trust, Manchester, UK (Chapter 8)
Andrew McPartlin, Department of Clinical Oncology, Christie Hospital, Manchester, UK (Chapter 19)
Valeria Mercadante, Department of Oral Medicine, University College London Eastman Dental Institute, London, UK (Chapter 16)
Zoe Merchant, Occupational Therapy and Prehab4Cancer Programme, Christie NHS Foundation Trust, Manchester, UK (Chapter 8)
Radu Mihai, Department of Endocrine Surgery, Oxford University Hospitals NHS Foundation Trust, Oxford, UK (Chapter 25)
Mufaddal T Moonim, Histopathology, Imperial College NHS Healthcare Trust, London, UK (Chapter 25)
John Moore, Division for Anaesthesia, Critical Care and Peri-operative Medicine, Manchester University Hospitals NHS Foundation Trust, Manchester, UK (Chapter 8*)
Paul Nankivell, Institute of Cancer and Genomic Sciences, University of Birmingham, Birmingham, UK, and Department of Otolaryngology, University Hospital Birmingham, Birmingham, UK (Chapter 5*)
Sonali Natu, Histopathology, University Hospital of North Tees and Hartlepool NHS Foundation Trust, Hartlepool, UK (Chapter 25)
A Nelson, Macmillan Head and Neck, Northwick Park Hospital, London North West University Healthcare NHS Trust, Harrow, UK (Chapter 12)
Pablo Nenclares, Oncology, Royal Marsden NHS Foundation Trust, London, UK (Chapter 29)
Kate Newbold, Oncology, Royal Marsden Hospital, London, UK (Chapter 25)
Carrie Newland, Oral and Maxillofacial Surgery, Royal Surrey NHS Foundation Trust, Guildford, UK, and School of Medicine, University of Surrey, Guildford, UK (Chapter 28)
Ailsa J Nicol, Restorative Dentistry, Newcastle Dental Hospital, Newcastle upon Tyne, UK (Chapter 13)
Iain J Nixon, ENT Department, NHS Lothian, Edinburgh, Scotland, UK (Chapter 25*)
Rupert Obholzer, Department of ENT, Guy's Hospital, London, UK (Chapter 22)
James T O'Hara, Department of ENT, Freeman Hospital, Newcastle upon Tyne, UK (Chapters 16* and 21*)
S Orr, Nursing, Colchester Hospital, East Suffolk and North Essex Foundation Trust, Colchester, UK (Chapter 12*)
Vinidh Paleri, Head and Neck Unit, Royal Marsden Hospital, London, UK (Chapters 26, 27* and 29)
James Palmer, Department of Oncology, Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, UK (Chapter 15)
Rachel S Parry, Palliative Medicine, Wirral University Teaching Hospitals NHS Trust, Birkenhead, UK (Chapter 15)
Claire Paterson, Beatson West of Scotland Cancer Centre, Glasgow, Scotland, UK (Chapters 5 and 17)
Gillian Patterson, Patient representative, UK (Chapter 29)
Joanne M Patterson, Liverpool Head and Neck Centre, School of Health Sciences, University of Liverpool, Liverpool, UK (Chapter 10*)
Miranda Payne, Medical Oncology, Oxford University Hospitals NHS Foundation Trust, Oxford, UK (Chapter 29)
L Pearson, Workforce and Education, Worcester Acute Hospitals NHS Trust, Worcester, UK (Chapter 12)
David N Poller, Department of Pathology, Queen Alexandra Hospital, Portsmouth, UK (Chapter 25)
Jonathan Pollock, Department of Plastic Surgery, Nottingham University Hospitals, Nottingham, UK (Chapter 28)
Stephen Ross Porter, Department of Oral Medicine, University College London Eastman Dental Institute, London, UK (Chapter 16)
Matthew Potter, Department of Plastic and Reconstructive Surgery, John Radcliffe Hospital, Oxford, UK (Chapters 22 and 23)
Robin J D Prestwich, Leeds Cancer Centre, St James's University Hospital, Leeds, UK (Chapters 4 and 19)
Ruth Price, Department of Physiotherapy, Liverpool University Hospitals NHS Foundation Trust, Liverpool, UK (Chapter 8)
Mani Ragbir, Department of Plastic and Reconstructive Surgery, Newcastle upon Tyne Hospitals NHS Foundation Trust, Newcastle upon Tyne, UK (Chapter 7*)
Meena S Ranka, Restorative Dentistry, Luton and Dunstable University Hospital, Luton, UK (Chapter 13)
Max Robinson, Department of Cellular Pathology, Royal Victoria Infirmary, Newcastle upon Tyne, UK (Chapters 3 and 18)
Justin W G Roe, Department of Speech, Voice and Swallowing, Royal Marsden NHS Foundation Trust, London, UK, and Complex Laryngology/Airways Service, National Centre for Airway Reconstruction, Imperial College Healthcare NHS Trust, London, UK (Chapter 10)
Tom Roques, Department of Oncology, Norfolk and Norwich University Hospital, Norwich, UK (Chapters 1, 4* and 27)
Aleix Rovira, Department of Otorhinolaryngology, Head and Neck Surgery, and Head, Neck Thyroid Cancer Unit, Guy's and St Thomas’ NHS Foundation Trust, London, UK (Chapter 20)
Sajid Sainuddin, Department of Oral and Maxillofacial Surgery, Manchester University NHS Foundation Trust, Manchester, UK (Chapter 7*)
I J Salmon, Macmillan Head and Neck, Bedford Hospital NHS Trust, Bedford, UK (Chapter 12)
Ann Sandison, Department of Head and Neck/Oral Pathology, Guy's and St Thomas’ NHS Foundation Trust, London, UK (Chapter 23)
Andy Scarsbrook, Leeds Institute of Health Research, University of Leeds, Leeds, UK, and Department of Radiology, Leeds Teaching Hospitals NHS Trust, Leeds, UK (Chapter 5)
Andrew G Schache, University of Liverpool Cancer Research Centre, Liverpool, UK (Chapters 6, 17* and 23)
A Scott, Macmillan Head and Neck, Mount Vernon Cancer Centre, East and North Hertfordshire NHS Trust, Northwood, UK (Chapter 12)
Diane Sellstrom, Department of Voice and Swallowing, Newcastle upon Tyne Hospitals NHS Trust, Newcastle upon Tyne, UK (Chapter 10)
Cherith J Semple, School of Nursing, Ulster University, Belfast, UK (Chapter 15)
Jagrit Shah, Radiology, Nottingham University Hospitals NHS Foundation Trust, Nottingham, UK (Chapter 23)
Praveen Sharma, Restorative Dentistry, University of Birmingham School of Dentistry, Birmingham, UK (Chapter 13)
Richard J Shaw, Oral and Maxillofacial/Head and Neck Surgery, University of Liverpool, Liverpool, UK (Chapter 16*)
Somiah Siddiq, Head and Neck Department, University Hospitals Birmingham, Birmingham, UK (Chapter 26*)
Priyamal Silva, Head and Neck Surgery, Oxford University Hospitals NHS Foundation Trust, Oxford, UK (Chapter 29)
Ricard Simo, Department of Otorhinolaryngology, Head and Neck Surgery, and Head, Neck Thyroid Cancer Unit, Guy's and St Thomas’ NHS Foundation Trust, London, UK (Chapter 20*)
Rabin P Singh, Maxillofacial, Head and Neck Surgery, University Hospital Southampton NHS Foundation Trust, Southampton, UK (Chapter 16*)
Maria Smith, Macmillan Head and Neck, Queen Elizabeth University Hospital, NHS Greater Glasgow and Clyde, Glasgow, Scotland, UK (Chapter 12*)
Rebekah Smith, Department of Nutrition and Dietetics, Queen Elizabeth Hospital, University Hospitals Birmingham NHS Foundation Trust, Birmingham, UK (Chapter 9)
Toby Oliver Smith, Warwick Clinical Trials Unit, University of Warwick, Coventry, UK, and School of Health Sciences, University of East Anglia, Norwich, UK (Chapter 11*)
Sanjai Sood, Otolaryngology, Head and Neck Surgery, Bradford Teaching Hospitals, Bradford, UK (Chapters 24* and 27)
Francis W Stafford, Otolaryngology – Head and Neck Surgery, Sunderland Royal Hospital, Sunderland, UK (Chapters 1* and 27)
Neil Steven, Medical Oncology, University Hospitals Birmingham, Birmingham, UK (Chapter 29)
Kay Stewart, Department of Palliative Medicine, Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, UK (Chapter 15)
Lisa Stoner, Physiotherapy, Norfolk and Norwich University Hospital NHS Foundation Trust, Norwich, UK (Chapter 11)
Steve Sweeney, Patient representative, Manchester, UK (Chapter 8)
Andrew Sykes, Department of Clinical Oncology, Christie Hospital, Manchester, UK (Chapter 28)
Carly L Taylor, Restorative Dentistry, University Dental Hospital of Manchester, Manchester, UK (Chapter 13)
Selvam Thavaraj, Head and Neck Pathology, Guy's Hospital, London, UK (Chapters 3* and 17)
David J Thomson, Department of Oncology, Christie NHS Foundation Trust, Manchester, UK, and Division of Cancer Sciences, University of Manchester, Manchester, UK (Chapters 4, 22 and 23)
Jane Thornton, Department of Speech and Language Therapy, Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, UK (Chapter 10)
Neil S Tolley, ENT Surgery, Imperial College NHS Healthcare Trust, London, UK (Chapter 25)
Nancy Turnbull, Melanoma Focus, Cambridge, UK (Chapter 29)
Sriram Vaidyanathan, Department of Nuclear Medicine, St James's University Hospital, Leeds, UK (Chapter 19)
Leandros Vassiliou, Oral and Maxillofacial Surgery, Royal Blackburn Hospital, Blackburn, UK (Chapter 1)
John Waas, Department of Endocrinology, Oxford University Hospital, Oxford, UK (Chapter 16*)
Kelly Wade-McBane, Department of Nutrition and Dietetics, Charing Cross Hospital, Imperial College Healthcare NHS Trust, London, UK (Chapter 9)
Donna Wakefield, Department of Palliative Medicine, University Hospital of North Tees, North Tees and Hartlepool NHS Foundation Trust, Stockton-on-Tees, UK (Chapter 15)
Amy Ward, Department of Oncology, Barking, Havering and Redbridge University Hospitals NHS Trust, Romford, UK (Chapter 4)
Laura Warner, Department of Otolaryngology, Head and Neck Surgery, Freeman Hospital, Newcastle upon Tyne Hospitals Trust, Newcastle upon Tyne, UK (Chapter 2*)
Laura-Jayne Watson, Speech and Language Therapy, South Tyneside and Sunderland NHS Foundation Trust, Sunderland, UK (Chapters 8* and 10)
H Watts, Macmillan Head and Neck, Northampton General Hospital NHS Trust, Northampton, UK (Chapter 12)
Christina Wilson, Beatson West of Scotland Cancer Centre, Glasgow, Scotland, UK (Chapters 4 and 17)
Stuart C Winter, Department of Otolaryngology, Head and Neck Surgery, Nuffield Department of Surgical Sciences, University of Oxford, Oxford, UK, and Blenheim Head and Neck Unit, Churchill Hospital, Oxford, UK (Introduction, and Chapters 16* and 18)
Winson Wong, Department of Otolaryngology, Head and Neck Surgery, Hull University Teaching Hospitals, Hull, UK (Chapter 18)
Chui-Yan Yip, Clinical Pharmacy, Wheatfield's Hospice and St Gemma's Hospice, Leeds, UK (Chapter 15)
Kent Yip, Clinical Oncology, Ipswich Hospital, East Suffolk and North Essex NHS Foundation Trust, Ipswich, UK (Chapter 29)
*Indicates chapter lead author
Section 1: Generic issues in head and neck cancer management
Chapter 1: Provision of head and neck cancer services for diagnosis and treatment centres (Sinclair M Gore, Jarrod J Homer, Tom Roques, Francis W Stafford* and Leandros Vassiliou) 9
Chapter 2: Clinical assessment, diagnosis and imaging (Katharine S Davies, Yatin Jain, Glyndwr W Jenkins and Laura Warner*) 13
Chapter 3: Head and neck cancer pathology (Guy Betts, Stephen Damato, Keith D Hunter, Max Robinson and Selvam Thavaraj*) 19
Chapter 4: Non-surgical head and neck cancer treatment (Kirsten Laws, Robin J D Prestwich, Tom Roques*, David J Thomson, Amy Ward and Christina Wilson) 24
Chapter 5: Follow up, surveillance and recurrent disease (Anthony Kong, Paul Nankivell*, Claire Paterson and Andy Scarsbrook) 29
Chapter 6: Epidemiology of head and neck cancer: definitions, trends and risk factors (David I Conway*, Andrew G Schache, Grant Creaney, Mark Gormley and Kate Ingarfield) 34
Chapter 7: Reconstructive considerations in head and neck surgical oncology (Brian Bisase*, Mani Ragbir* and Sajid Sainuddin*) 36
Section 2: Patient support
Chapter 8: Patient preparation for treatment and enhanced recovery (Val Bryant, Linda Cantwell, Clare Conlon, Samantha Holmes, Pádraig McDonnell, Zoe Merchant, John Moore*, Ruth Price, Steve Sweeney and Laura-Jayne Watson*) 42
Chapter 9: Nutritional management in the treatment of head and neck cancer (Florence Cook*, Lesley Freeman, Deborah Howland, Rebekah Smith and Kelly Wade-McBane) 48
Chapter 10: Speech, voice and swallowing rehabilitation for head and neck cancer (Grainne C Brady, Gemma Clunie, Margaret Coffey, Roganie Govender, Joanne M Patterson*, Justin W G Roe, Diane Sellstrom, Jane Thornton and Laura-Jayne Watson) 58
Chapter 11: Physiotherapy and exercise (Harriet Finze, Beth Fordham, Victoria Gallyer, Toby Oliver Smith* and Lisa Stoner) 63
Chapter 12: The clinical nurse specialist role in head and neck cancer care (Joanne Greedy, Andrea Nelson, Sarah Orr*, Louise Pearson, Ian J Salmon, Audrey Scott, Maria Smith* and Hollie Watts) 69
Chapter 13: Restorative dentistry and orofacial rehabilitation for patients with head and neck cancer (Pallavi Gaitonde, Sachin Jauhar, Sarra Jawad, William Keys, Lorna K McCaul*, Ailsa J Nicol, Meena S Ranka, Praveen Sharma and Carly L Taylor) 74
Chapter 14: Psychological management in head and neck cancer (Gerald Humphris*) 80
Chapter 15: Palliative care in head and neck cancer (Elizabeth C Abbey, Matthew R Cooper, Mohammed M A S Foridi, Katherine E Frew, Catriona R Mayland*, James Palmer, Rachel S Parry, Cherith J Semple, Kay Stewart, Donna Wakefield and Chui-Yan Yip) 83
Chapter 16: Management of treatment effects and complications (Peter Brennan*, Declan Costello*, Joe Davies, Mark R D Edmond, Stefano Fedele*, Kristina Isand, Anastasios Kanatas*, Emma V King*, Valeria Mercadante, James T O'Hara*, Stephen Ross Porter, Richard J Shaw*, Rabin P Singh*, John Waas* and Stuart C Winter*) 95
Section 3: Site-specific guidelines
Chapter 17: Oral cavity and lip cancer (Jonathan M Bernstein, Sinclair M Gore, Cyrus J Kerawala, Claire Paterson, Andrew G Schache*, Selvam Thavaraj and Christina Wilson) 107
Chapter 18: Oropharyngeal squamous cell carcinoma (Eleanor Aynsley, Arun Batra, Ben Cosway, Shane Lester*, Max Robinson, Stuart C Winter and Winson Wong) 114
Chapter 19: Nasopharyngeal carcinoma (Melvin L K Chua, Andrew McPartlin, Robin J D Prestwich* and Sriram Vaidyanathan) 120
Chapter 20: Hypopharyngeal cancer (Aina Brunet, Teresa Guerrero Urbano, Anthony Kong, Aleix Rovira and Ricard Simo*) 125
Chapter 21: Laryngeal cancer (Guy Betts, Bernadette H Foran, David W Hamilton, Yatin Jain and James T O'Hara*) 129
Chapter 22: Management of lateral skull base cancer (Patrick Axon, Navin Mani*, Rupert Obholzer, Matthew Potter and David J Thompson) 135
Chapter 23: Nasal/paranasal sinus and anterior skull base cancer (Peter Clarke, Jason C Fleming*, Samuel C Leong, Matthew Potter, Ann Sandison, Andrew G Schache, Jagrit Shah and David J Thomson) 142
Chapter 24: Salivary gland cancer (Jonathan M Bernstein, Preetha Chengot, James McCaul and Sanjai Sood*) 149
Chapter 25: Management of thyroid cancer (Reshma Agrawal, Ashish Chandra, Neil J A Cozens, Fraser W Gibb, Jarrod J Homer, Radu Mihai, Mufaddal T Moonim, Sonali Natu, Kate Newbold, Iain J Nixon*, David N Poller and Neil S Tolley) 156
Chapter 26: Management of neck metastases in head and neck cancer (Hiba Aga, Alina D Dragan, Daljit K Gahir, Vinidh Paleri and Somiah Siddiq*) 165
Chapter 27: Management of head and neck squamous cell carcinoma of unknown primary (John C Hardman*, Kevin Harrington, Vinidh Paleri*, Tom Roques, Sanjai Sood and Francis W Stafford) 172
Chapter 28: Non-melanoma skin cancer (Stephen Keohane, Paolo L Matteucci*, Carrie Newland, Jonathan Pollock and Andrew Sykes) 176
Chapter 29: Head and neck mucosal melanoma (Derfel ap Dafydd, Izhar N Bagwan, Donna Begg, Mark R D Edmond, Kevin Harrington, Cyrus J Kerawala, Emma V King*, Ken Lingley, Pablo Nenclares, Vinidh Paleri, Gillian Patterson, Miranda Payne, Priyamal Silva, Neil Steven, Nancy Turnbull and Kent Yip) 186
Chapter 30: Paraganglioma of the head and neck (Takhar Arunjit, Jonathan M Bernstein*, Colin Bicknell, Yong Du, Dorothy Gujral, Jarrod J Homer, Richard Irving, Fiona Lalloo, Miles Levy and Gitta Madani) 189
Appendices
Appendix 1: List of reviewers 195
Appendix 2: Management of head and neck squamous cell carcinoma of unknown primary (Chapter 27) – initiative collaborators 196
*Indicates chapter lead author
Key recommendations
• Head and neck cancer services should be organised around specialist multidisciplinary teams (MDTs), comprising health professionals within a head and neck cancer treatment centre including surgical and oncological services that can provide comprehensive treatment and management for patients with head and neck cancers (essential (E))
• Diagnostic services outside of the head and neck cancers should have agreed protocols with the referral centre regarding the diagnosis and referral of patients with head and neck cancer (E)
• In-reach models of service delivery in which surgeons primarily located at a peripheral diagnostic unit travel to a centre to perform cancer surgery is not recommended (desirable (D))
• Where geographically possible, there should be the maximum consolidation into as few head and neck treatment centres as possible. In these circumstances, a head and neck treatment centre should comprise a single surgical centre and a single oncological centre (D)
Introduction and current service provision
The initial development of head and neck cancer services in UK was based upon the National Institute for Health and Care Excellence (NICE) Improving Outcomes Guidelines 2004.1 Subsequent UK guidelines, standards, quality assessment metrics and commissioning standards essentially all mirror the structure outlined in the original NICE Improving Outcomes Guidelines.
Head and neck cancer services in the UK are generally based on a hub-and-spoke model of treatment centres (surgery and oncology) (Figure 1), in which MDTs are centrally based, and peripheral ‘spoke’ diagnostic units offer diagnostic services, as well as, ideally, ongoing post-treatment support and follow up, closer to patients’ homes. There is variation in the nature of this set-up. Diagnostic units are important, potentially maximising the proportion of the patient journey closer to home.
Figure 1. Hub-and-spoke model of treatment centres.
Diagnostic units can be entirely separate from treatment centres, provided that patient-centred communication and co-ordination is not compromised, and the pathway remains streamlined (Figure 2). This requires clinicians working to agreed MDT diagnostic and follow-up guidelines. However, the alternative arrangement is that MDT surgeons from the treatment centre may ‘out-reach’ to provide diagnostic and follow-up services in peripheral units. Less commonly and less ideally, surgeons from a peripheral unit ‘in-reach’ to the treatment centre to perform surgery on patients who are diagnosed with cancer in their unit. Many have concerns over this model, with a limited presence in the main surgical treatment centre and disjointed care models. Similar in-reach and out-reach models can be applied to other disciplines, such as allied health and specialised nursing professionals.
Figure 2. In-reach and out-reach model of treatment centres.
There is variation in this model in the UK, and variation in MDT size (e.g. as defined by the population served). In England, in 2017, one region had a population of 2.4 million with 1 MDT and an annual caseload of 548 new patients. Another with a population of 2.7 million had 6 MDTs and a caseload of 584, with only 2 of the units seeing more than 100 new cases. Some MDTs have more than one surgical and radiotherapy centre. In addition, the incidence of head and neck cancers in the UK varies significantly. The UK average is 18.8 per 100 000 per year, higher in Scotland and Wales.2
The economic and clinical value in various different set-ups is unclear, and may vary according to population density, incidence, and geographical area and spread.
Diagnostics and follow-up units
The purpose of diagnostic units is to provide early diagnosis and appropriate investigation, as well as staging, for patients with head and neck cancer in the locality which they serve. Additionally, where appropriate, patient follow up can be delivered in these units, at least in part, maximising care nearer to home. Each head and neck cancer centre will also act as the diagnostic unit for its locality or secondary referral population. The process of diagnostic investigation is discussed in Chapter 2 of these guidelines. Essential and desirable specifications are shown below (Table 1).
Table 1. Diagnostic units: essential and desirable
* The facility to request a positron emission tomography/computed tomography (PET-CT) scan (rather than provide PET-CT scans). H&N = head and neck; USS = ultrasound scanning; FNAC = fine needle aspiration cytology; CT = computed tomography; MRI = magnetic resonance imaging; MDT = multidisciplinary team; SLT = speech and language therapist
Head and neck cancer treatment centres
A treatment centre must provide comprehensive management for all patients with head and neck cancer. The head and neck cancer MDT is at the heart of this, comprising all health professionals and allied staff required to provide this and support patients. Suggested core and extended MDT members are listed in Table 2. Core members are required to provide input into patients’ care, and should be available for every patient who needs their expertise or input. Whether this takes the form of physical attendance at the MDT meeting and/or at an MDT clinic with the patients will vary, but, in general, core members are expected to attend (physically or virtually) the weekly MDT meeting (see below). Extended members are professionals whose input is less regular, and who would not be expected to routinely attend MDT meetings and clinics.
Table 2. H&N MDT members
H&N = head and neck; MDT = multidisciplinary team; SLT = speech and language therapist
Multidisciplinary team meetings
Head and neck MDT meetings improve decision-making and diagnosis, impacting positively on both staging and treatment.Reference Bergamini, Locati, Bossi, Granata, Alfieri and Resteghini3–Reference Wheless, McKinney and Zanation8 Effective MDTs reduce the time from diagnosis to definitive treatment. An MDT management approach is associated with improved survival rates after treatment for head and neck cancer.Reference Shang, Feng, Gu, Hong, Hong and Hou9
The MDT meetings to record a patient's care will take place at different stages during a cancer journey. This includes:
• Upon initial diagnosis, to ensure diagnosis and staging is accurate, that no further investigations are required, and to determine appropriate treatment options according to agreed national and local protocols
• After surgery, to discuss histopathology and options for post-operative adjuvant treatment
• After primary chemoradiotherapy, where there is concern for residual disease on post-treatment imaging
• For suspected or confirmed recurrence, or second primary cancers, after initial diagnosis and treatment
The MDT should meet at least weekly. Every meeting should have appropriate input from core members (see above). External imaging is generally reviewed by an MDT radiologist. Some or all external pathology findings may also need reviewing according to local policy.
The MDT meeting offers an opportunity to collect data and to offer clinical trials, where appropriate, for eligible patients.
Each head and neck cancer MDT should have an operational policy setting out the roles and responsibilities of its core members, including attendance required at MDT meetings. These will vary between specialist groups: all surgeons and oncologists should be expected to attend most meetings, whilst radiologists and pathologists may attend according to a rota. Most core member groups need at least two individuals, to allow for cover in case of leave.
The MDT meeting should have an appropriately trained chair who leads the meeting; this individual also ensures that a clear summary is documented after each case is discussed and that recorded data are accurate. The agreed diagnosis and staging, relevant discussions, and treatment plans should be documented in real time and circulated to the MDT members immediately afterwards, along with the person responsible for confirming the outcome to the patient. The MDT co-ordinator role is crucial to the preparation and smooth running of the meeting, and in disseminating information after the meeting. Enquiries for the co-ordinator should be sent to a generic monitored e-mail address, so that absence is covered. The co-ordinator(s) are also responsible for uploading data to cancer waiting times databases and cancer registries.
An annual report documenting MDT activity, including workload, outcomes, significant events such as staff changes, waiting time targets, critical incidents, and ‘never events’ should be written.
The head and neck MDT should have formal links with inter-dependent MDTs for other related tumour types, such as skull base or neurosurgical, sarcoma, skin cancer and melanoma, thyroid, teenage and young adult, cancer of unknown primary, and lung MDTs.
Following MDT discussions, patients should be reviewed in an MDT clinic, in which core members can provide input for each patient depending on their needs in one visit. This would typically include, for example:
• Primary clinician(s) (surgical or oncological) offering prime treatment or explaining treatment options
• Clinical nurse specialist
• Allied health professionals (speech and language therapy, or dietetics), as required
• Dental assessment including orthopantomagram X-ray
• Anaesthetic assessment (ideally) for patients needing surgery
• Clinical trial nurse if clinical trial option is available
Surgery and radiotherapy within treatment centre
Surgery
A comprehensive head and neck cancer centre should be able to offer all patients the surgical treatment they require. It is accepted that not every head and neck cancer surgical centre will offer treatment for rarer cancers, or in niche areas, such as skull base cancer surgery.
Surgeons should have job plans reflecting their commitment to specialist practice. Subspecialty training in head and neck oncology is essential, ideally including a fellowship after completion of training.
Staffing numbers in all aspects of care need to be large enough to provide a sustainable service 24 hours a day, 7 days a week, allowing cover for holidays and unexpected absences. Case numbers need to be large enough to maintain individual expertise amongst all the personnel, whilst providing case experience for trainees.
The essential and desirable services that the main head and neck cancer surgical centres should offer are shown in Table 3 below. This assumes the provision of the MDT members listed above.
Table 3. Essential and desirable services that the main H&N cancer surgical centre (the ‘hub’) should provide
H&N = head and neck; CT = computed tomography; MRI = magnetic resonance imaging; USS = ultrasound scanning; AHP = allied health professional; SLT = speech and language therapist; 24/7 = 24 hours a day, 7 days a week; RIG = radiologically inserted gastrostomy; PEG = percutaneous endoscopic gastrostomy; 3D = three-dimensional
Radiotherapy
Radiotherapy is already centralised in 62 UK centres, most but not all of which treat head and neck cancer. Case numbers are relatively small – there were 5987 courses of curative head and neck cancer radiotherapy delivered in England in 2017, compared with 30 708 for breast cancer.10 In 2020, there were 149 UK clinical oncologists treating head and neck patients.11
Head and neck radiotherapy is one of the most complex treatments given by clinical oncologists. The 2019 NHS England Service Specification for Radiotherapy Operational Delivery Networks states that each consultant clinical oncologist should be responsible for at least 25–50 cases of radical radiotherapy per year for each tumour site treated.12 The ‘25–50’ number is empirical and not supported by strong evidence. As different head and neck subsites are also managed differently, it could be argued that nasopharyngeal cancer, for example, is so rare in the UK and so different to other tumour subsites that it should be super-specialised and treated by a very small number of clinical oncologists.
Rather than trying to establish an ideal number of cases to be treated by a radiotherapy team, it is perhaps more sensible for commissioners to examine surrogate indicators of radiotherapy quality which are likely to reduce unwarranted variation between clinical teams and centres (Table 4).
Table 4. Surrogate indicators of radiotherapy service quality
H&N = head and neck; NICE = National Institute for Health and Care Excellence; BAHNO = British Association of Head and Neck Oncologists; RCR = Royal College of Radiologists; SLT = speech and language therapy; RT = radiotherapy; NCRN = National Cancer Research Network
Follow up and data collection
The purpose of interactions with patients following treatment for head and neck cancer is not just about tumour surveillance. Restoring quality of life is as important as cancer cure. Follow-up services should have speech and language therapists, dietitians, clinical nurse specialists, and restorative dentistry at their core, delivered as close to home as possible. Psychological support is often less readily available.
The collection of survival metrics, patient-reported outcome and experience measures is key to assess and improve care. Each MDT should be able to collect and publish its outcomes. Follow-up services should be designed and funded with the ability to collect such data from every patient. Completing simple, validated digital tools before a clinic visit can enable more patient-focused appointments, and their analysis can help drive improvements in the quality of the clinical service.
The future: evidence for minimal treatment volumes and further centralisation?
The importance of concentrating patient volumes in specialised head and neck services through reconfiguration and centralisation has been illustrated in reports from Canada, USA, UK and Europe. The weight of opinion has moved away from the 100 new cases recommended by the NICE Improving Outcomes Guidelines in 2004.
The questions that arise are: what is the minimum or optimal size of a head and neck cancer centre?; and is it practical and affordable? The difficulty in addressing these within the UK is exemplified by the lack of any update on guidance since 2004.
There are consistent data that support a relationship between hospital or institutional volume or caseload and various measures of quality, mainly improved patient survival and a reduction in complications.Reference Eskander, Irish, Groome, Freeman, Gullane and Gilbert13–Reference Robertson, Robertson, Soutar, Burns, Hole and McCarron17 However, the definition of what defines a low-volume and high-volume unit are often empirical. The largest analysis of 351 052 head and neck squamous cell carcinoma cases treated in 1229 centres found that low-volume units treating less than 54 cases per year had significantly higher mortality rates than moderate volume units (54–164 cases per year), and high-volume units with over 165 cases per year were even better.Reference Torabi, Benchetrit, Kuo, Cheraghlou, Savoca and Tate18
The recent Getting It Right First Time review of oral and maxillofacial surgery servicesReference Morton19 suggested that the figure of 250 new cases per MDT was optimal.
Figures taken from analysis of the 2013 Data for Head and Neck Oncology (‘DAHNO’) audit, and presented by the initial head and neck surgery clinical reference group to NHS England, formed the basis of estimated caseload, and effect upon MDT numbers was considered if alterations were made to the required caseload. The data were illustrative, rather than definitive, and are shown in Table 5.
Table 5. Projected number of surgical treatments based on annual H&N MDT caseload
H&N = head and neck; MDT = multidisciplinary team
If it is assumed that a centre must have a comprehensive free-flap service, then a minimum number of free flaps per year (and per reconstructive surgeon) could be considered. There is no UK consensus on this. Canadian guidelines stipulate 20 free flaps per surgeon per year. Assuming 3 surgeons are required to support a 24 hours a day, 7 days a week service, but perhaps factoring in more than 1 reconstructive surgeon per case, this would suggest an annual case load of at least 300 might be required (Table 5). The figures in Table 5 (essentially 1 free flap per 10 new patients with head and neck cancer) are consistent with the known incidence of head and neck cancer in England (average of 10 053 cases in 2016–2018)2 and coded free flaps per year for head and neck cancer (average of 1184 cases in 2003–2013).Reference Nouraei, Middleton, Hudovsky, Branford, Lau and Clarke20
Hence, it is difficult to define a metric that defines minimal caseload or throughput. The variance in cancer incidence and geographical spread or population density within the UK also needs to be factored in.
A reasonable conclusion is that, where geographically possible, the provision of head and neck cancer services should follow the evidence, i.e. have as high volume as possible. Within cities and conurbations, there should be the maximum consolidation into as few head and neck treatment centres as possible. In these circumstances, a head and neck treatment centre should comprise a single surgical centre and a single oncological centre (and co-location is ideal). For more rural, less populated areas, a practical compromise might entail having smaller centres within a service offering restricted services (e.g. no microvascular surgery) linked to a large centre with the same operational policies and linked MDT meetings.
Introduction
Head and neck malignancy is the fourth commonest reason for referral on a cancer pathway, yet the eighth commonest cancer diagnosed in England.21 Diagnosis at an early stage is a challenge in head and neck cancer, as many patients arise from lower socio-economic groups and late presentation is common. Best practice timed pathways in secondary care aim to expedite and streamline diagnosis. However, the lack of patient engagement with primary care services remains a fundamental issue.
This chapter outlines suggested best practice, and summarises the available evidence concerning the initial assessment and diagnosis of patients with suspected head and neck cancer, from primary care referral through to diagnosis. The focus is predominantly upon upper aerodigestive tract head and neck squamous cell carcinoma (SCC) and salivary cancers. Diagnosis of thyroid and skin pathology are covered in dedicated chapters.
The process of a patient with potentially suspicious symptoms can be broken down into three steps, culminating with a patient confirmed with cancer and appropriately investigated and staged (Figure 1). Each of these steps has its challenges.
Figure 1. Process for a patient with potentially suspicious symptoms. GP = general practitioner; MDT = multidisciplinary team
Public awareness and referral from primary to secondary care
Recommendations
• Co-ordinated public awareness campaigns to boost recognition of the early signs of head and neck cancer should be promoted (desirable (D))
• Improved primary care awareness of the early signs of head and neck cancer is needed to reduce time to diagnosis and emergency late-stage presentation (D)
• Collaboration between cancer alliances and primary care is required to agree upon clear and appropriate two-week-wait referral criteria (essential (E))
• Use and adoption of validated risk calculators is recommended to assist with urgent referral (accepted standard of care (evidence-based recommendation (R))
On average 100 000 new head and neck referrals are seen in each year, with only 2.8 per cent on the cancer waiting time pathway resulting in a cancer diagnosis; yet only 35–56 per cent of patients diagnosed with cancer present via this route in England.21,22
These data demonstrate deficiencies in the current referral process for head and neck cancer. Whilst urgent clinics are full of patients without cancer, nearly half of head and neck malignancies present via other routes, too frequently with advanced disease. The negative impact of delayed diagnosis upon survival and quality of life is well proven.Reference Seoane, Takkouche, Varela-Centelles, Tomás and Seoane-Romero23
The National Institute for Health and Care Excellence (NICE) refined their referral guidelines in 2015 to include symptoms with a positive predictive value above 3 per cent (Table 1).24 These adjustments were made in response to consistent data proving no discernible difference between the numbers of patients being diagnosed through the urgent and non-urgent pathways. Regrettably, the detection rate remains low despite these changes.Reference Gao, Qin, Freeman, Oskooee and Hughes25
Table 1. NICE guidelines 2015
NICE = National Institute for Health and Care Excellence
An alternative list of referral criteria has been proven to correlate well with a head and neck cancer diagnosis (Table 2).Reference Tikka, Pracy and Paleri26 This has formed the basis for a risk calculator to guide primary care referral or triage (www.orlhealth.com/risk-calculator.html). This has been refined and validated, and shown to be effective for triage during the coronavirus disease 2019 (Covid-19) pandemic. It uses symptoms such as intermittent hoarseness and globus sensation as negative predictors of malignancy.Reference Tikka, Kavanagh, Lowit, Jiafeng, Burns and Nixon27,Reference Hardman, Tikka and Paleri28
Table 2. Suggested referral criteria after Tikka et al.Reference Tikka, Pracy and Paleri26
FOSIT = feeling of something in throat
More fundamentally, early diagnosis is contingent on patients seeking medical attention with suspicious symptoms. Public awareness campaigns have successfully increased numbers of patients with lung cancer referred and diagnosed at an early stage. The same is required for head and neck cancer, particularly in deprived populations.
Assessment in secondary care and best timed pathways
Recommendations
• Use of validated risk calculators is recommended to assist initial assessment and triage (evidence-based recommendation (R))
• Trusts should commit to achieving the faster diagnosis standard, with communication of the outcome to patients and general practitioners (essential (E))
• Patients with suspected cancer should be seen in a dedicated ENT, oral and maxillofacial surgery or neck lump specialist clinic (R)
• Best timed pathways centred around one-stop clinics for diagnosis of head and neck cancers should be implemented, enabling same-day investigations and expedient booking of diagnostic imaging and general anaesthesia procedures (R)
• Equipment for endoscopy, and biopsy in clinic and dental procedures must be readily available (R)
• Video monitors and facilities for photographic documentation should be available (good practice point (G))
• Narrow band imaging or optical florescence imaging can assist with the identification of malignancy and dysplasia (R)
• Diagnostic imaging should be performed prior to biopsy under anaesthesia if possible, but biopsy in clinic should not be delayed and should be carried out initially if possible (G)
• Patients with cervical lymphadenopathy require fibre-optic nasopharyngolaryngoscopy (R)
• A neck lump clinic should include same-day ultrasound and ultrasound-guided biopsy (fine needle aspiration cytology (FNAC) or core biopsy) (G)
• Ultrasound-guided biopsy should be performed by an experienced head and neck radiologist, ultrasonographer, or appropriately trained surgeon capable of performing diagnostic ultrasound imaging and both FNAC and core needle biopsy (G)
• Immediate FNAC adequacy checking improves the rate of diagnostic sampling (R)
• Rapid cytopathological assessment of FNAC samples is desirable (G)
The National Health Service's (NHS's) long-term plan aims to achieve 55 000 extra patients surviving five years after treatment, with an early-stage diagnosis for 75 per cent of patients (with a commitment to achieving a ‘faster diagnosis standard’, either confirming malignancy or the exclusion thereof, within 28 days of referral).29
Alternative methods of assessment using tele- and video-consultations, straight-to-test pathways, and so on were implemented in response to Covid-19 and may prove to be effective post-pandemic. Although originally intended to assist with the selection of cases for urgent referral from primary care, the refined and validated risk calculator (www.orlhealth.com/risk-calculator.html) has been shown to be effective for initial telephone triage or assessment during the Covid-19 pandemic.Reference Hardman, Tikka and Paleri28 This appears to be a safe and effective means of selecting those patients who should remain on an urgent diagnostic pathway and determining those who can be seen routinely. At this stage, allocation to the appropriate clinic can be carried out (e.g. ENT clinic, oral and maxillofacial surgery clinic, neck lump clinic). Hence, assessment in secondary care may comprise any of the following steps:
• Triage
• Initial telephone assessment
• Investigations before clinic attendance
• Clinic attendance
• Investigations during or after clinic attendance
For those patients on an urgent diagnostic pathway in secondary care, best practice or optimal pathways should be adopted and implemented. This involves early, appropriate investigations and a reduction of unnecessary investigations, together with the simplification of diagnostic pathways. Such pathways may vary and be tailored to the facilities and practices within the hospital concerned.
Out-patient assessment
In addition to pertinent head and neck symptoms and risk factors, co-morbidities, performance status and social circumstances should be recorded for those patients who have or may have cancer. This is vital information to inform multidisciplinary team (MDT) discussion and treatment planning.
Detailed description of examination findings should be documented. Photographic documentation is recommended, to aid MDT discussion and disease monitoring.
Patients with oral cavity lesions should be seen in a clinic with the capabilities for dental radiography and local anaesthetic biopsy. Palpation of lesions clinically is useful to estimate thickness and depth. Pre-malignant-looking oral lesions should be investigated as early oral cavity cancers.
Patients with pharygolaryngeal or sinonasal symptoms and those with neck lumps that may be metastatic lymph nodes require flexible nasopharyngolaryngoscopy. Video equipment offers higher definition images, which are helpful for training and peer review of appearances. Narrow band imaging and optical fluorescence imaging may improve the detection of subtle malignant and pre-malignant lesions.Reference van Schaik, Halmos, Witjes and Plaat30
Transnasal oesophagoscopy can be considered; it enables enhanced views of the hypopharynx and upper oesophagus compared to conventional flexible laryngoscopy, and offers benefit when excluding oesophageal pathology in patients with upper dysphagia or globus symptoms.Reference Abou-Nader, Wilson and Paleri31
A representative biopsy in clinic should be performed for accessible lesions (oral cavity, and some nasal or oropharyngeal lesions). This allows for early confirmed diagnosis and greater understanding of the tumour biology, for example depth of invasion, peri-neural or lymphovascular invasion, which inform pertinent management decisions. Whilst diagnostic imaging prior to biopsy may be preferable to avoid distortion of anatomy, the impact upon accurate disease staging is inconclusive,Reference Howe, Khurram, Hunter, Martin and Fry32 and it is recognised that diagnostic time pressures may impact upon this ideal pathway.
Biopsy of laryngeal and pharyngeal lesions with transnasal oesophagoscopy or a channelled flexible laryngoscope offers a safe and effective means of gaining a histological sample in out-patients, avoiding general anaesthetic in a significant proportion.Reference Schutte, Takes, Slootweg, Arts, Honings and van den Hoogen33 The depth of biopsy can be a limiting factor in some cases.
Neck lumps
The key features of assessment are:
• Complete head and neck examination, including fibre-optic nasopharyngolaryngoscopy and skin inspection, if the lump is possibly a lymph node
• Ultrasound (as part of a one-stop clinic ideally)
• Ultrasound-guided tissue sampling (FNAC or core biopsy) (see also Chapter 3 on pathology)
The evidence for ultrasound guidance in needle biopsy sampling of neck masses to reduce sampling errors is compelling.Reference Cozens34
Adequacy check and rapid on-site evaluation
On-site assessment of the adequacy of cytopathology samples by biomedical scientists reduces pathway delays due to non-diagnostic specimens. Rapid on-site evaluation by cytopathologists goes one step further. Not only are non-diagnostic sampling rates further reduced,Reference Breeze, Poller, Gibson, Tilley, Cooke and Soar35 but also same-day diagnosis offers unmeasurable advantages from the patient's perspective, and speeds up subsequent steps on the diagnostic or staging pathway. Furthermore, ancillary tests such as core biopsy for suspected lymphoma can then be performed immediately, if necessary.
Fine needle aspiration cytology versus core needle biopsy
Core biopsy, although more invasive, may offer increased accuracy rates over FNAC for immunohistochemical staining (and prevents the need for incision or excision biopsy), particularly when lymphoma is suspected, in cases of salivary gland tumoursReference Douville and Bradford36 and for unusual tumours (see also Chapter 3 on pathology).
Preference between core biopsy and FNAC may depend on local cytopathological expertise. Because core biopsy may not be suitable for smaller lesions or those within proximity to neurovascular structures, it is recommended that the personnel sampling the neck masses are familiar with both techniques.
Many lateral cystic neck masses in patients aged over 40 years prove to be malignant, most commonly metastatic human papillomavirus (HPV)-associated SCC from an oropharyngeal primary.Reference Goldenberg, Begum, Westra, Khan, Sciubba and Pai37 Achieving diagnosis by FNAC or core biopsy can be difficult when the lymph node is mainly cystic. Targeting the cyst wall can help optimise accuracy (see Chapter 27, dealing with unknown primary cancer).
Open biopsy (by incision or excision) should be a last resort to achieve histological diagnosis. The FNAC and core biopsy should be performed prior to this, as well as imaging to identify a possible primary in the case of suspected lymph node metastasis (suspected carcinoma of unknown primary).
Unknown primary: metastatic squamous cell carcinoma
This topic is covered in more detail in the dedicated chapter (Chapter 27). Most patients with carcinoma of unknown primary SCC will present to neck lump clinics and should be managed as outlined above.
Panendoscopy or examination under anaesthesia and biopsy of primary cancer
Many lesions will not be amenable to clinical staging and biopsy in clinic. This should be carried out after other investigations and imaging.
When examination under anaesthesia and biopsy is required, a panendoscopy, examination of all of the upper aerodigestive tract (oral cavity, pharynx, larynx and cervical oesophagus), should ideally be performed, although the detection rate of second primary malignancies is very low through this.
For carcinoma of unknown primary, positron emission tomography/computed tomography (PET-CT) should be carried out before this step and requested as soon as carcinoma of unknown primary is suspected. Chapter 27 discusses what should be biopsied and how.
Imaging in head and neck cancer
Recommendations
• Magnetic resonance imaging is recommended for supra-hyoid primary lesions, and for cases of carcinoma with an unknown primary (good practice point (G))
• Computed tomography is recommended to assess bony involvement where required (evidence-based recommendation (R))
• Computed tomography thorax is mandated for systemic staging in advanced disease and may be used to exclude synchronous primary lung malignancies in the early stage of disease (R)
• Positron emission tomography/CT is advocated for the assessment of distant metastasis in advanced tumour (T) stage (T4) nasopharyngeal and hypopharyngeal malignancies, and in examination of nodal (N) stage N3 cancers (R)
• Positron emission tomography/CT is recommended for the post-treatment assessment of nodal disease following non-surgical treatment (R)
• Post-treatment assessment of primary tumours should follow the original modality used for primary staging (G)
The roles of imaging for head and neck cancers are: to assess and characterise a suspected lesion, and to stage local, regional and distant metastatic disease; and to detect second primary malignancy. The Royal College of Radiologists published guidelines in 2014.38
Imaging modality and primary tumour considerations
Computed tomography
Advantages:
• Widely available
• Short scanning time
• Excellent anatomical resolution
• No contraindications through ferrous implants
• Limited problems for patients with claustrophobia
Limitations:
• Radiation exposure
• Risk of contrast-induced nephropathy
• Distortion from dental amalgam artefact
• Limited contrast resolution inherent to the technique means that magnetic resonance imaging (MRI) is often preferred, particularly in the suprahyoid neck, salivary glands, sinonasal areas, nasopharynx and skull base, but increasingly for the larynx and hypopharynx. However, CT is still extremely valuable for patients in whom MRI is contraindicated, to provide bony detail and when rapid scan acquisition is required.Reference Batuwitage, Hanlon and Charters39
A variety of techniques can be used with CT to improve visualisation in certain subsites, including: shallow free breathing for laryngeal lesions; e-phonation for lesions of the laryngeal ventricle, anterior commissure and aryepiglottic folds; and the ‘puffed cheek’ technique for buccal mucosa lesions. Slice thickness depends upon scanning capability, but, in general, sections are acquired at 0.625–1.25 mm and reformatted no greater than 2.5 mm for viewing. Intravenous contrast improves the delineation of tumour extent and the detection of lymph nodes. Traditionally, a scan delay of 50–75 seconds was used to allow adequate enhancement of primary tumour and accurate differentiation of cervical lymph nodes from vessels, although a delay of 90 seconds is thought to improve this further and is now used widely.Reference Bartz, Case, Srinivasan and Mukherji40
Magnetic resonance imaging
Advantages:
• Superior contrast resolution
• No radiation
• The risks of intravenous gadolinium-based contrast are also lower than with iodinated CT-based contrastReference Batuwitage, Hanlon and Charters39
Limitations:
• Longer scanning times
• More susceptible to image degradation and therefore technically more challenging
• Magnetic resonance imaging may also be contraindicated in the presence of metallic implants and cardiac devices
• Difficult for patients with claustrophobia
With superior contrast resolution, MRI is the preferred modality for multiple subsites.
Scanning times are longer causing restricted availability, and MRI therefore may be reserved for select cases when CT findings are inconclusive.
Protocols vary, but they include axial and coronal T1- and T2-weighted imaging with further post-contrast T1-weighted sequences. Diffusion-weighted imaging may increase the conspicuity of small lesions, but has a role mainly in post-therapy imaging. A variety of new MRI sequences are being introduced to help in certain situations, for example for better evaluation of peri-neural tumour spread.Reference Touska and Connor41
Ultrasound scanning
Ultrasound scanning is particularly useful in the evaluation of neck nodes. All imaging modalities use size criteria for the evaluation of lymph nodes (generally, a short axis of more than 10 mm), but there is no perfect size threshold. Morphological ultrasound characteristics such as extracapsular spread, non-hilar vascular pattern, parenchymal granular echoes, necrosis, as well as the clustering of nodes are features that suggest malignancy. Ultrasound scanning with ultrasound-guided fine needle aspiration remains the most accurate test for the differentiation of benign from malignant nodal disease.Reference Richards and Peacock42
Ultrasound assessment of primary tumour sites is limited, with high resolution ultrasound occasionally performed at a small number of centres to aid in tumour (T) staging and to determine the depth of invasion of some oral cavity or laryngeal tumours.Reference Marchi, Filauro, Iandelli, Carobbio, Mazzola and Santori43
Positron emission tomography/computed tomography
Combined 18F-fluoro-2-deoxyglucose (18F-FDG) PET-CT is useful for:
• The detection of the unknown primary cancer (see Chapter 27)
• The detection of metastatic disease in the setting of locoregional advanced cancers with a high risk for distant metastatic disease
• Post-treatment imaging (particularly after chemoradiotherapy)
Combined 18F-FDG PET-CT is also useful to exclude distant metastatic disease when contemplating salvage surgery for recurrent cancer.
In the setting of a cancer with an unknown primary, accurate identification of occult primary sites is clearly important. The 18F-FDG positron emission tomography (PET) scan detects 25 per cent of tumours not apparent after conventional investigation.Reference Rusthoven, Koshy and Paulino44
The PET-CT has enhanced sensitivity (83 per cent) and specificity (96 per cent) for the detection of metastatic disease in patients with head and neck cancer; the corresponding values for conventional imaging methods are 44 per cent and 96 per cent, respectively. These results show the advantage of using 18F-FDG PET-CT over conventional imaging methods to detect distant metastatic disease.Reference Xu, Li, Zuo and Li45
Current NICE guidelines indicate the use of PET-CT for T4 cancers of the hypopharynx and nasopharynx and N3 cancers which are most likely to have distant metastases.46
Metastatic disease
Cervical lymph node metastases:
• Regional lymph nodal involvement is typically determined with the cross-sectional imaging modality used for delineation of the primary site
• Where uncertainty exists, ultrasound scanning (with or without needle biopsy) is advocated
Diffusion-weighted MRI shows promise in the detection of malignant nodes, but there is some difficulty in the translation of apparent diffusion co-efficient values between MRI systems, and the results are variable.Reference Chen, Hagiwara, Givi, Schmidt, Liu and Chen47
Positron emission tomography/CT does not demonstrate significant improvement in sensitivity and specificity in comparison to conventional cross-sectional imaging, limited by small size of some metastatic nodes and inflammatory uptake.Reference Mukherji and Bradford48
Distant metastasis
The lung is the commonest site of distant metastasis, representing 66 per cent of haematogenous metastases, with mediastinum, bone and liver sites also recognised.Reference Ferlito, Shaha, Silver, Rinaldo and Mondin49 The NICE guidelines suggest CT of the chest for the detection of metastatic lung disease in higher risk tumours, excluding T1N0 and T2N0 tumours.46 A dilemma exists regarding the use of CT of the thorax as lung screening, given the similar risk factors and high instance of synchronous lung cancer in head and neck patients.Reference Piersiala, Akst, Hillel and Best50 The National Comprehensive Cancer Network classifies head and neck cancer patients with a 20 pack-year smoking history as high risk and recommends annual screening. Many centres therefore follow a pragmatic approach of routinely performing CT of the chest in patients with head and neck carcinoma, irrespective of primary stage.
As above, PET-CT should be used for T4 cancers of the hypopharynx and nasopharynx and N3 cancers which are most likely to have distant metastases.46 It could also be considered for unusual tumours at higher risk of non-lung distant metastasis.
Post-treatment imaging
Post-treatment imaging is primarily aimed at identifying residual or recurrent disease but also detects late effects of treatment and second primary malignancies.
Positron emission tomography/CT is advocated for patients who have been treated with radical chemoradiotherapy for oropharyngeal, hypopharyngeal or laryngeal primaries with nodal involvement, following the outcomes of the ‘PET-NECK’ trial,Reference Mehanna, Wong, McConkey, Rahman, Robinson and Hartley51 typically at 10–12 weeks post treatment.
For other tumours, imaging usually follows the original modality used for staging, to allow for comparison. In some of these situations, for example in laryngeal cancer without nodal disease treated non-surgically, PET-CT demonstrates non-inferiority in comparison to direct laryngoscopy, as shown within the ‘RELAPS’ (REcurrent LAryngeal carcinoma PET Study) trial.Reference de Bree, van der Putten, van Tinteren, Wedman, Oyen and Janssen52
Having a post-treatment baseline scan is also useful for the comparison of future imaging, should it be needed; thus, the effects of surgery and radiotherapy on imaging can be factored in through reference to the baseline imaging.
A summary of recommended imaging is shown in Table 3. More specific detail can be found in the relevant site-specific chapters.
Table 3. Summary of recommended imaging
* Positron emission tomography/computed tomography (PET-CT) is recommended instead for nodal stage N3 disease, tumour stage T4 cancer of the nasopharynx and T4 cancer of the hypopharynx. MRI = magnetic resonance imaging; CT = computed tomography; CUP = carcinoma of unknown primary
Staging of head and neck cancer
In conjunction with the Union for International Cancer Control (‘UICC’), the eighth edition of the TNM Classification of Malignant Tumours was released in December 2016.Reference Brierley, Gospodarowicz and Wittekind53 The major changes from the seventh edition are as follows:
• A new classification group for HPV-associated oropharyngeal cancers
• Carcinoma of unknown primary: stage groupings now differentiate between non-viral, HPV-related and Epstein–Barr virus (EBV)-related
• A new N3b stage reflecting the prognostic importance of extra-nodal extension
• Oral cavity T-staging: the impact of depth of invasion
• Thyroid cancer staging reflects survival rather than risk of recurrence
The detail of the staging for each site can be found in the relevant chapter.
The eighth edition of the TNM Classification of Malignant Tumours acknowledges the distinct biological behaviour of HPV-associated oropharyngeal cancers with a new classification group. The main difference relates to regional metastasis, where there is recognition that multiple ipsilateral lymph nodes are frequently encountered in HPV-associated disease and do not carry the poor prognosis seen in non-HPV-related cancers. It also differentiates the clinical and post-surgical nodal staging.
Carcinoma of unknown primary stage groupings now differentiate between non-viral, HPV- and EBV-related. Alterations reflect differences in prognosis, with non-HPV-related carcinoma of an unknown primary including only stages III and IV, and EBV-related carcinoma of an unknown primary including stages II–IV, to acknowledge the poorer prognosis when compared to HPV-associated disease which includes stages I–IV. In other non-viral related cancers there are new clinical and pathological nodal staging categories, where the prognostic implication of extra-nodal extension is recognised with a new N3b stage.
Changes in oral cavity T-staging recognise the impact of depth of invasion, and in nasopharyngeal cancers amendments to both the primary tumour and nodal staging are seen. The changes to the staging of differentiated and anaplastic thyroid malignancies in the eighth edition of the TNM Classification of Malignant Tumours aims to represent the risk to life from a thyroid cancer diagnosis, as opposed to risk of recurrence as seen in previous staging and risk stratification systems.
Exclusion of second primary malignancies
Recommendations
• Patients with new and recurrent upper aerodigestive tract cancers should undergo comprehensive clinical examination, including endoscopic nasopharyngolaryngoscopy, as part of diagnostic investigation (evidence-based recommendation (R))
• Cross-sectional imaging, appropriate for tumour site, should be performed to screen for upper aerodigestive tract and thoracic second primary cancer (good practice point (G))
Head and neck cancers have the highest propensity of all malignancies for synchronous and metachronous cancers,Reference Tanjak, Suktitipat, Vorasan, Juengwiwattanakitti, Thiengtrong and Songjang54 with reported incidences of around 13 per cent.Reference Coca-Pelaz, Rodrigo, Suárez, Nixon, Mäkitie and Sanabria55 Second head and neck primaries are commonest, followed by oesophageal cancers and lung cancers.Reference Kuhlin, Kramer, Nefas, Rotter and Aderhold56
The rate of synchronous second primary malignancies, i.e. present at or around the time of diagnosis of the index tumour, is around 5 per cent.Reference Coca-Pelaz, Rodrigo, Suárez, Nixon, Mäkitie and Sanabria55
These sites are usually encompassed by imaging performed for the primary diagnosis and staging (including the thorax).
Panendoscopy under general anaesthetic offers a low diagnostic yield in the absence of significant symptoms in patients with head and neck cancers,Reference McGarey, O'Rourke, Owen, Shonka, Reibel and Levine57 and has fallen out of favour as a screening tool in this population. Incidental second oesophageal primaries are more likely to be better detected with out-patient endoscopy and imaging alone. However, transnasal oesophagoscopy could play an important role in the exclusion of both second head and neck and oesophageal primaries.Reference Abou-Nader, Wilson and Paleri31
Best timed pathways
Best timed pathways (also known as optimal pathways) put together all the elements described above, with a view to fast diagnosis in as few hospital visits as possible. They are somewhat idealised and can be difficult to adhere to (e.g. when an initial biopsy is non-diagnostic).
Examples of these have been described by NHS Wales.58
Simplified examples for neck lumps and upper aerodigestive tract symptoms can be found in Figure 2. The selection of the investigations will depend on the clinical situation, and this has been described in earlier sections.
Figure 2. Twenty-eight-day best practice timed pathway for neck lumps and upper aerodigestive tract symptoms. GP = general practitioner; GDP = general dental practitioner; OMF = oral and maxillofacial; MRI = magnetic resonance imaging; OPG = orthopantomogram; ROSE = rapid on-site evaluation; CT = computed tomography; LA = local anaesthetic; EUA = examination under anaesthesia; GA = general anaesthetic; MDT = multidisciplinary team; PET = positron emission tomography; OPA = out-patient appointment; CNS = clinical nurse specialist; FDS = faster diagnosis standard; pre-op = pre-operative
Important issues to be answered
These include:
• Refinement and development of cost-effective methods to improve rates of early-stage diagnosis
• Optimisation of systems to discern suspicious head and neck symptoms from those suggestive of benign disease
• Understanding of the role of transnasal oesophagoscopy in diagnosis and detection of second primary malignancies
Clinical trials due to report
‘EVEREST-HN’ (2022–2028): EVolution of a patiEnt-REported symptom-based risk stratification sySTem to redesign the suspected Head and Neck cancer referral pathway (https://fundingawards.nihr.ac.uk/award/NIHR202862).
Key recommendations
• The head and neck multidisciplinary team (MDT) should understand the scope and limitations of histopathology and cytopathology in order to inform multidisciplinary discussion. This requires effective communication between clinicians and pathologists within the MDT (evidence-based recommendation (R))
• A clinically suspected diagnosis of malignancy should be confirmed by biopsy or cytology prior to oncological treatment (R)
• Participation in the National Head and Neck Histopathology External Quality Assurance Scheme is essential (R)
• The current World Health Organization (WHO) dysplasia grading system for oral cavity and larynx should be used (R)
• The meaning of the descriptors ‘carcinoma in situ’ and ‘severe epithelial dysplasia’ should be well understood between the reporting pathologist and the MDT (good practice point (G))
• Correct orientation of resection specimens and accompanying clinical information are essential (R)
Introduction
Pathologists have critically important roles in confirming or excluding specific diseases, assessing the adequacy of treatment, recognising key predictive and prognostic factors, and contributing to the evidence-based stratification of clinical outcomes for audit and research purposes. This document is aimed at providing a broad overview of the key elements of pathology services required to underpin best practice for the multidisciplinary management of head and neck cancer patients. These guidelines also aim to provide salient aspects of the pathology report that surgeons, oncologists and allied health professionals should consider when discussing the implications of a diagnosis and management options with patients and the MDT. The recommendations for pathology practice are based on published evidence as well as accepted standard practice, and have been endorsed by the Royal College of Pathologists. These guidelines are largely based on the WHO Classification of Head and Neck Tumours, and the histopathology reporting datasets published by the Royal College of Pathologists and the International Collaboration on Cancer Reporting.59–66
Head and neck pathology services
Quality management for head and neck pathology services
Robust internal and external quality assurance programmes are inherent in any head and neck pathology service. The pathology service should be embedded within a medical laboratory that is accredited against the international standard ISO15189:2012 by the UK Accreditation Service and integrated within local Cancer Alliances.67–69 Local departmental consensus reporting for all borderline malignant or equivocal cases is encouraged.
Multidisciplinary team working
Histopathologists and cytopathologists are core members of cancer MDTs, and are essential to the provision of a successful service. Participation in the National Head and Neck Histopathology External Quality Assurance Scheme is essential.69–Reference Schache, Kerawala, Ahmed, Brennan, Cook and Garrett71 The MDT should be able to review external pathology, in accordance with a risk-based approach, particularly for patients who have had diagnostic biopsies performed at non-specialist centres. For all newly diagnosed thyroid malignancies (including Thy4 (abnormal, suspicious of malignancy) and Thy5 (diagnostic of malignancy) cytology specimens), a central cancer network pathology review by a specialist thyroid MDT pathologist is necessary.72,73 Evaluation of excision completeness may not be possible for resections specimens that have been processed at other centres.
Tissue pathways
Patient management should be guided primarily by pre-operative biopsy and/or fine needle aspiration cytology (FNAC).
Frozen sections
The sensitivity of frozen sections for margin assessment is suboptimal, but there may be clinical utility in certain intra-operative scenarios, including determining the extent of margin involvement, and confirmation or exclusion of lymph node metastasis.Reference Kain, Birkeland, Udayakumar, Morlandt, Stevens and Carroll74 When required, frozen sections should be pre-arranged with the histopathology department. Clear annotation of the anatomical location of frozen sections enables correlation with the main resection specimen for final margin evaluation. The indications of frozen sections for intra-operative diagnosis are highly limited and its routine use is discouraged.
Resection specimens
Some specimens may require sampling of fresh tissue for purposes such as biobanking, special diagnostic tests and research. Following sampling of fresh tissue, specimens should be fully submerged in 10 per cent neutral buffered formalin in at least four times the volume of the specimen as soon as possible, to avoid denaturing. The site, laterality, clinical stage and nature of each specimen should be clearly indicated on the request form. The form must include the clinical indication for the operation, pre-operative radiotherapy or chemotherapy, details of previous biopsies or cytological investigations, and relevant biochemistry (particularly for thyroid diseases). Correct orientation of the specimen is of paramount importance and needs to be clearly communicated to the pathologist, preferably with the use of labelled diagrams, sutures or other markers on important structures, and peri-operative clinical photographs. Macroscopic photography of specimens is highly beneficial when communicating post-surgical pathological findings at MDT meetings.Reference Simo, Morgan, Jeannon, Odell, Harrison and Almeida75
Resection specimens containing calcified tissue
Optimal decalcification is a balance between preservation of histomorphology and timeliness. Decalcification end-point testing requires experienced laboratory technical staff with optimised standard operating procedures. In resection specimens containing bone, at least one non-decalcified tissue block containing tumour should be obtained. This is to preserve the quality of antigens and nucleic acids for immunohistochemistry and molecular testing, should these be required. Decalcification may take several days or weeks. If necessary, a provisional report on the soft tissue components of the specimen may be issued for adjuvant treatment planning purposes.
Neck dissection specimens
Neck dissections should be orientated with all important structures (e.g. internal jugular vein) clearly labelled, and the nodal groups indicated, preferably with a diagram.63,Reference Woolgar and Triantafyllou76 Avoiding errors in the interpretation of post-operative neck levels is challenging.Reference Varley, Howe, Hunter and Smith77
Ancillary testing
Immunohistochemistry
Immunohistochemistry plays an important role in the correct diagnosis of primary head and neck cancers, particularly for the less common entities. The addition of novel diagnostic antibodies to the repertoire should be part of the laboratory quality assurance programme. Except for p16, the prognostic use of immunohistochemistry in head and neck cancer is less established.
In situ hybridisation
DNA and RNA in situ hybridisation is used to determine the presence of human papillomavirus (HPV) and Epstein–Barr virus (EBV) in head and neck cancers.
Human papillomavirus testing
The HPV testing should be undertaken on all primary squamous cell carcinomas (SCCs) arising in the oropharynx and neck metastases from carcinomas of unknown primary.62,78 The sensitivity of DNA in situ hybridisation testing on fine needle aspiration (FNA) material is suboptimal.Reference Bernadt and Collins79 Routine HPV testing is not recommended for head and neck SCC outside the oropharynx.
Epstein–Barr virus testing
The EBV testing should be undertaken on all suspected primary nasopharyngeal carcinomas and neck metastases from carcinomas of unknown primary in which primary nasopharyngeal cancer is possible.62 The EBV testing may be indicated in certain lymphomas, but should always be performed as part of a wider lymphoma testing panel.
Carcinoma of unknown primary
Morphologically similar poorly differentiated carcinomas arising in the oropharynx and nasopharynx, and their nodal metastases, may be distinguished by the presence of HPV and EBV infection, respectively. In patients with metastatic malignancy in cervical lymph nodes without evidence of primary disease, the morphological features of the metastatic tumour may be useful, e.g. thyroid and salivary neoplasms. Immunocytochemical investigation of FNA or biopsy material does not reliably distinguish between primary SCC sites, but may be helpful in identifying metastatic carcinomas originating from the lungs, breast, gastrointestinal tract, mediastinum, genitourinary tract or prostate. Clinicians should note that immunohistochemical markers are very rarely specific for particular tissues, and that opinions on likely primary sites are based on the assessment of a panel of different markers, the availability of patient history and the balance of probabilities. Imaging studies and correlation with clinical features are essential for accurate multidisciplinary assessment of these patients. Molecular genetic profiling of head and neck cancers is not currently recommended outside the research setting.
Molecular testing
Access to accredited molecular testing facilities is essential. Tissue preparation and transfer between cellular and molecular pathology laboratories should adhere to standard operating procedures.Reference Cree, Deans, Ligtenberg, Normanno, Edsjo and Rouleau80 Gene mutation, rearrangement, amplification and deletion studies are important in some thyroid, salivary gland and sinonasal neoplasms, carcinomas with unknown primary, or when the nature of malignancy is unclear. A directory for genomic tests under NHS England and NHS Scotland describes the available tests.81,82 All molecular testing results should be interpreted in the context of clinical, histomorphological and immunohistochemical findings, and integrated into the final pathology report.Reference Cree, Deans, Ligtenberg, Normanno, Edsjo and Rouleau80
PD-L1
The PD-L1 (programmed death-ligand 1) testing should only be undertaken on cases requested by the MDT following clinical assessment of suitability for immunotherapy. The MDT should clearly state which immunotherapy drug is being considered, and convey whether the combined percentage score or tumour percentage score is required.83,84 Specific companion diagnostic antibodies and automated platforms are matched to the immunotherapy drug being considered. Diagnostic validation following a period of training is required prior to combined percentage score and tumour percentage score reporting. Because inter-observer variation is recognised, it is prudent that a combined percentage score or tumour percentage score close to the 1 per cent cut-off be the consensus scored by two pathologists.Reference Crosta, Boldorini, Bono, Brambilla, Dainese and Fusco85
Squamous cell carcinoma of upper aerodigestive tract
Squamous cell carcinoma, conventional type
Practical problems that may preclude definitive diagnosis on diagnostic biopsies include incomplete clinical information, poor orientation, necrotic or inflammatory debris, small samples containing few cells, and crush artefact. The edges of laser resection specimens often show thermal artefacts, making detailed interpretation impossible. Extensive scarring, radiation-associated nuclear atypia and loss of the normal anatomical landmarks in post-radiation specimens may make assessment difficult. A good chemo-radiotherapeutic response may leave a mass of necrotic tissue containing degenerate keratinocytes. Viable carcinoma may be difficult to identify even after extensive histological sampling, and immunohistochemistry may be of use in this setting to detect viable keratinocytes.
Morphological variants of squamous cell carcinoma
Some variants of SCC are associated with particular difficulties in diagnosis and clinical assessment, but should be managed, stage for stage, in line with conventional SCC.Reference El-Naggar, Chan, Grandis, Takata and Slootweg65
Papillary squamous cell carcinoma
Papillary SCC is typified by an exophytic growth pattern, with fronds of fibrovascular tissue covered by squamous epithelium showing marked atypia; areas of invasive carcinoma are often small and limited in extent. Definitive invasion may not be demonstrable in diagnostic biopsies despite a bulky tumour, and close correlation with clinical impression and radiological features are required. The prognosis is relatively good because of the limited invasive component.
Verrucous squamous cell carcinoma
Verrucous SCC has exophytic and endophytic components. It is formed by well-differentiated squamous epithelium, with minimal atypia and abundant surface keratin. The defining criterion of verrucous SCC is a broad ‘pushing’ invasive front, extending deeper to adjacent non-neoplastic surface epithelium. It may not be possible to make the diagnosis of verrucous SCC on superficial diagnostic biopsies, which do not include the deep invasive front. Repeated biopsies, and appreciation of the discrepancy between a clinically obvious malignancy and minimal microscopic atypia are sometimes required to make a diagnosis of verrucous SCC.
Spindle cell (sarcomatoid) squamous carcinoma
Spindle cell carcinomas typically present as polypoid tumours with an ulcerated surface, and are formed by sheets of atypical spindle cells, often raising the possibility of sarcoma. Sarcomas of mucosal origin are extremely rare in adults, but a definitive diagnosis of spindle cell carcinoma may only be possible on resection specimens when small areas of surface dysplasia or more typical invasive carcinoma are identified. Immunohistochemistry only identifies squamous epithelial differentiation in approximately 60–70 per cent of cases.
Basaloid squamous cell carcinoma
Basaloid SCC forms sheets and rounded nests of basaloid cells with palisaded nuclei of peripheral cells. Comedo-necrosis and stromal hyalinisation are frequently present. While most tumours are submucosal, continuity with dysplastic surface epithelium should be demonstrable. As morphological squamous differentiation may be absent, immunohistochemistry may be necessary to distinguish basaloid SCC from adenoid cystic (solid type) and neuroendocrine carcinomas. When located in the oropharynx, the term ‘basaloid SCC’ should not be used interchangeably with ‘conventional HPV-associated, non-keratinising SCC’.Reference Lewis86
Adenosquamous carcinoma
Adenosquamous carcinoma is a rare variant of SCC arising from surface epithelium, characterised by biphasic squamous and glandular elements. This variant is clinically more aggressive than conventional SCC. The main differential diagnosis is mucoepidermoid carcinoma; a combination of cytogenetic testing and demonstration of local surface origin may be useful in distinguishing between these entities.
Histopathology reporting
Diagnostic biopsies
The histological features from diagnostic biopsies may be limited, but it should normally be possible to determine whether any carcinoma is invasive or in situ. For invasive carcinomas, a provisional estimate of the degree of differentiation and the growth pattern should be made, but the MDT should appreciate that diagnostic sampling may not be representative, and the final grade and pattern of invasion are proffered based on the definitive resection specimen. In the oral cavity, the depth of invasion or tissues involved (e.g. muscle or bone) may guide the extent of surgery.
Resection specimens
Resection specimens provide sufficient tissue for pathological staging and to describe the full range of prognostic information.Reference Amin, Edge, Greene, Byrd, Brookland and Washington87 The evidence base for this prognostic information is provided in guidelines published by the Royal College of Pathologists, and varies between anatomical head and neck subsites.59–Reference Roya64,72,73 All Royal College of Pathologists dataset items should be included in the histopathology reports of resection specimens. The International Collaboration on Cancer Reporting may be used for reporting odontogenic carcinomas, mucosal melanomas and malignant neoplasms of the ear and temporal bone.Reference Odell, Baumhoer, Carlos, Hunter, Mosqueda-Taylor and Richardson88–Reference Thompson, Franchi, Helliwell, Müller and Williams90
Tumour–node–metastasis classification
For the purposes of local treatment protocols and continuity in clinical trial stratification, both the seventh and eighth editions of the TNM Classification of Malignant Tumours should be referred to. It is important to note the difference in clinical and pathological neck staging in p16 positive oropharyngeal cancer (Table 1),Reference Brierley, Gospodarowicz and Wittekind91 which reflects the need for different stratification of patients having primary radiotherapy (with or without chemotherapy) versus the need for adjuvant treatment in patients receiving primary surgery.
Table 1. Differences between clinical and pathological neck staging for p16 positive oropharyngeal cancer
Other important advances in the seventh and eighth editions of the TNM Classification of Malignant Tumours are:
• The incorporation of the depth of invasion for tumour (T) category in oral cavity carcinoma
• Separate staging systems for p16 positive and p16 negative oropharyngeal carcinoma
• The inclusion of extra-nodal extension for neck staging in all head and neck cancer except mucosal melanoma, nasopharyngeal and p16 positive oropharyngeal carcinoma
• A separate staging system for carcinoma of unknown primary, head and neck skin cancerReference Schutte, Takes, Slootweg, Arts, Honings and van den Hoogen33, Reference Cozens34
Neck dissections
The presence of extra-nodal extension (formerly termed ‘extracapsular spread’) is considered by many to be the most important prognosticator in the majority of head and neck cancer subtypes.Reference de Juan, Garcia, Lopez, Orus, Esteller and Quer93 Extra-nodal extension may be further subdivided into minor or major categories, based on whether the metastasis has extended less than 2 mm or more than 2 mm beyond the lymph node capsule, respectively.Reference Amin, Edge, Greene, Byrd, Brookland and Washington87 As the presence of extra-nodal extension informs adjuvant treatment, consensus reporting of incipient extra-nodal extension is recommended, as suboptimal inter-observer reproducibility has been reported.Reference van den Brekel, Lodder, Stel, Bloemena, Leemans and van der Waal94,Reference Lewis, Tarabishy, Luo, Mani, Bishop and Leon95
Dysplasia
Squamous cell carcinoma results from a combination of genetic alterations, some of which may manifest as precursor lesions characterised by morphological changes in epithelial cells, collectively referred to as dysplasia. An increasing degree of dysplasia is positively correlated with a greater risk of transformation to carcinoma. The various commonly used grading systems utilise the cumulation of microscopic architectural and cytological features to provide a continuous spectrum. The current WHO grading system for dysplasia of the oral cavity and larynx is recommended, summarised in Table 2.Reference El-Naggar, Chan, Grandis, Takata and Slootweg65
Table 2. Summary of WHO grading system for oral cavity and laryngeal dysplasia
WHO = World Health Organization
Management of dysplasia should take account of the microscopic grade of the lesion and its clinically assessed extent. While dysplasia grading remains subjective, with suboptimal inter-observer agreement, consensus reporting is likely to enhance diagnostic reliability.Reference Kujan, Khattab, Oliver, Roberts, Thakker and Sloan96–Reference Speight, Abram, Floriano, James, Vick and Thornhill98 Clear communication between the pathologist and clinician is necessary to convey the degree of concern regarding malignant transformation. ‘Carcinoma in situ’ is often used interchangeably with ‘severe epithelial dysplasia’, and may be appropriate in certain circumstances if the meaning is well understood between the reporting pathologist and the MDT. It is important to note the small but significant risk of concurrent invasive carcinoma in lesions biopsied and reported as severe epithelial dysplasia. The presence of severe epithelial dysplasia at resection margins should be included in pathology reports of resection specimens, as it may predict local recurrence.59,61
Diagnosis and management of neck lumps
Fine needle aspiration and core biopsy
Fine needle aspiration cytology is an important first-line investigation for mass lesions in the head and neck, and it can also be useful as part of staging procedures for patients with known head and neck cancer.99 The sample collected should be high-quality, cellular, well-spread (if it is smear preparation) and not overly contaminated with blood. Aspirates should be obtained under ultrasound guidance unless very superficial. Depending on the clinical circumstance, rapidly air-dried direct smear preparations and/or needle washings into preservative solution may be required; the latter are useful for ancillary tests such as immunostaining. Rapid on-site evaluation of specimen adequacy by a biomedical scientist can reduce the number of non-diagnostic aspirates, but this has workload and cost implications.Reference Moberly, Vural, Nahas, Bergeson and Kokoska100,Reference Schmidt, Witt, Lopez-Calderon and Layfield101
Reasonable cytological expertise should be available for interpreting the findings and for recognition of the diagnostic pitfalls. In the assessment of lymphadenopathy, FNAC shows high diagnostic specificity for granulomatous lymphadenitis, metastatic carcinoma, high-grade lymphoma and Hodgkin's lymphoma. Immunohistochemistry for p16 should be performed on FNAC specimens that show SCC. Cytology can be useful in the evaluation of cystic neck masses, particularly if the cyst wall is targeted, but definitive characterisation is not always possible, e.g. the differentiation between a lympho-epithelial cyst and cystic SCC metastasis. Such cases require close clinical and radiological assessment. The diagnostic sensitivity and specificity for low-grade lymphoma on cytological morphology is low, and ancillary investigations such as flow cytometry and immunocytochemistry on cytology specimens are not universally available.
The choice between FNAC and core biopsy may depend on the extent of cytopathological expertise, and the site and nature of the lesion. Core biopsy may be preferred, for example, for salivary gland tumours and lesions that require an immunohistochemistry panel or molecular testing, such as lymphoma or malignancies that require further characterisation.
Lymphoma, sarcoma, skin cancer and mucosal melanoma
The diagnostic and tissue pathways for lymphoreticular neoplasms, sarcomas, skin cancers and mucosal melanomas should be subject to local and cancer network standard operating procedures.
Role of pathologist in clinical trials
The Standard Protocol Items: Recommendations for Interventional Trials (‘SPIRIT’) statement provides evidence-based recommendations for the minimum content of clinical trial protocols, and is widely endorsed by the research community.Reference Chan, Tetzlaff, Gotzsche, Altman, Mann and Berlin102 The Cellular Molecular Pathology Initiative (‘CM-Path’), hosted by the National Cancer Research Institute, has recently published an extension to the Standard Protocol Items: Recommendations for Interventional Trials statement, to provide the details required to effectively incorporate pathology support into clinical trial protocols; the guidance is called ‘SPIRIT-Path’.Reference Kendall, Robinson, Brierley, Lim, O'Connor and Shaaban103
Important questions to be answered
Head and neck pathology is a rapidly evolving field that is continuously being shaped by technological advances. Developments likely to influence the practice of head and neck pathology in the near future include the following points.
• Routine use of diagnostic digital pathology. This is likely to facilitate streamlining of the diagnostic pathway and rapid pathology review across cancer networks. High throughput digital pathology will also advance artificial intelligence algorithms for computer-assisted diagnosis and the development of biomarkers. Incorporation of whole slide images linked to tissue biobanks, with associated clinical metadata, will underpin the drive to greater refinement of personalised medicine.
• Characterisation of the tumour microenvironment. Better understanding of the function of interplay between the tumour, immune cells and the extracellular matrix is likely to lead to the identification of individuals most likely to benefit from immunomodulatory oncology drugs.
• Whole genome sequencing. Routine whole genome sequencing of fresh tumour samples in selected tumour types may identify actionable mutations in head and neck cancer.104
• Integration of pathology support in clinical trials. There is a growing need for clinical trials with pathology support for biomarker-driven stratification and embedded translational studies, to develop novel molecular-based diagnostic, prognostic and predictive biomarkers.
• Multiplex risk stratification. The integration of morphological, molecular and digital pathology signatures with radiomic, microbiome and serological biomarkers can be used to improve the stratification of patients with a risk of tumour recurrence and determine progression.
Introduction
The ability of ionising radiation to kill tumour cells makes it an ideal curative treatment for head and neck cancers where the rates of metastasis are relatively small. Many of the developments in radiotherapy techniques over the last 20 years have been applied to great effect in head and neck cancers to target smaller treatment volumes and reduce side effects. For many years, cisplatin has been combined with radiotherapy to increase cure rates and used in palliative chemotherapy regimens. The promise of newer drugs such as immunotherapies is now beginning to be realised in head and neck cancers.
Curative radiotherapy
Recommendations
• Use intensity-modulated radiotherapy techniques and international consensus contouring guidance when treating head and neck cancers with curative intent (evidence-based recommendation (R))
• Use Royal College of Radiologists’ recommended dose fractionation schedules e.g. 70 Gy in 35 fractions over seven weeks and 65–66 Gy in 30 fractions over six weeks (R)
• Review patients weekly during treatment by a multidisciplinary team (MDT), with particular focus on nutrition and swallowing assessment and therapy (R)
• Dental assessment with extraction if necessary should be carried out before radiotherapy for dentate patients (R)
• Smoking cessation management is a key component of radiotherapy treatment (R)
Definitive or ‘radical’ radiotherapy treatment offers the potential for cure with organ preservation for many head and neck cancer anatomical sites.
Radiotherapy planning initially involves fabrication of a mask to keep the patient still and precisely positioned for each fraction of treatment, followed by a planning computed tomography (CT) scan performed whilst in the mask, usually with the administration of intravenous contrast. There is then a period of time whilst planning takes place before treatment is ready to be commenced. For a course of radical radiotherapy, a clinician will contour target volumes using diagnostic imaging such as magnetic resonance imaging (MRI), CT and positron emission tomography (PET)-CT, sometimes co-registered to improve the accuracy of target delineation. This is followed by the generation of a radiotherapy plan by dosimetrists or physicists using advanced computing.
Intensity-modulated radiotherapy is now the standard of care – highly conformal dose distributions are achieved by shaping or modulating multiple beams, allowing the sparing of nearby organs at risk. It has been demonstrated that intensity-modulated radiotherapy offers superior disease control for nasopharyngeal carcinoma and a reduction in radiotherapy-induced xerostomia.Reference Zhang, Mo, Du, Wang, Liu and Wei105,Reference Nutting, Morden, Harrington, Urbano, Bhide and Clark106 Intensity-modulated radiotherapy allows the delivery of different doses to differing target volumes (e.g. tumour-bearing tissues and elective lymph node regions) within each fraction of treatment. For each fraction, a patient will typically be in the treatment room for 10–20 minutes. Image-guided verification of treatment accuracy is a key part of modern radiotherapy treatment with the recently updated UK guidelines.107
Recommended dose fractionation schedules are summarised in the Royal College of Radiologists’ guidance.108 Schedules of 70 Gy in 35 fractions over seven weeks and 65–66 Gy in 30 fractions over six weeks are radiobiologically similar, and represent standard schedules in the UK for definitive radiotherapy. For primary non-surgical treatment of locally advanced disease, combining radiotherapy with concurrent chemotherapy is standard for patients aged less than 70 years with a good performance status.Reference Lacas, Carmel, Landais, Wong, Licitra and Tobias109 For patients unsuitable for chemotherapy, modest acceleration of radiotherapy (e.g. delivering six rather than five fractions per week) has been shown to offer some improvement in local control but not overall survival, with the benefit diminishing with increasing age.Reference Lacas, Bourhis, Overgaard, Zhang, Grégoire and Nankivell110 For early glottic cancer, hypofractionated schedules are superior to conventional fractionation e.g. 55 Gy in 20 fractions over four weeks.
Radiotherapy often has both short-term (acute) and long-term (late) side effects, which are dependent upon the dose delivered, target volumes, and patient fitness and co-morbidity. Short-term side effects gradually build up during a course of fractionated treatment, and can last for weeks or months after treatment. These include skin reaction, mucositis, xerostomia, dysgeusia, throat secretions, risk of aspiration pneumonia and voice changes. Common toxicity criteria are often used to grade acute toxicity. Table 1 summarises grading for acute radiation-related skin, mucositis and dysphagia toxicity.Reference Trotti, Byhardt, Stetz, Gwede, Corn and Fu111 Long-term side effects are highly variable, and can include long-term swallowing dysfunction, osteoradionecrosis, xerostomia, dysgeusia, lymphoedema, hypothyroidism, skin and subcutaneous fibrosis, and radiotherapy-induced cancer in the treated area many years later.
Table 1. Summary of Radiotherapy Therapy Oncology Group Common Toxicity Criteria 2.0 scoring for skin, oral mucositis and pharynx
IV = intravenous
The provision of supportive care prior to, during and after a course of treatment is a key component of the delivery of radiotherapy. This includes restorative dental, nursing, dietetic, and speech and language therapy input. Some patients may suffer with significant treatment-related anxiety and claustrophobia, and can receive individualised support. Continued smoking during head and neck radiotherapy has been shown to approximately double the risk of locoregional treatment failure and the risk of mortality, along with increased late toxicity.Reference Smith, Nastasi, Tso, Vangaveti, Renison and Chilkuri112 Smoking cessation management is a key component of radiotherapy treatment.
Unscheduled interruptions in radiotherapy for head and neck cancers can allow tumours to repopulate, with a detrimental impact upon cure rates. The Royal College of Radiologists has published guidance on the appropriate management of unavoidable gaps in treatment, with options including weekend or bi-daily treatment administration.113
Response to (chemo)-radiotherapy is often best assessed clinically. A PET-CT performed three months after treatment is often used in node-positive disease to assess the need for a neck dissection, as evidenced by the ‘PET-NECK’ trial.Reference Mehanna, Wong, McConkey, Rahman, Robinson and Hartley51 It can be helpful to repeat diagnostic imaging (e.g. MRI) three months after radiation, but it can be difficult to distinguish radiotherapy-induced oedema from tumour recurrence with imaging alone.
Post-operative radiotherapy
Recommendations
• Discuss all surgical resection histology at an MDT meeting with input from a clinical oncologist, pathologist and operating surgeon, to decide on suitability for post-operative radiotherapy (evidence-based recommendation (R))
• Commence post-operative radiotherapy within six weeks of surgery if the patient has recovered well enough (R)
Radiotherapy should be considered after surgery when there is a significant risk of locoregional recurrence. In practice, this usually means considering post-operative radiotherapy in pathologically staged T3/4 tumours, node-positive disease or combinations of other risk factors. There are some specific cases where radiotherapy is also recommended in early stage disease, for example adenoid cystic cancer. As a rule of thumb, locoregional recurrence rates are halved by post-operative radiotherapy, but with relatively little effect on overall survival.
Tumour histology should be discussed at an MDT meeting with the surgeon present. The team should consider indications for irradiating the primary site, dissected neck and undissected neck. Absolute and relative indications for post-operative radiotherapy are summarised in Table 2 and discussed further in the relevant site-specific chapters.
Table 2. Absolute and relative indications for post-operative radiotherapy
Some patients may not feel able to undertake six weeks of adjuvant radiotherapy having just recovered from a major operation if the benefit of radiotherapy is relatively small. The decision to administer post-operative radiotherapy may depend as much on patient fitness, their support network and their preferences as on the tumour factors listed above. If patients present with disease that is locally advanced (e.g. T4 disease or with obvious extra-nodal extension on diagnostic imaging) a plan to offer surgery and post-operative radiotherapy should be discussed with the patient and oncologist before the operation. Putting someone through a major resection for locally advanced disease when they cannot complete post-operative radiotherapy may not be to their advantage.
Post-operative radiotherapy should ideally commence within six weeks of surgery. This can be challenging to achieve, particularly if there have been post-operative complications. It generally takes several weeks from a decision to offer adjuvant radiotherapy to commencing treatment, so efficient MDT organisation is important in order to review post-operative pathology promptly and to allow the oncologist to meet the patient in a timely fashion. Sometimes it may be advantageous to organise post-operative radiotherapy before histology results are available (e.g. in a tumour that was T4 at diagnosis), or to start radiotherapy planning, but not treatment itself, before a fistula or wound is fully healed. The absence of a tumour to contour and the altered anatomy after surgery often make radiotherapy contouring more challenging for the oncologist than in definitive radiation cases. Marginal and out-of-field recurrences are more common than following definitive radiotherapy, and hence target volume delineation needs to be generous.Reference Waldram, Taylor, Whittam, Iyizoba-Ebozue, Murray and Frood114
As with curative radiotherapy, the support of an MDT, comprising dietitians, speech and language therapists, clinical nurse specialists, and therapy radiographers, is critical to ensure patients complete the treatment course.
Palliative radiotherapy
Recommendations
• Consider palliative radiotherapy to ameliorate local symptoms in people who are not able to receive curative treatment (evidence-based recommendation (R))
Primary palliative radiotherapy can be used as a first-line treatment in those who are not fit enough to undergo curative-intent treatment with surgery or radiation. The aims of this include longer-term local control of an incurable tumour, and short-term palliation of distressing symptoms such as pain, bleeding or fungation. Typical dose fractionation schedules can range from a single fraction to 30 Gy in 10 fractions or even longer schedules.108
Radiation toxicity during and after treatment can temporarily make symptoms worse, so patient selection is key, and the overall aims of treatment need to be carefully considered. Doses should be chosen to minimise toxicity and travelling where possible. Close working with palliative care teams can be invaluable.
Palliative re-irradiation can be considered for symptomatic local recurrences after a prior definitive or adjuvant radiotherapy treatment, though the increased availability of systemic treatments such as immune checkpoint inhibitors means other treatment options may be preferable. The role of stereotactic ablative radiotherapy in this context remains uncertain and it is not routinely commissioned within the National Health Service (NHS).
Palliative radiotherapy has a role in the management of symptomatic systemic metastatic disease, especially for painful bone metastasis. Stereotactic ablative radiotherapy is increasingly used in oligometastatic disease. This is defined as the presence of one to three sites of metastatic disease, typically within the lung, bone, lymph nodes or liver, presenting six months or more after primary treatment. This has been shown to be well tolerated, with local control rates of 80–90 per cent, and the promise of improved overall survival in other cancer types.Reference Tree, Khoo, Eeles, Ahmed, Dearnaley and Hawkins115,Reference Palma, Olson, Harrow, Gaede, Louie and Haasbeek116
Proton beam therapy
Recommendations
• Consider proton beam therapy for adults with head and neck cancer who meet appropriate commissioning criteria, or as part of randomised, controlled trials (evidence-based recommendation (R))
Proton beam therapy is routinely used in preference to standard photon radiotherapy for children and young adults, where the reduced integral dose lessens effects on organ growth and lowers the risk of secondary cancers.Reference Xiang, Chang and Pollom117 Because of the complex anatomy and proximity of sensitive organs at risk in the head and neck region, proton beam therapy may also benefit adult patients, by reducing early and late treatment toxicities, or by improved optimisation of target volume coverage in areas of dose-limiting structures, for example the base of the skull.
The distinct physical properties of protons result in a sharp distal fall-off in dose beyond the target volume, with minimal ‘exit-dose’.Reference Newhauser and Zhang118 Compared with standard treatment using photons, there is a reduction in the low-to-intermediate dose delivered to normal tissues. Proton treatment planning is complex; it factors in possible variation in the relative biological effectiveness (protons compared to photons),Reference Paganetti, Niemierko, Ancukiewicz, Gerweck, Goitein and Loeffler119 as well as the nature of tissues the beam passes through, changes in patient weight or variation in air cavity, all of which can alter the dose distribution. Changes in patient anatomy are carefully monitored for alteration in the dose distribution, with re-planning routinely needed during head and neck cancer treatments.
There are observational data to support the benefits of proton beam therapy for the treatment of head and neck cancers. In oropharynx cancer, patients were less likely to require a feeding tube during treatment;Reference Gunn, Blanchard, Garden, Zhu, Fuller and Mohamed120 or to develop grade 3 weight loss or require a feeding tube at three months (odds ratio = 0.44; 95 per cent confidence interval (CI) = 0.19–1.0; p = 0.05) or one year after treatment (odds ratio = 0.23; 95 per cent CI = 0.07–0.73; p = 0.01).Reference Blanchard, Garden, Gunn, Rosenthal, Morrison and Hernandez121 Patients with nasopharynx cancer were less likely to develop any grade 2 or higher acute adverse events (odds ratio = 0.15; 95 per cent CI = 0.03–0.60; p = 0.01)Reference Li, Kitpanit, Lee, Mah, Sine and Sherman122 or require feeding tube placement (20 per cent vs 65 per cent; p = 0.02).Reference Holliday, Garden, Rosenthal, Fuller, Morrison and Gunn123 In a systematic review of non-comparative sinonasal cancer observational studies, subgroup analysis showed higher disease-free survival at five years (relative risk = 1.44, 95 per cent CI = 1.01–2.05; p = 0.045) and locoregional control at the longest follow up (relative risk = 1.26, 95 per cent CI = 1.05–1.51; p = 0.011) using proton beam therapy.Reference Patel, Wang, Wong, Murad, Buckey and Mohammed124
NHS England now has high-energy proton beam therapy centres in Manchester and London, and is well-placed to systematically evaluate the potential benefits of proton beam therapy, through a combination of randomised clinical trials,Reference Price, Hall, West and Thomson125 pre-commissioning studies and outcome monitoring.Reference Burnet, Mackay, Smith, Chadwick, Whitfield and Thomson126 Rigorous assessments are needed to justify the routine use of a specialist and constrained resource, the increased treatment costs, and the inconvenience to patients and their families in accessing a centralised service.
Chemotherapy and other drugs concomitant with radiotherapy
Recommendations
• Offer concomitant cisplatin with curative radiotherapy to patients who are young and fit enough to tolerate the likely increased toxicity when they have T3/4 tumours and/or node-positive disease (evidence-based recommendation (R))
• Offer concomitant cisplatin with post-operative radiotherapy to patients with R1 (presence of microscopic cancer cells) or R2 (presence of macroscopic residual tumour) resections, or extracapsular nodal extension (R)
Concomitant chemotherapy given with radiotherapy is now a standard of care in advanced disease, both in the primary and adjuvant setting. It has been shown to improve locoregional control, and adds a small survival benefit to radiotherapy alone but with the downside of greater toxicity. The mechanism of action is still not fully understood. Chemotherapy may act as a radiosensitiser by impairing pathways to repair tumour DNA damage caused by radiotherapy. It may also have a role in sterilising microscopic metastatic disease cells.Reference Boeckman, Trego and Turchi127
The Meta-Analysis of Chemotherapy in squamous cell Head and Neck Cancer (‘MACH-NC’) database demonstrated that concomitant chemotherapy improved overall survival. The 2021 update assessed 19 805 patients in 107 trials (excluding nasopharyngeal cancer) and showed an absolute survival benefit of 6.5 per cent at 5 years and 3.6 per cent at 10 years for concomitant chemotherapy compared to radiotherapy alone. The benefit decreased with increasing age, with no significant benefit for patients aged over 70 years (hazard ratio of 0.78 for those aged less than 50 years, compared with 0.97 for those aged 70 years or more).Reference Lacas, Carmel, Landais, Wong, Licitra and Tobias109 Concomitant cisplatin should be considered for patients who are young and fit enough to tolerate the likely increased toxicity when curative radiotherapy is used for T3/4 tumours and/or with node-positive disease.
Concomitant chemotherapy in the post-operative setting has also been shown to improve local control and survival in those with poor prognostic factors and a high recurrence risk, particularly if there are involved resection margins or extracapsular nodal extension. In a separate meta-analysis, the five-year absolute benefit rate with concomitant chemotherapy versus radiotherapy alone was 7.9 per cent.Reference Winquist, Oliver and Gilbert128,Reference Dauzier, Lacas, Blanchard, Le, Simon and Wolf129
The international standard treatment regimen is cisplatin 100 mg/mReference Nutting, Morden, Harrington, Urbano, Bhide and Clark106 three-weekly, but because of the intensiveness and morbidity of treatment, chemotherapy modification (delays, omission and dose modification) is not uncommon. Weekly cisplatin 30–40 mg/mReference Nutting, Morden, Harrington, Urbano, Bhide and Clark106 is preferred by some centres in the belief that it is better tolerated and more flexible in delivery, as occasional weeks can be missed with less risk of compromising total dose. Regardless of treatment schedule, it has been shown in studies that a cumulative dose of around 200 mg/mReference Nutting, Morden, Harrington, Urbano, Bhide and Clark106 gives the best therapeutic benefit.
A meta-analysis published in 2017 concluded that, although no difference in treatment efficacy was seen for three-weekly versus weekly cisplatin regimens, the studies analysed were often flawed in terms of balance and study numbers, and few trials were prospective and randomised.Reference Szturz, Wouters, Kiyota, Tahara, Prabhash and Noronha130,Reference Szturz, Wouters, Kiyota, Tahara, Prabhash and Noronha131 Subsequently, Noronha et al. reported a prospective phase 3 randomised study of 300 patients (93 per cent adjuvant), and showed a two-year locoregional control benefit rate of 58.5 per cent for weekly cisplatin versus 73.1 per cent for three-weekly.Reference Noronha, Joshi, Patil, Agarwal, Ghosh-Laskar and Budrukkar132 Other studies have reported conflicting results and so conclusions remain difficult to draw.Reference Helfenstein, Riesterer, Meier, Papachristofilou, Kasenda and Pless133
Cetuximab has also been extensively investigated in combination with radiotherapy. Although the initial major study showed a benefit for cetuximab compared to radiotherapy alone, subsequent studies have shown cetuximab to be inferior to cisplatin, particularly in human papillomavirus (HPV)-positive groups.Reference Bonner, Harari, Giralt, Cohen, Jones and Sur134–Reference Mehanna, Robinson, Hartley, Kong, Foran and Fulton-Lieuw136
Immunotherapy (discussed in the next section for recurrent and metastatic disease) may have a place in the radical setting in combination with surgery and/or radiotherapy. The global multicentre, multicohort phase I/II trial ‘Checkmate-358’ assessed the safety of neo-adjuvant nivolumab prior to surgery, and the ‘CompARE’ (Comparing Alternative REgimens for escalating treatment of intermediate and high-risk oropharyngeal cancer) trial is currently recruiting patients with intermediate and high-risk oropharyngeal tumours receiving radical (chemo)radiotherapy with durvalumab as one of the investigation arms. There are many more studies open or in development that are focused on assessing the benefit of immune checkpoint inhibitors; this is an exciting era for concomitant systemic therapy in radical head and neck cancer.
Palliative systemic therapies
Recommendations
• Explain the potential benefits of different systemic therapies and their possible negative effects (side effects, time taken for treatment, likelihood of treatment not working) to patients and families before agreeing a treatment plan (evidence-based recommendation (R))
• Measure PD-L1 (programmed death-ligand 1) biomarkers in all patients who are considering systemic palliative therapies (R)
• Offer palliative immunotherapy as an alternative to chemotherapy, in line with current NHS commissioning guidance* (R)
*NHS England currently funds single-agent pembrolizumab for untreated metastatic or unresectable recurrent head and neck squamous cell carcinoma (SCC), provided the combined positive score is 1 or more. Nivolumab is funded for patients who have progressed during or within six months of platinum-based chemotherapy.
Palliative chemotherapy can be considered in cases of recurrent or metastatic head and neck SCC that are not amenable to local therapies and where the patient continues to have a good World Health Organization performance status classification, of 0–2 (ranging from asymptomatic to symptomatic and in bed less than 50 per of the day). In recurrent and metastatic disease, overall survival is generally poor, and measured between 6 and 12 months in most series. Responses to systemic therapies can be of limited duration and toxicity rates can be high, so it is important to discuss the risks of treatment compared to best supportive care alone, particularly taking into account a patient's fitness for treatment and their preferences and wishes when life expectancy is short. Medications and strategies aimed at improving symptom control are essential, along with good palliative care and clinical nurse specialist and allied health professional support. It is important to consider clinical trials if available.
Chemotherapy and cetuximab
Traditional first-line palliative chemotherapy is platinum and 5-fluorouracil, in combination with cetuximab, a monoclonal antibody directed against epidermal growth factor receptor (the ‘EXTREME’ regimen). The inclusion of cetuximab in this combination only has NHS England approval for oral cavity tumours. This combination produces an overall response rate of 36 per cent and an overall survival of 10 months, in comparison to 7.4 months for chemotherapy alone.Reference Vermorken, Mesia, Rivera, Remenar, Kawecki and Rottey137 It is, however, associated with substantial toxicity, including skin toxicity, hypomagnesaemia and sepsis, and so patients should have a good performance status. Simpler schedules of similar drugs may offer reduced toxicity and reduced duration of day-unit or in-patient stay, but have less evidence to support them (e.g. carboplatin and capecitabine).
If the patient remains fit after progression, second-line chemotherapy options include carboplatin and paclitaxel, which demonstrates an overall response rate of 25.9 per cent and median overall survival of eight months, with a reasonable toxicity profile.Reference Ferrari, Fiore, Codecà, Di Maria, Bozzoni and Bordin138 Further options include single-agent cetuximab, taxanes or methotrexate, each of which can be associated with response rates of 10–27 per cent, but with no overall survival benefit.Reference Vermorken, Trigo, Hitt, Koralewski, Diaz-Rubio and Rolland139,Reference Guardiola, Peyrade, Chaigneau, Cupissol, Tchiknavorian and Bompas140
Immunotherapy
Immune checkpoints modify immune responses to prevent autoimmune reactions. One such check point is the PD1 protein (programmed cell death protein 1), expressed by T cells. This ligates with PD-L1 on cancer cells, which often over-express PD-L1, therefore providing a means of tumour immune escape by suppressing the immunological response of the T cell.
Anti-PD1/PD-L1 immunotherapy agents, such as pembrolizumab and nivolumab, block this signalling through the PD-L1 pathway and thereby enhance immune activity.
Immune check point inhibitors have recently been shown to improve outcomes in recurrent or metastatic head and neck SCC, in both the first-line setting in comparison to traditional chemotherapy, and in platinum-resistant tumours.
Pembrolizumab as a single agent or in combination with platinum and 5-fluoruacil is now the most effective first-line treatment in recurrent or metastatic head and neck SCC cases. The PD-L1 biomarker testing should be routinely assessed in patients with recurrent or metastatic head and neck SCC in order to determine eligibility for pembrolizumab, using the combined positive score or tumour positive score measure (combined positive score of more than 1). Up to 85 per cent of head and neck SCC patients have a PD-L1 combined positive score of 1 or more.
Combination pembrolizumab and chemotherapy demonstrates an overall survival of 13.6 months, compared to 10.4 months for standard chemotherapy.Reference Burtness, Harrington, Greil, Soulières, Tahara and de Castro141 Single-agent pembrolizumab also demonstrates an improved overall survival of 12.3 months. Single-agent pembrolizumab is generally well tolerated, with a favourable safety profile in comparison to standard chemotherapy or the pembrolizumab chemotherapy combination (7 per cent had grade 3–5 treatment-related adverse events, compared to 39–47 per cent for the combination or standard treatment). NHS England currently funds single-agent pembrolizumab for untreated metastatic or unresectable recurrent head and neck SCC, provided the combined positive score is 1 or more, but not the combination with chemotherapy. Typical potential toxicity includes fatigue, endocrine toxicity including hypothyroidism, gastrointestinal disturbance, and anaemia. A subgroup of patients demonstrates prolonged survival when treated with immunotherapy, with a survival rate at 24 months of 38 per cent for pembrolizumab as a single agent and 29 per cent for pembrolizumab in combination with chemotherapy.
Monotherapy with both nivolumab or pembrolizumab has demonstrated efficacy in patients who have progressed during or after platinum-based chemotherapy compared to standard second-line single-agent chemotherapy, with an improved toxicity profile.Reference Cohen, Soulières, Le Tourneau, Dinis, Licitra and Ahn142,Reference Ferris, Blumenschein, Fayette, Guigay, Colevas and Licitra143
Current trials that may change practice and future directions
The National Cancer Research Institute supports a wide variety of clinical trials in head and neck cancer. Studies that are open to recruitment or in the set-up stage can be viewed here via: https://www.ncri.org.uk (search portfolio maps). Major USA trials undertaken by the NRG co-operative group can be viewed here: https://www.nrgoncology.org/Clinical-Trials/Protocol-Search. There are perhaps three major research areas that may change practice within the next five years.
In HPV-associated oropharyngeal cancer, there are many studies exploring different de-intensification strategies: reducing radiation dose, treatment volumes or concomitant therapies in order to maintain high cure rates with less toxicity. There is almost a risk that the plethora of trials with different strategies will make the optimal treatment paradigm difficult to assess. Key UK phase III trials that should report in the next few years are ‘PATHOS’ (a trial of risk-stratified, reduced-intensity adjuvant treatment in patients undergoing transoral surgery for HPV-positive oropharyngeal cancer) and Torpedo (a trial of intensity-modulated proton beam therapy vs intensity-modulated radiotherapy for multi-toxicity reduction in oropharyngeal cancer).
The cure rates for other head and neck SCC have not changed much for decades, so trials are also exploring improving survival rates in these cancers, often by adding new treatment options such as immunotherapy to existing protocols. In the UK, the ‘CompARE’ trial has almost completed accrual.
Systemic therapy for many cancers has changed hugely in the last decade, with major advances being made in melanoma and lung cancer in particular. It is hoped that continued immunotherapy and targeted-therapy research will also improve survival rates in head and neck cancers.
The head and neck oncology community worldwide has produced an excellent series of practical guidelines for radiotherapy target volume contouring and selection in the last decade.Reference Grégoire, Ang, Budach, Grau, Hamoir and Langendijk144–Reference Biau, Lapeyre, Troussier, Budach, Giralt and Grau146 These have been supplemented with the Royal College of Radiologists’ head and neck cancer consensus statements, and other associated UK national guidance on topics such as radiotherapy contour quality assurance and dose fractionation.107,108,147 If implemented in every radiotherapy centre, such guidelines have huge potential to reduce variation and improve outcomes. Future guidelines, for example on post-operative radiotherapy contouring, are eagerly awaited.
Introduction
Follow up is an essential component of the management of patients who have undergone radical treatment for head and neck cancer. The rationale for follow up is multifactorial:
• Monitoring of treatment effectiveness
• Early detection of recurrent disease
• Surveillance for second primary tumours
• Assessment of treatment morbidity and functional deficit
• Education of patients and caregivers
• Risk factor modification
Approximately 25 per cent of patients treated for head and neck cancer develop cancer recurrence, most of which is locoregional, and the majority of recurrences occur within the first two years after treatment.Reference Pagh, Grau and Overgaard148
Patients with head and neck cancer are also at risk of developing metachronous second primary tumours, with the risk of a second primary tumour doubled (compared to patients without head and neck cancer). Over 10 years, around 17 per cent of patients will develop a second primary tumour, mostly in the head and neck or lung.Reference Morris, Sikora, Patel, Hayes and Ganly149
Most protocols adopt shorter intervals between clinic visits in the first two years (when the risk of locoregional recurrence is known to be at its highest), with gradually lengthening intervals through to five years.
Despite the accepted importance of follow up, current strategies, including imaging protocols, are largely based on consensus opinion rather than prospective data, with significant variation in practice.150
This chapter covers the principles of follow up and surveillance, and the management of disease recurrence. Site- and tumour-specific guidance can be found in the relevant chapters. Imaging is discussed in more detail in Chapter 2 and non-surgical treatment for recurrence is described in Chapter 4. The aspects of patient support and survivorship – crucial parts of follow up – are discussed in more detail in Chapters 9–14.
Follow up
Recommendations
• Patients should undergo follow up on at least a two-monthly basis for two years, then three- to six-monthly thereafter for a minimum of five years (good practice point (G))
• Patients should be able to access urgent clinical assessment for suspicious symptoms at any time during follow up (evidence-based recommendation (R))
• Follow up and clinical examination with imaging, if indicated, should aim to identify cancer recurrence as early as possible and to detect second primary tumours (R)
• Patients should undergo follow up in clinics at, or linked to, a head and neck treatment centre, and have access to the wider multidisciplinary team (MDT), including clinical nurse specialists, speech and language therapists, and dieticians (R)
• Patients receiving radiotherapy or chemoradiotherapy for stage 3 and 4 disease should undergo post-treatment surveillance imaging to assess for disease response, as well as for baseline imaging (R)
• Baseline imaging should be considered after primary surgery in which anatomy is altered (G)
• Patients should undergo regular thyroid function monitoring after treatment for head and neck cancer (R)
Detection of recurrent disease
Recurrent head and neck cancer poses significant challenges for treatment and is associated with a poor prognosis. However, early detection of recurrence means that the patient has a greater chance of being a candidate for surgical salvage,Reference Tan, Giger, Auperin, Bourhis, Janot and Temam151 as well as for active treatments such as radiotherapy if not already employed, and systemic therapies including immunotherapy.
Of the 25 per cent of patients who will develop cancer recurrence, around 65 per cent of recurrences are in the primary site, 36 per cent in regional lymph nodes and 22 per cent are distant.Reference Pagh, Grau and Overgaard148 The majority occur within the first two years after treatment (62 per cent in first year, 82 per cent within two years).
Hence, standard follow-up protocols with short intervals in the first two years, followed by less frequent follow up, would seem appropriate. However, it is not clear what impact regular scheduled follow up has on either the detection of recurrence or on overall survival. In a study of 4839 patients with locoregional recurrent head and neck cancer, more than 60 per cent presented at an advanced stage (stage III and IV disease).Reference Chang, Wu, Yuan, Wu and Wu152 In another study, only 27 per cent of patients with detected recurrence were suitable to undergo salvage surgery.Reference Mabanta, Mendenhall, Stringer and Cassisi153
There are consistent data showing that most recurrences detected in clinic arise when a patient has become symptomatic and therefore expedited their clinic appointment.150 For example, one study showed that the detection rate rose from 0.2 per cent in asymptomatic patients to 56 per cent in those with symptoms.Reference Kothari, Trinidade, Hewitt, Singh and O'Flynn154
In the detection of the response to treatment, post-treatment response evaluation using 18Fluorine-fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET-CT) has become the standard of care to evaluate patients after (chemo)radiotherapy and to determine the need for neck dissection.Reference Mehanna, Wong, McConkey, Rahman, Robinson and Hartley51 Yet the positive predictive value in this setting is low, and is affected by both treatment and tumour biology (being lower in human papillomavirus (HPV)-positive tumours).Reference Hanemaaijer, Fazzi, Steenbakkers, Dorgelo, van der Vegt and Witjes155 Therefore, changes to the timing of post-treatment imaging for HPV-positive tumours, from three to four months, has been proposed, to reduce the rates of equivocal scans from involuting disease, which is known to be slower in HPV-associated disease.Reference Slevin, Subesinghe, Ramasamy, Sen, Scarsbrook and Prestwich156
For the detection of recurrence (rather than persistence), clinical assessment remains the mainstay of cancer surveillance. Baseline post-treatment imaging, usually by magnetic resonance imaging (MRI), typically performed three to four months after the completion of treatment, is recommended for any patient having undergone surgical or non-surgical treatment that may alter anatomy or imaging. This serves as a means of comparison if further imaging is later required.
When recurrence is clinically suspected, axial imaging should be organised and compared to baseline imaging (using the same modality).
Ultrasound provides rapid assessment of regional nodal recurrence and facilitates accurate concurrent pathological sampling. It may be available at the time of the clinic appointment. It may also have a role for monitoring an untreated neck with a significant chance of cervical lymph node metastases, although such cases should generally have had elective treatment or sentinel node biopsy in cases of oral cavity cancer.
Imaging may be the only means of assessing the recurrence status of clinically inaccessible tumours, e.g. the maxilla or midface after reconstruction, or the anterior and lateral skull base. However, there is no consensus on the frequency of such imaging.
18Fluorine-fluorodeoxyglucose PET-CT offers potential advantages in the detection of recurrences, with evidence suggesting a role for FDG PET-CT imaging at one year post treatment, outperforming other imaging modalities and showing additional benefit over clinical examination alone.Reference Ho, Tsao, Chen, Shen, Kaplan and Colevas157
Head and neck adenoid cystic carcinoma is a rare indolent cancer, but is characterised by late recurrence, both locally and to lungs and bone. Long-term interval imaging with low-dose computed tomography (CT) is proposed as a means of early metastasis detection. However, the evidence for improved survival with metastasectomy is mixed and the cumulative radiation risk from serial imaging, especially in younger patients, needs to be considered.Reference Girelli, Locati, Galeone, Scanagatta, Duranti and Licitra158 Hence, there is no evidence-based or consensus position on this.
Detection of second primary disease
Patients with head and neck cancer are not only at risk from recurrence of the index tumour, but also from the development of metachronous second primary tumours. The standardised incidence ratio of second primary tumours is estimated to be more than double in patients who have already had head and neck cancer, with an excess absolute risk of 167.7 per 10 000 per year (essentially, 17 second primary tumours in 100 treated head and neck cancer patients over 10 years of follow up).Reference Morris, Sikora, Patel, Hayes and Ganly149 This risk remains stable over time and relates to previous and continued exposure to known carcinogens, such as tobacco and alcohol, leading to widespread genomic instability or field change.Reference Slaughter, Southwick and Smejkal159,Reference Do, Johnson, Doherty, Lee, Wu and Dong160
Overall, most second primary tumours are lung cancers, although other oral cavity cancers are the commonest for patients with an index oral cavity cancer. The anatomical subsite of the first tumour affects the risk of second primary tumours, with the risk being highest for hypopharyngeal tumours and lowest for laryngeal tumours.Reference Morris, Sikora, Patel, Hayes and Ganly149
While the risk of a second primary tumour from oropharyngeal cancer appears to have decreased in the era of HPV-related disease, the presence of any HPV-induced malignancy (including head and neck) exposes patients to a significantly increased risk of a second HPV-induced tumour.Reference Gilbert, Wakeham, Langley and Vale161
As for recurrent disease, second primary tumours in the head and neck will often give rise to symptoms that may be reported by a patient and may then be detected via clinical examination or by imaging. However, as many second primary tumours may be both asymptomatic and arise outside the head and neck (in particular, the lung and oesophagus), the role of prolonged imaging surveillance is discussed, yet remains controversial. Low-dose CT appears more effective than chest X-ray alone, yet the outcome for patients with secondary lung primaries is poor and therefore raises the question of its effectiveness in this setting.Reference Cramer, Grauer, Sukari and Nagasaka162 The PET-CT scan may offer advantages over CT and MRI, with an increasing sensitivity and specificity the longer after treatment completion it is used.Reference Prestwich, Arunsingh, Zhong, Dyker, Vaidyanathan and Scarsbrook163,Reference Sheikhbahaei, Taghipour, Ahmad, Fakhry, Kiess and Chung164 Currently, no robust prospective data have yet confirmed a survival advantage from regular surveillance imaging.
Monitoring and managing symptom burden
Because of their anatomical location, head and neck tumours and their treatment result in significant morbidity, including problems with speech and voice, swallowing, pain, and disfigurement. Consequently, head and neck cancer has one of the highest disease burdens of any cancer type.Reference Semple, Dunwoody, George Kernohan, McCaughan and Sullivan165 In addition, neurological morbidity and endocrinopathies may pose long-term challenges to rehabilitation.
Follow up provides an opportunity to both assess symptom burden in the early phase after treatment, but also monitor for late and long-term effects. Examples of late effects include xerostomia, dysphagia, dental problems, osteoradionecrosis and lethargy. Some late effects of treatment, such as dysphagia, may have a period of stability for several years before late deterioration.Reference Aggarwal, Goepfert, Garden, Garg, Zaveri and Du166 The wide range of morbidity experienced by head and neck cancer survivors demonstrates the need for a multidisciplinary input into follow-up services. This includes access to speech and language therapy, dieticians, clinical psychology or counselling, physiotherapy, and dental rehabilitation. Longer-term follow up also provides an opportunity to support patients with smoking and alcohol cessation, as well as psychological support regarding fear of cancer recurrence (see also Chapters 9–16).Reference Ghazali, Cadwallader, Lowe, Humphris, Ozakinci and Rogers167
Endocrinopathies are not uncommon after treatment for head and neck cancer.Reference Barroso-Sousa, Barry, Garrido-Castro, Hodi, Min and Krop168 Particular attention should be paid to the development of hypothyroidism. This may occur in 25–40 per cent of patients undergoing radiation treatment, but rises to greater than 60 per cent in patients undergoing total laryngectomy and radiotherapy, with an average time to detection of eight months.Reference Sinard, Tobin, Mazzaferri, Hodgson, Young and Kunz169
Follow-up guidelines
Following from the issues and evidence explained above, it is possible to arrive at guidelines for the follow up of patients after treatment for head and neck cancer.
Setting
Patients should be followed up in specialist head and neck oncology clinics, where they can be seen by clinicians with a specialty interest in head and neck cancer, and with access, in clinic, to an MDT, dependent on an individual patient's needs. Studies of specialist nurse counselling and intervention after treatment, in conjunction with regular clinical follow up, have demonstrated improvements in health-related quality of life and depressive symptoms.Reference van der Meulen, May, de Leeuw, Koole, Oosterom and Hordijk170 Continuity of care is an important issue to patients.Reference Lewis, Neal, Hendry, France, Williams and Russell171 There should be access to all aspects of patient support, i.e. clinical nurse specialists, dieticians, speech and language therapists, physiotherapy, dentistry, and psychological support. There should also be access to smoking cessation services.
There should be access to urgent CT and MRI, if required, when recurrence is suspected. Ultrasound can ideally be performed in clinic.
There may be a tension between all follow up taking place at a head and neck treatment centre and travel or access issues, but models of service delivery with out-reach peripheral spokes can provide the required level of MDT input, as close as possible to home (see Chapter 1).
Frequency
Current UK national guidelines recommend that patients undergo clinical follow up every two months for the first two years after treatment, and then every three to six months for the next three years, consistent with the aim to align with the risk of locoregional recurrence.
The fact that most recurrences are detected because the patient has symptoms (rather than through routine examination) has led to the consideration of alternative flexible patient-led follow-up strategies. It is not yet clear if they will be any more effective in detecting recurrences at an earlier stage,Reference Lorenc, Wells, Fulton-Lieuw, Nankivell, Mehanna and Jepson172 or whether these will replace or add to traditional follow up.
The use of patient-reported outcome measures may assist patients with describing, recording and temporally monitoring relevant symptoms within this setting.Reference Faria S, Hurwitz, Kim, Liberty, Orchard and Liu173 In the absence of high-quality prospective trial data, follow-up schedules should remain in line with international standards, although current evidence would support additional patient-initiated follow up.
The use of patient-reported outcome measures and holistic need assessments can also assist with the identification of patients requiring more support.
Duration
Patients should undergo follow up for a minimum of five years, a point in time at which the risk of recurrence is very low and tends to plateau in many cancers. Many clinicians extend beyond this, and patient preference will play a role in the shared decision-making after the five-year period, especially in the presence of significant ongoing fear of cancer recurrence. Other situations that may benefit from prolonged follow up include where patients experience ongoing symptom burden or when delayed recurrence is known to occur. For example, distant failures can occur much later in HPV-positive cancers.Reference Huang, Perez-Ordonez, Weinreb, Hope, Massey and Waldron174 How the early detection of late distant metastases can best be achieved has yet to be defined.
Nature
Follow up should include:
• Post-treatment imaging after chemoradiotherapy, usually by PET-CT scan at three to four months
• Baseline axial imaging for patients with altered anatomy or imaging as a result of treatment
• Clinical examination including, when indicated, flexible nasopharyngolaryngoscopy
• Access to a wider MDT for patient support
• Access to urgent imaging for suspected recurrence
Recurrent disease
Recommendations
• Consider curative, palliative and supportive options for patients with recurrent disease (evidence-based recommendation (R))
• Biopsy is required before active treatment and should include PD-L1 (programmed death-ligand 1) testing for immunotherapy (R)
• The decision-making process is highly complex and requires multidisciplinary input (R)
• Patients and their families should be aware of the prognosis, chance of treatment efficacy and complications when making decisions about possible treatment (R)
• Consider PET-CT before active treatment (good practice point (G))
General principles
The development of recurrent disease after head and neck cancer is common and, on the whole, carries a poor prognosis. In general, management options include:
• Surgery for local or regional recurrence
• Chemoradiotherapy (if not already given) for local or regional recurrence
• Re-irradiation (see Chapter 4)
• Chemotherapy (generally palliative, see Chapters 4 and 15)
• Immunotherapy (see Chapter 4)
• Local targeted therapies (e.g. photodynamic therapy or electrochemotherapy)
• Best supportive care
Many patients will be offered, or choose, best supportive care alone, either through having disease that is not amenable to treatment, or by being too frail to tolerate therapy and jeopardising their quality of life. However, there are patients who can benefit from surgical and non-surgical treatment in the recurrent setting. Careful patient selection is fundamental, taking into consideration specific tumour, patient and prognostic factors, as well as the effects of prior treatment. The complexity of recurrent cancers necessitates management by an MDT that can offer a full range of both surgical and non-surgical treatments, as well as significant functional support and rehabilitation. The early involvement of palliative and supportive care services is important in all patients with recurrent disease.
Assessment
Patients with recurrent disease require careful evaluation, essentially repeating the same assessment processes used for primary disease, but with greater emphasis on co-morbidities, functional limitations from prior treatment and current performance status, to help determine the patient's level of frailty and their ability to tolerate therapy. Crucially, the social context and support systems in place for patients also need to be considered. Continued smoking or significant alcohol intake should be addressed at this stage, as these are likely to increase treatment complication rates.Reference Kaka, Zhao, Ozer, Agrawal, Kang and Rocco175
Biopsy is mandatory if further treatment is to be considered. This should include molecular profiling in this era of evolving systemic therapies (e.g. combined positive score and PD-L1 for immunotherapy – see Chapter 4).
Imaging
The role of imaging is similar to the assessment of primary disease. Baseline imaging (after primary treatment, before recurrence) helps in the delineation between distorted anatomy, inflammation, fibrosis and recurrent tumour. There is evidence that techniques such as diffusion-weighted MRI may help with this.Reference Ailianou, Mundada, De Perrot, Pusztaszieri, Poletti and Becker176
18Fluorine-fluorodeoxyglucose PET-CT imaging can struggle to differentiate metabolic activity and inflammation from tumour recurrence. However, it has two significant advantages in the recurrent setting. The first is a high negative predictive value, both for disease in the primary site and the neck, which persists even at 12 and 24 months.Reference Sheikhbahaei, Taghipour, Ahmad, Fakhry, Kiess and Chung164 In addition, it may detect occult recurrence including distant metastases, absent on clinical examination, as well as second primary tumours.
Salvage surgery
Selection
Organ preservation protocols have emerged as a standard of care for locally advanced head and neck cancer. Approximately 25 per cent of patients develop a locoregional recurrence. In this setting, salvage surgery provides the best opportunity for long-term survival.Reference Tan, Giger, Auperin, Bourhis, Janot and Temam151 Even then, the effectiveness of salvage surgery has been reported at only 39 per cent in a meta-analysis of 1080 patients.Reference Goodwin177 Some anatomical subsites offer a better chance of salvage, for example a five-year survival rate of 83 per cent for early larynx cancer recurrence. Yet salvage surgery also comes with significant and serious risks – complications after salvage surgery have been reported to be as high as 68 per cent.Reference Putten, Bree, Doornaert, Buter, Eerenstein and Rietveld178 These two factors of poor overall control and high morbidity to the patient mandate open and detailed discussions with patients before committing to surgery.
Efforts have been made to identify prognostic factors in order to better select those patients with the highest chance of success with salvage surgery. Broadly, these can be divided into: (1) patient factors; (2) prior treatment factors; and (3) tumour factors.
Patient factors: Patient age and co-morbidity are significant factors in determining outcome after salvage surgery. In a study of 191 patients, the pre-salvage Charlson–Age Comorbidity Index (‘CACI’) was identified as an independent risk factor for death at one year post salvage surgery.Reference Kim, Kim, Albergotti, Choi, Kaplan and Abberbock179 The Eastern Cooperative Oncology Group (‘ECOG’) performance status may also be used to determine general health. Continued smoking and excessive alcohol use lead to higher complication rates and should be addressed before surgery.Reference Kaka, Zhao, Ozer, Agrawal, Kang and Rocco175 Nutritional deficits are likely to be present, either because of ongoing swallowing difficulties from previous treatment or because of the disease process itself. These will need correcting through dietetic and nutritional support. Other pre-operative optimisation to aid tissue healing will include normalising thyroid function, improving diabetic control, and the cessation of any immunosuppressive agents where possible.
Prior treatment factors: Previous treatment with radiotherapy or chemotherapy is associated with poor outcomes after salvage surgery. A study of 39 patients with recurrent oral cavity cancer undergoing salvage surgery had a 43 per cent overall five-year survival rate. Yet this was reduced to 10 per cent in those who had undergone previous irradiation.Reference Tam, Araslanova, Low, Warner, Yoo and Fung180 Patients with a short duration between primary treatment and recurrence have a poor prognosis. If this interval is less than six months, it is likely that this represents persistent rather than recurrent disease. Conversely, a longer disease-free interval has been demonstrated to lead to a lower risk of death.Reference Mucke, Wagenpfeil, Kesting, Holzle and Wolff181 This is likely reflective of the inherent aggressiveness of the biology of the disease.
Tumour factors: Advanced stage of either the primary tumour or the recurrence negatively impacts prognosis, which appears to be independent of anatomical site.Reference Goodwin177,Reference Borsetto, Higginson, Aslam, Al-Qamachi, Dhanda and Marioni182 Other tumour factors considered poor prognosticators include positive margins, recurrence in the neck and locoregional recurrence (as opposed to local recurrence only). Human papillomavirus positive recurrent disease has a better prognosis than HPV-negative disease.
Combining prognostic factors to develop stratification scores for post salvage surgery survival has been attempted. In a study of 38 patients undergoing salvage surgery, initial advanced stage, and concurrent local and regional failures, were demonstrated to be independent poor predictors for decreased survival. Two-year overall survival rates for patients with two, one or none of these predictive factors were 0 per cent, 49 per cent and 83 per cent, respectively.Reference Tan, Giger, Auperin, Bourhis, Janot and Temam151
Neck recurrence
Treatment failure in the neck may occur in isolation, or combination with primary site recurrence. As with other recurrent disease, meticulous assessment is required to exclude distant disease, and determine the extent of nodal disease, the presence of extra-nodal extension and the involvement of adjacent structures. Progression of neck disease may lead to fungation, and subsequently poor quality of palliation for a patient. Therefore, even in the presence of limited distant disease, with careful consideration given to patient expectations and wishes, there may be a rationale for considering salvage neck surgery to control regional disease.
Traditionally, salvage neck dissection has been undertaken with either a radical or modified radical approach. In this situation, the occurrence of complications, including wound infection, dehiscence, chyle leak or bleeding, are common. In the era of HPV-positive disease, there has been a move towards a more selective, and even super-selective approach to the extent of nodal levels dissected. The evidence for this is not yet definitive.Reference Okano, Hayashi, Matsuura, Shinozaki and Tomioka183,Reference Robbins, Doweck, Samant and Vieira184
Oligometastatic disease
The prognosis of metastatic head neck cancer is poor. The current standard of care is palliative intent treatment with systemic therapies, or best supportive care. Hellman and Weichselbaum proposed a transitional state between minimal detectable metastatic burden (more recently thought of as five lesions or fewer) and more widespread disease – the concept of oligometastasis.Reference Hellman and Weichselbaum185 Patient selection for the treatment of oligometastasis should consider similar factors as those discussed for recurrent disease. Treatment of the metastatic disease should not be undertaken without concurrent treatment of locoregional disease. Broadly, the options are between surgery (metastasectomy) or stereotactic body radiation therapy. There are too few studies to accurately compare the effectiveness of the two approaches in head and neck cancer.Reference Thomas T, Packianathan, Bhanat, Albert, Abraham and Gordy186 Few data exist of treatment of oligometastasis from the oral cavity, or laryngeal or hypopharyngeal cancer. In nasopharyngeal cancer and HPV-positive oropharyngeal cancer, there are data to support improved overall survival in patients with oligometastatic disease treated aggressively, compared to those receiving systemic treatment alone.Reference Thomas T, Packianathan, Bhanat, Albert, Abraham and Gordy186
Targeted local therapies
Photodynamic therapy
Photodynamic therapy is an ablative treatment that relies on localised cell and tissue destruction by the activation of a photosensitising agent. It largely preserves connective tissue, minimising disfigurement and maintaining function. Importantly, it is repeatable, even in previously irradiated areas. Photodynamic therapy may have a role in patients with locoregional recurrent disease. A study of 128 patients undergoing photodynamic therapy, for whom standard multimodality treatment had failed, was conducted to evaluate the overall clinical benefit, demonstrating a complete response rate of 16 per cent. This rose to 30 per cent in those who had favourable characteristics, namely smaller, more superficial tumours amenable to surface illumination.Reference Lustig, Vogl, Fromm, Cuenca, Alex Hsi and D'Cruz187 Patients with a complete response had a much greater one-year survival rate compared to those with a non-complete response (73 per cent vs 32 per cent). These results were subsequently validated in a study of 39 patients unsuitable for further salvage treatment. Here, a higher complete response rate of 49 per cent was demonstrated, with the difference in overall one-year survival again being significantly different between the complete response and non-complete response groups (86 per cent vs 28 per cent).Reference Tan, Dolivet, Ceruse, Vander Poorten, Roest and Rauschning188
Electrochemotherapy
Electrochemotherapy is a localised therapy whereby cytotoxic agents are applied either locally or topically, and entry to tumour cells is gained through the application of pulsed electrical currents. These electric pulses temporarily depolarise the cell membrane, thereby increasing the permeability of tumour cells. A systematic review of its use in mucosal head and neck cancer only identified a total of 128 patients who had received electrochemotherapy in the palliative setting.Reference Strojan, Groselj, Sersa, Plaschke, Vermorken and Nuyts189 In this group, a response rate (either complete or partial) of 73.1 per cent was demonstrated, with little apparent deterioration in quality of life scores after treatment. More prospective data are required before the effectiveness of electrochemotherapy can be judged.
Important questions to be answered and future developments
Remote consultation
Prior to the emergence of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), it is unlikely that either patients or clinicians would have been widely accepting of remote clinical review. Alongside telephone assessment of urgent head and neck cancer referrals, follow up also moved to a model of remote consultations via telephone or video call initially, with few face-to-face appointments. This was further refined, and evidence from UK centres indicates that care has now moved to a hybrid model of both remote and face-to-face consultations based on risk stratification.Reference Kanatas, Coffey, Spellman, Twigg, Lowe and Rogers190
Patient-initiated follow up
Limited evidence for the effectiveness of routine follow-up schedules, increasing pressure on head and neck cancer services, and patients’ desires for a more flexible system, mean that alternative strategies are being explored. Patient-initiated follow up, a model where patients have more control over when and where their care is delivered, has been demonstrated to improve satisfaction and quality of life, without deleterious effects on outcome.Reference Taneja, Su'a and Hill191 This approach may be particularly effective when used in conjunction with the stratification of follow up based on an individual's risk of recurrence (e.g. via imaging with PET-CT, or associated clinical or pathological factors). A retrospective study examining different intensities of follow up, stratified by the patient's risk of recurrence based on PET-CT imaging, found that the time to recurrence detection, overall survival, and proportion of salvageable recurrences were similar between the two cohorts. A PET-CT stratified follow up reduced the mean number of visits and led to a significant cost saving per patient (£2738 over five years of follow up).Reference Shah, Te Marvelde, Collins, De Abreu Lourenco, D'Costa and Coleman192 Although these strategies have not yet been widely evaluated in head and neck cancer, clinicians appear willing to engage and test this hypothesis.Reference Lorenc, Wells, Fulton-Lieuw, Nankivell, Mehanna and Jepson172
Liquid biopsies
Technologies to develop blood- or saliva-based biomarkers, in the form of either circulating fragments of DNA or circulating whole tumour cells, have developed rapidly. These liquid biopsies allow serial monitoring and may overcome the issue of tumour heterogeneity – a significant factor in head and neck cancer. There is early evidence that in HPV-positive disease, HPV circulating tumour DNA may be able to complement imaging to predict and detect recurrent disease.Reference Tanaka, Takemoto, Horie, Takai, Fukusumi and Suzuki193 Prospective trial data will be required before the incorporation of liquid biopsies into surveillance strategies.
Management of early metastatic disease
Earlier detection through imaging and liquid biopsies is likely to open new considerations about how best to manage patients with early metastatic disease, especially those with oligometastatic disease. Limited data currently exist on how best to manage this in head and neck cancer patients.
Studies due to report
‘PETNECK 2’ trial
The efficacy and cost-effectiveness of FDG PET-CT to guide follow up after treatment for head and neck cancer has been posed as an outstanding research recommendation by the National Institute for Health and Care Excellence.194 The ‘PETNECK2’ trial has been funded by the National Institute for Health Research and will assess an alternative active surveillance strategy (using PET-CT-guided, patient-initiated, symptom-based follow up), compared to the current standard-of-care routine regular clinical follow up. This randomised, controlled trial will complete in April 2026.
Key points
• Both globally and in the UK, head and neck cancer incidence is increasing, and is projected to continue to rise, largely driven by increases in oropharyngeal cancer.
• Mortality rates in the UK over the last decade have started to increase, reflecting the rising incidence and static survival rates.
• The major risk factors for oropharyngeal cancer are: tobacco smoking, alongside alcohol consumption and tobacco used in combination; betel chewing in Southeast Asian populations; and human papillomavirus (HPV).
• Head and neck cancers are clearly socio-economically patterned, with individuals from the most deprived backgrounds having the greatest burden and poorest survival outcomes, and this socio-economic risk is not entirely explained by smoking and alcohol consumption behaviours.
• Head and neck cancer incidence is higher among men than women, and is more common in older age groups, although oropharyngeal cancer incidence peaks around 10 years younger at around 60–65 years.
Definitions of head and neck cancer
Approximately 90 per cent of head and neck cancers are squamous cell carcinoma that arise from the epithelial lining of the oral cavity, pharynx and larynx.Reference Sanderson and Ironside195 There are many types of head and neck cancer, which are discretely categorised on the basis of their anatomical location using the International Classification of Diseases, 10th revision (‘ICD-10’), from the World Health Organization.196 In addition, the subsites that are included under the definition of ‘head and neck cancer’ often vary across studies, particularly with relation to the oral cavity and oropharynx.Reference Conway, Purkayastha and Chestnutt197,Reference Kaste, Dolecek, Zavras and Radosevich198 Because of differences in definition, it is important that anatomical subsites are clearly specified when reviewing the epidemiological literature (ideally by using corresponding International Classification of Diseases codes).
Global incidence trends in head and neck cancer
Head and neck cancer is the seventh most common cancer globally, accounting for more than 660 000 new cases and 325 000 deaths annually. The overall incidence of the disease continues to rise, with a predicted increase of 30 per cent (over one million) new cases annually by 2030.Reference Sung, Ferlay, Siegel, Laversanne, Soerjomataram and Jemal199,Reference Johnson, Burtness, Leemans, Lui, Bauman and Grandis200 This increase in incidence has been recorded across both developed and developing countries.Reference Bravi, Lee, Hashibe, Boffetta, Conway and Ferraroni201 Southeast Asia and Asia-Pacific regions have particularly high incidences of oral cancer, which are strongly associated with chewing of the areca nut (betel quid), with or without tobacco.Reference Shield, Ferlay, Jemal, Sankaranarayanan, Chaturvedi and Bray202 Oral cancer is therefore expected to rise within Southeast Asia, in line with population growth.Reference Cheong, Vatanasapt, Yi-Hsin, Zain, Kerr and Johnson203 The increasing rates of head and neck cancer in the USA and Europe have been attributed to a rise in oropharyngeal cancer, linked to HPV infection (Figures 1 and 2).Reference Mehanna, Beech, Nicholson, El-Hariry, McConkey and Paleri204,Reference Gillison, Chaturvedi, Anderson and Fakhry205 Over the next 20 years, it is expected that the majority of head and neck cancers will be HPV-positive, with projections that in some European countries, such as the UK, oropharyngeal cancer incidence will overtake cancer of the oral cavity.Reference Conway, Purkayastha and Chestnutt197
Figure 1. Global age-standardised incidence rates (ASR) of head and neck cancer. Reprinted with permission from the World Health Organization International Agency for Research on Cancer ‘Cancer Today – Data visualization tools for exploring the global cancer burden in 2020’ (in: http://gco.iarc.fr/today, accessed August 2021). The map was generated using the Global Cancer Observatory (‘Globocan’) website mapping tool by selecting the ‘lip, oral cavity’, ‘oropharynx’, hypopharynx’ and ‘larynx’ cancer sites. Estimated age-standardised rates of head and neck cancer incidence worldwide are shown for both sexes.
Figure 2. Age-standardised incidence rates (ASR) of head and neck cancer, by continent. Reprinted with permission from World Health Organization International Agency for Research on Cancer ‘Cancer Today – Data visualization tools for exploring the global cancer burden in 2020’ (in: http://gco.iarc.fr/today, accessed August 2021). The map was generated using the Global Cancer Observatory (‘Globocan’) website mapping tool by selecting the ‘lip, oral cavity’, ‘oropharynx’, hypopharynx’ and ‘larynx’ cancer sites. Estimated age-standardised rates of head and neck cancer incidence by continent are shown for both sexes.
Worldwide, laryngeal cancer incidence and prevalence have increased by 12 per cent and 24 per cent, respectively, during the past three decades.206 However, age-adjusted rates for new laryngeal cancer cases have been falling in countries with a higher sociodemographic index, perhaps reflecting changes in smoking and alcohol drinking behaviours.206 Overall, head and neck cancer affects males two to four times more than females, with estimates reaching over 20 per 100 000.Reference Miranda-Filho and Bray207 For men in developing countries, lip and oral cavity cancer is the second most common cancer (10 per 100 000). Male incidence of oral and oropharyngeal cancer has declined over recent years in France (−12.6 per cent), Slovakia (−4.0 per cent), Spain (−10.8 per cent), Brazil (−26.7 per cent) and Hong Kong (−10.5 per cent), while it increased in the UK (18.8 per cent), Australia (8.7 per cent), Japan (21.3 per cent) and in the USA (3.7 per cent).Reference Bosetti, Carioli, Santucci, Bertuccio, Gallus and Garavello208 The risk of head and neck cancer increases with age across populations, with the majority of cases diagnosed in those aged over 50 years.Reference Warnakulasuriya209 Similarly, there has been a rise in cases amongst females, predominantly in European countries, which may be explained by sex-specific patterns of tobacco and alcohol consumption.Reference Miranda-Filho and Bray207,Reference Bosetti, Carioli, Santucci, Bertuccio, Gallus and Garavello208
Trends in head and neck cancer in the UK
As described previously, head and neck cancer rates are also rising in the UK. Studies have shown that from 1995 to 2011, oropharyngeal cancer incidence increased by 7.3 per cent for males and 6.5 per cent for women in England, with oral cavity cancer showing a 2.8 per cent rise in men and 3.0 per cent rise in women over the same period.Reference Louie, Mehanna and Sasieni210 Incidence rates are highest in Scotland, where oropharyngeal cases were shown to have increased by 85 per cent from 2011 to 2012.Reference Purkayastha, McMahon, Gibson and Conway211 These rates are continuing to rise according to most recent UK Cancer Registry Data, which show a 34 per cent increase in total cases diagnosed across the four nations from 2011 to 2018. The burden of head and neck cancer is strongly socio-economically patterned, with the highest rates being observed among people living in the most socio-economically deprived communities.Reference Purkayastha, McMahon, Gibson and Conway211,212 These cancer registry data also demonstrate that the majority of head and neck cancers are diagnosed at an advanced stage.213 For the UK as a whole, 58.5 per cent of head and neck cancers with a known stage are diagnosed at an advanced stage, III or IV, in accordance with the TNM Atlas, seventh edition.Reference Brierley, Asamura, van Eycken and Rous214 The highest rates of advanced disease are found in Scotland and Northern Ireland, where 67.6 per cent are at an advanced stage. Stage IV is the most common stage at diagnosis for oral and oropharyngeal cancer, whereas stage I is most common for laryngeal cancer.213
Mortality and survival trends in head and neck cancer in the UK
In 2018, there were 4078 deaths attributable to head and neck cancer in the UK, accounting for approximately 2 per cent of all cancer deaths annually.215 Variation in national head and neck cancer mortality rates (European age-standardised rate per 100 000 population) was apparent between nations of the UK: Scotland (rate = 8.7) and Northern Ireland (rate = 8.4) had worse outcomes than England (rate = 6.2) and Wales (rate = 5.8). Age-specific mortality that is attributable to head and neck cancer rises from the fifth decade of life onwards, towards a peak mortality in those aged over 90 years – a phenomenon most pronounced in males. In the UK, since the early 1970s, the combined head and neck cancer mortality for men and women has fallen by 11 per cent overall (age-standardised rate per 100 000 population = 7.3 in 1971 and 6.5 in 2018); however, the last decade has seen a gradual rise in mortality rates from a low in 2006 (age-standardised rate per 100 000 population = 5.6), possibly reflecting the changes in disease incidence and static survival rates.215
Survival rates can vary significantly according to geographical location, tumour site, HPV association (associated with increased chances of survival),Reference Du, Mazul, Farquhar, Brennan, Anantharaman and Abedi-Ardekani216 and, most prominently, stage at diagnosis. Those diagnosed with advanced disease have notably poorer outcomes than those with early disease. Analysis of a large cohort in the USA showed that patients with HPV-positive cancers had a better chance of long-term survival compared to those with non-HPV cancers, confirming previous studies also suggesting this phenomenon.Reference Du, Mazul, Farquhar, Brennan, Anantharaman and Abedi-Ardekani216 There are several studies within the recently established Head and Neck Cancer in South America and Europe (‘HEADSpAcE’) international consortium217 that are exploring the factors associated with relatively poor survival among people with head and neck cancer. Analysis of routinely collected data and from large prospective cohort studies, such as Head and Neck 5000,Reference Ness, Waylen, Hurley, Jeffreys, Penfold and Pring218 have demonstrated that significant socio-economic inequalities in head and neck cancer survival exist in the UK, not all of which can be explained by behavioural factors.Reference Ingarfield, McMahon, Hurley, Toms, Pring and Thomas219
Risk factors associated with head and neck cancer
Tobacco smoking and alcohol consumption are well established risk factors for head and neck cancer. However, a detailed understanding of these somewhat complex behaviours in terms of precise estimates of risk, recognising the joint tobacco–alcohol effect, the dose–response, and the benefits of quitting both smoking and alcohol, remains less well established. The role of other risk factors in head and neck cancer risk, such as smokeless tobacco, betel chewing, diet, oral health and hygiene, and hormonal, genetic, occupational and socio-economic status, is also poorly understood. A major challenge in elucidating detailed information from the epidemiological literature is the heterogeneity in study designs and populations from often small observational studies. The dominant effects of tobacco smoking and alcohol drinking also overshadow other minor risk factors.
A recent umbrella review of systematic reviews and meta-analyses (Conway et al.Reference Conway, Hashibe, Boffetta, Wunsch-Filho, Muscat and La Vecchia220) has been combined with pooled analyses at the individual level data from studies from around the world, by The International Head And Neck Cancer Epidemiology (‘INHANCE’) Consortium.221 These data confirm that tobacco smoking and alcohol drinking behaviours, separately and in combination, are major risk factors for head and neck cancer, accounting for 72 per cent of cases when used in combination.Reference Hashibe, Brennan, Chuang, Boccia, Castellsague and Chen222 Recent evidence has shown an independent causal effect of alcohol consumption, when controlling for smoking (odds ratio = 2.1), suggesting that the role of alcohol may have been previously underestimated.Reference Gormley, Dudding, Sanderson, Martin, Thomas and Tyrrell223
High-risk HPV, especially HPV type 16, is a major risk factor for oropharyngeal cancer,Reference Hobbs, Sterne, Bailey, Heyderman, Birchall and Thomas224 thought to be sexually transmitted via oro-genital contact.Reference Gillison, Chaturvedi, Anderson and Fakhry205 Smoking has been shown to interact with HPV and increase risk.Reference Anantharaman, Muller, Lagiou, Ahrens, Holcátová and Merletti225 Those who have HPV-negative oropharyngeal tumours are more likely to be heavier smokers, with an increased risk of death for every additional pack-year, compared to HPV-positive cases.Reference Ang, Harris, Wheeler, Weber, Rosenthal and Nguyen-Tan226 Genetic susceptibility to head and neck cancer has also been investigated,Reference Vukovic, Stojanovic, Vecchioni, Pastorino and Boccia227 with the largest genome-wide association study of oral and pharyngeal cancer (6034 cases and 6585 controls from Europe, North America and South America), detecting seven unique loci.Reference Lesseur, Diergaarde, Olshan, Wunsch, Ness and Liu228 Genetic variants in alcohol-metabolising genes, such as alcohol dehydrogenase (ADH), are associated with increased head and neck cancer risk. This study also found a strong protective association at a chromosome within the human leukocyte antigen class II region in oropharyngeal subgroup analysis.Reference Lesseur, Diergaarde, Olshan, Wunsch, Ness and Liu228 Going forward, this could help explain why some individuals are more at risk of developing the disease following HPV infection.
Key recommendations
• Microsurgical free-flap reconstruction should be the primary reconstructive option for most defects of the head and neck that need tissue transfer (good practice point (G))
• Composite free tissue transfer should be offered as the first choice to all patients needing mandibular reconstruction (evidence-based recommendation (R))
• Free flaps should be offered as the first choice of reconstruction for all patients needing circumferential pharyngoesophageal reconstruction (R)
• Free-flap reconstruction should be offered for patients with class III or higher defects of the maxilla (R)
• Patients undergoing salvage total laryngectomy should be considered for vascularised flap reconstruction, to reduce pharyngocutaneous fistula rates (R)
• For reconstruction involving the upper and lower jaws, or involving rhinectomy or orbital exenteration, pre-operative multidisciplinary decision-making should include restorative dentistry or dental prosthodontists (R)
• Tubing over and use of a salivary bypass tube appear to decrease complication rates with anterolateral thigh and radial forearm free flaps (G)
• Each head and neck centre should have a protocol for free-flap monitoring and rescue (G)
Introduction
Reconstructive surgery for head and neck cancer defects can be complex and challenging. Priorities of reconstruction include restoring oral and upper aerodigestive tract lining, maintaining oral competence along with the function of speech and swallowing, and providing an acceptable aesthetic result.
These guidelines have been divided into the management of defects in the oral cavity soft tissues, mandible, maxilla and midface, oropharynx, laryngopharynx, and neck soft tissue. Where tumour ablation or reconstruction involves the possible need for orofacial implants and oral rehabilitation or prosthetics, close collaboration with the consultant restorative dentist or oral rehabilitation team is required. This is discussed in detail in Chapter 13. Options for facial palsy are discussed in Chapter 22 (lateral skull base), and skull base defects are also discussed in Chapters 22 (lateral) and 23 (anterior).
There is little evidence relating to the optimal reconstruction of head and neck defects. While many mandibular and soft tissue upper aerodigestive tract reconstruction techniques are fairly standard, some controversy remains regarding the midface and maxilla because of the complexity of the defects and the possibility of using a dental or facial prosthesis, especially with the advent of three-dimensional (3D) digital planning and printing. Flap selection is usually determined based on the expertise and experience of the individual surgical teams, as well as on patient co-morbidities, the exact nature of the surgical defect, any future possible treatments including radiotherapy, and donor site morbidity.
The ‘work horse’ microvascular and pedicled flaps that a head and neck surgical centre should provide, and which serve the majority of defects in the head and neck, are summarised in Table 1.
Table 1. Principal microvascular and pedicled flaps for head and neck reconstruction
Most reconstructions are performed primarily following tumour ablation. Modern techniques aim for one-stage reconstruction utilising vascularised tissues, with a high success rate and good overall results. However, secondary reconstructions are also undertaken to treat problems such as fistulae, osteoradionecrosis or previously obturated maxillary defects.
Oral cavity soft tissues
Oral soft tissues include the tongue, floor of mouth, buccal mucosa and the retromolar trigone extending to the tonsillar area. Because of the close proximity of these areas, cancers of these sites occasionally cross over to one another. Reconstructive access is usually determined by the extent and access of the surgical resection.
Free tissue transfer provides the mainstay of oral soft tissue reconstruction, as it allows importation of large volumes of healthy tissue from sites distant to prior surgical or radiotherapy fields. The radial artery forearm flap and anterolateral thigh perforator flap remain the ‘go to’ options for most significant oral soft tissue defects. Other options exist, as detailed in Table 2.Reference Chalmers, Rahman, Young, Kennedy, Endersby and Adams229,Reference Green, Rahman, Owen, Paleri, Adams and Ahmed230
Table 2. Mainstream and promising options for oral cavity soft tissue reconstruction
Regional flaps such as pectoralis major, submental island artery, supraclavicular artery island and nasolabial flaps can be effective in importing tissue, but are not optimal choices.
Mandible
Reconstruction of the mandible must address the site and size of the bony defect, associated soft tissue loss and the desirability of dental rehabilitation. Free tissue transfer is the mainstay of mandibular reconstruction, as it allows the importation of bone which can be tailored to fit the desired shape, is well vascularised and is amenable to osseointegration. The main flap options are:
• Fibula flap
• Deep circumflex iliac artery flap
• Scapular flap
Dental rehabilitation is a key part of most mandibular reconstructions, and pre-operative liaison with the restorative team, including consideration of 3D planning and osseo-integrated implants, is recommended (see Chapter 13).
The fibular flap allows harvest of a long piece of bone that is of adequate height for osseointegration and can be osteotomised several times for contouring. This is now made easier with the availability of 3D software to plan the osteotomies at the mandible and on the fibula prior to transfer. It is relatively easy to harvest as an osseous or osteoseptocutaneous flap, with or without muscle. This versatility means it is the workhorse for mandibular reconstruction in most centres. One drawback of the flap is its relative lack of height.
The deep circumflex iliac artery flap provides for a high bony segment, and the natural curve of the ilium lends itself to lateral mandibular defects where an osteotomy may not be necessary. The donor site defect can be problematic (pain restricting mobility, hernia and mesh infection), and its skin paddle is usually reserved for external use although muscle can be incorporated for oral reconstruction.
The scapular flap allows for harvest of a relatively small amount of bone. The main advantage of this flap is the large volume of skin and muscle (latissimus dorsi), which can be used in a chimeric fashion. The bone is a good height, but two-team flap harvesting is generally not possible.
A composite radial forearm flap is rarely used for bone reconstruction as only a small volume of bone of low height can be harvested. There is also a risk of subsequent fracture of the radius, which can be debilitating.
A new classification of the mandibular defect has been described based on the four corners of the mandible, which are both angles and both canines (Figure 1).Reference Brown, Barry, Ho and Shaw231
Figure 1. Classification of mandibular defects.Reference Brown, Barry, Ho and Shaw231
Although not in the official classifications in Figure 1, posterior mandibular defects distal to the second lower molar, involving the ramus of the mandible but sparing the subcondylar/condylar segment, may be amenable to a soft tissue only flap with a reinforced titanium plate. This is on the premise that dental rehabilitation is not required, but it risks plate fracture or extrusion. In order to minimise the latter, a robust chimeric flap such as an anterolateral thigh flap with vastus muscle would be recommended.
Maxilla and midface
The level of evidence is very weak in all areas of reconstruction, but more particularly in the maxilla and midface because of the differing complexity of the defects, and the potential for skull base involvement.
All cases involving the loss or ablation of the maxilla and/or midface should be discussed in a multidisciplinary setting. The choice of reconstruction or prosthetics requires discussion among the ablative and reconstructive teams, prosthodontist, maxillofacial technician, patient and family. There are clear advantages in simplifying the surgery and using prosthetic options, but this choice becomes more difficult to deliver, and makes it more difficult for the patient to cope, as the defect becomes larger and more complex.
The use of the maxillectomy defect classification is recommended (Figure 2).Reference Brown and Shaw232
Figure 2. Classification of the maxillary and midface defects. Classes I–VI relate to the vertical component of the defect, including orbitomaxillary (class V) and nasomaxillary (class VI) defects, when often the palate and dental alveolus are intact. Classes a–d relate to the increasing size of the palatal and dento-alveolar parts of the defect, indicating increasing difficulty in obtaining good results with obturation.Reference Brown and Shaw232
The choice of a prosthetic option or reconstruction depends on the nature of the defect. In class I and II defects, an obturator is a reasonable option, but this becomes less favourable with orbital adnexae involvement (class III), orbital exenteration (class IV), and when the midface defects are of an orbitomaxillary (class V) or nasomaxillary (class VI) nature. In addition to the vertical component, the extent of the dental or alveolar part of the resection is relevant to the prosthodontist in deciding on appropriate obturation. A summary of the options can be found in Table 3.
Table 3. Recommended reconstruction method, according to midface and maxillectomy defect classificationReference Brown and Shaw232
Letters (a, b, c) refer to the horizontal classification (Figure 2). *Supplies the iliac crest. †Supplies the scapula tip. + = recommended; − = not recommended
Class I: Resections of the alveolar bone not resulting in an oroantral fistula can either be left to granulate or treated with a local flap. Palatal defects can be obturated or reconstructed with a soft tissue flap. For larger defects not requiring implants, a vastus lateralis muscle flap based on the descending branch of the lateral circumflex femoral artery may be used. This has a low donor site morbidity, and matures with shrinkage and some fibrosis to mimic the pre-morbid hard palate.
Class II: This is the standard hemi-maxillectomy not involving the orbital floor or adnexae. Obturation is often very successful for this form of defect, as the orbit does not require support, and if the defect is small enough for retention and stability of the prosthesis. In more extensive cases (classes IIc–d), the options are an implant-retained prosthesis or composite flap. Reconstruction with the fibula flap has also shown good outcomes. The deep circumflex iliac artery flap, with greater height, and which includes the iliac crest and internal oblique muscle, will give better support to the peri-nasal area. The scapula flap can be supplied by the circumflex scapular artery that supplies the lateral scapula (scapula flap) through periosteal perforators along its length, or the angular branch of the thoracodorsal artery which supplies the scapula tip. The advantage of the scapula tip option is that the pedicle is considerably longer than the circumflex scapula artery option, which is a great advantage in the maxilla and midface as the recipient vessels are more distant.
Class III: In these cases, there is loss of the orbital support, and often a part of the nasal bones may also require reconstruction. There is good consensus in the literature that the restoration of orbital support with vascularised tissue (pedicled or free flap) is essential to ensure healing of the bone graft, and to reduce soft tissue problems such as epiphora and ectropion. The deep circumflex iliac artery with internal oblique provides the best solution if an implant-retained prosthesis is planned, but the scapula tip flap using latissimus dorsi muscle is also a good option with a more reliable pedicle. The fibula is also described for this defect, but considerable skill in the adaptation of this flap for the defect is required, with variable results. Obturation alone will result in facial collapse, poor support of the orbit, and a high risk of vertical orbital dystopia and ectropion. In children, the scapula tip will probably be the best option, as the iliac crest has a cartilaginous cover and the vessels are much smaller.
Class IV: Reasonable results can be achieved with a soft tissue flap alone such as rectus abdominus or vastus lateralis, but this will result in poor definition of the orbital defect and some facial collapse. The choice is similar to class III in that the iliac crest with internal oblique offers better implant options, but the scapula tip and fibula flaps are also good options.
Class V: In the orbitomaxillary defect, the main aim is not to obturate the orbital space with too much soft tissue, to allow space for an orbital prosthesis. The temporalis or temporoparietal flap are ideal, but in more extensive defects it is worth considering the radial or anterolateral thigh flap in a thinner patient. Some patients may prefer the natural ‘eye patch’ option provided by a thicker flap to a prosthesis. Thicker flaps will atrophy in time and can be thinned secondarily.
Class VI: If there is loss of the facial skin between the orbits and nasal bones, then free tissue transfer is probably essential. The composite radial artery forearm flap can be ideal if harvested with fascia to line the nasal side of the radial strut and the skin to restore the face. This can be augmented with a glabella or forehead flap.
A classical rhinectomy can be rehabilitated with a prosthesis, and of course the surgeon can check the margins of resection and resect more tissue if required. There are very successful full rhinectomy reconstructions performed, which can give a permanent biological solution if preferred. In this defect, attention must be paid to the restoration of the cartilaginous scaffolding and nasal bones with vascularised tissue, to prevent complications during and following radiotherapy. Frequently used combinations are radial artery forearm flap (inner lining), non-vascularised auricular cartilage and bone grafts, and a paramedian forehead flap (external surface).
Skull base reconstruction
The primary intentions are to seal the cranial cavity off and prevent cerebrospinal fluid leaks. This critical defect requires careful planning by all teams and surgeons involved in the case. This is discussed in more detail in Chapters 22 (lateral skull base) and 23 (anterior skull base).
Oropharyngeal reconstruction
Most tumour ablation involving the oropharynx is now transoral, after which there is usually no requirement for reconstruction. However, transoral surgery for tumour recurrence after radiotherapy may require reconstruction, for example, if the carotid sheath is left exposed, as the vitality of remaining tissue will have already been compromised. Other indications include soft palate reconstruction and tongue base reconstruction.
When required, radial artery forearm flap can be used and inset transorally following transoral tumour ablation. Anterolateral thigh flaps might be useful when more bulk is required. A posteromedially based musculomucosal flap, facial artery musculomucosal flap or radial artery forearm flap microvascular flaps can be used for soft palate reconstruction.Reference Meccariello, Montevecchi, Sgarzani, De Vito, D'Agostino and Gobbi233
When open surgery is performed, microvascular options are preferred, mainly radial artery forearm or anterolateral thigh flaps.
Pharyngolaryngectomy reconstruction
Non-circumferential (partial) pharyngeal defects
Defects that result in insufficient pharyngeal mucosa for a primary repair will require a ‘patch’ flap. Options include the pectoralis major myocutaneous flap, and the supraclavicular artery island flap which does not have the bulk of the pectoralis major muscle. Free flaps, such as radial artery forearm flap, anterolateral thigh flap and medial sural artery perforator flap, may also be used. The use of a salivary bypass tube appears to decrease fistula rates.Reference Marijić, Grasl, Grasl, Faisal, Erovic and Janik234
A pectoralis major myocutaneous flap or myofascial flap is also a good option for persistent fistula after total laryngectomy.
If the pharyngeal mucosal remnant is very narrow (less than 1 cm in width), then it is often better to excise the remnant and undertake a total circumferential reconstruction.
Total circumferential pharyngolaryngectomy defects
There are several options for reconstruction following circumferential pharyngolaryngectomy, summarised with an historic context by Patel et al.Reference Patel, Goldstein, Brown, Irish, Gullane and Gilbert235 The main options are, however:
• Tubed* anterolateral thigh flap
• Tubed* radial artery forearm flap
• Jejunal free flap
*Tubing around a salivary bypass tube with two-layered closure (fascia and skin) appears to decrease fistula rates.Reference Marijić, Grasl, Grasl, Faisal, Erovic and Janik234
Expertise and experience with these options may vary, particularly concerning jejunal free flaps. However, most published experience with regard to voice and swallowing, as well as morbidity, favours cutaneous flaps rather than a jejunal free flap.Reference Patel, Goldstein, Brown, Irish, Gullane and Gilbert235 Problems due to hyper-peristalsis and a ‘wet’ sounding voice are common with the jejunal free flap, which also carries a morbidity rate associated with abdominal complications (approximately 5 per cent). For cutaneous flaps, the choice between an anterolateral thigh flap and radial artery forearm flap may reflect the size of the patient's legs and their forearms. Around 30 per cent of patients may need a subsequent dilatation for stricture.
The use of a pectoralis major myocutaneous flap is not generally applicable, except as a last resort. It is very difficult to tube, although it can be used as a 270-degree flap with the pre-vertebral fascia.
For salvage circumferential pharyngolaryngectomy after chemoradiotherapy, especially for extensive defects with poor tissue vitality, additional options may include gastro-omental free flaps. Limited case series suggest that these may have an advantage associated with the availability of the omentum. This can be wrapped around the anastomotic site to decrease the possibility of leakage and improve the overlying skin quality. However, the complication rate is significant, although may in part reflect patients in whom this option is considered.Reference Viñals, Rodrigues, Lopez, Payro, Porté and Sildenikova236,Reference Patel, Makitie, Goldstein, Gullane, Brown and Irish237 Additional vascularised tissue can be included with the anterolateral thigh as a chimeric flap to resurface the neck, in cases where there is poor quality skin or contracted skin that would not safely close post-operatively.
When oesophageal resection is significantly intrathoracic, there may be insufficient access for the lower anastomosis, even with various forms of manubriumectomy. In such cases, a gastric pull-up may be used. This technique carries significant morbidity and mortality associated with the need to enter three visceral cavities. However, the mortality rate associated with gastric pull-up has dropped to less than 10 per cent in more modern case series.Reference Butskiy, Rahmanian, White, Durham, Anderson and Prisman238 Colonic transposition is an alternative but is rarely used in the modern era.
Vascularised tissue after salvage laryngectomy
Pharyngocutaneous fistulae are known to occur in nearly one-third of patients who undergo salvage total laryngectomy after chemoradiation. Recent meta-analyses suggest that there is an advantage in using vascularised tissue from outside the radiation field in the laryngectomy defect, either as a buttress or to augment the circumference of the pharynx.Reference Paleri, Drinnan, van den Brekel, Hinni, Bradley and Wolf239 This may be in the form of myocutaneous inset of free or pedicled soft tissue (e.g. anterolateral thigh or pectoralis major respectively), or myofascial onlay, usually a pectoralis major myofascial flap. This intervention reduces the risk of pharyngocutaneous fistulae by one-third to a half.
Reconstruction of soft tissue neck defects
After extended neck dissections for neck disease with skin involvement, with loss of skin, a pectoralis major myocutaneous flap should be considered. Most of these cases will also involve loss of the sternocleidomastoid muscle, and this provides both skin cover and protection of the otherwise exposed carotid artery through its volume and muscle. A supraclavicular artery island flap can also be used for skin cover, but does not have the bulk.
In cases without a skin defect, when the neck dissection has included the sacrifice of the sternocleidomastoid muscle, and either chemoradiotherapy is planned or the neck dissection is for salvage treatment of recurrent disease after chemoradiotherapy, a pectoralis major myofascial flap can be considered in order to provide vascularised carotid artery cover and lessen the risk of breakdown causing catastrophic haemorrhage.
Care of patients after microvascular free-flap surgery
Microvascular reconstruction using free tissue transfer is standard practice in head and neck reconstruction. The literature demonstrates success rates in excess of 95 per cent.Reference Walia, Lee, Jackson, Hardi, Bollig and Graboyes240,Reference Pohlenz, Klatt, Schön, Blessmann, Li and Schmelzle241 The early detection of and intervention for flap failure forms an important part of ensuring low failure rates.
There is no universally agreed protocol regarding the duration and method of flap monitoring. However, the following should be considered:
• The critical period for flap vascular complications is 24–48 hours post-operativelyReference Pohlenz, Klatt, Schön, Blessmann, Li and Schmelzle241
• The probability of flap salvage is inversely related to the duration of ischaemia, and successful flap salvage is highly unlikely after 12 hoursReference Pohlenz, Klatt, Schön, Blessmann, Li and Schmelzle241
• Clinical flap monitoring includes using capillary refill, skin colour, cutaneous temperature and arterial Doppler signal. It is highly accurate and effective, with 85–95 per cent flap success rates using clinical assessment aloneReference Han, Guo, Liu, Li, Zhou and Wei242
• The use of an implantable Doppler may be considered for buried flaps or to augment clinical monitoring, with some improvement in reducing flap failure but possibly at the expense of false alarmsReference Han, Guo, Liu, Li, Zhou and Wei242
A typical protocol for monitoring is:
• One-hourly observation on post-operative day 1
• Two-hourly observation on post-operative day 2
• Four-hourly observation on post-operative day 3
While late flap vascular complications are reported, even at day 7 post-operatively, both the probability of their occurrence and their salvage is low, as these are usually related to intrinsic flap problems, which are not easily correctable.Reference Zoccali, Molina and Farhadi243
Key points
• Patients and family members should be at the centre of their care. They should receive education and information regarding prehabilitation and enhanced recovery after surgery early in their pathway, so that they can provide informed consent and engage in preparation.
• Prehabilitation and enhanced recovery after surgery relies on interdisciplinary teams involving a range of professionals, including those who may traditionally fall outside healthcare, e.g. voluntary sector, health and leisure centres, to ensure that the optimum patient outcome is achieved.
• Healthcare workers require the appropriate training, governance and operational support to be able to enact this guidance and embed the recommendations into their practice. It is recommended that clinical services identify champions within their organisation to effectively implement organisational change with appropriately resourced audit data collection.
• There may be barriers present for some head and neck cancer patients that prevent them from engaging in prehabilitation and enhanced recovery after surgery programmes; for example, difficulty in changing maladaptive behaviours, and/or psychological factors including the lived experience of trauma. All attempts should be made to gain awareness of such barriers, and to support patients in overcoming these issues in order to achieve the benefits offered from prehabilitation and enhanced recovery after surgery programmes.
• Prehabilitation and enhanced recovery after surgery programmes are more effective when delivered across a pathway that also includes rehabilitation.
Introduction
The following guidelines cover the provision of interdisciplinary team prehabilitation and enhanced recovery after surgery services required for head and neck cancer patients. An overview of the current literature in prehabilitation and enhanced recovery after surgery is provided, alongside an overview of each element. Recommendations for services are divided into key time points across the patient journey. Each section is then separated into recommendations that are considered ‘essential’ and ‘desirable’.
It is recommended that this chapter is cross-referenced with the following other chapters available in these guidelines, including those on: nutritional management in the treatment of head and neck cancer; physiotherapy and exercise; psychological management; speech, voice and swallowing rehabilitation; and the clinical nurse specialist's role.
Additionally, the following resources provide further information for both head and neck cancer clinicians and patients involved in prehabilitation and enhanced recovery after surgery programmes:
• Mouth Cancer Foundation, in: https://www.mouthcancerfoundation.org/
• The Swallows, in: https://www.theswallows.org.uk/
• Heads2gether, in: https://www.heads2gether.net/home/
• Prehab4cancer, in: www.prehab4cancer.co.uk
• ‘ERAS+’ (enhanced recovery after surgery plus), in: www.erasplus.co.uk
• SafeFit, in: https://safefit.nhs.uk/
• National Health Service – Exercise, in: https://www.nhs.uk/live-well/exercise/
• The Royal Marsden – Exercise at home, in: https://www.royalmarsden.nhs.uk/your-care/living-and-beyond-cancer/exercise-home
• ‘CURE’ (treating tobacco addiction), in: www.thecureproject.co.uk
• MacMillan Cancer Support – The building-up diet, in: https://www.macmillan.org.uk/cancer-information-and-support/stories-and-media/booklets/the-building-up-diet; Swallowing, in: https://www.macmillan.org.uk/dfsmedia/1a6f23537f7f4519bb0cf14c45b2a629/1601-10061/swallowing-tcm9-355063; and Dry mouth and changes in saliva after head and neck cancer treatment, in: https://www.macmillan.org.uk/cancer-information-and-support/impacts-of-cancer/dry-mouth-and-changes-in-saliva-after-head-and-neck-cancer-treatment
• Maggie's cancer support, in: https://www.maggies.org/cancer-support/our-support/
• ‘PINNT’ (Patients on Intravenous and Naso-gastric Nutrition Treatment), in: https://pinnt.com/Home.aspx
Prehabilitation
Cancer prehabilitation is an intervention commonly defined as occurring from the time of diagnosis to the commencement of acute treatment. Macmillan prehabilitation guidance (2019)244 outlines a set of principles to empower people with cancer to prepare for treatment, through promoting healthy behaviours and needs-based prescribing of exercise, nutrition and psychological interventions. We would suggest that cancer prehabilitation is best viewed as a continuum throughout the patients’ journey, from diagnosis to treatment, into rehabilitation and to survivorship.Reference Silver and Baima245
The treatments for head and neck cancer are often multimodal and complex, and include primary surgery and reconstruction, and/or adjuvant radiotherapy with or without chemotherapy. These treatments are associated with both acute nutrition impact symptoms and long-term side effects, including dysphagia, pain, taste changes, mucositis and dry mouth, which result in a decline in nutritional and physical function that negatively affects quality of life (QoL) and survivorship.Reference Langius, van Dijk, Doornaert, Kruizenga, Langendijk and Leemans246,Reference Langendijk, Doornaert, Verdonck-de Leeuw, Leemans, Aaronson and Slotman247 In addition, many patients present with higher rates of smoking and alcohol consumption, low socio-economic status, and many co-morbidities; these result in low physical, nutritional and emotional well-being, which can negatively impact treatment outcomes. On diagnosis, up to 60 per cent of patients have been identified as malnourished or are at risk of malnutrition.Reference Jager-Wittenaar, Dijkstra, Dijkstra, Bijzet, Langendijk and van der Laan248 Subsequent treatment often serves to worsen malnutrition status, and leads to sarcopenia development in over 50 per cent of head and neck cancer patients, which is associated with worse outcomes.Reference Findlay, White, Lai, Luo and Bauer249
Despite the downward trend of smoking, the incidence of head and neck cancer in the UK continues to rise, with approximately 50 per cent of oropharyngeal cases related to human papillomavirus.250,Reference Schache, Powell, Cuschieri, Robinson, Leary and Mehanna251 With this increasing prevalence, and the significant patient burden of head and neck cancer disease and treatment, interventions such as prehabilitation are particularly attractive, as they aim to improve the patient's physical, nutritional and mental resilience. This process aims to mitigate against the impact of treatment-related side effects, and to improve QoL as well as longevity.
A review of the literature identified several systematic reviews demonstrating the positive impact of prehabilitation amongst cancer patients.Reference Boereboom, Doleman, Lund and Williams252–Reference Gillis, Buhler, Bresee, Carli, Gramlich and Culos-Reed256 Prehabilitation that includes an exercise component can improve physical function prior to surgery, and demonstrates positive effects on length of hospital stay and post-operative surgical complications.Reference Boereboom, Doleman, Lund and Williams252,Reference Singh, Newton, Galvao, Spry and Baker255 One review, which aimed to identify whether prehabilitation improves health outcomes (physical function, nutritional status and patient-reported outcome measures) after more than 30 days post-treatment, concluded that prehabilitation improved gait, cardiopulmonary function, urinary continence, lung function and mood.Reference Faithfull, Turner, Poole, Joy, Manders and Weprin253 In a study of older adults with cancer that examined the efficacy of nutritional and exercise interventions on health-related QoL, the majority of interventions were exercise only, with variations in exercise modalities, duration and location.Reference Forbes, Swan, Greenley, Lind and Johnson254 Despite this heterogeneity, improvements in QoL were identified, highlighting the potential benefits of prehabilitation in the older adult population. Further robust and personalised interventions are recommended.
Implementing prehabilitation strategies commonly requires motivation, self-efficacy support and behaviour change in patient cohorts. Eating As Treatment (‘EAT’) is a psychological intervention that aims to improve the nutritional status in patients undergoing radiotherapy.Reference Britton, Baker, Wolfenden, Wratten, Bauer and Beck257 Dietitians were trained to include motivational interviewing and cognitive behavioural therapy strategies into their dietary counselling. This resulted in positive significant differences in nutritional status, as measured by the Patient-Generated Subjective Global Assessment (‘PG-SGA’) Short Form, weight loss and treatment interruptions. In addition, although not significant, global QoL was reported to have improved.
Currently, recommendations for head and neck cancer prehabilitation and its optimal delivery are limited by the volume of published research. This is likely because of the difficulties posed by the more complex treatment pathways in head and neck cancer when compared with abdominal and thoracic cancer resection surgery, which have tended to dominate the large prehabilitation trials. There is also a focus on more urgency in head and neck cancer surgery, potentially giving less time for prehabilitation interventions to be included in patient pathways. Additionally, there can often be difficulties associated with delivering treatment across increased geographical areas, as the head and neck cancer tertiary centre model may make the distance for patients to travel appear impractical.
Despite these challenges, there is confirmed safety and efficacy for exercise and nutrition interventions during chemo/radiotherapy in head and neck cancer patients.Reference Bye, Sandmael, Stene, Thorsen, Balstad and Solheim258 Exercise interventions included a combination of strength and aerobic exercises and were performed up to five times per week. Nutritional interventions were predominantly dietary counselling personalised to meet estimated energy and protein requirements, with oral nutritional supplements being used to meet any deficit. Additionally, a 7-day prehabilitation intervention in head and neck cancer and abdominal surgery patients, including a nutritional (oral nutritional supplements), exercise (step targets), breathing technique and skin cleaning regimen, has confirmed the feasibility of short interventions by demonstrating significant improvements in post-operative mobility and improved prevention of pulmonary morbidity.Reference Moore, Scoggins, Philips, Egger, Tennant and Little259
Despite the paucity of published research specifically in head and neck cancer, it is clear from the burden of the disease and its treatment that this group of patients can significantly benefit from many of the processes inherent in prehabilitation. These recommendations have been developed by drawing upon the current evidence, clinical expertise in head and neck, knowledge from established prehabilitation programmes, and the lived experience of head and neck cancer proton pump inhibitor (PPI) groups.
Pre-treatment (general to all head and neck cancers)
The recommendations are summarised in Table 1.Reference Grimmett, Bradbury, Dalton, Fecher-Jones, Hoedjes and Varkonyi-Sepp260–Reference Gillis, Li, Lee, Awasthi, Augustin and Gamsa262
Table 1. Prehabilitation: pre-treatment recommendations general to all head and neck cancers
* See chapter on nutritional management in the treatment of head and neck cancer for further details. †See chapter on psychological management for further details. ‡See chapter on speech, voice and swallowing rehabilitation for head and neck cancer for further details. OMFS = oral and maxillofacial surgery; PG-SGA SF = Patient-Generated Subjective Global Assessment Short Form; NRS = nutrition risk screening; MUST = Malnutrition Universal Screening Tool; SARC-F = Strength, Assistance with walking, Rising from a chair, Climbing stairs, and Falls questionnaire; BIA = bioelectrical impedance analysis; CT = computed tomography; DEXA = dual-energy X-ray absorptiometry; HbA1c = haemoglobin A1c; IAPT = improving access to psychological therapies; SLT = speech and language therapy; EORTC = European Organization for the Research and Treatment of Cancer; QLQ-30 = 30-item Quality of Life Questionnaire Core Module; QLQ-H&N35 = 35-item Quality of Life Questionnaire Head and Neck Module; PCI = Patient Concerns Inventory; SF-36 = 36-item Short Form Health Survey; QoL = quality of life
During treatment
Surgery-specific
The recommendations are summarised in Table 2.
Table 2. Prehabilitation: intra-treatment recommendations specific to surgery
Chemo/radiotherapy-specific
The recommendations are summarised in Table 3.Reference Capozzi, Nishimura, McNeely, Lau and Culos-Reed263,Reference Starmer, Yang, Gourin, Kumar, Jones and McNutt264
Table 3. Prehabilitation: intra-treatment recommendations specific to chemo/radiotherapy
Chemo/RT = chemo/radiotherapy; SLT = speech and language therapy
Post-treatment (general to all head and neck cancers)
The recommendations are summarised in Table 4.
Table 4. Prehabilitation: post-treatment recommendations general to all head and neck cancers
QoL = quality of life; SMART goals = specific, measurable, achievable, relevant and time-bound goals
Enhanced recovery
Enhanced recovery after surgery pathways are used in some surgical pathways with the aim of improving post-operative recovery by optimising pre-operative function and by reducing the post-operative stress response.Reference Melnyk, Casey, Black and Koupparis265 Whilst novel in its application to head and neck cancer surgical pathways, enhanced recovery after surgery has been implemented for decades in both general and colorectal surgery, with demonstrated long-term improvements in patient outcomes.Reference Muller, Zalunardo, Hubner, Clavien and Demartines266 Established enhanced recovery after surgery pathways in these specialties recommend principles that could be applied to head and neck cancer patients, including pre-operative counselling, nutrition optimisation, standardised analgesic and anaesthetic regimens, and early post-operative mobilisation for all surgical patients.
Despite this proven evidence for benefit in other cancers and surgical specialties, current research exploring enhanced recovery after surgery in head and neck cancer surgery remains limited in number and variable in study design.Reference Watson and Ewers267 The key concepts that have been explored to date focus on early oral feeding following major head and neck cancer surgery,Reference Seven, Calis and Turgut268–Reference Prasad, Sreedharan, Dannana, Prasad and Chandra271 early tracheostoma fistula closure, and aspects of service delivery including predictors of complications.Reference Brookes, Seikaly, Diamond, Mechor and Harris272–Reference Farwell, Reilly, Weymuller, Greenberg, Staiger and Futran276 Dort and colleagues’ (2017)Reference Dort, Farwell, Findlay, Huber, Kerr and Shea-Budgell273 systematic review of recommendations for peri-operative care in head and neck cancer surgery with free-flap reconstruction identified 17 topic areas, and reported their evidence levels. Ten of these topic areas are applicable to all head and neck cancer patients, and are supported by other research.Reference Worrall, Tanella, DeMaria and Miles277 It is for this reason that they are included in this guideline.
Whilst there are acknowledged limitations in the current literature, findings from the most recent studies are encouraging;Reference Dort, Farwell, Findlay, Huber, Kerr and Shea-Budgell273,Reference Worrall, Tanella, DeMaria and Miles277 and it must be recognised that the concept of enhanced recovery after surgery has potential to positively influence head and neck cancer patient outcomes and service provision.Reference Watson and Ewers267 For this to be achieved, audit cycles of changes in clinical practice should be conducted in all head and neck cancer services adopting enhanced recovery after surgery protocols in the UK.
As we know, there are a wide range of head and neck cancer surgical procedures offered to patients in the UK, ranging from ‘minor’ to ‘major’ surgery. Some enhanced recovery after surgery principles may vary depending on the extent of surgery (e.g. timing of re-commencing oral intake); however, the general principles of enhanced recovery after surgery should be applicable to all head and neck cancer surgery patients, irrespective of their surgery.
Therefore, the following recommendations are generalised to all head and neck cancer patients undergoing surgical procedures, pre-treatment, during treatment and post-treatment care. These guidelines have been developed based on critical literature appraisal, expert consensus and PPI involvement. It is recommended that they be used as a model for clinical services and applied to patients on an individualised basis. We recommend referring to the speech and language therapy chapter for more specific recommendations on laryngectomy, if required.
Pre-treatment
The recommendations are summarised in Table 5.
Table 5. Pre-treatment recommendations for enhanced recovery
* See chapter on nutritional management of head and neck cancer for further information. †An anaesthetist who specialises in the management of difficult airways. SLT = speech and language therapy; pre-op = pre-operative; COPD = chronic obstructive pulmonary disease; ERAS+ = enhanced recovery after surgery plus; intra-op = intra-operative; post-op = post-operative
During treatment
The recommendations are summarised in Table 6.
Table 6. Intra-treatment recommendations for enhanced recovery
* Indicates the use of more than one modality to control pain. Intra-op = intra-operative; VTE = venous thromboembolism; peri-op = peri-operative; H&N = head and neck; post-op = post-operative
Post-treatment
The recommendations are summarised in Table 7.
Table 7. Post-treatment recommendations for enhanced recovery
post-op = post-operative; SLT = speech and language therapy; UTI = urinary tract infection
Research
The evidence base for both prehabilitation and enhanced recovery after surgery in head and neck cancer is emerging, with promising outcomes. In addition, trials that aim to investigate the effectiveness of prehabilitation within the head and neck cancer population are underway.278
However, there is a continuing need for high-quality studies with replicable methods and statistical analysis. This will support the implementation and evaluation of evidence-based standardised prehabilitation programmes and peri-operative enhanced recovery after surgery pathways for head and neck cancer patients. This is reinforced by a number of National Cancer Research Institute (‘NCRI’) key research priorities.279
In order to evaluate the effectiveness of developed programmes or pathways, there are key research questions and areas that should be addressed in future studies:
• Patient-reported outcomes or experiences are not currently routinely represented in enhanced recovery after surgery literature. These should be explored in both prehabilitation and enhanced recovery after surgery services, and outcomes should be incorporated into any future pathway development to measure outcomes, ensuring the patient report and experience are at the core
• Clinician-reported experiences are not currently represented in the literature. Future work should explore this, as a deeper understanding of clinician perspectives could benefit implementation through increasing compliance with enhanced recovery after surgery pathways
• There should be understanding of cost-effectiveness and workforce provision of prehabilitation and enhanced recovery after surgery interventions, to support services in providing prehabilitation and enhanced recovery after surgery
• Facilitator and barriers to the integration of prehabilitation with enhanced recovery after surgery programmes should be investigated
• Prehabilitation that includes patients undergoing primary (chemo)radiotherapy treatment should be explored
• Comprehensive multimodal prehabilitation trials should be conducted, which include a combination of nutrition, exercise, behaviour and/or swallowing interventions
• The impact of delaying treatment to enable prehabilitation should be assessed
• The impact of including peer, caregiver and family support within prehabilitation and enhanced recovery after surgery programmes should be evaluated
Future research into both prehabilitation and enhanced recovery after surgery must demonstrate quality improvement, efficiency and embed patient-reported outcomes. This will enable prehabilitation and enhanced recovery after surgery models to become the standard of care, and allow thorough evaluation of the services offered to head and neck cancer patients in the UK. This also supports a number of National Cancer Research Institute key research priority questions, including: (1) How can the short-term, long-term and late effects of cancer treatments be (a) prevented, and/or (b) best treated and managed?; and (2) What specific lifestyle changes (e.g. diet, exercise and stress reduction) help with recovery from treatment, restore health and improve QoL?
Future aspects of prehabilitation and enhanced recovery after surgery
Prehabilitation and enhanced recovery after surgery programmes have the potential to benefit long-term patient outcomes and experience, as well as cost-effectiveness of clinical services. Participants of the Greater Manchester Prehab4Cancer and recovery programmeReference Moore, Scoggins, Philips, Egger, Tennant and Little259 provided clear positive feedback around being able to ‘take back control’ of their cancer journey. Many patients anecdotally felt stronger than the cohort of patients with similar conditions who did not take part. One patient said, ‘I was out of bed faster than everyone on my ward’. They were unanimous about the positive mental benefits of the exercise programme. Engagement with personal trainers was critically important, as well as feeling part of a group within an environment in which they could participate and share experiences with other patients. The group universally would recommend an exercise dose as part of their cancer journey, and felt this should be offered to all relevant cancer patients. This should be considered in future aspects of both prehabilitation and enhanced recovery after surgery services.
Head and neck cancer services in the UK will need to be expanded to achieve optimum prehabilitation and enhanced recovery after surgery services. These services should have interdisciplinary working at their core, so that the optimum patient outcome is achieved. This recommendation is echoed by feedback (from a head and neck cancer PPI group), with one person saying: ‘I think setting bespoke individual goals is an excellent idea if constructed in an achievable way. I like goals, things to aim for but emphasis must be on the positive’.
Therefore, both prehabilitation and enhanced recovery after surgery services should have access to:
• Community well-being and leisure services that support engagement in exercise
• Patient and public involvement groups
• Digital healthcare technology
• Social prescribing and other primary care personalised care initiatives
• Third sector provision such as Macmillan and Maggie's charities, for information and support
Key points
• Specialist dietitians should be part of head and neck cancer multidisciplinary teams (MDTs) throughout patients’ continuum of care.
• Clear pathways between primary, secondary and tertiary care across organisational boundaries should be maintained, reviewed and monitored to ensure seamless delivery of dietetic support.
• Patients with head and neck cancer are at a high risk of malnutrition because of the impact of disease and subsequent treatment.
• At diagnosis, all patients should be screened for malnutrition using validated tools, e.g. Malnutrition Universal Screening Tool (‘MUST’). Any patient at risk of malnutrition, or likely to become malnourished as a result of treatment should be referred to a dietitian for early intervention, and assessed for malnutrition using validated tools, e.g. Patient-Generated Subjective Global Assessment (‘PG-SGA’). Malnutrition screening and assessment should be repeated at various intervals throughout patients’ continuum of care.
• All patients should receive a pre-treatment appointment prior to any treatment that is expected to impact on the ability to maintain nutritional status. This should include counselling on enteral tube feeding options, where appropriate.
• Nutritional requirements can be estimated using evidence-based equations; however, patients’ anthropometry and tolerance of nutrition support should be monitored to ensure adequacy of intake.
• Prophylactic gastrostomy placement should be considered on an individualised basis, where the MDT should take account of the following: performance status and social factors, baseline nutritional status, tumour stage, tumour site, pre-existing dysphagia, and impact of planned treatment.
• Patients having nasogastric tubes in the community should all have a nasogastric tube risk assessment completed.
• All patients undergoing a surgical intervention should be provided with carbohydrate loading and have tube feeding initiated within 24 hours of surgery (where oral intake is contraindicated) as part of enhanced recovery after surgery protocols.
• All patients at severe nutritional risk, who are not meeting nutritional requirements, should receive nutrition support for 10–14 days prior to surgery. Delaying surgery to achieve this may be necessary but has to be weighed against the risk of delaying treatment.
• All patients undergoing radiotherapy should receive a dietetic review at least once weekly during treatment, fortnightly for six weeks after treatment, and as appropriate thereafter for up to a minimum of three to six months.
• Dietitians can become advanced clinical practitioners and extend their scope of practice within their MDTs. This includes becoming non-medical supplementary prescribers with appropriate supervision in place from MDT members.
• All patients having palliative treatment should have access to a dietitian. The advantages and disadvantages of nutrition support and/or artificial feeding should be discussed with the MDT based on a goal of maintaining patients’ quality of life.
• Patients who have completed rehabilitation and are no longer at risk of malnutrition should be offered cancer prevention and healthy eating advice. Where appropriate, this should include encouraging physical activity as per national recommendations.
• Dietitians should offer telehealth appointments where appropriate to support service flexibility.
• Dietetic provision in proton beam therapy services should mirror existing intensity-modulated radiation therapy services.
Head and neck cancer dietetic service provision and delivery
Nutritional intervention and support is integral in the management of head and neck cancer patients. It is estimated that 35–75 per cent of patients present malnourished at diagnosis.Reference Ravasco, Monteiro-Grillo, Marques Vidal and Camilo280,Reference Hébuterne, Lemarié, Michallet, de Montreuil, Schneider and Goldwasser281 In addition, the obesity epidemic has led to more patients presenting as overweight or obese at diagnosis, and recent research suggests that these individuals have increased rates of malnutrition and sarcopenia during treatment.Reference Brown, Ross, Jones, Hughes and Banks282
Dietetic services within head and neck cancer units should be seamless, and ensure access at each stage of the patient pathway from diagnosis to survivorship and/or palliation. As part of the National Health Service (NHS) ‘Long Term Plan’, ‘Cancer Plan’ and ‘Cancer Reform Strategy’, head and neck cancer services in the UK have been restructured to facilitate the centralisation of services, MDT unification and the streamlining of surgical treatments at tertiary centres.283 Dietitians are key stakeholders in ensuring that the process of centralisation allows access to adequate nutritional screening, support and intervention, to optimise patients’ experience and outcomes (Table 1). Furthermore, it is integral that pathways are maintained and reviewed, to ensure continuity of care between primary, secondary and tertiary facilities across organisational boundaries in order to prevent fragmented service delivery.Reference Melnychuk, Vindrola-Padros, Aitchison, Clarke, Fulop and Levermore284
Table 1. Recommendations for dietetic provision in head and neck cancer services
HNC = head and neck cancer; MDT = multidisciplinary team
The British Association of Head and Neck Oncologists recommend that all head and neck cancer units have a specialist dietitian with at least 50 per cent of their time dedicated to head and neck cancer.Reference Schache, Kerawala, Ahmed, Brennan, Cook and Garrett71 Early, frequent dietetic counselling is associated with improvements in nutrition, patient-centred outcomes, quality of life, compliance and tolerance to treatment (Table 1).Reference Ho, Yeh, Hsueh, Hung, Lu and Tsang285
Malnutrition and body composition screening
Emerging research highlights the importance of body composition on identifying malnutrition, sarcopenia and/or cachexia. These can affect treatment outcomes, hence should be taken into consideration by the MDT when planning treatment.Reference Findlay, Brown, De Abreu Lourenço, White and Bauer286
Malnutrition within cancer is the negative energy balance and skeletal muscle loss driven by a combination of reduced food intake and metabolic derangements. It is adversely associated with survival, morbidity, mortality, quality of life and treatment response, impacting on clinical outcomes, cost and patient experience.Reference Muscaritoli, Arends, Bachmann, Baracos, Barthelemy and Bertz287 Table 2 depicts how malnutrition can be generally identified, and the consequences, but it is important that screening and assessment for this is undertaken. Patients with head and neck cancer are at risk of malnutrition because of the site of their cancer, the impact of the disease process and treatment, and lifestyle factors.288,Reference Müller-Richter, Betz, Hartmann and Brands289 More recently, it has been reported that patients who are overweight or obese at baseline are at higher risk of becoming malnourished during and after head and neck cancer treatments.Reference Brown, Ross, Jones, Hughes and Banks282
Table 2. Identification and consequences of malnutrition
BMI = body mass index
Unintentional weight loss (independent of presenting body mass index) of 10 per cent or more in the six months preceding diagnosis288,Reference Gourin, Couch and Johnson290 can lead to the adverse effects detailed in Table 2.Reference Müller-Richter, Betz, Hartmann and Brands289–Reference Cederholm, Jensen, Correia, Gonzalez, Fukushima and Higashiguchi291
Cancer cachexia is difficult to diagnose as it is multifactorial in nature. It is characterised by the loss of weight, skeletal muscle and adipose tissue, alongside an imbalance of metabolic regulation and reduced food intake. It negatively affects treatment outcomes and quality of life. Management is difficult in light of limited agreed treatment options. Evidence to support pharmacological treatments remains inconclusive. Improvement has been reported with fish oil supplements; however, palatability limits compliance.Reference Ni and Zhang292
Sarcopenia can be defined by the combination of low grip strength plus low muscle mass.Reference Cruz-Jentoft, Baeyens, Bauer, Boirie, Cederholm and Landi293 Sarcopenia and myosteatosis (infiltration of muscle with fat) can be significant negative predictors of overall survival, and muscle status evaluation should be considered during treatment planning.Reference Findlay, Brown, De Abreu Lourenço, White and Bauer286 Methods for identifying sarcopenia include the Strength, Assistance with walking, Rising from a chair, Climbing stairs and Falls (‘SARC-F’) questionnaire, computed tomography (CT) defined assessment, anthropometry, and bioelectrical impedance. Treatment for sarcopenia requires a combined approach, including physical activity and nutrition support, which should be offered as part of a prehabilitation programme (see ‘Patient preparation for treatment and enhanced recovery’ chapter for further information).
Screening and assessment
Malnutrition screening tools are used to identify patients promptly who are at risk of malnutrition and require further dietetic intervention, and should be used throughout the patient's pathway. Subsequently, malnutrition assessment tools can be used to diagnose malnutrition.Reference Muscaritoli, Arends, Bachmann, Baracos, Barthelemy and Bertz287,Reference Mendes, Barros, Rosa and Franceschini294 Some examples are illustrated in Table 3.Reference Muscaritoli, Arends, Bachmann, Baracos, Barthelemy and Bertz287 Several of these tools have been validated, but this may differ for various parameters including age and clinical settings. The most highly recommended assessment tool is the Patient-Generated Subjective Global Assessment. Screening should be agreed locally using a validated screening tool for the patient population. Following this, any patient identified at risk of malnutrition should be referred to a dietitian for full nutritional assessment (detailed in Table 4) using a validated nutritional assessment tool.
Table 3. Malnutrition screening and assessment toolsReference Muscaritoli, Arends, Bachmann, Baracos, Barthelemy and Bertz287
MNA-SF = Mini Nutritional Assessment Short-Form; PG-SGA = Patient-Generated Subjective Global Assessment; HNC = head and neck cancer; NIS = nutritional impact symptoms
Table 4. Nutritional screening and assessment recommendations
HNC = head and neck cancer; MUST = Malnutrition Universal Screening Tool; PG-SGA = Patient-Generated Subjective Global Assessment
Patients who are malnourished, or at risk of malnutrition, may be identified by various members of the MDT, and should be referred to the dietitian for assessment. Dietitians should attend MDT meetings and MDT head and neck cancer clinics to support rapid referral (Table 5).
Table 5. MDT and combined head and neck clinic recommendations
MDT = multidisciplinary team; HNC = head and neck cancer; SLT = speech and language therapist
Nutrition support
Interventions can have a profound impact on the ability to eat and drink or take adequate nutrition orally, related to loss of function and treatment side effects such as nausea, vomiting, xerostomia, pain, mucositis and dysphagia.
Pre-treatment
All patients scheduled to undergo treatment that is likely to impact on nutritional status should be offered a dietetic pre-treatment appointment in order to assess baseline nutritional status, and provide dietary intervention as appropriate to maintain nutritional status and prevent decline (Table 6).Reference Findlay, Rankin, Shaw, White, Boyer and Milross295,Reference Talwar and Findlay296 The British Dietetic Association care pathway (Table 7) should be used to collect relevant baseline information alongside a malnutrition assessment tool.297
Table 6. Pre-treatment assessment recommendations
HNC = head and neck cancer; SLT = speech and language therapist; MDT = multidisciplinary team; NG = nasogastric
Table 7. Nutritional assessment and dietetic care pathway recommendations297
*Denotes ideal body weight. †Denotes body mass index (BMI) of 18.5–30 kg/m2. ‡Denotes BMI of 18.5–30 kg/m2. Adjust for obesity (use 75 per cent value for BMI over 30 kg/m2; 65% value for BMI over 50 kg/m2; start at upper end of range for BMI less than 18.5 kg/m2 and monitor regularly). BMI = body mass index; HNC = head and neck cancer; MDT = multidisciplinary team; SLT = speech and language therapy; ESPEN = European Society for Parenteral and Enteral Nutrition; BDA = British Dietetic Association
The appointment should also provide counselling to assist informed decision-making and manage expectations on the impact of upcoming treatment on nutritional status. This includes possible changes to dietary textures, the requirement for enteral tube feeding in the short and long term, and side effects of treatments affecting the ability to take nutrition.Reference Parr, Johnson, Langley and Richardson298
Prehabilitation
Prehabilitation promotes healthy behaviours such as exercise, optimal nutrition and psychological support, to maximise resilience to treatment and improve long-term health. All patients should be counselled on the importance of prehabilitation, and services should be developed to facilitate access for all head and neck cancer patients.Reference Loewen, Jeffery, Rieger and Constantinescu299 Further information can be found in the ‘Patient preparation for treatment and enhanced recovery’ chapter.
Estimation of nutritional requirements
Total energy expenditure and protein requirements can be estimated using a variety of calculations, detailed in Table 7.Reference Muscaritoli, Arends, Bachmann, Baracos, Barthelemy and Bertz287,297,Reference McCurdy, Nejatinamini, Debenham, Álvarez-Camacho, Kubrak and Wismer300–Reference Todorovic and Mafrici302
Requirements will vary depending on many factors including: infection; biochemical and physical parameters indicating metabolic stress and reactions to treatment; body composition; activity levels; clinical condition; and type of treatment.
Equations may be less accurate, and may be inadequate, for patients who are severely malnourished or morbidly obese. Therefore, it is essential to monitor anthropometric changes to ensure adequacy.Reference McCurdy, Nejatinamini, Debenham, Álvarez-Camacho, Kubrak and Wismer300
Previous guidance recommended 25–35 kcal/kg, aiming for at least 30 kcal/kg. Updated European Society for Parenteral and Enteral Nutrition (‘ESPEN’) guidance recommends 25–30 kcal/kg.Reference Muscaritoli, Arends, Bachmann, Baracos, Barthelemy and Bertz287 It is important to note that this recommendation is for all cancer diagnoses, and may not be accurate for head and neck cancer patients, where equations often underestimate requirements.Reference McCurdy, Nejatinamini, Debenham, Álvarez-Camacho, Kubrak and Wismer300 Thus, it is recommended that estimated requirements for head and neck cancer patients currently undergoing treatment should start at 30 kcal/kg, and increase as required, as radiotherapy with or without chemotherapy and surgery can increase energy requirements significantly. Alternatively, the British Dietetic Association Parenteral and Enteral Nutrition Group head and neck cancer guidelines can be used, with a physical activity level of 1.4, as this gives similar results to previous guidelines.Reference Todorovic and Mafrici302
Vitamins and minerals should be provided as per daily recommendations, unless considered deficient. Particular attention should be given to patients on long-term home enteral feeding and ensuring nutritional completeness of enteral feed.
Refeeding syndrome
Refeeding syndrome consists of metabolic disturbances that occur as a result of the reintroduction of nutrition (namely carbohydrates) to patients who are severely malnourished. Head and neck cancer patients may be at a high risk of refeeding syndrome, and it can occur irrespective of nutrition route.Reference da Silva, Seres, Sabino, Adams, Berdahl and Citty303
Refeeding syndrome is characterised by hypophosphataemia, hypokalaemia and hypomagnesaemia, as well as abnormal sodium and fluid balance, changes in glucose, protein and fat metabolism, and thiamine deficiency.Reference Todorovic and Mafrici302,Reference da Silva, Seres, Sabino, Adams, Berdahl and Citty303
Management for refeeding syndrome includes the gradual reintroduction of nutrition, supplementation of thiamine/B vitamins, and monitoring, with or without replacement of electrolytes. Latest guidance from the American Society of Parenteral and Enteral Nutrition (‘ASPEN’) have been included (Tables 8 and 9), but clinicians should also be guided by locally agreed policies for identification, avoidance and management. Identification and management will differ for paediatric patients; guidance is available from the American Society of Parenteral and Enteral Nutrition.Reference da Silva, Seres, Sabino, Adams, Berdahl and Citty303
Table 8. Refeeding syndrome identification for adults*
* American Society of Parenteral and Enteral Nutrition (‘ASPEN’) consensus criteria for identifying adult patients at risk for refeeding syndrome, adapted and reproduced with permission.Reference da Silva, Seres, Sabino, Adams, Berdahl and Citty303 †Please note electrolytes may be normal despite total body deficiency, which is believed to increase the risk of refeeding syndrome. ‡These include: acquired immunodeficiency syndrome; chronic alcohol or drug use disorder; dysphagia and oesophageal dysmotility (e.g. eosinophilic oesophagitis, achalasia, gastric dysmotility); eating disorders (e.g. anorexia nervosa); food insecurity and homelessness; failure to thrive, including physical and sexual abuse and victims of neglect (particularly children); hyperemesis gravidarum or protracted vomiting; major stressors or surgery without nutrition for prolonged periods of time; malabsorptive states (e.g. short-bowel syndrome, Crohn's disease, cystic fibrosis, pyloric stenosis, maldigestion, pancreatic insufficiency); cancer; advanced neurological impairment or general inability to communicate needs; post-bariatric surgery; post-operative patients with complications; prolonged fasting (e.g. individuals on hunger strikes, anorexia nervosa); refugees; protein malnourishment. BMI = body mass index
Table 9. Refeeding syndrome management for adults*
*American Society of Parenteral and Enteral Nutrition (‘ASPEN’) consensus recommendations for avoidance and treatment of refeeding syndrome in at-risk adults, adapted and reproduced with permission.Reference da Silva, Seres, Sabino, Adams, Berdahl and Citty303 †Management will depend on local refeeding syndrome guidelines and policies; thus, these recommendations should be used as a general guide only. IV = intravenous
Nutrition support
Nutrition support should be considered where malnutrition is diagnosed following screening and assessment, to optimise nutritional status. There are three main methods of nutrition support: oral, enteral and parenteral. Parenteral nutrition is rarely used in head and neck cancer, but should be considered if clinically indicated e.g. chyle leak, post-operative ileus, no enteral access route.
Oral nutrition support
Oral nutritional support aims to address deficiencies and minimise further nutritional compromise. Nourishing dietary advice including food-first fortification should be recommended, but may not meet the nutritional deficit alone because of the impact of disease and treatment. In order to meet these deficits, more intensive support such as nutritionally complete oral nutritional supplements can also be prescribed. There are a wide variety of products available, and choice will depend on patient preference, compliance and local policy.
Enteral nutrition
Enteral feeding is often required to support patients in meeting their nutritional requirements, to prevent weight loss, and to maintain good nutritional status on a short- and long-term basis. Enteral feeding tubes include the nasogastric tube, nasojejunal tube, orogastric tube, oesophageal-fistulae tube, gastrostomy tube (with or without jejunal extension) and jejunostomy tube. The type of tube placed should account for tumour type and size, treatment plan, anticipated length of enteral feeding, and patient choice, with patients being included in the decision-making process.Reference Williams, White, Sen and Prestwich304
The most common approaches for head and neck cancer patients with tube feeding in the community are prophylactic gastrostomies and reactive nasogastric tubes. The impact of tube type and timing in head and neck cancer is controversial, leading to variation in practice. Controversy exists because of a lack of consensus on whether a prophylactic or reactive approach leads to improved patient outcomes.
Prophylactic approaches mainly consist of gastrostomy placement prior to head and neck cancer treatment where it is likely that enteral feeding will be required for a long-term period. ‘Long-term’ is not always well defined; within the UK, the National Institute for Health and Care Excellence (NICE) suggests gastrostomy tubes should be considered when enteral feeding is required for more than four weeks.288 The advantages of prophylactic gastrostomy include improved nutritional outcomes and quality of life, in addition to reduced incidences of malnourished patients, hospital admissions or treatment interruptions when compared with reactive feeding approaches.Reference Brown, Banks, Hughes, Lin, Kenny and Bauer305 The disadvantages include the risk that some tubes are not used,Reference Madhoun306 but patient compliance and appropriateness of decision-making for tube placement in such studies are unclear. Furthermore, it has been argued that prophylactic gastrostomies lead to poorer swallowing outcomes related to prolonged tube use. These conclusions are limited because of a lack of high-quality studies including a nutritional outcome analysis. Recent studies have not shown a relationship between long-term swallowing dysfunction and prophylactic gastrostomy.Reference Axelsson, Silander, Nyman, Bove, Johansson and Hammerlid307
Gastrostomy placement is generally considered a safe procedure, but can result in complications; however, major complications are rare. Placement can be endoscopic, radiological or surgical, and no nationally agreed selection criterion on placement method currently exists. All patients should be screened and assessed for suitability and optimal method, accounting for local resources and policy. Endoscopic placement is generally associated with fewer complications compared to radiological gastrostomy; however, the former may be contraindicated in head and neck cancer (e.g. tumour location, trismus), and gastrotomy tubes are often retained by an internal bumper which requires endoscopic removal.Reference Strijbos, Keszthelyi, Bogie, Gilissen, Lacko and Hoeijmakers308 Radiologically placed tubes are usually retained with a balloon filled with water, which can be deflated and removed at bedside. This is advantageous for enabling prompt removal, but corresponds with an increased risk of unintentional tube dislodgement.
Reactive approaches involve the placement of a gastrostomy or nasogastric tube during or after treatment. Advantages of this approach include avoidance of: the risk of placing a tube that may not be used and gastrostomy-related complications. Nasogastric tubes have the advantage of being quick to insert and remove, and thus may remain in place for less time than a gastrostomy tube. Dietitians can be trained to become competent with tube insertions, to streamline services. Disadvantages of this approach include tube visibility, dislodgement, discomfort, fear or worry of tube displacement, pharyngeal irritation, and the public reaction, which has been reported negatively by service users.Reference Williams, White, Sen and Prestwich304 In addition, complications associated with tube management in the community are not well documented (e.g. tube displacement requiring emergency department attendance, with or without tube replacement and chest X-ray).
In the UK, the discharge of patients with a nasogastric tube from the acute to community setting is not always permitted by some NHS trusts because of the potential risk of feeding into the lungs causing a ‘never event’. This has resulted in many district nursing services refusing to support patients with nasogastric tubes in the community, but continuing to provide support for gastrostomy tubes. Therefore, patients who require a nasogastric tube at home are often expected to self-manage or have family support. This can complicate discharge planning, thus early identification is crucial. The NHS requires all patients discharged with a nasogastric tube to have an MDT-supported nasogastric tube risk assessment completed, and a care plan in place to assess suitability and mitigate any risks.309 Success with prolonged nasogastric tube use (more than 28 days) in the community has been reported,Reference Espeli, Vergotte, Dietrich, Pichard and Siano310 with some units developing out-patient nasogastric tube services, which is advantageous in reducing in-patient admissions.Reference McCloskey, Duffy, Brown, Irwin and Faloon311 These factors highlight the importance of dietetic pre-treatment appointments to aid informed decision-making on feeding options, accounting for individualised needs and preference at the point of diagnosis.Reference Hazzard, Gulliver, Walton, McMahon, Milosavljevic and Tapsell312
In summary, the evidence base remains inconclusive, with insufficient evidence to suggest an optimal feeding route.Reference Hazzard, Gulliver, Walton, McMahon, Milosavljevic and Tapsell312–Reference McClelland, Andrews, Chaudhry, Teckie and Goenka314 Guidelines have been developed based on predictors of prolonged tube use to aid clinicians with appropriate decision-making for placing prophylactic gastrostomy tubes. These include overall clinical stage, tumour site, clinical tumour (T) and nodal (N) stage, and patient age.Reference Brown, Spurgin, Ross, Tripcony, Keller and Hughes315,Reference Brown, Getliffe, Banks, Hughes, Lin and Kenny316 The NICE guidelines suggest individualised multidisciplinary decision-making at diagnosis, accounting for predictors as detailed in Table 10.317
Table 10. Gastrostomy selection criteria in head and neck cancer*
* Adapted from National Institute for Health and Care Excellence with permission.317 MDT = multidisciplinary team; BMI = body mass index
There are a wide variety of enteral nutrition products available, and the volume and type will depend upon the patient's requirements, tolerance and local contractual agreements.288
Monitoring nutrition support
The monitoring of enteral feeding regimens is essential given issues of: potential inaccuracies of nutritional requirement equations, compliance, and tolerance. Dietitians should be informed by any member of the MDT if signs of intolerance are reported, to ensure an optimal regimen is in place, and adjustments made accordingly.
On treatment
Nutritional considerations during surgery
Pre-operative nutrition
Inadequate oral intake for more than 14 days is associated with a higher mortality. These patients should receive nutrition support prior to major surgery. Delaying surgery to achieve this may be necessary but has to be weighed against the risk of delaying treatment.Reference Stableforth, Thomas and Lewis318
Enhanced recovery after surgery
Surgery is associated with reduced muscle function, prolonged fatigue, poor wound healing, increased morbidity and a longer length of hospital stay.Reference Bilku, Dennison, Hall, Metcalfe and Garcea319 Enhanced recovery after surgery protocols help ameliorate these effects; for example, increased carbohydrate utilisation leads to an anabolic state promoting enhanced recovery, which is well established in many centres. Nutritional interventions include pre-operative carbohydrate loading of 100 g carbohydrate the night before surgery and 50 g carbohydrate 2 hours before surgery. In order to avoid harm, carbohydrate loading should not be given for patients with uncontrolled diabetes or gastroparesis, and is unlikely to be beneficial for those with type 1 diabetes.Reference Weimann, Braga, Carli, Higashiguchi, Hübner and Klek320 Further information can be found in the ‘Patient preparation for treatment and enhanced recovery’ chapter.
Post-operative nutrition
Early post-operative enteral tube feeding (within 24 hours) should be commenced in patients if early oral nutrition cannot be initiated.Reference McClave, Taylor, Martindale, Warren, Johnson and Braunschweig321 Early oral feeding in the absence of contraindications after primary total laryngectomy and free-flap surgery has been reported to reduce length of stay, without increasing peri-operative complications. Further research is required to support the uptake of this approach, and studies vary in their definition of ‘early oral feeding’.Reference Aires, Dedivitis, Petrarolha, Bernardo, Cernea and Brandao270,Reference Kerawala, Riva and Paleri322
Chyle leaks
If a chyle leak is suspected, it can be confirmed by testing for triglycerides and chylomicrons. Nutritional interventions (Table 11) are often used alongside medical options, such as somatostatin analogues, pressure dressings and suction drainage. (Further information can be found in the ‘Complications of treatment’ chapter.)
Table 11. Nutritional management recommendations for surgery, including suspected and confirmed chyle leak cases
ERAS = enhanced recovery after surgery; HNC = head and neck cancer; H&N = head and neck; MDT = multidisciplinary team; SLT = speech and language therapists
Radiotherapy
Dietary counselling during radiotherapy for head and neck cancer patients with nutrition support is recommended to manage treatment-induced toxicities (e.g. mucositis, pain, increased secretions, odynophagia, dysphagia), to prevent weight loss (Table 12). Significant weight loss (more than 10 per cent) cannot be completely prevented by nutritional counselling and intervention alone,Reference Weimann, Braga, Harsanyi, Laviano, Ljungqvist and Soeters301 and can lead to re-planning of radiotherapy fields and interruptions in treatment. Patients should be reviewed by a dietitian at least once a week during radiotherapy, and some centres offer twice-weekly review.
Table 12. Recommendations for management during radiotherapy
SLT = speech and language therapist; NG = nasogastric; MDT = multidisciplinary team; HNC = head and neck cancer; PPI = proton pump inhibitor
Patients should be encouraged with optimising the nutrition and tolerance of oral diet for as long as safely possible during radiotherapy, in conjunction with speech and language therapy advice. Enteral feeding should be considered if treatment impacts on swallowing and the ability to meet full nutritional requirements orally for weight maintenance. The timing and type of enteral feeding is controversial (discussed in the ‘enteral feeding’ section). The choice of tube should be MDT led, and should account for local resources, patient preference and high-risk factors of prolonged tube use (oral plus bilateral chemoradiotherapy, midline oropharyngeal/nasopharyngeal/pharyngeal plus chemoradiotherapy, dysphagia and/or severe malnutrition at presentation).317 Patients should be counselled on an individualised basis, with the risks and benefits of both reactive and prophylactic approaches discussed before treatment.
Post-treatment care
All head and neck cancer patients should be reviewed in the first three months post radiotherapy or major surgery to prevent decline in nutritional status, and thereafter on an individual basis. The goal of dietetic input post treatment is to reduce reliance on enteral feeding and/or nutritional supplements if possible, whilst assisting a return to healthy eating, accounting for possible chronic changes, e.g. texture modification (Table 13). It should be noted that patients’ experiences after treatment will vary greatly, with some returning to pre-treatment function rapidly and others needing long-term input.
Table 13. Post-treatment care recommendations
SLT = speech and language therapist/therapy; QoL = quality of life; PINNT = Patients on Intravenous & Nasogastric Nutrition Therapy
Head and neck cancer patients can experience treatment-related side effects in the short and long term that may impact quality of life and functional status. These include xerostomia, dysphagia, trismus, dental problems, difficulty chewing, taste alterations and mucositis.Reference Brook323 Patients may need ongoing nutrition support to manage these symptoms, and weight loss has been shown to continue for up to a year post treatment.Reference Vlooswijk, van Rooij, Kruize, Schuring, Al-Mamgani and de Roos324
Late effects of treatment such as progressive dysphagia, pharyngoesophageal stenosis and osteoradionecrosis can lead to ongoing poor nutritional intake, and food avoidance may lead to vitamin and mineral deficiencies.Reference Crowder, Douglas, Yanina Pepino, Sarma and Arthur325,Reference Crowder, Najam, Sarma, Fiese and Arthur326 This may necessitate further dietetic input, and nutrition screening should continue in order to identify these patients if they have been previously discharged from dietetics.Reference Vlooswijk, van Rooij, Kruize, Schuring, Al-Mamgani and de Roos324
In the absence of disease recurrence or late effects, cancer survivors who are no longer at risk of malnutrition should be offered healthy eating and physical activity advice, to prevent long-term morbidity. This should be tailored on an individualised basis given the complexity of head and neck cancer.Reference Muscaritoli, Arends, Bachmann, Baracos, Barthelemy and Bertz287
Palliative care
Patients may be offered palliative chemotherapy, radiotherapy or immunotherapy to reduce symptoms caused by cancer. Patients should be nutritionally assessed early in this pathway, and the aim of nutritional interventions should be to maintain quality of life and reduce the symptoms, with early anticipation of any deterioration (Table 14). Dietitians should discuss the advantages and disadvantages of artificial feeding with patients and their family, as nutrition is likely to be affected by treatment and disease progression.
Table 14. Palliative care management recommendations
QoL = quality of life; MDT = multidisciplinary team
Palliative care patients can develop symptoms as a result of the tumour, and may also be suffering from the long-term side effects of previous cancer treatment.Reference Goldstein, Genden and Morrison327 Dietitians should work collaboratively with other MDT members to ensure that informed decisions are made regarding palliative and end-of-life care, including the appropriateness of commencing artificial feeding and/or gastrostomy tube placement within the context of risk, benefit and prognosis.Reference Sciubba328 Further information is discussed in the palliative care chapter.
Future aspects and important research questions
Immunonutrition
Immune-enhanced nutrition are feeds containing amino acids, nucleotides and lipids. The evidence base for outcome measures including length of stay, wound infection, mortality and feed tolerance is lacking.Reference Howes, Atkinson, Thomas and Lewis329 A systematic review of cancer patients receiving immunonutritionReference Yu, Zheng, Wang, Liu, Li and Yu330 has suggested a reduced risk of post-operative infectious complications, a decreased risk of anastomotic leakage and a reduced hospital stay. Current studies are low in quality, at a high risk of bias and vary in regard to supplemental regimens, and thus should be interpreted with caution. Further high-quality and larger studies are required to justify routine usage.Reference Lyra, Meira, Guedes and Bueno331
Immunotherapy
Immunotherapy is a fast-emerging treatment used for people with head and neck cancer. The nutritional management of complications and individualised assessment are required. Further research is needed in this area.
Proton beam therapy
Proton beam therapy has recently been implemented within NHS services on two UK sites (University College London Hospital and The Christie in Manchester) following the strategic outline case published by the Department of Health. Eligibility is via a national panel, and dietitians are essential members of the MDT for managing patients during their treatment pathway.
Emerging research suggests that patients undergoing proton beam therapy may have reduced toxicity related side effects affecting ability to take adequate nutrition, and are therefore likely to have better nutritional outcomes and reduced tube dependency.Reference Blanchard, Garden, Gunn, Rosenthal, Morrison and Hernandez332 The Torpedo trial is currently underway, and aims to compare proton beam therapy with intensity-modulated radiotherapy in regard to late treatment-related toxicities in patients with locally advanced oropharyngeal squamous cell carcinoma. Patient-reported toxicity, feeding tube dependence and severe weight loss at 12 months post treatment will all be measured.Reference Price, Hall, West and Thomson125
Telehealth
The delivery of dietetic services is changing, as indicated by the NHS ‘Long Term Plan’, enhancing the use of digital technologies through the provision of new service models supported by the Royal College of Physicians (2018).283,333,334 The unprecedented coronavirus disease 2019 pandemic has accelerated this, with rapid adoption of telehealth across the NHS.
Telehealth can assist in optimising healthcare resource utilisation and efficiency, whilst enabling improved flexibility, patient satisfaction and experience. However, limitations include access to equipment from service users, especially in older and low-income populations.Reference Aapro, Bossi, Dasari, Fallowfield, Gascón and Geller335
Encouraging results have been reported that support the use of virtual clinics for head and neck cancer patients, from prehabilitation through to post treatment and supportive care.Reference Pfeifer, Keeney, Bumpous, Schapmire, Studts and Myers336 Feasibility studies have investigated a home-based telehealth model for the delivery of speech and language therapist and dietetic reviews.
Extended scope of practice
There is opportunity for dietitians to become advanced clinical practitioners and extend their scope of practice within their MDTs (Table 15). Some options include post-graduate training for non-medical supplementary prescribing, assisting with enteral tube placement and care, and prehabilitation. The agreed scope of practice should be documented formally within job plans, with a clear framework, mentorship and supervision in place.
Table 15. Future considerations for dietitians working in head and neck cancer services
PBT = proton beam therapy; IMRT = intensity-modulated radiation therapy; SGA = Subjective Global Assessment; PG-SGA = Patient-Generated Subjective Global Assessment; MDT = multidisciplinary team
Non-medical supplementary prescribing may include reviewing symptoms of acute radiotherapy toxicity, alongside their impact on nutritional intake, and prescribing medication where appropriate. This aims to alleviate side effects within the agreed patient clinical management plan, whilst liaising closely with the medical team.334
Future research
Further research questions associated with the nutrition management of head and neck cancer include:
• The reliability and validity of dietetic assessment and monitoring tools that can easily be adapted to virtual consultations
• Standardised outcome measures that can be obtained virtually
• Patient satisfaction and experience with telehealth
• Use of imaging technology (e.g. positron emission tomography/CT) to measure body composition change
• Use of diet counselling skills (e.g. cognitive based therapy) to improve patient outcomes
• Investigation of diet-based management of taste and smell training to improve taste changes
• High-quality studies to investigate the routine use of immunonutrition in head and neck cancer
Key points
• Patients with speech, voice or swallowing dysfunction should be identified, assessed, and offered rehabilitation by a speech and language therapist with specialist skills.
• All multidisciplinary teams (MDTs) should have rehabilitation patient pathways covering all stages of the patient's journey, including multidisciplinary and pre-treatment clinics, with clear mechanisms for re-entry.
• Rehabilitation needs to begin early in the care pathway, involving comprehensive evaluation and tailored evidence-based interventions, from pre-treatment and including end-of-life care.
• Instrumental evaluation is integral to assessment and rehabilitation pathways given the complex relationships between function and altered anatomy and pathophysiology in head and neck cancer patients.
• Services should strive for robust collection of functional outcome measures at specified time points to incorporate clinician-rated and patient-reported tools.
• Healthcare professionals need to remain vigilant for signs of late treatment effects where the downward trajectory of speech, voice and swallowing function, and airway compromise pose complex challenges and involve careful MDT management.
• Although speech and language therapists hold a key role, MDT discussion is essential to attain optimal patient outcomes. This may involve collaboration with other healthcare professionals, as appropriate, to optimise care, e.g. oral hygiene and laryngectomy stoma care.
• The support of carers is an important part of the speech and language therapist's role when gathering and delivering information, developing a tailored rehabilitation package, and/or guiding patients in a decision-making process.
• Patient involvement in decision-making is crucial, with high importance being attributed to functional outcomes when weighing up treatment and management options. Speech and language therapists engage in eliciting patient priorities and values, and give realistic information about alterations to function and the impact on daily living and health.
• Speech and language therapists are involved in generating and contributing towards research and audit for improved patient outcomes and service delivery.
• Clear pathways for access and re-referral to speech and language therapists need to be in place, with excellent communication between central and locality-based services with a robust training and supervision programme for specialist knowledge and skill development and maintenance.
Introduction
Head and neck cancer treatment can have a major deleterious impact on communication and swallowing function. Poor functioning can result in low mood, distress, reduced quality of life, and difficulties returning to work and socialising.Reference Wilson, Carding and Patterson337 Furthermore, dysphagia can have serious medical consequences, such as malnutrition, dehydration and aspiration-associated pneumonia.Reference Dziewas, Beck, Clave, Hamdy, Heppner and Langmore338 A conservative estimate of dysphagia prevalence is between 50 and 60 per cent, with evidence of further deterioration over time.Reference Hutcheson, Lewin, Barringer, Lisec, Gunn and Moore339,Reference Hutcheson, Yuk, Holsinger, Gunn and Lewin340 The presence of dysphagia extends hospital stay and increases associated costs by 60 per cent.Reference Allen, Greene, Sabido, Stretton and Miles341 Emerging evidence suggests that early speech and language therapist and dietetic intervention may contribute to reduced health and patient costs for some head and neck cancer patients.Reference Martino, Ringash, Durkin, Greco, Huang and Xu342,Reference VanderMolen, VanRossum, Burkhead, Smeele, Rasch and Hilgers343
The following guidelines cover the provision of voice, speech and swallowing services, delivered by speech and language therapists with specialist competencies. They are divided into key timepoints across the patient journey, with a dedicated section for laryngectomy rehabilitation. Each section is separated into recommendations that are considered ‘essential’ and ‘desirable’. The delivery and measurement of outcomes associated with speech and language therapy is an essential component of delivering effective and high-quality services; recommendations for the selection and timing of measures are described in the final section.
In addition to these guidelines, the following resources provide further support to both speech and language therapists and patients with head and neck cancer:
• Macmillan head and neck cancer information and support, in: https://www.macmillan.org.uk/cancer-information-and-support/head-and-neck-cancer
• Maggie's Centres, in: https://www.maggies.org
• Cancer Research UK, in: https://www.cancerresearchuk.org
• Head & Neck Cancer UK, in: http://hancuk.org
• The Swallows Head & Neck Cancer Support Group, in: https://theswallows.org.uk/ (@swallowsgroup, X, formerly Twitter)
• Saving Faces UK, in: https://savingfaces.co.uk
• National Association of Laryngectomee Clubs, in: https://www.laryngectomy.org.uk
Pre-treatment
The majority of individuals undertaking head and neck cancer treatment are likely to experience some level of difficulty or alteration to their swallowing and/or communication during their cancer care journey, with as many as 60 per cent presenting with problems at diagnosis.Reference Russi, Corvò, Merlotti, Alterio, Franco and Pergolizzi344 The impact of oncological interventions on function and the likely trajectory of recovery can reasonably be predicted,Reference Patterson, McColl, Carding, Hildreth, Kelly and Wilson345,Reference Robbins, Levine, Wood, Roecker and Luschei346 which therefore offers an opportunity for early intervention. A pre-treatment consultation should thus include informational counselling about the upcoming cancer treatment, and about the impact on swallowing and communication, and, where indicated, could provide a window for prehabilitation that may include prophylactic swallowing exercises (Table 1).Reference Govender, Smith, Barratt, Gardner and Taylor347
Table 1. Recommendations for MDT meetings and combined head and neck clinic
MDT = multidisciplinary team; SLT = speech and language therapist; HNC = head and neck cancer
Early post-operative phase
Surgery results in loss of tissue, with changes to the anatomy and structural relationships. Post-operative voice, speech and swallowing impairments can change over time, with oedema and/or pain being more prevalent in the early post-operative period, and atrophy or scarring with a reduced range of motion being more common in the long term.Reference Patterson, Brady and Roe348 Surgery can also result in nerve damage, leading to motor and sensory deficits. The severity of voice, speech and swallowing impairment depends on factors such as tumour site, the volume resected and the nature of the reconstruction. The inevitable functional impairments coupled with aesthetic changes, in addition to activity limitations and participation restrictions, can have a profound psychological impact on the patient. Rehabilitation, including very early engagement and interaction, can be both healing and therapeutic for patients (Table 2).Reference Dawson, Pracy, Patterson and Paleri349
Table 2. Recommendations for early post-operative management
MDT = multidisciplinary team; SLT = speech and language therapist; IDDSI = International Dysphagia Diet Standardisation Initiative
During radiotherapy
Painful and uncomfortable mucositis side effects, xerostomia, dysgeusia, and increased secretions are frequently experienced by patients during radiation treatment. There is an increasing body of evidence investigating strategies introduced before or during radiation treatment to reduce the incidence and severity of dysphagia.Reference Perry, Lee, Cotton and Kennedy350,Reference Greco, Simic, Ringash, Tomlinson, Inamoto and Martino351 Patients are encouraged to continue to eat and drink throughout radiotherapy, where safe, avoiding periods of nil by mouth status.Reference Patterson, Brady and Roe348 On-treatment rehabilitation is generally provided in an MDT setting to ensure that analgesia and mouth care are optimised, to allow ongoing oral intake and exercises as indicated (Table 3).Reference Starmer, Ayoub, Byward, Kizner, Le and Hara352–Reference Occomore and Knight354
Table 3. Recommendations for management during radiotherapy
SLT = speech and language therapy/therapist; MDT = multidisciplinary team
Rehabilitation following head and neck cancer treatment
Rehabilitation of swallowing should aim to optimise oral intake, reduce reliance on enteral feeding and supplements, and facilitate psychosocial adjustment. Studies support the use of exercises to improve swallowing function.Reference Malandraki and Hutcheson355–Reference Van Daele, Langmore, Krisciunas, Lazarus, Pauloski and McCulloch358 Postures such a chin tuck and head rotation can be effective in controlling bolus flow, to reduce or eliminate aspiration.Reference Pauloski359 Manoeuvres including the super supraglottic swallowReference Logemann, Pauloski, Rademaker and Colangelo360 and effortful swallowReference Clark and Shelton361 can also be utilised to control specific aspects of the oropharyngeal swallow, improving function. The use of expiratory muscle strength training may improve airway protection in those with chronic radiation-induced aspiration.Reference Hutcheson, Hammer, Rosen, Jones and McCulloch362 Combining cognitive behavioural therapy with swallowing therapy has highlighted the importance of addressing the emotional, behavioural and cognitive components of dysphagia alongside the physical impairment.Reference Patterson, McColl, Carding and Wilson363
Voice can be impaired by both radiotherapy and chemotherapy,Reference Radhakrishna, Yamini, Kadam, Shivashankar, Vishwanathan and Javarappa364,Reference Lazarus365 and by laser surgery.Reference Aaltonen, Rautiainen, Sellman, Saarilahti, Mäkitie and Rihkanen366 Phonatory changes as a result of head and neck cancer treatment may include reduced vocal intensity, impaired pitch, compromised breath support, as well as roughness, breathiness, hoarseness and vocal fatigue. Specific therapy techniquesReference Verdolini-Marston, Burke, Lessac, Glaze and Caldwell367,Reference Stemple, Lee, D'Amico and Pickup368 utilised with non-cancer patients may be considered to enable the patient to achieve the optimum voice quality.
Speech may be impaired as a result of surgical resection, or chemoradiotherapy side effects such as xerostomia and trismus. Intelligibility will largely be dependent on the site of the lesion, the extent of the resection and the flexibility of tissue. In some instances, referral for dental or prosthetic enhancement may optimise communication and/or swallowing function (Table 4).
Table 4. Recommendations for rehabilitation
Total laryngectomy
Laryngectomy surgery results in permanent anatomical and physiological changes that affect communication,Reference Sharpe, Camoes Costa, Doubé, Sita, McCarthy and Carding369 swallowing,Reference Coffey, Tolley, Howard, Drinnan and Hickson370,Reference Lee, Govender, Roy, Vaz and Hilari371 breathing,Reference Macri, Bogaardt, Parrilla, Minni, D'Alatri and de Vincentiis372 olfactionReference Longobardi, Parrilla, Di Cintio, De Corso, Marenda and Mari373 and appearance.Reference van Sluis, Kornman, van der Molen, van den Brekel and Yaron374 Intervention from speech and language therapy is essential along each stage of the pathway, from diagnosis to long-term management, to support individuals in reaching and adjusting to their new optimum functional potential (Table 5).
Table 5. Recommendations for total laryngectomy
SLT = speech and language therapy; MDT = multidisciplinary team; HME = heat and moisture exchanger; FEES = fibre-optic endoscopic evaluation of swallowing; SVR = surgical voice restoration
Late treatment effects
Progressive functional deterioration following treatment for head and neck cancer needs careful MDT assessment and management. Where disease recurrence is excluded, these symptoms are likely to represent ‘late-stage’ treatment effects. The gradual and insidious development over a period of time can make early identification of patients challenging,Reference Patterson375 but healthcare professionals need to remain vigilant. Education of the wider MDT and community teams is key, so they are alert to ‘red flag’ symptoms like chest infections, dehydration, weight loss and airway difficulties.
Patients may report increased swallowing difficulty, weight loss, recurrent chest symptoms, as well as speech, voice and breathing changes.Reference Dong, Ridge, Li, Lango, Churilla and Bauman376,Reference Davies-Husband, Murphy, Kelly, Drinnan and Paleri377 The sequelae can be devastating on both physical and emotional well-being,Reference Hutcheson, Lewin, Barringer, Lisec, Gunn and Moore339,Reference Aylward, Abdelaziz, Hunt, Buchmann, Cannon and Lloyd378 as complex choices between airway, swallowing and voice functions can be required (Table 6).Reference Clunie, Kinshuck, Sandhu and Roe379
Table 6. Recommendations for late effects of treatment
SLT = speech and language therapy
Palliative care
Patients with advanced incurable head and neck cancer have a high number and diverse range of complex symptoms. Difficulty eating and weight loss are some of the most frequently reported problems.Reference Cocks, Ah-See, Capel and Taylor380,Reference Mayland, Ho, Doughty, Rogers, Peddinti and Chada381 These symptoms can result in numerous hospital visits, which is particularly problematic with the centralisation of head and neck cancer services.
Early identification of functional changes is essential to providing high-quality and pre-emptive speech and language therapy services, with good communication between central and locality-based therapists and the wider palliative care team (Table 7).
Table 7. Recommendations for palliative care
SLT = speech and language therapy/therapist; HNC = head and neck cancer
Outcome measures
A combination of clinician- and patient-reported outcome measures should be collected at pre-determined time intervals. Systems should be developed to enable data to be collected, stored and protected in a robust manner. At a local level, data can be used for clinical purposes to: highlight rehabilitation needs for individuals; inform the MDT of functional outcomes post treatment; direct future service developments; and provide stakeholders with information regarding effectiveness. On a larger scale, quality data can identify trends in functional changes. This is especially relevant for new and emerging treatment regimens where gaps in understanding can be identified, providing direction for future research topics. Patient-reported outcome measures provide a subjective evaluation of symptom burden.Reference Schache, Kerawala, Ahmed, Brennan, Cook and Garrett71 Quality of life patient-reported outcome measures are integral to a complete dataset, and can be collected by any member of the head and neck cancer MDT (Table 8).Reference Hirano382–Reference Starmer, Drinnan, Bhabra, Watson and Patterson398
Table 8. Recommendations for outcome measures
SLT = speech and language therapy; GRBAS scale = grade, roughness, breathiness, asthenia, strain scale; PSS-HN = Performance Status Scale for Head and Neck Cancer Patients; PROM = patient-reported outcome measure; MDADI = MD Anderson Dysphagia Inventory; TOMs = therapy outcome measures; DIGEST = Dynamic Imaging Grade of Swallowing Toxicity
Important research questions
Over the next few years, results from national and international trials will further our knowledge on whether alterations to treatment modalities – including intensity-modulated radiation therapy versus dysphagia optimised intensity-modulated radiation therapy,Reference Petkar, Bhide, Newbold, Harrington and Nutting399 de-escalation of post-operative adjuvant (chemo)radiotherapy for oropharyngeal cancer,Reference Owadally, Hurt, Timmins, Parsons, Townsend and Patterson400 intensity-modulated proton beam therapyReference Price, Hall, West and Thomson125 – impact on swallowing outcomes. Secondary analysis of these data will inform our understanding of swallowing outcome measures. Findings may then be translated into treatment selection, improved shared decision-making tools and more accurate patient information.
Further work is indicated in the arena of speech and language therapy interventions, to include: exploration of novel technologies; investigation of specific time points, e.g. early post-operative and late effects of dysphagia; examination of defined populations, e.g. recurrent disease; and consideration of impairment, as well as the psychosocial sequelae of speech, voice and swallowing problems. It is also important to take a holistic view, being inclusive of the needs of carers. These interventions are defined as ‘complex’; specifically, they have a number of interacting components, require new behaviours by those receiving the intervention and can have a variety of outcomes (e.g. patient-reported outcomes, measurements of physiology). For this reason, it is imperative that we understand issues of take-up, adherence and retention, in conjunction with testing for effectiveness.
The context of coronavirus disease 2019 has bought many challenges in the delivery of speech and language therapy services and rehabilitation. In response to this, future research needs to address:Reference Patterson, Govender, Roe, Clunie, Murphy and Brady401
• Reliability and validity of non-invasive voice and swallowing screening and assessment tools, including those that can be conducted remotely
• Collection of standardised outcome measures, utilising digital technologies where possible
Service delivery and models of care need evaluation and further development as to how we can best support patients, the workforce and the National Health Service as a whole. Examples of priorities include:
• Provision of remote highly specialist care for our most vulnerable patients
• Models of telehealth for ENT speech and language therapy services
• Implementation of ENT speech and language therapist-led clinics and triaging
Conclusion
The majority of head and neck cancer patients incur changes to their speech, voice and swallowing function as a result of their disease and/or treatment. The role of the speech and language therapist can therefore be pertinent at any stage from presentation to palliation. Even in those patients with good survival outcomes, treatment sequelae can result in a return to the care of the speech and language therapist for management of complex late side effects. Working collaboratively with MDT colleagues will ensure patients receive and reach optimal potential. Consistent recording of an agreed set of outcome measures is needed to capture the burden of disease and thus inform services of existing patient needs, and identify gaps in understanding.
Key points
• The evidence base is of low-quality or very low-quality evidence, based on underpowered trials that are of high risk of bias.
• There is insufficient evidence to understand whether and how pre-operative rehabilitation should be provided to patients prior to neck dissection for head and neck cancer.
• There is low-quality evidence to justify the role of physiotherapy during the in-patient stay, and uncertainty over what exercise prescription should be offered to patients following neck dissection. Given the heterogeneity in the clinical presentation and functional impairments of these individuals, each patient should be assessed for the potential for post-operative complications, and exercises should be prescribed, tailored to the individual, which aim to prevent complications, optimise range of motion and symptom management, and enhance function.
• There is very low-quality evidence supporting the provision of exercises for people who undergo neck dissection following hospital discharge. There is uncertainty regarding what these exercises should be, but there is some promise that progressive resistance and strengthening exercises may offer benefit over simple range of motion exercises. There is evidence that out-patient physiotherapy may offer benefit over self-directed exercises, but this is based on underpowered trials.
• Further evidence is required to understand what components of treatment are required for this population, and to determine when they should be offered.
• There remains uncertainty as to whether some patients are at greater risk of poor outcome and would benefit more from physiotherapy. It is suggested that people who experience an intra-operative spinal accessory nerve injury may be at risk. Further evidence is required to determine whether this population should routinely be reviewed in out-patient physiotherapy, and explore whether certain patients are of lower risk of poor outcome and may be better managed through a self-directed approach.
Introduction
Head and neck cancer affects 700 000 people worldwide and over 11 000 in the UK annually.402–Reference Deschler, Richmon, Khariwala, Ferris and Wang404 Whilst the incidence of head and neck cancer is increasing, prognosis and survival in the UK continues to improve.405,Reference Louie, Mehanna and Sasieni210 As such, the proportion of people living with the effects of this cancer and its treatment is increasing.
Post-operative complications are common following neck dissection.Reference Cappiello, Piazza, Giudice, De Maria and Nicolai406–Reference Shone and Yardley409 Early complications can include shoulder pain and infection. Late complications may not appear until three months post treatment, and can continue to present over five years.Reference Chan, Wong, Chan and Wei410,Reference Guru, Manoor and Supe411 These late complications include shoulder movement dysfunction, and speech, swallowing and musculoskeletal problems, such as cervical contracture and muscle wastage.Reference Chan, Wong, Chan and Wei410 Psychosocial complications are also highly prevalent post-operatively – predominantly fatigue, anxiety, depression, sleep disturbance and social isolation. Shoulder dysfunction and psychosocial complications are strongly associated with reduced return to work, with up to 50 per cent of patients ceasing working because of shoulder disability alone.Reference Shone and Yardley409,Reference Chaplin and Morton412
The treatment pathway for head and neck cancer is complex given the varied anatomical sites of disease and the needs of the patient. Treatment for head and neck cancer requires treatment of the primary site, e.g. tonsil and larynx, as well as the neck. The neck is included when there is spread to the lymph nodes or a high probability of spread. Treatment involves surgery, radiotherapy or chemoradiotherapy, used as single treatments or in combination. Treatment of the neck requires a neck dissection or inclusion in the radiotherapy fields. Side effects from surgery can be significant, including swallowing problems, neck and shoulder girdle problems, difficulties sleeping, fatigue, and anxiety.Reference Roerink, Coolen, Schenning, Husson, Smit and Marres413 Accordingly, physiotherapy and exercise prescription are considered interventions that may manage these surgical complications.
There has been no agreement on what UK practice for head and neck cancer neck dissection rehabilitation consists of.Reference Robinson, Ward, Mehanna, Paleri and Winter414 The role of physiotherapy in the in-patient and out-patient (post-discharge phase) broadly includes: reducing the risk of early or later complications, including respiratory, musculoskeletal and neurological complications; optimising normal recovery and healing; restoring functional and occupational capabilities; and maximising psychosocial enhancement and outcomes.
Accordingly, usual in-patient physiotherapy frequently commences day 1 post-operatively and involves:
• Early mobilisation to reduce risk of post-operative pulmonary complications
• Additional respiratory support for airway clearance and alveolar recruitment, as indicated
• Prescription of a personalised exercise programme, including neck and shoulder range of motion and progressive shoulder strengthening exercises, to minimise the risk of post-surgical contracture, optimise neck and shoulder function, and potentially help optimise movement of muscles used in swallowing
• Education on body positioning to reduce pressure and pull on the shoulder girdle, protection advice for the eyes and mouth in the presence of facial nerve palsy and pain management, and pacing activities to optimise levels of comfort and function
• Assessment of the spinal accessory nerve
Post-operatively, there remains, in the UK, less consistency of physiotherapy provision.Reference Robinson, Ward, Mehanna, Paleri and Winter414 However, two key themes exist:
• Assessment of the need for post-discharge physiotherapy, to minimise complications such as reduced shoulder and neck range of motion, strength and function. This may be particularly important for those who have experienced an intra-operative spinal accessory nerve injury or have pre-operative reduced joint range of movement and function.
• Provision of advice and guidance (with or without supporting educational materials in the form of a paper-based leaflet or online) on post-operative self-management strategies, including exercise, pain management and return to work, and activities of daily living.
As acknowledged, currently there is no national standard best practice for rehabilitation following head and neck cancer. Physiotherapy practice varies across the UK.Reference Robinson, Ward, Mehanna, Paleri and Winter414 Rehabilitation, in the form of physiotherapy, is not routinely available to patients with head and neck cancer, in either in-patient or out-patient settings.Reference Roerink, Coolen, Schenning, Husson, Smit and Marres413 Nonetheless, rehabilitation was the focus of 1 of the 22 key questions in the 2016 National Institute for Health and Care Excellence (NICE) Clinical Guideline415 on the management of head and neck cancer. The guideline recommends clinicians ‘consider progressive resistance training for people with impaired shoulder function, as soon as possible after neck dissection’. The review highlighted that the evidence was from small trials with a high risk of bias. The NICE guideline concluded that a prospective randomised trial was required to understand how best to promote recovery following head and neck cancer.415 Uncertainty therefore remains regarding the effectiveness of physiotherapy in reducing post-operative complications following neck dissection for head and neck cancer.
In addition to these guidelines, the following resources provide further support to both physiotherapists and patients with head and neck cancer:
• MacMillan Cancer Support – recovering from head and neck cancer surgery (www.macmillan.org.uk)
• UK National Health Service – head and neck cancer (www.nhs.uk)
• American Head and Neck Society – neck dissection (www.ahns.info)
• UK NICE – cancer of the upper aerodigestive tract: assessment and management in people aged 16 and over (www.nice.org.uk)
Pre-operative physiotherapy and exercise prescription
No studies were identified that assessed the effectiveness of pre-operative physiotherapy interventions for people before neck dissection for head and neck cancer. This is a major limitation to the evidence base. Table 1 outlines the essential and desirable pre-operative physiotherapy and exercise recommendations, based solely on clinical recommendations and transferable evidence in other surgical specialties.
Table 1. Recommendations for pre-operative physiotherapy and exercise intervention
Based on other surgical specialties such as general surgery, there is moderate quality evidence that a multimodal prehabilitation intervention, including respiratory and whole-body exercises, education and advice, and psychological support, may be beneficial for patients in terms of reducing length of hospital stay (mean difference = 3.68 days; 95 per cent confidence interval (CI) = 0.92, 6.44).Reference Waterland, McCourt, Edbrooke, Granger, Ismail and Riedel416 There was moderate quality evidence that this may improve pre-operative functional capacity, as measured in terms of 6-minute walk test distance (mean difference = 33.09 metres; 95 per cent CI = 17.69, 48.50), but this may not necessarily translate into a reduced risk of post-operative complications (odds ratio = 0.81; 95 per cent CI = 0.55, 1.18) or post-operative mortality (odds ratio = 0.95; 95 per cent CI = 0.43, 2.09).Reference Waterland, McCourt, Edbrooke, Granger, Ismail and Riedel416
Across the pre-operative (non-head and neck cancer) literature, there is evidence that pre-operative education is important for patients and their family members to prepare for surgery. This may concern shaping expectations of the hospital admission and post-hospital discharge recovery, and particularly to inform of the risks of spinal accessory nerve injury during surgery, and associated neck and shoulder complications.Reference Gane, Michaleff, Cottrell, McPhail, Hatton and Panizza417 This may also include practical recommendations on ‘normal’ recovery, preparing the home and living arrangements for after surgery, or liaising with workplaces and other people important in the patients’ home and work life to prepare for before and after hospital.
There is insufficient evidence on how and when prehabilitation interventions should be delivered for people undergoing neck dissection for head and neck cancer.Reference Thomas, Tahir, Bongers, Kallen, Slooter and van Meeteren418 Given the timescales between listing for surgery and the hospital admission, it may not be feasible to provide prehabilitation interventions weeks in advance of surgery. However, there is potential that patients may receive support pre-operatively, with tailored guidance on their personal prehabilitation programme before surgery.
Post-operative physiotherapy and exercise prescription – hospital admission
The recommendations for a post-operative physiotherapy and exercise intervention commencing during the in-patient hospital phase for patients undergoing neck dissection for head and neck cancer are presented in Table 2. Two trials have investigated physiotherapy interventions prescribed to patients following neck dissection surgery in the hospital setting.Reference Steegmann, Bartella, Kloss-Brandstatter, Kamal, Holzle and Lethaus419,Reference Takamura, Miyauchi, Tomoda, Uruno, Ito and Miya420
Table 2. Recommendations for post-operative physiotherapy and exercise intervention commenced during the in-patient hospital phase
Takamura et al.Reference Takamura, Miyauchi, Tomoda, Uruno, Ito and Miya420 assessed the prescription of a stretching and range of motion exercise programme to improve and optimise shoulder and neck motion. This programme was monitored by a nurse or doctor whilst in hospital, and was supported with an exercise leaflet. Outcomes were compared to a comparator group who were encouraged to return to normal movement, but without a specific stretching or exercise programme. There was low-quality evidence, downgraded because of risk of bias and imprecision, that those who received the supported exercise programme were less likely to experience pain at 1 month (odds ratio = 0.55; 95 per cent CI = 0.36, 0.84), 6 months (odds ratio = 0.47, 95 per cent CI = 0.28, 0.78) or 12 months (odds ratio = 0.47; 95 per cent CI = 0.27, 0.83) post-operatively. There was no difference in the probability of experiencing a complication between the groups (odds ratio = 1.17; 95 per cent CI = 0.39, 3.53). There was no substantial difference between the groups in terms of length of hospital stay (mean difference = 0.50 days; 95 per cent CI = 1.02, 0.02).
In the study by Steegmann et al.,Reference Steegmann, Bartella, Kloss-Brandstatter, Kamal, Holzle and Lethaus419 both the comparator and experimental groups received range of motion shoulder and neck exercises; patients allocated to the experimental group received those exercises and an individual cardiovascular exercise programme. Based on the GRADE (Grading of Recommendations Assessment, Development and Evaluation) approach, there was low-quality evidence (downgraded because of risk of bias and imprecision) of a significant difference between the groups in terms of length of hospital stay, with those who received mobilisation and general exercise demonstrating a mean difference of 5.40 days (95 per cent CI = 8.08, 2.72) compared to the control group. Whilst there was no difference in complication score (assessed using a 0–10 visual analogue scale) for neurological, cardiovascular, respiratory, musculoskeletal, sleep, concentration or anxiety complications (p > 0.05), there was low-quality evidence (downgraded because of risk of bias and imprecision) that those patients randomised to the general exercise group experienced fewer complications regarding their digestion (mean difference = 2.10 points; 95 per cent CI = 3.48, 0.72) and less fatigue (mean difference = 1.60 points; 95 per cent CI = 2.98, 0.22) compared to the range of motion alone exercise group.
There was no evidence identified concerning the requirement for a respiratory assessment or the need to assess for factors for a safe discharge to the post-hospital residence. The review authors recommend that these aspects are considered, in addition to an assessment of whether patients require post-discharge physiotherapy, to ensure safe and effective hospital care. A recent survey of UK provision of physiotherapy following neck dissection surgery acknowledged that 78 per cent of the nine regional centres surveyed offered post-discharge physiotherapy for those who had symptoms.Reference Robinson, Ward, Mehanna, Paleri and Winter414 The assessment of symptoms is therefore important within the pre-discharge assessment, and may be particularly important for those with spinal accessory nerve neuropraxia. Whilst there remains insufficient evidence justifying the essential recommendation for physiotherapists to ensure that patients are given advice regarding goal-setting, pacing, pain management, scar management and limb positioning (when appropriate), these are considered potentially valuable skills that can be taught to patients prior to discharge.Reference Lauchlan, McCaul and McCarron421 It is therefore considered desirable that these skills are shared with patients, particularly those who are experiencing or at risk of complications following discharge.Reference Lauchlan, McCaul and McCarron421
Post-operative physiotherapy and exercise prescription – post-hospital admission
The recommendations for physiotherapy and exercise interventions for patients undergoing neck dissection for head and neck cancer following hospital discharge are presented in Table 3. A number of different physiotherapy programmes for people who have undergone neck dissection surgery for head and neck cancer, which begin after hospital discharge, have been evaluated and reported in the literature. These include: the assessment of progressive resistance exercises, muscle energy technique exercise programmes, the use of acupuncture, the provision of self-help interventions, the prescription of yoga and the use of a multimodal rehabilitation programme.
Table 3. Recommendations for post-operative physiotherapy and exercise intervention commenced following hospital discharge
Three trials (reported across four papers) have assessed the outcome of progressive shoulder resistance exercise training compared to active shoulder range of motion exercise programmes.Reference McNeely, Parliament, Courneya, Seikaly, Jha and Scrimger422–Reference McNeely, Parliament, Seikaly, Jha, Magee and Haykowsky424 On meta-analysis that pooled outcomes, there were no differences between those who received progressive shoulder resistance exercises compared to active shoulder range of motion exercises in terms of flexion range of motion at 12 weeks (mean difference = 9.27 degrees; 95 per cent CI = −2.97, 17.51) or abduction at 12 weeks (mean difference = 15.93 degrees; 95 per cent CI = −0.92, 32.78), although those who were randomised to the progressive resistance exercise programme demonstrated significantly greater external rotation range at 12 weeks (mean difference = 12.0 degrees; 95 per cent CI = 1.56, 22.44). McGarvey et al.Reference McGarvey, Hoffman, Osmotherly and Chiarelli425 reported 12-month data. There were no differences between the two exercise programmes at this timepoint in terms of flexion (mean difference = 1.70 degrees; 95 per cent CI = −14.10, 17.50) or abduction (mean difference = −4.50; 95 per cent CI = −31.28, 22.28). There was no difference in health-related quality of life when measured using the Functional Assessment of Cancer Therapy – General (‘FACT-G’) questionnaireReference Cella, Tulsky, Gray, Sarafian, Linn and Bonomi426 at 12 weeks (mean difference = 7.18 points; 95 per cent CI = −0.85, 15.24) or at 12 months (mean difference = 1.50 points; 95 per cent CI = −7.20, 10.20). This was the same for pain at 12 weeks, which was assessed in McNeely et al.Reference McNeely, Parliament, Courneya, Seikaly, Jha and Scrimger422 (mean difference = 1.60 points; 95 per cent CI = −17.50, 20.70), and for shoulder function both at 12 weeks (standardised mean difference = 0.01; 95 per cent CI = −0.35, 0.38) and 12 months (standardised mean difference = 0.18; 95 per cent CI = −0.45, 0.82). However, for all outcomes, this evidence was judged as very low-quality based on the GRADE assessment, downgraded because of risk of bias, imprecision and inconsistency.
Two trials investigated outcomes comparing physiotherapy when delivered in a supervised out-patient setting compared to being self-directed by patients at home, after receiving instructions on hospital discharge.Reference Lauchlan, McCaul and McCarron421,Reference Su, Chen, Leong, Huang, Chiang and Chen427 There was very low-quality evidence, based on the GRADE assessment (downgraded because of risk of bias, imprecision and inconsistency), that those randomised to out-patient-based physiotherapy had significantly greater neck and shoulder function at 6 weeks (mean difference = 29.68 points; 95 per cent CI = 15.27, 44.09) but no difference at 12 weeks (mean difference = −9.81 points; 95 per cent CI = −23.18, 3.56). However, there were no differences in shoulder range of motion in respect to: flexion at 6 weeks (mean difference = 1.40 degrees; 95 per cent CI = −9.78, 12.58) or 12 weeks (mean difference = 3.65 degrees, 95 per cent CI = −7.80, 15.10); abduction at 6 weeks (mean difference = 4.08 degrees; 95 per cent CI = −17.54, 25.70) or 12 weeks (mean difference = 8.09 degrees; 95 per cent CI = −11.63, 27.81); or external rotation at 6 weeks (mean difference = 3.87; 95 per cent CI = −2.61, 10.35) or 12 weeks (mean difference = −2.98; 95 per cent CI = −9.40, 3.44). There was no difference between the groups in respect to pain scores at 6 weeks (mean difference = 1.11 points; 95 per cent CI = −0.36, 2.58) or 12 weeks (mean difference = 1.20 points; 95 per cent CI = −0.15, 2.55).
Thomas et al.Reference Thomas, D'Silva, Mohandas, Pais and Samuel428 compared the prescription of a muscle energy technique exercise programme for the shoulder versus an active range of motion exercise programme for this population. They reported that whilst there was no difference between the exercise programmes at 10 days for abduction (mean difference = −1.28 degrees; 95 per cent CI = −11.12, 8.56) or internal rotation range (mean difference = 1.06 degrees; 95 per cent CI = −3.70, 5.82), those who received the muscle energy technique exercises demonstrated greater flexion (mean difference = 15.35 degrees; 95 per cent CI = 6.34, 24.36) and external rotation (mean difference = 5.94 degrees; 95 per cent CI = 0.82, 11.06). There was no difference between the groups in pain scores at 10 days (mean difference = 0.13 points; 95 per cent CI = −0.79, 0.53). However, the evidence was judged as very low-quality given a high risk of bias and imprecision, being based on a single, underpowered trial.
One trial was identified that investigated the use of acupuncture following neck dissection, assessing outcomes at six weeks post-intervention.Reference Pfister, Cassileth, Deng, Yeung, Lee and Garrity429 They reported very low-quality evidence (downgraded three levels because of risk of bias and imprecision) indicating that, whilst there was no difference between people who received acupuncture versus those who did not and followed the usual care for shoulder function at six weeks (mean difference = 6.30 points; 95 per cent CI = −3.73, 16.33), the acupuncture patients demonstrated significantly lower pain scores at this timepoint (mean difference = 2.20; 95 per cent CI = −3.41, −0.99).
One trial (reported in two papers) compared the outcomes of prescribing a self-help exercise programme and self-care education programme versus a self-care education programme alone.Reference Jansen, Eerenstein, Cnossen, Lissenberg-Witte, de Bree and Doornaert430,Reference Jansen, Coupe, Eerenstein, Cnossen, van Uden-Kraan and de Bree431 They reported that, whilst there was no difference in health-related quality of life, measured using the European Organization for the Research and Treatment of Cancer 30-item Quality of Life Questionnaire (‘EORTC QLQ-C30’) (global health status)Reference Aaronson, Ahmedzai, Bergman, Bullinger, Cull and Duez432 at three months (mean difference = 4.80; points; 95 per cent CI = −4.39, 13.99), those who received self-directed education and exercise demonstrated greater health-related quality of life compared to people who only received self-directed education (mean difference = 8.00 points; 95 per cent CI = 0.48, 15.52). Similarly, people who received both self-directed exercises and education demonstrated better pain outcomes compared to those who only received education, both at three months (mean difference = −14.60 points; 95 per cent CI = −25.44, −3.76) and at six months follow up (mean difference = −12.20 points; 95 per cent CI = −22.75, −1.65). There was also a significant benefit in favour of people who received the education and exercise intervention for neck and shoulder function when measured using the European Organization for the Research and Treatment of Cancer Quality of Life Questionnaire (functional scale)Reference Aaronson, Ahmedzai, Bergman, Bullinger, Cull and Duez432 at three months (mean difference = 10.40 points; 95 per cent CI = 3.39, 17.41) and at six months (mean difference = 8.80 points; 95 per cent CI = 0.54, 17.06). Whilst providing support for the provision of exercises with education, the evidence for these outcomes was based on low-quality evidence, downgraded for risk of bias and imprecision. It is interesting to note that only one-quarter of these patients were randomised within six months of their operation, suggesting patients might still be able to benefit from physiotherapy input later post-operatively.
One trial reported the outcomes of shoulder range of motion, pain and anxiety in people who were prescribed a yoga programme for eight weeks compared to those who did not and followed usual care following neck dissection for head and neck cancer.Reference Adair, Murphy, Yarlagadda, Deng, Dietrich and Ridner433 The data were presented as medians and interquartile ranges, precluding the ability to perform meta-analysis. The paper reported significantly greater shoulder range of motion for active and passive abduction and external rotation for those who receive the yoga compared to those who did not (p < 0.05). They also reported significantly lower pain interference scores and anxiety scores, both at four weeks and eight weeks, compared to the usual care group (p < 0.05). The evidence was judged low in quality because of a high risk of bias and being underpowered.
Chen et al.Reference Chen, Lin, Liang and Huang434 used a multimodal intervention that included pain management, scar massage, stretching, active and passive shoulder range of motion and scapular setting exercises, with education on scapular setting and positioning. Their control group was the same multimodal programme, but without the scapular setting component. Both groups commenced their intervention post-operatively for the first three weeks post-discharge. The authors reported a significant improvement in shoulder abduction within the scapular setting group compared to the general shoulder exercise group (p < 0.05). Whilst pain significantly decreased in both groups over the three-week follow-up period, there was no significant difference between the groups for pain score (p > 0.05).
Important research questions
As highlighted, the current evidence base presents a number of issues. Recommendations are largely based on clinical reasoning rather than robust evidence. There is therefore a major gap in knowledge that urgently requires addressing. Key research themes arising from the current state of the evidence base are listed below.
Prehabilitation for people scheduled for neck dissection to treat head and neck cancer
There is currently uncertainty as to what interventions should be included in a prehabilitation programme for this population. The composition, timing, frequency, duration, personnel and setting to deliver such an intervention remain unknowledge. There is a gap in knowledge regarding the effectiveness of such an intervention, and it is unclear whether effectiveness would vary for those with different stages of cancer and different surgical requirements, and for concomitant treatments with radiotherapy or chemotherapy.
In-patient and out-patient physiotherapy requirements for people undergoing neck dissection to treat head and neck cancer
Current physiotherapy and exercise management for people who undergo neck dissection to treat head and neck cancer is geographically varied. It remains unclear from the evidence whether there is a subgroup of patients who have greater requirement for in-patient or out-patient physiotherapy, or whether all patients who undergo physiotherapy have the potential to benefit. Given the diverseness of the population, there is potential heterogeneity in terms of the need for physiotherapy, the composition of a programme and dosage (frequency, duration, intensity). Future research should be prioritised to determine who has the potential to benefit from such an intervention, and whether a flexible model should be adopted to account for varying clinical presentations within hospital and community settings. Assessing the effectiveness of such a programme and approach to treatment is then required, with a sufficiently powered and robust pragmatic clinical trial.
Outcome measures for people recovering from neck dissection conducted to treat head and neck cancer
The literature review identified that whilst pain and active shoulder flexion, abduction and external rotation are frequently reported outcome measures in trials of physiotherapy interventions for this population, there is no consistency on outcomes reported, or on the outcome instruments or tools reporting these domains. There is no core outcome set for trials assessing physiotherapy interventions in people recovering from neck dissection conducted to treat head and neck cancer. Accordingly, outcome domains that may be important to patients, healthcare professionals and other stakeholders (including health utilisation or psychosocial outcomes such as anxiety, depression, return to work or social pursuits) have not been reported in the literature. The development of such a core outcome set would be valuable to aid standardisation of outcome reporting, allowing comparison between interventions across trials, whilst also facilitating meta-analysis once the evidence base develops in this area.
Future aspects of physiotherapy
There is evidence that patients value physiotherapy following neck dissection conducted to treat head and neck cancer.Reference Gallyer, Smith, Fordham, Dutton, Chester-Jones and Lamb435 However, the provision of such treatment is restricted by service provision and capacity. This may be overcome by healthcare providers if a stronger evidence base existed to justify the clinical and cost-effectiveness of physiotherapy for these patients. The role of physiotherapy for these patients may therefore develop as the evidence develops.
Neck dissection surgery is frequently undertaken in a smaller number of specialist centres in the UK.Reference Robinson, Ward, Mehanna, Paleri and Winter414 This is particularly the case for more complex head and neck cancer surgery. In such an instance, patients are frequently required to travel distances for follow-up care and, if provided, specialist physiotherapy. With the growing interest in offering telerehabilitation through video conferencing and computer platforms, physiotherapy for this patient group may develop through a more virtual approach.Reference Gilbert, Jones, Stokes and May436 However, this may present with challenges, particularly when assessing neuromusculoskeletal function where face-to-face examination may be preferable, and when patients present with communication challenges post-operatively that may be exacerbated in a virtual rather than face-to-face approach. Consideration of the approaches of such a platform may be examined in the future.
The population who are treated with neck dissection for head and neck cancer is changing. Historically, this was an older population whose risk of developing head and neck cancer was increased through smoking and alcohol consumption. More recently, people affected by and surviving head and neck cancer in the UK are younger and more active than previous generations,Reference Louie, Mehanna and Sasieni210 attributed to human papillomavirusReference Schache, Powell, Cuschieri, Robinson, Leary and Mehanna251 being an increasing cause of the disease. With this change in demographic, there has been a change in recovery expectation, with patients now being more physically active, with social and occupational pursuits, interests and requirements. However, still approximately 50 per cent of patients are unable to return to their work post-surgery.Reference Rogers, Ferlito, Pellitteri, Shaha and Rinaldo408,Reference Roerink, Coolen, Schenning, Husson, Smit and Marres413 The role of physiotherapy has adapted accordingly. Therefore, a greater emphasis on vocational rehabilitation and a return to occupational goals is required to support the personal needs of our patients. Consideration on how we deliver this, and how this relates to the timing of other interventions such as chemotherapy or radiotherapy, should be considered.Reference Gallyer, Smith, Fordham, Dutton, Chester-Jones and Lamb435
Conclusion
Physiotherapy for people who have undergone neck dissection in the management of head and neck cancer may be beneficial. This is largely based on clinical recommendation and a low-quality evidence base. With a changing demographic of patients, determining who can benefit from physiotherapy, and when and in what form physiotherapy and exercise should be provided, is a research priority. Through developing the evidence, it is anticipated that physiotherapists, and the wider head and neck cancer multidisciplinary team, will be able to justify the provision of this intervention, which may offer considerable benefits to the health and well-being of this growing patient population.
Key points
All multidisciplinary teams (MDTs) should make adequate provision of clinical nurse specialist posts or individuals to ensure the following essential care elements can be provided at all times:
• Clinical nurse specialist contact with individuals at the time of diagnosis
• Clinical nurse specialist involvement in all MDT meetings
• Psychosocial support and care co-ordination for all individuals being managed within the MDT (to varying extents) by all modalities of treatment
• Availability to provide co-ordination of multiprotection care pathways
• Accessibility as a point of contact for expert advice to primary care physicians, district nurses, specialist palliative care teams etc. (e.g. tracheostomy and gastrostomy care, wound care)
• Co-ordinated, tailored approach to the provision of support for individuals living with the effects of cancer and long-term consequences of cancer treatment. This should include end of treatment summaries, health and well-being advice, signposting to resources available, and appropriate follow up
Introduction
Head and neck cancers are complex, and are managed with a range of arduous treatments. Head and neck cancer and its treatment is physically debilitating, and the psychological impact is immense. Patients and carers often require assistance and support, from diagnosis and treatment through to long-term support, to help them live well with the impact of the disease.
Specific aspects of clinical nurse specialist role
The role of the clinical nurse specialist is diverse, involving the co-ordination and signposting of patients and carers, for advice, information and support, from diagnosis through to the later stages of disease. Clinical nurse specialists act as the gate-keeper to the patients' cancer pathway, to deliver a seamless journey.Reference Dempsey, Orr, Lane and Scott437
There are specific core aspects of the clinical nurse specialist role; these are described within this chapter (Table 1).438–442 It is essential that the clinical nurse specialist works closely with the speech and language therapy and dietetic teams as significant others, in order to deliver a multidisciplinary approach in which to achieve the best possible outcomes. Whilst the roles may overlap, their reliance on one another is imperative.
Table 1. Specific aspects of clinical nurse specialist role
HPV = human papillomavirus; NHS = National Health Service; SLT = speech and language therapy
Surgery
Surgery is one of the key modalities used in head and neck cancer treatment. Surgery can have permanent effects on structures essential for normal human activities, therefore affecting speech, mastication and breathing.443
Head and neck cancer is complex and life-changing, often having a traumatic and devastating impact on the patient and their family, both physically and psychologically, because of the impact on highly important functions.Reference Iftikhar, Islam, Shepherd, Jones and Ellis444
Support is required before, during and after treatment, to help the patient adjust to the physical, social and emotional effects, sometimes with disfigurement, on quality of life (Tables 2 and 3).Reference Davidson and Williams445
Table 2. Role of clinical nurse specialist pre-treatment (pre-surgery)*
* Multidisciplinary team (MDT) meeting and combined head and neck clinic. CNS = clinical nurse specialist; AHP = allied health professional; HNC = head and neck cancer; GP = general practitioner
Table 3. Role of clinical nurse specialist during admission for surgery
The head and neck cancer clinical nurse specialist plays a pivotal role throughout the disease trajectory. For example, information given at diagnosis often includes working uniquely with the MDT (speech therapist, dietician, physiotherapist, oncologist etc.), setting realistic expectations with patients and their families before ‘life-changing surgery or treatment’, and this is often a challenging process. In addition, managing disease symptoms and treatment side effects requires complex care management with issues such as complex airway management, or bleeding that could signify a life-threatening situation and which needs to be urgently assessed and addressed. This highlights the importance of advocacy, which is an MDT responsibility.Reference Dawson, Roe, Starmer, Brady, Nund and Coffey446
The clinical nurse specialist can develop a unique relationship with the patients and their carers. This is established through using expert clinical skills, decision-making skills and judgement, together with experience and intuition. Practising within this field can be both challenging and rewarding.
From surgery through to post-treatment care, the clinical nurse specialist helps to build an appropriate care pathway for each patient given their everyday challenges. By providing rationale and advice for each treatment option, the patient is included in the decision-making process and feels engaged in their care.Reference Robson447
‘It has been recognised that care co-ordination individualised to the patient during and after treatment is vital to deliver appropriate person-centred care’.Reference Dempsey, Orr, Lane and Scott437
Prehabilitation is vital in the preparation for potentially life-changing surgery and treatment. It is a continuum to rehabilitation, and focuses on personal empowerment by improving physiological function and psychological well-being, therefore improving resilience to the effects of cancer treatment.Reference Dawson, Roe, Starmer, Brady, Nund and Coffey446
Prehabilitation, described in detail in a separate dedicated chapter, should be implemented as soon after diagnosis as possible to enable maximal benefit in advance of treatment. These interventions aim to reduce post-operative complications and minimise functional decline following surgery.
Macmillan Cancer Support published ‘Principles and guidance for prehabilitation within the management and support of people with cancer’ in 2019.448 This document sets out principles with which people with cancer can be prepared for treatment, through a multimodal approach that promotes healthy behaviours, exercise prescription, nutrition and psychological interventions appropriate to their needs.
Early intervention, engagement and interaction can be both healing and therapeutic for patients post-surgery.Reference Dawson, Adams and Fenlon449 Regular patient reviews for support and information are vital in the recovery, as this is often a particularly anxious time for the patient and family as they adjust to the post-operative changes. Recent research has also indicated that the prevalence of mental health disorders in people with head and neck cancer increases by 10 per cent after diagnosis.Reference Dawson, Adams and Fenlon449
The head and neck nursing team also play an active role in educating junior nursing staff, student nurses, allied health professionals and medical staff whilst reviewing patients in the clinical environment. The clinical nurse specialist's experience, mentoring and educational support of colleagues is a valuable resource.
In 2020, the British Association of Head and Neck Oncologists set out clear guidance for the management of head and neck cancer patients, both in the hospital and community setting, as described below.Reference Schache, Kerawala, Ahmed, Brennan, Cook and Garrett71
In-patient nursing staff
(1) The nurse in charge on each shift should have a specialist qualification in a related discipline and a minimum of five years of experience.
(2) Two other nurses on the staff should have, or be preparing for, a specialist qualification in related disciplines.
(3) Nursing staff, including healthcare assistants, should have competencies associated with altered airway management and major haemorrhage in the head and neck setting.
Nurses should be informed and aware of ongoing clinical research projects, audits and clinical trials.
Crisis planning
• All units and hospices managing individuals with head and neck cancer should adhere to local guidelines for tracheostomy blockage and major haemorrhage.
• All specialist head and neck ward nurses should be aware of these protocols.
• Individuals (and their carers) at risk of these crises should be made aware of the warning signs in all cases, unless the patient has expressed a wish to not be provided with this information.
Role of clinical nurse specialist during oncology treatment
The role of the oncology clinical nurse specialist is equally as important (Table 4). This can be a difficult period for patients as they move from one treatment modality to another. Anxiety is often high and information is key.
Table 4. Role of clinical nurse specialist during oncology treatment
MDT = multidisciplinary team; SLT = speech and language therapist; CNS = clinical nurse specialist; NCI = National Cancer Institute; CTCAE = Common Terminology Criteria for Adverse Events; HNC = head and neck cancer
Supportive care, appropriate information and individualised care planning is key to improving the experience of the patient and the carer.Reference Dempsey, Orr, Lane and Scott437
This is also often a point of change of key worker. It is vital the oncology clinical nurse specialist meets the patient and carer at the start of the oncology treatment period. This starts at the pre-treatment clinics, where there is an opportunity for patients and carers to meet the clinical nurse specialist and other allied health professionals prior to treatment. It allows the giving of information, the explanation of treatment and the implementation of a health needs assessment. A health needs assessment ensures that the patients’ and carers’ physical, emotional and social needs are met in a timely and appropriate way, and that advice and support are available from the right source at the right time.450
Radiotherapy or chemoradiotherapy side effects from treatment are often debilitating. Up to 89 per cent of patients having treatment develop oral mucositis and require analgesia.Reference Muzumder, Srikantia, Udayashankar, Kainthaje and John Sebastian451 Patients should have a clear explanation of what to expect during this period. Side effects build during treatment to a peak on the final week and the week after completion. In general, side effects develop from week two onwards, and treatment toxicity should be assessed and scored with a recognised grading tool. This can help guide consultations and required interventions. Patients should have weekly reviews to ensure compliance with medications and supportive measures. Effective MDT communication will be vital in supporting the patients through this period. Offer onward referral for psychological support if required and if the patient is struggling with treatment.
The clinical nurse specialist should always be a source of information and comfort for the patient undergoing treatment. They should be contactable within working hours and available to review the patient outside of the clinic setting if needed (either face to face or by telephone).
Those patients undergoing chemotherapy or immunotherapy only are generally on treatment with palliative intent. In addition to toxicity management, the clinical nurse specialist should assess the level of support required and arrange community palliative care support if required.
Living with cancer
An individual approach to living with cancer (last 12 months of life), including clear documentation and communication with each relevant team providing palliative care, must be employed for all individuals. This may include liaison with primary care and specialist palliative care teams. Clear offers must be made to discuss and implement advance care plans with all patients. These may reasonably include patients’ preferred place of care, agreed thresholds or ceilings of treatment.
For those recognised to be dying (last days of life), an individualised care plan should support care for all individuals.
Late effects of treatment
The role of the clinical nurse specialist incorporates support with late effects of treatment, in relation to chemotherapy, radiotherapy, surgery or combined modality treatment (Table 5).440–442,Reference Gunn, Gilbert, Nenclares, Soliman, Newbold and Bhide452 Macmillan defines ‘late effects’ as side effects that do not disappear after treatment, or that do not materialise until months or even years after treatments have taken place.Reference Dawson, Roe, Starmer, Brady, Nund and Coffey446 Such effects differ between patients, both in terms of the effects and the timing at which they occur.
Table 5. Late effects of treatment and role of clinical nurse specialist
NHS = National Health Service; SLT = speech and language therapy
The clinical nurse specialist works closely with other team members in supporting patients with late effects, such as lymphoedema teams, late effects radiographers and core members of the MDT.
Community, palliative and supportive care
The head and neck clinical nurse specialist is instrumental in ensuring that support, advice and appropriate signposting is available to patients and their carers when a referral to the acute setting for suspected cancer is made (Table 6).Reference Dempsey, Orr, Lane and Scott437,443,450
Table 6. Palliative and supportive care
HNC = head and neck cancer; CNS = clinical nurse specialist; GP = general practitioner
If discussions at the MDT meeting conclude that it is not possible to cure a patient of their cancer, a less radical course of treatment known as ‘treatment with palliative intent’ may be offered to help control symptoms and slow down progression of the disease. For some patients, it may be recommended that no active treatment be given, if the MDT feels treatment would not be beneficial or may worsen their current symptoms; this is known as ‘supportive care’. Whilst a diagnosis for palliative treatment or supportive care may be given at initial presentation to the acute service, a patient may also be given this diagnosis at any later stage of the pathway, even if the initial intention was for cure. In any case, the clinical nurse specialist acts as the patient's advocate at the MDT, and ensures that sufficient support and advice are available whilst the patient remains under the care of the hospital team.443
Palliative and supportive care
Depending on local service provision and a patient's wishes, specialist palliative care support can be provided, no matter what the care setting, dependent on complexity of needs.Reference Cocks, Ah-See, Capel and Taylor453 It is important that the clinical nurse specialist recognises when further input from a specialist palliative care team is required, and that they ensure a timely referral is made so that a seamless transition of care takes place. Involvement from these different teams and other allied health professionals (such as dietitians, speech therapists and physiotherapists) will require the clinical nurse specialist to have a sound knowledge of local service provision and an understanding of how an onward referral to these teams may occur.
Table 6 lists some essential and desirable skills for the clinical nurse specialist to possess when dealing with a patient being treated with palliative intent.
Community care
As a patient’s condition deteriorates, there is frequently an emphasis on managing palliative patients within the community setting (Table 7). This concerns those patients with distant metastasis, late-stage disease or cancer that is life-limiting, and/or those with a prognosis of 6–12 months. Given that head and neck cancer is a very specialised area of care, some localities provide hospital out-reach services for their patients, which can be beneficial in supporting community staff to manage more complex and specialised situations.
Table 7. Community care
In order to facilitate greater community working, the clinical nurse specialist should have the skills listed in Table 7.
Psychology in the management of head and neck cancer
A diagnosis of head and neck cancer is often described by patients as physically and psychologically debilitating, with significant psychosocial implications on physical, emotional, spiritual, financial and interpersonal interactions (Table 8). Unlike other tumour sites, head and neck cancer and its treatment cannot be hidden, often because of disfigurement, altered anatomy, and lasting effects on eating and communication. Whilst some patients adapt to their altered anatomy with little support, others experience a loss of confidence in social interactions because of self-image,Reference Nayak, Pai and George454 and go through significant periods of adjustment to obtain an acceptable level of quality of life. Windon et al.Reference Windon, D'Souza, Faraji, Troy, Koch and Gourin455 identified that head and neck patients experienced feelings of regret post treatment, particularly following multiple treatment modalities despite curative treatment intention. This signifies the crucial role that psychological support has within the complex and challenging management of head and neck cancer.
Often patients share a fear of recurrence as a primary concern, whilst milestones such as post-treatment investigations, follow-up appointments and physical reminders of treatment frequently trigger patients back into psychological turmoil. Collectively, head and neck cancer patients display an increased frequency of depression associated with the multifaceted consequences of cancer and the effects of treatment decreasing quality of life for patients.Reference Lin, Starmer and Gourin456 Declines in performance status and functional ability collectively impact on many aspects of daily living, including employment, in the head and neck patient.Reference Yilmaz, Yener, Yollu, Akil, Haciyev and Yargic457 Interviews have shown that survivors of head and neck cancer call themselves the ‘visible minority’, stemming from their noticeable disfigurement.Reference Henry, Ho, Lambert, Carnevale, Greenfield and MacDonald458 The need for psychological support is supported by the author's reflection on a case study, whereby, despite the patient's remission, he was left looking in the mirror at a reflection that overwhelmed him with feelings of shock and regret. This patient acknowledged that the late and physically permanent effects of cancer treatment were undoubtedly worse than the initial cancer diagnosis itself.
As key worker and patient advocate, the clinical nurse specialist is level 2 trained in providing timely psychological interventions, utilising advanced communication skills. The clinical nurse specialist ensures frequent opportunities for one-to-one supportive conversation to meet patients’ information needs around their cancer diagnosis and potential treatment options, but also identifying and responding to psychological needs. This undoubtedly helps to inform therapeutic relationships, and enhance holistic patient-centred care and choice. Involving patients in decisions around their care will help to enable a sense of partnership between the patient and healthcare professional. Screening for psychological distress using a recognised tool and providing ongoing psychological support throughout the entire pathway are imperative in improving the quality of life of those diagnosed with head and neck cancer.
Future roles and development
Advanced nursing roles such as clinical nurse specialist advanced practitioner and advanced nurse practitioner are clinical expert roles that require academic degrees. Healthcare delivered by nurses in these advanced roles has proven to impact important care quality factors such as patients’ experiences, safety, symptom burden and cancer care co-ordination. By improving communication between and within the teams, advanced nursing care can reduce re-admissions. Advanced nursing roles can contribute to improving clinical practice and patient centeredness, through education, developing guidelines and spanning organisational boundaries, to progress the patient through the system.
The clinical nurse specialist is the clinical nursing role with in-depth knowledge of cancer care and symptom management, supporting both the patient and their families through the cancer journey. The clinical nurse specialist's knowledge enables them to offer expert care to patients with all stages of cancer, including screening for early detection, making a diagnosis, administrating treatments, and discussing survivorship. Most clinical nurse specialists perform advanced tasks; these can be nurse-led clinics for pre-treatment, on treatment and after treatment, with clinical nurse specialists often seeing the patient weekly to support with symptom management. The clinical nurse specialist is recognised as the first point of contact for patients.
Some clinical nurse specialists offer additional advanced roles, which include nurse-led tracheostomy tube changes for patients requiring long-term airway management, and joint valve clinics with speech and language therapists. In order to address local demands, nurses working in partnership with physicians may select a group of patients that can be seen independently within a risk-stratified nurse-led clinic, which could include nasendoscopy follow up and working alongside ultrasound clinics for rapid access.
Key points
• Consultants in restorative dentistry are core multidisciplinary team (MDT) members. Their close collaboration with the MDT is essential. The restorative dentistry consultant should be proactive in the MDT, contribute to research and audit, and act as an advocate for optimum oral health outcomes.
• Consideration of oral rehabilitation needs to begin early in the surgical and non-surgical treatment pathways. By managing the pre-treatment pathway optimally, the restorative dentistry consultant can facilitate prevention or reduction of complications.
• Patient involvement in decision-making regarding the oral and dental care plans is essential, and should be underpinned by written and verbal information regarding the expected impact of treatment on appearance, ability to speak, eat and chew. A clear understanding of patient expectations and priorities is essential to providing individual counselling on likely outcomes and personalised plans.
• The possibility of implant rehabilitation should always be considered early, as timely implant rehabilitation can improve patient outcomes, reduce overall treatment times and decrease costs where appropriate.
• Referral pathways for management post treatment should be clear.
Introduction
Impact on patients
Patients diagnosed with head and neck cancer are referred to a specialist head and neck cancer MDT for treatment, which may include surgery, chemotherapy or radiotherapy, or a combination of these. These treatment modalities can have significant and long-lasting adverse effects on orofacial and dental function, appearance and quality of life (QoL). In a recent study of the Patient Concerns Inventory, ‘dental health and teeth’ was the second most commonly selected concern at the baseline clinic, alongside ‘fear of the cancer coming back’. Oral health-related issues were selected in four of the top five issues at this stage in the pathway. Of the health professionals that patients wanted to see at baseline, the dentist was the most selected professional.Reference Rogers, Allmark, Bekiroglu, Edwards, Fabbroni and Flavel459
Thus, the impact on patients can be devastating, adding to the trauma of a cancer diagnosis. Some of these effects are preventable or can be minimised by early intervention from the specialist restorative dentistry team.
The management of long-term oral and dental complications can have a protracted, often lifelong pathway, with attendant costs. There is a marked increase in consumption and costs for dental care in the first two years following diagnosis compared with those for patients without head and neck cancer.Reference Lexomboon, Karlsson, Adolfsson, Ekbom, Naimi-Akbar and Bahmanyar460 In the UK, some of these costs currently fall out of National Health Service (NHS) provision and, consequently are borne by patients themselves.
Standardised, specialist delivery of oral and dental prehabilitation and rehabilitation
Predicting and managing oral and dental complications is complex and highly specialised. For this reason, it is recommended that, at a minimum, each MDT should have at least one consultant in restorative dentistry as a core member of the team.461–463 Specialist restorative dentistry is for patients who have complex dental problems requiring multidisciplinary, specialist dental care.464
The restorative dentistry consultant functions as implant surgeon, maxillofacial prosthodontist and ‘dental oncologist’. Outside the UK, the term ‘dental oncologist’ is used to describe specialists in the dental side effects of head and neck cancer non-surgical treatment. Such multiplicity of roles facilitates a clinically effective pathway for patients treated surgically or non-surgically, and optimises service delivery.
The incorporation of the restorative dentistry consultant to UK head and neck cancer MDTs has developed significantly in the last 15 years. The increase in human papillomavirus associated disease in younger, usually dentate patients, who are expected to survive for longer, the development of new technologies in relation to osseointegrated implants and new approaches to radiotherapy cement the importance of this specialist input.
Outline of requirements for a service
All head and neck cancer services must have continuous service provision by a consultant-led restorative dentistry team, and should have a dedicated specialist dental hygienist and maxillofacial prosthodontic technician or reconstructive scientist technical support.
Digitally planned implant placement in head and neck cancer patients is highly complex, and requires ready access to cone beam computed tomography, digital planning software, optical scanning devices, dedicated software and three-dimensional (3D) printing facilities.
Guideline principles
This paper provides guidelines on planning and treatment for oral and dental prehabilitation and rehabilitation for patients having treatment for head and neck cancer. They were devised with consensus meetings from members of the Restorative Dentistry UK (‘RD-UK’) Head and Neck Cancer Clinical Excellence Network from eight major treatment centres across the UK.
These guidelines cover oral rehabilitation planning and management for patients undergoing radiotherapy or surgery or multimodality treatment. Input is needed at key times: before, during and after cancer treatment. Therefore, the paper is set out along these lines (Figure 1). Each section is separated into recommendations that are considered either ‘essential’ or ‘desirable’. Future roles, and areas for research and audit are described in the final section. In addition to these guidelines, the guidelines produced by Restorative Dentistry UK465 will provide further information for the MDT.
Figure 1. Restorative dentistry consultant role in the patient pathway. MDT = multidisciplinary team; HNC = head and neck cancer; CRD = consultant restorative dentist
Pre-treatment
Patients whose cancer treatment will affect oral and dental function and appearance will require oral rehabilitation planning (Tables 1 and 2).461,462,465 This generally includes patients scheduled for surgical intervention that alters oral anatomy, patients requiring radiotherapy where the treatment field includes any part of the maxilla, mandible or salivary glands, and patients with specific dental concerns or pre-existing conditions.
Table 1. Pre-oral and maxillofacial surgery and ENT surgery planning
MDT = multidisciplinary team; CRD = consultant in restorative dentistry; OMFS = oral and maxillofacial surgery; 3D = three-dimensional
Table 2. Pre-radiotherapy planning
MDT = multidisciplinary team; CRD = consultant in restorative dentistry; SLT = speech and language therapist
Clinically edentulous patients may have retained roots, buried teeth or local bony pathology, and should also be considered for prehabilitation. This stage is often referred to as ‘screening’ or ‘assessment’, which belies a highly complex planning and initial treatment stage with time pressures. It should be more accurately viewed as planning or oral and dental prehabilitation. It presents challenges, as the outcomes of cancer treatment for the individual patient are not yet known. Risk assessment for adverse oral and dental outcomes, therefore, is key.
Some dental pathology is preventable if appropriate early interventions are put in place. Primary implant placement should always be considered as part of surgical planning. For patients having dual modality treatment, this may be the only opportunity to place implants. Secondary implant placement may extend the rehabilitation pathway, but allows more time for comprehensive planning and assessment of any post-treatment challenges. However, implants may not be appropriate in some cases. Patients should be counselled by the restorative dentistry consultant regarding expected prosthodontic oral rehabilitation outcomes from the start.
Success in implant rehabilitation is improved by early digital planning. Digital workflows can greatly facilitate the planning and execution of implant based orofacial rehabilitation, allowing rapid treatment completion.Reference Levine, Bae, Soares, Brecht, Saadeh and Ceradini466,Reference Witjes, Schepers and Kraeima467 The restorative dentistry consultant should be involved in planning with surgical colleagues from the outset.
Software used for digitally planned osseous reconstruction following resection allows dental implant planning with guided surgical stents, and should be undertaken jointly with the restorative dentistry consultant and maxillofacial surgeons during the early planning stage utilising a prosthetically driven approach to optimise rehabilitation outcomes.Reference Rodby, Turin, Jacobs, Cruz, Hassid and Kolokythas468 This helps achieve more predictable outcomes in terms of function, biology and aesthetics,Reference Schubert, Schweiger, Stimmelmayr, Nold and Guth469 which contribute to improving QoL.
Dental extractions for head and neck cancer patients can be a traumatic, highly emotive experience. Extraction of teeth, if indicated, should be organised as early as possible after the cancer treatment plan is known, to maximise healing time and expedite the pathway. However, care should be taken to avoid unnecessary dental extractions, especially where the cancer treatment plan is not yet clarified. Where multimodality treatment is definite or where gross dental pathology exists, extraction during primary surgery should be considered.
The essential and desirable aspects of the pre-treatment restorative planning appointment before surgery and before radiotherapy are outlined in Table 1.
Peri-treatment
Osseointegrated implants can improve the support and retention of prostheses, help raise self-esteem and body image, and improve overall QoL.Reference Rogers, McNally, Mahmoud, Chan and Humphris470 Primary implant placement in head and neck cancer patients involves placing dental implants at the time of the ablative surgery, to allow osseointegration to take place prior to any necessary adjunctive radiotherapy. It facilitates more rapid rehabilitation,Reference Patel, Antov and Nixon471 while avoiding further surgery and further in-patient treatment costs. Disadvantages include reduced planning time, and the risk of implants not being used because of tumour recurrence or changed anatomy during or following surgery.Reference Alberga, Vosselman, Korfage, Delli, Witjes and Raghoebar472
Peri-surgical maxillectomy defect management can involve prosthetic obturation or surgical reconstruction of the defect. Dental implants can help to retain an obturator or can support a fixed dental bridge in conjunction with a free flap. The latter may involve placing implants into native bone, composite free flaps or remote anchorage in the zygomatic buttress. Overall, the literature fails to demonstrate the superiority of obturation or reconstruction, as a result of unique patient presentation and a lack of data reporting standardisation.Reference dos Santos, de Caxias, Bitencourt, Turcio, Pesqueira and Goiato473 Limited evidence suggests that surgical reconstruction may offer improved QoL over prosthodontic rehabilitation.Reference Sharaf, Ibrahim, Eskander and Shaker474 Individual studies have demonstrated comparable masticatory function with implant supported obturators and surgical reconstruction, both superior to conventional obturation in low-level defects.Reference Buurman, Speksnijder, de Groot, Kessler and Rieger475,Reference Buurman, Speksnijder, Engelen and Kessler476 Implants can also be used to help retain facial prostheses, obviating the need for tissue adhesives. There are multiple case reports and case series on the use of zygomatic implants for orofacial and nasal prostheses, with good results.Reference King, Abbott, Dovgalski and Owens477
Zygomatic implants have been used for over 20 years. They can be used splinted or un-splinted. There are multiple implant designs for post-resection orofacial rehabilitation.Reference Branemark478 A recent review highlighted current evidence for the use of zygomatic implants in the midface and maxillary rehabilitation of patients with head and neck cancer.Reference Hackett, El-Wazani and Butterworth479 Overall survival rates of 77–100 per cent were reported, with few complications.
Zygomatic implant positioning for intra-oral prostheses should be restoratively driven.Reference Pellegrino, Basile, Relics, Ferri, Grande and Tarsitano480 Involvement of the restorative dentistry consultant in planning and surgical placement should reduce the risks of malposition and improve the potential for rehabilitation (Table 3).Reference Ramezanzade, Yates, Tuminelli, Keyhan, Yousefi and Lopez-Lopez481
Table 3. Peri-treatment
OMFS = oral and maxillofacial surgery; CRD = consultant in restorative dentistry; SLT = speech and language therapist
Post-treatment
Some of the most commonly reported issues for patients during the early post-treatment phase are dry mouth, problems with chewing or eating, dental health and teeth issues, fear of recurrence, and salivation problems.Reference Rogers, Alvear, Anesi, Babin, Balik and Batstone482 As the late effects of treatment develop, oral and dental concerns are a noted priority for patients in most patient-related outcome measures. Ongoing management of dental issues is, therefore, essential.Reference Rogers and Barber483
For some patients, placement of osseointegrated implants can positively impact health-related QoL outcomes.Reference Korfage, Schoen, Raghoebar, Roodenburg, Vissink and Reintsema484
Post-surgery
Following maxillofacial surgery for head and neck cancer, most patients will have residual dental or orofacial derangement in their anatomy (Table 4).Reference Dholam and Gurav485 This may be a result of pre-surgical dental extractions, and/or the removal of teeth and hard or soft tissues to achieve clear margins. Prosthetic replacement of teeth and associated hard and soft tissue may be achieved using conventional or implant-retained prostheses. Dental implants have been shown to be an appropriate treatment modality.Reference Javed, Al-Hezaimi, Al-Rasheed, Almas and Romanos486 Smokers, patients who cannot maintain adequate oral hygiene and those with a history of previous osteoradionecrosis are less suitable for implant treatment because of an increased risk of complications.
Table 4. Post-maxillofacial surgery
OMFS = oral and maxillofacial surgery; CRD = consultant in restorative dentistry
A post-resection assessment should be undertaken – ideally in formal joint restorative dentistry and maxillofacial clinics – to discuss prosthetic oral rehabilitation, and to plan any pre-prosthetic surgery such as vestibuloplasty, the release of tethered flaps and the use of free gingival grafts needed to facilitate prosthetic rehabilitation. Second-stage surgery to expose primary implants will also be planned at this stage.
Where primary implants have not been placed, secondary intra-oral implants may now be planned and placed in native (maxillary, zygomatic or mandibular) bone, or in grafted bone. Implants may also be used extra-orally for orofacial prostheses.
In a scoping literature review, the pooled five-year survival rate for primary placed implants was 92.8 per cent (95 per cent confidence interval (CI) = 87.1–98.5), and for secondary placed implants it was 86.4 per cent (95 per cent CI = 77.0–95.8), demonstrating a higher rate for primary placement. A higher survival rate has been reported for primary implants when compared to secondary implants;Reference Alberga, Vosselman, Korfage, Delli, Witjes and Raghoebar472,Reference In 't Veld, Schulten and Leusink487 however, a confounding factor may be that less complex cases are selected for primary implant placement, whereas delayed placement is favoured in more complex cases or more extensive tumours, which inadvertently affects survival rates. In a systematic review by Barber et al.,Reference Barber, Butterworth and Rogers488 primary implants had a survival rate of 96–100 per cent, with a follow up ranging from 15 to 96 months.
Implants placed in grafted bone have been shown to require further surgical soft tissue manipulation because of the lack of keratinised mucosa around the implants,Reference Patel, Antov and Nixon471 but still have a survival rate of 82–100 per cent up to 12 years later.Reference Zhang, Ding, Liu, Sun, Xie and Zhou489
Dental implants can be placed before or after radiotherapy. The survival of dental implants placed 12 months before or after radiotherapy showed no significant difference,Reference Chrcanovic, Albrektsson and Wennerberg490 although those placed 6–12 months after radiotherapy have shown a minimally greater risk of failure than those delayed further.Reference Claudy, Quevedo Miguens, Celeste, Parente, Gonzalez Hernandez and da Silva491 Studies prior to 2007 show a difference in survival between implants placed in radiated and non-radiated bone, but more recent studies fail to show a difference.Reference Schiegnitz, Al-Nawas, Kaemmerer and Grotz492 In irradiated bone, the survival rates are 74–97 per cent.Reference Nobrega, Santiago, de Faria Almeida, dos Santos, Pellizzer and Goiato493–Reference Zen Filho, Tolentino and Silva Santos495 Radiotherapy doses over 55 Gy have been shown to impact on implant success,Reference Ettl, Junold, Zeman, Hautmann, Hahnel and Kolbeck496,Reference Nooh497 but bone grafting, smoking and implant positioning have an even greater impact, with higher failure rates in irradiated grafted bone.Reference Nooh497 Radiotherapy should therefore not be considered a frank contraindication to implant placement,Reference Zen Filho, Tolentino and Silva Santos495 although doses over 60 Gy should be considered with caution, especially when placing implants in the maxilla.Reference Shugaa-Addin, Al-Shamiri, Al-Maweri and Tarakji494
Placing secondary implants in irradiated patients carries the risk of osteoradionecrosis development. The results of the ‘HOPON’ (Hyperbaric Oxygen for the Prevention of Osteoradionecrosis) trial do not recommend consideration of hyperbaric oxygen for dental extractions or implant placement in irradiated mandibles.Reference Shaw, Butterworth, Silcocks, Tesfaye, Bickerstaff and Jackson498
Because of the increased time to rehabilitation in secondary implant casesReference Patel, Antov and Nixon471, and the lack of evidence that radiotherapy post-placement results in implant failure or osteoradionecrosis,Reference Koudougou, Bertin, Lecaplain, Badran, Longis and Corre499 there is an increasing trend to support primary placement where this is logistically possible. However, costs and resources such as 3D scanning and implant planning software can be a limiting factor.Reference Negreiros, Teixeira, Peixoto and Regis500
Post-(chemo)radiotherapy
Patients who were seen prior to radiotherapy will be reviewed again soon after treatment is completed given the increased risk of dental disease in the immediate post-(chemo)radiotherapy phase (Table 5). The main oral side effects of trismus, xerostomia, caries and osteoradionecrosis are assessed. Periodontal therapy prior to (chemo)radiotherapy, and periodontal maintenance thereafter, is advised.Reference Hong, Hu, Haverman, Stokman, Napeñas and Braber501
Table 5. Post-(chemo)radiotherapy
CRD = consultant in restorative dentistry; SLT = speech and language therapist
Treatment of head and neck cancer and its associated side effects can adversely affect patients’ health-related QoL.Reference Vartanian, Rogers and Kowalski502 Health-related QoL is integral for patient careReference Rogers, Semple, Babb and Humphris503 and should be regularly assessed. Information from health-related QoL questionnaires can help improve patient care and can be used in treatment decisions.Reference Vartanian, Rogers and Kowalski502 The patient will remain under the care of the restorative dentistry consultant until oral side effects have stabilised and are manageable by the patient. Discharge to primary care can then take place, with clear instructions on providing care for the patient as well as information on when to refer back, if needed.Reference Beech, Robinson, Porceddu and Batstone504
Research
Predicting and managing the oral and dental complications of head and neck cancer treatment is complex, and requires the input of highly specialised clinicians. Following the NHS Getting It Right First Time approach, Restorative Dentistry UK have produced clear guidance on best practice.465
The quality of patient care and outcomes can be improved by developing clinical networks. The Restorative Dentistry UK Head and Neck Cancer Clinical Excellence Network aims to connect consultants and specialty trainees in restorative dentistry across the UK in order to work at improving outcomes, facilitating multicentre research and audit, and reducing variation, so that the quality of, and access to, patient care are improved. Crucially, the Restorative Dentistry UK Head and Neck Cancer Clinical Excellence Network works with the other MDT clinical specialties, including ENT and maxillofacial surgery, oncology, cancer specialist nurses, dietitians, and speech and language therapists, so that outcomes are meaningful and not produced in a clinical vacuum.
Now that 100 per cent of Scottish and Welsh MDTs and over 80 per cent of English MDTs have restorative dentistry consultant input, the development of multicentre studies is possible. As part of holistic, patient-centred care, future research and audit should aim to improve dentally focused pre-surgical planning, as supported by patient-related outcome measures with the use of the Patient Concerns Inventory. A focused Patient Concerns Inventory in relation to oral and dental health would be useful. The introduction of minimum dataset collection – including the number of decayed, missing and filled permanent teeth (‘DMFT’), and implant placement and rehabilitation – will help inform future research. Service delivery and workforce data from various centres will help develop and deliver best treatment to patients of this cohort.
In the future, the role of the restorative dentistry consultant will continue to evolve to address oral impacts of emerging technologies such as robotic surgery, immune therapy and proton beam therapy. Ongoing engagement with consultants in restorative dentistry will be essential to maximise the benefit of these technologies for patients.
Key points
• Develop information services for patients and carers. Consider introducing new technology to collect routine patient self-report data on health behaviour, psychological responses to care received, outlining of key messages and outcome assessments.
• Develop decision-making tools (such as explanatory tablet applications, e.g. Patient Concerns Inventory) to aid patients in entering into discussion with the multidisciplinary team (MDT) to agree on a treatment plan.
• Collect routine psychological assessments at key points during the course of care. These indicators must be supported with dedicated and tailored interventions to prevent neglecting identified psychological distress or depression.
• Focus on the level of support and intervention that the current team can realistically provide with the current level of resources; remain cautious when introducing change, but strengthen and build upon supports already available.
• Develop more comprehensive support services by improving generic communication skills training for current staff, and ensure consistency of message-giving to patients and/or carers across the MDT.
• Introduce staff training to assist with the management of potential burnout in MDT staff; consider flexible responses including secondments, study breaks and peer-support programmes.
• Audit current psychological services applied in the head and neck cancer service; identify current usage and gaps in service, and develop forward plans to address these gaps.
• Assess current capability of specialist clinical nurse skills to support head and neck cancer patients psychologically, and introduce dedicated training and supervision programmes.
• Actively search for clinical psychology service input and negotiate improved access and response time; estimate likely demand of service.
• Consider appointing sessional input of a clinical or counselling psychologist or psychotherapist to the cancer network.
• Identify liaison psychiatry service, and negotiate referral pathway and response time.
Introduction
The patient with head and neck cancer and their carers have considerable challenges to overcome.Reference Wells, Cunningham, Lang, Swartzman, Philp and Taylor505 First, the initial meeting of the patient with the care team is regarded with foreboding by the patients and family. An MDT approach to concentrate diagnostic data collection has shown considerable advantages for patients in enhancing speed of curative or palliative interventions.Reference van Huizen, Dijkstra, van der Laan, Reintsema, Ahaus and Bijl506 A study to predict the variation of delay in initial treatment has shown that there is no simple systematic factor or sets of factors responsible, other than possibly the severity of illness.Reference Schoonbeek, de Vries, Bras, Plaat, van Dijk and Halmos507 The second major challenge is the psychological experience of the patient with head and neck cancer. This has been closely described in a systematic review and meta-synthesis.Reference Lang, France, Williams, Humphris and Wells508
In addition to the negative psychological effect of a diagnosis and treatment of this cancer, there is a recognisable economic burden, with wider implications for the patient, their family and health service when suffering mental duress.Reference Van Beek, Wijnhoven, Holtmaat, Custers, Prins and Verdonck-de Leeuw509 Although many patients appear to cope surprisingly well, a sizeable minority experience considerable psychological effects, including uncertainty about the return of cancer, disruption to daily life, a diminished self, attempts to understand the changes that occur and finding a plan forward. Treatment recovery may be hampered by mood changes, whereas longer-term psychological states may feature some months and even years following initial treatment.Reference Hammerlid, Ahlner-Elmqvist, Bjordal, Biörklund, Evensen and Boysen510 The field of clinical and health psychology has expanded in the past five years, and provides both firmer evidence and more diverse approaches for care teams to explore and incorporate enhanced service features. This chapter expands and supports the previous edition.
Communication of diagnosis and treatment
Evidence from areas of treating cancer at other sites has demonstrated clearly that the way in which the diagnosis is presented to the patient is important to their psychological response to the disease and treatment.Reference Badr, Carmack and Diefenbach511,Reference Back, Arnold, Baile, Fryer-Edwards, Alexander and Barley512 It is important that the patient is told explicitly that they have a cancer and its nature is described, and that all treatment available is presented to them in an unambiguous manner. This information needs to be relayed consistently by all members of the team, so that the patient and carer are able to adapt, especially to be sensitive in the relaying of ‘bad’ news. This needs to be closely exercised, as this is often the first contact the patient has with the head and neck team. The initial contact has great impact, over and above the actual time spent. Evidence shows that delivering information without interruption, avoiding jargon and showing appropriate empathy are important features of the diagnostic interview that help prevent illness concerns developing.Reference Back, Arnold, Baile, Fryer-Edwards, Alexander and Barley512 Decision-making and designing of tools to improve communication between clinician and patient are improving rapidly, and highlight an important growth area for the future of head and neck cancer care where complex choices are discussed and commitments made with patients.Reference Newton513,Reference Austin, Mohottige, Sudore, Smith and Hanson514 Surgery, a major treatment modality, has received considerable attention regarding how to assist patients in coping with procedures. Providing early psychological support and frequent distress screening were features identified for improving outcomes.Reference Grimmett, Heneka and Chambers515
Delivering information about treatment and recovery
Considerable efforts have been expended to determine the information needs of head and neck cancer patients.Reference Chen, Lai, Liao, Chang and Lin516,Reference Humphris and Ozakinci517 Poor satisfaction with information supplied by the team was predictive of patients’ lowered mood and quality of life (QoL) in the longer term.Reference Llewellyn, McGurk and Weinman518 More information was required on financial advice, support groups and ability to return to work. Virtually no studies have been reported on patients’ desire to be involved in treatment decision-making. The nature of the disease and its complex profile of mixed treatment methods have favoured the MDT's sole authority to determine treatment regimens. However, large datasets of ‘normative’ QoL estimates linked to various treatment options have been compiled, which enable the team to start sharing the potential risks and benefits of certain treatment packages, and tailoring to patient preferences of retained functions on recovery.Reference Ethunandan, Rennie, Hoffman, Morey and Brennan519
Managing psychological distress
The use of routine assessments for psychological distress such as the Distress Thermometer and the Hospital Anxiety and Depression Scale are being considered as means to identify those patients who may suffer during the process of treatment preparation, the treatment itself, the initial stages of recovery and the follow-up out-patient appointments.Reference Mitchell520 These assessments have the ability to capture those patients who would not necessarily be identified by the MDT as needing psychological support.Reference Semple, Dunwoody, Kernohan and McCaughan521 Two issues are raised, however: an increased number of patients in need of assistance; and screening measures that may indicate substantial distress when there is none because of measurement error. Hence, it is recommended as essential that service heads organise links with local health service providers to input directly into the MDT and create a referral pathway.
The types of psychological distress require attention and definition. The classical typology of mental distress includes anxiety and depression. In addition, assessments of recurrence fears (the most frequently reported concern of head and neck cancer patients), facial disfigurement, body image, loneliness and sexual dysfunction may also be compiled within an MDT assessment profile library for occasional use when required.Reference Hodges and Humphris522,Reference Hagedoorn and Molleman523 Recurrence fears have been found to be linked closely to depression in patients, and some evidence exists that patients can stimulate these fears in their carers.Reference Rogers, Scott, Lowe, Ozakinci and Humphris524 Furthermore, it is now recognised that high recurrence fears promote more requests for medical services, incurring higher treatment and surveillance costs.Reference Williams, Pearce and Smith525 Acknowledgement of the patient's experience of the severity and longevity of these fears is important, and more in-depth approaches may be required to alleviate debilitating distress.Reference Groß, Nitzsche, Gloede, Ansmann, Street and Pfaff526 It is widely accepted that the patient's treatment and tumour characteristics are not good predictors of who will experience high fear of cancer recurrence levels.Reference Casswell, Gough, Drosdowsky, Bressel, Coleman and Shrestha527,Reference Simard, Thewes, Humphris, Dixon, Hayden and Mireskandari528 The exception to this generality is patient age. The young patient reports a greater reported fear of cancer recurrence level.Reference Lim and Humphris529 Gender may also be implicated, but less so.Reference Pang and Humphris530,Reference Rogers, Monssen, Humphris, Lowe and Kanatas531 Two types of patient experience of fear of cancer recurrence have been found in the six months following diagnosis. The first is a ‘low declining’ group and a sizeable minority (20 per cent) of ‘high stable’ fear of cancer recurrence sufferers.Reference Deuning-Smit, Custers, Miroševič, Takes, Jansen and Langendijk532 This fear of cancer recurrence characterisation of head and neck cancer patients strongly suggests that regular assessment of fear of cancer recurrence using a brief assessment is indicated, especially as patients tend not to volunteer their concern unless explicitly questioned.Reference Ozakinci, Swash, Humphris, Rogers and Hulbert-Williams533
The profile of staff expertise and skills needs close inspection to enable a flexible and tailored matching of needs to professional training of support or specialist staff. The MDTs need to plan their services to provide an escalating level of care according to the specific psychological difficulties presented by the patient. Stepped-care approaches are being developed and tested.Reference Butow, Lebel, Shaw and Humphris534 Initial support and educational approaches can be offered widely by the MDTs, with brief structured interventions provided by staff with additional training or a mental health qualification (counselling) to those patients with an identifiable psychological problem. More extensive interventions for patients with complex psychological difficulties can be offered, usually by referral to clinical psychologists, psychotherapists and liaison psychiatrists. A recent review and meta-analysis of fear of cancer recurrence structured interventions has been conducted.Reference Tauber, O'Toole, Dinkel, Galica, Humphris and Lebel535 The effect size of the Adjustment to the Fear, Threat or Expectation of Recurrence (‘AFTER’) intervention designed for patients with head and neck cancer is comparable to other well developed but more general programmes (‘Conquer Fear’ and Survivors’ worries of recurrent disease (‘SWORD’) interventions). The services offered would consist of complex psychotherapeutic approaches. A simple example of a stepped-care pathway is presented (Figure 1) for those patients with moderate or extensive fears of cancer recurrence. The pathway is triggered by simple and regular assessments using a four-question patient-rated outcome measure called the Fear of Cancer Recurrence scale (‘FCR4’) (Figure 2).Reference Humphris, Watson, Sharpe and Ozakinci536
Figure 1. Example of a stepped-care pathway for managing patients’ fear of cancer recurrence (FCR). FCR4 = four-item Fear of Cancer Recurrence scaleReference Humphris, Watson, Sharpe and Ozakinci536
Figure 2. ‘FCR4’ – four-item Fear of Cancer Recurrence scale.Reference Humphris, Watson, Sharpe and Ozakinci536
The changes in service delivery precipitated by the coronavirus disease 2019 (Covid-19) pandemic, some of which may continue, can produce feelings of isolation and abandonment for those diagnosed with head and neck cancer.Reference Dermody and Shuman537 Close attention is needed to offset the use of triage procedures due to the virus or delays in diagnostic reports. Successful communication of the team members becomes even more vital to provide consistency of messaging to patients and carers.
Family and social support
Developments are progressing to design interventions that assist communication between patient and carers, with initial results indicating success.Reference Gremore, Brockstein, Porter, Brenner, Benfield and Baucom538 It is important for the MDT to raise survivorship issues with patients.Reference Badr, Yeung, Lewis, Milbury and Redd539 Not only does the patient remain watchful for indicators and symptoms that may raise concern for life-reducing disease processes, but also to maintain function for as long as possible. Two areas are pertinent here. First, carers and spouses should be encouraged to use techniques to enhance the adherence of follow-up MDT recommendations. Second, and closely related, is the use of social media to link other members of the local community with similar health conditions and survivorship concerns, who can share information and provide extended social support outside the hospital boundaries. Finally, use of brief telephone contact to assist fear of cancer recurrence concerns, for example, may be considered cost-effective,Reference Davidson, Malloch and Humphris540 and, in some cases, online psychological interventions may be offered as a way to reduce face-to-face contact.Reference Yang, Zhang, Li, Liu, Liu and Li541
End-of-life issues
Communication with the patient assumes even greater importance when curative treatment options are not available and care focuses towards a palliative approach.Reference Sciubba542 Areas such as assessing patient preferences concerning life expectancy and control of pain, and managing fears of uncertainty and family reactions, are features of these discussions with the staff of the MDT and palliative care services.
Pressure on multidisciplinary teams and staff burnout
The psychological burden to staff requires recognition, supervision and training. A recent qualitative study identified themes that describe staff experiences in providing a comprehensive service to patients.Reference Gibson, O'Connor, White, Baxi and Halkett543 The authors recommended advanced communication skills training, trauma sensitivity training and self-compassion. Excellent leadership qualities are essential in running such MDTs.Reference Brenner, Hickson, Boothman, Rushton and Bradford544 This field requires further study and development to enable staff to maintain their exemplary service levels.Reference Alabi, Hietanen, Elmusrati, Youssef, Almangush and Mäkitie545
Key points
The key points concerning the role of specialist palliative care teams within the provision of palliative care are:
• Integration of a multidisciplinary, person-centred approach to care (good practice point (G))
• Liaison with both primary care and specialist palliative care teams (G)
• Clear communication of treatment options to facilitate decision-making on the treatment pathway (G)
‘Palliative care is an approach that improves the quality of life of patients and their families who are facing problems associated with life-threatening illness. It prevents and relieves suffering through the early identification, correct assessment and treatment of pain and other problems, whether physical, psychosocial or spiritual’. (World Health Organization (WHO), 2020)546
Introduction
The WHO's encompassing definition of palliative care outlines the aim to achieve the best quality of life for patients and their families.546 This is especially pertinent for those with head and neck cancer who often have numerous and complex palliative care needs, including a high degree of symptom burden. Every healthcare professional has a role and responsibility towards providing palliative care. In order to offer a collaborative approach, each member of the multidisciplinary team (MDT), including community healthcare practitioners, should have core palliative care knowledge and skills.
The following guidelines provide an overview of the key palliative care issues related to those with head and neck cancer. These include:
• The role of specialist palliative care teams within the provision of palliative care
• Management of common physical symptoms for those with incurable head and neck cancer
• Bleeding, airway and wound management
• Principles involved in advance care planning including resuscitation and care for the dying patient
• An overview of the key areas where further research is needed
Within these guidelines, we have used the American Society of Clinical Oncology definition of ‘advanced cancer’, namely ‘those with distant metastasis, late-stage disease, cancer that is life limiting, and/or with prognosis of 6 to 24 months’ as a general reference point.Reference Ferrell, Temel, Temin, Alesi, Balboni and Basch547
How palliative care should be delivered
Palliative care for those with head and neck cancer, in its broadest remit, should be delivered within a multi-professional context. As a minimum, this should include a surgeon, oncologist, specialist palliative care clinician, clinical nurse specialist, dietitian, speech and language therapist, emotional support team, and community team, all embedded within a clear communication framework across the acute and community settings. Developing effective working relationships, pathways and closer integration of specialist palliative care teams and the referring surgical and oncology teams is imperative.Reference Ferrell, Temel, Temin, Alesi, Balboni and Basch547–Reference Quill and Abernethy549 Models within the in-patient and out-patient setting have been developed,Reference Ferrell, Temel, Temin, Alesi, Balboni and Basch547,Reference Hui, Mori, Watanabe, Caraceni, Strasser and Saarto550 e.g. ‘co-rounding’ of specialist palliative care and oncology teams.Reference Meijuan Yang, Kanesvaran, Hui-Shan Neo, Chung Pheng Yee, Block and Bun Cheung551 When there is a shift in treatment intent from curative to incurable, this should correlate with a directional move of care delivery from the surgical and oncology team to include the specialist palliative care MDT.
Open and honest communication is fundamental to enable optimal delivery of palliative care, as advanced head and neck cancer can be unpredictable, necessitating complex decisions with uncertain outcomes. Discussions should be centred on prognostication information, and the establishment of treatment goals based on patient and family priorities. Research has demonstrated that head and neck cancer patient preferences and clinicians’ priorities are often out of line,Reference Lin, Kang, Donermeyer, Teknos and Wells-Di Gregorio552 which accentuates the need for skilful communication by the healthcare team to explicitly inform patients of anticipated benefits and burdens of available treatments. Palliative care delivery should be underpinned by the principles of person-centred care, shared decision-making and respect for patients, given the vulnerability of this patient population.Reference North, Carson, Sharp, Patterson and Hamilton553
When specialist palliative care teams should be involved
The importance of timely identification of head and neck cancer patients who may benefit from specialist palliative care services is widely recognised.Reference Lin, Kang, Donermeyer, Teknos and Wells-Di Gregorio552,Reference Mayland, Ho, Doughty, Rogers, Peddinti and Chada381 The key challenges for clinical teams are identifying who should be referred to specialist palliative care services and determining when is the optimal time (Table 1). When referrals are too late, head and neck cancer patients can be denied the full benefit of specialist palliative care, including timely symptom management and advance care planning conversations.Reference Hoesseini, Offerman, van de Wall-Neecke, Sewnaik, Wieringa and Baatenburg de Jong554 Conversely, referrals that are too early may result in patients with few concerns being assessed, thus inappropriate use of a specialist resource.Reference Hui, Mori, Watanabe, Caraceni, Strasser and Saarto550
Table 1. Reasons to prompt referral for specialist palliative care team involvement
Common physical symptoms
Within this section, we have focused on pain, nausea and vomiting, constipation, and the management of secretions. This is not an exhaustive list; for other details about specific symptoms and medications, we would advise reference to texts such as the Palliative Care Formulary Reference Wilcock, Howard and Charlesworth555 and British National Formulary.Reference :556 As the evidence base has limitations, we have provided advice for commonly used medications. Consultation with the local pharmacy and specialist palliative care teams is advisable in case of regional variations. Within specialist palliative care teams, medications that are ‘off-label’ or used in unlicensed ways are generally accepted. Note that ‘unlicensed medicine’ refers to a medicinal product that does not have a UK marketing authorisation;Reference Wilcock, Howard and Charlesworth555 ‘off-label use’ refers to the use of a medicine with a UK marketing authorisation for an indication outside of its licensing.Reference Wilcock, Howard and Charlesworth555
Pain
Key recommendations for the management of pain:
• A detailed, individualised pain history should be taken (evidence-based recommendation (R))
• Awareness of the signs and symptoms of opioid toxicity is important. This may include drowsiness, confusion, vivid dreams, hallucinations, myoclonus and pin-point pupils, occurring before respiratory rate reduces and level of consciousness decreases (good practice point (G))
• Consider the addition of adjuvant medications, local analgesics and disease-modifying treatments when managing pain (G)
• First-line adjuvants include gabapentin, pregabalin and amitriptyline (R)
• Review and reassess effectiveness (G)
An estimated two-thirds of patients with advanced head and neck cancer have severe pain requiring management with opioids (Table 2).Reference Chua, Reddy, Lee and Patt557 The pain is often multifactorial, relating directly to the tumour and/or occurring as a result of treatment. Initiating analgesia can be guided by using the WHO analgesic ‘pain ladder’, a simple three-step approach to managing cancer pain (Figure 1).558,Reference Fallon, Dierberger, Leng, Hall, Allende and Sabar559
Figure 1. The World Health Organization analgesic ladder.558 *Note that recent evidenceReference Fallon, Dierberger, Leng, Hall, Allende and Sabar559 supports that step 2 may be omitted. An individualised approach should be taken, tailored to the patient.
This approach, however, must be used as a general guide only, alongside an individualised assessment of pain and a management plan tailored to the individuals’ specific needs. In practice, especially for those with severe uncontrolled cancer pain, it would be appropriate to progress directly to step 3 and prescribe a strong opioid, with steps 1–2 being omitted.Reference Caraceni, Hanks, Kaasa, Bennett, Brunelli and Cherny560 A recent open-label randomised, controlled trialReference Fallon, Dierberger, Leng, Hall, Allende and Sabar559 supported a two-step approach (omitting the weak opioid step) as an alternative option for cancer pain management, which is associated with few side effects and is cost-effective.
Assessing pain
A detailed pain assessment should be conducted (Table 2).Reference Hui and Bruera561 This can help identify the primary cause of the pain and in turn indicate which treatment(s) might be most effective, including the use of non-pharmacological interventions and disease-modifying treatments. Consideration should be given to the overall holistic needs of the patient, as additional factors such as psychological, social and spiritual problems can influence the patient's pain experience and response to treatment, a concept referred to as ‘total pain’.Reference George, Minello, Allano, Maindet, Burnod and Lemaire562
Types of pain and their management
Nociceptive pain may be present as a result of direct tumour invasion of soft tissue and bone. Oral morphine would usually be the first choice because of familiarity and availability (Tables 3 and 4).Reference Hanks, Conno, Cherny, Hanna, Kalso and McQuay563,564 It is important to explain clearly to patients the difference between the immediate release and modified release preparations, as confusion can lead to opioid toxicity (Table 3).
Table 3. Commonly prescribed short-acting opioids
Note: when initiating opioids, make patients aware of the associated driving regulations.Reference Prichard, Norton and Bharucha574 CSCI = continuous subcutaneous infusion; IV = intravenous; eGFR = estimated glomerular filtration rate (in ml/minute/1.73 m2)
Table 4. Commonly prescribed long-acting opioids*
Note: when initiating opioids, make patients aware of the associated driving regulations.564 *Including syringe drivers, which although are short-acting preparations, are administered over 24 hours and so provide long-acting pain relief. NG = nasogastric; CSCI = continuous subcutaneous infusion
Neuropathic pain may be present because of the compression or infiltration of nerves. Opioids may provide limited relief for such pain, and the use of an adjuvant (a drug that has a primary indication for something other than pain, but which can have an analgesic effect) may be required. First-line adjuvants would be amitriptyline, gabapentin or pregabalin (Table 5).Reference Mishra, Bhatnagar, Goyal, Rana and Upadhya565,566 The choice of medication can be influenced by factors such as side effect profile, contraindications, patient co-morbidities and dosing regimen.
Table 5. Commonly prescribed neuropathic pain agents
o.d. = once a day; t.d.s. = three times a day; b.d. = twice a day
Oral mucosal pain is common in head and neck cancer patients, but there is a lack of robust evidence supporting the management of oral mucositis.Reference Saunders, Rouleau, Cheng, Yarom, Kandwal and Joy567 Treatment should focus on the underlying cause, maintaining oral hygiene, providing pain relief, protecting ulcerated areas and treating any secondary infection. Topical anaesthetics such as benzydamine and/or oxetacaine can be used. The regular application of coating agents (such as Orabase™ and Gelclair™) may provide some short-term relief. The use of topical opioids can be helpful, although immediate release morphine (oral solution, which contains alcohol) may exacerbate oral pain. Some centres can prepare a special-order morphine, which can be held in the mouth for a local analgesic effect. Otherwise, consider the use of oxycodone liquid, as this is alcohol-free.
Patients may have other types of pain related to their cancer, including visceral pain from metastatic disease. The same approach to pain assessment should be taken, with additional consideration given to specific types of pain. For example, patients with liver capsule pain may benefit from a trial of dexamethasone. Pain management can be complex and supported through discussion with the specialist palliative care team, including escalation to medications used under specialist supervision such as methadone or ketamine, if required.
Nausea and vomiting
Key recommendations for the management of nausea:
• Identification of the underlying cause(s) can help direct management (good practice point (G))
• Non-pharmacological and pharmacological methods should be used to manage nausea and vomiting (G)
• Choose the most suitable anti-emetic depending on the likely cause(s) (G)
• Consider the most appropriate route of administration, formulation and dose (G)
Nausea is defined as an ‘unpleasant sensation of the need to vomit’, presenting with or without vomiting,568 and often in conjunction with other autonomic symptoms, e.g. cold sweats, tachycardia and diarrhoea. Persistent nausea and frequent vomiting significantly reduce quality of life, and can result in complications like dehydration and malnutrition.Reference Leach569 In a similar approach to pain management, comprehensive assessment is needed to identify and treat the cause(s), and both non-pharmacological (Table 6) and pharmacological methods should be used.
Table 6. Non-pharmacological interventions to manage nausea and vomiting
Many different causes can contribute to nausea and vomiting in those with head and neck cancer (Table 7).568–570 Identification of the underlying cause(s) can help direct pharmacological management (Table 8).Reference Wilcock, Howard and Charlesworth555,568,Reference Leach569,Reference Smyth571 We have not specifically mentioned chemotherapy- or radiotherapy-induced nausea and vomiting, or post-operative causes, as local management guidelines are often in place.
Table 7. Choosing most suitable anti-emetic by possible cause of nausea and vomiting
Table 8. Common choices of anti-emetics used in pharmacological management of nausea and vomiting
t.d.s. = three times a day; NG = nasogastric; 5-HT = 5-hydroxytryptamine; IV = intravenous; CSCI = continuous subcutaneous infusion; max = maximum; o.d. = once a day; GABA = γ-aminobutyric acid
Generally, the oral route of administration may not be available or appropriate (if there is persistent nausea or vomiting); therefore, the subcutaneous route and continuous subcutaneous infusions can be used. Intravenous and intramuscular administration is usually avoided because of issues with venous access and pain on or after administration. In patients unable to swallow or who have enteral feeding tubes, the NEWT GuidelinesReference Smyth571 can be used to help identify alternative routes and administration methods.
Constipation
Key recommendations for the management of constipation:
• Assess stool frequency and consistency, to determine whether stimulant and/or softener required (good practice point (G))
• Assess patient's previous experience of laxatives and tolerability (taste, volume, frequency) (G)
• When initiating opioids, also prescribe a regular low-dose stimulant or osmotic agent (G)
Constipation is a distressing symptom, and those with head and neck cancer are particularly vulnerable because of dietary factors (struggling to obtain adequate fibre or hydration), and iatrogenic causes (opioid analgesia, serotonin antagonists for nausea, and chemotherapeutic agents). Favourable management of constipation is preventative, ensuring patients are adequately provided with laxatives if requiring constipating therapies.Reference Larkin, Cherny, La Carpia, Guglielmo, Ostgathe and Scotte572,Reference Muller-Lissner, Bassotti, Coffin, Drewes, Breivik and Eisenberg573 Individual patient needs should be considered, so if a patient has prior experience with laxatives, the preferred agent may be revealed. Otherwise, consider factors such as available routes and tolerance of volume for your patient. Generally, continue with one agent from each class and titrate (Table 9). It is worth noting that an additional class of ‘bulk-forming’ laxatives exist, but are not generally recommended in palliative care. Suppositories may be required, followed by enemas, to prevent faecal impaction in severe cases.
Table 9. Commonly prescribed types of laxatives
b.d. = twice a day; o.d. = once a day
Opioid-induced constipation is common, so, when initiating opioid treatment, also start a regular low-dose stimulant or osmotic agent, and titrate to affect.Reference Prichard, Norton and Bharucha574 If constipation persists beyond a few days despite regular laxatives, a rectal examination with or without enema is the next step. If these methods fail, specialist palliative care advice can be sought for use of peripheral opioid antagonists such as naloxegol and methylnaltrexone.Reference Muller-Lissner, Bassotti, Coffin, Drewes, Breivik and Eisenberg573,Reference Prichard, Norton and Bharucha574
Secretions
Key recommendations for the management of secretions:
• Identification of the underlying cause(s) of secretions will help identify potentially reversible causes of the secretions (good practice point (G))
• Detailed assessment of the types of secretions is essential, as thick or thin or secretions at the end of life are managed differently (G)
• Non-pharmacological methods should be considered to help manage all types of secretions (G)
Although xerostomia (dry mouth) is common in head and neck cancer patients, excess secretions and/or the inability to swallow or clear secretions are often troublesome.Reference Bomeli, Desai, Johnson and Walvekar575,576 Potentially reversible causes should be considered (Table 10).Reference Wilcock, Howard and Charlesworth555,576,577
Table 10. Choosing most suitable treatment by possible cause of secretions
IV = intravenous; H2 = histamine type 2; PPI = proton pump inhibitors
Reduced salivary production, xerostomia and thick salivary secretions are the most commonly observed complication of radiotherapy.Reference Bomeli, Desai, Johnson and Walvekar575 The management of thick, tenacious secretions involves largely conservative measures, such as suctioning, using pineapple juice as a lytic agent, employing a cough assist device, and reducing or stopping the medication potentially causing the secretions if appropriate.Reference Bomeli, Desai, Johnson and Walvekar575,Reference McGeachan and McDermott578 Humidification, sodium chloride nebules and mucolytics can be considered as treatment for these secretions. Sodium chloride nebules can help loosen the secretions; however, they are unsuitable for patients who cannot expectorate, as it can increase production of liquid sputum. Mucolytics can reduce the viscosity of the secretions, especially in the context of chronic obstructive airways disease (Table 11).576,Reference McGeachan and McDermott578,579
Table 11. Choice of medication to help with thick, tenacious secretions
q.d.s. = four times a day; t.d.s. = three times a day; b.d. = twice a day; o.d. = once a day
Drooling or sialorrhoea
The role for palliative care in the management of drooling or sialorrhea will be primarily for patients who have not previously received radiotherapy. Despite the burden of excessive saliva accumulation (sialorrhea) and subsequent drooling having a significant negative impact on the patient's quality of life, there is little research to guide management.Reference Bomeli, Desai, Johnson and Walvekar575 Pharmacological management would include antimuscarinics or tricyclic antidepressants (Table 12).Reference Bomeli, Desai, Johnson and Walvekar575 Occasionally, under specialist supervision, botulinum toxin injection could be considered if standard treatment is ineffective.Reference Bomeli, Desai, Johnson and Walvekar575
Table 12. Choice of medication to help with sialorrhoea
Secretions at end of life
Noisy secretions caused by fluid collecting in the upper airways, also known as the ‘death rattle,’ occur in about 50 per cent of dying patients.Reference Wilcock, Howard and Charlesworth555,Reference Wee and Hillier580 If unconscious, this is unlikely to be distressing to the patient, but can cause distress to family and friends witnessing it, so reassurance is important.Reference Wilcock, Howard and Charlesworth555,576,Reference Wee and Hillier580 Non-pharmacological treatment such as positioning, suctioning and mouthcare can help.576,Reference Wee and Hillier580 Antibiotics may be indicated for symptom relief if the secretions are caused by profuse purulent sputum in a semi-conscious patient.Reference Wilcock, Howard and Charlesworth555,576 Pharmacological treatment should be initiated as soon as the first sign of secretions occurs, as they reduce further secretion production, but do not dry out existing secretions.Reference Wilcock, Howard and Charlesworth555,576 A Cochrane review suggests subcutaneous antimuscarinics are of limited benefit, but established practice accepts their use for patients approaching the end of life (Table 13).Reference Wilcock, Howard and Charlesworth555,576,Reference Wee and Hillier580 If the rattle is associated with distressing breathlessness in a semi-conscious patient, supplement the recommendations above with an opioid and an anxiolytic sedative.Reference Wilcock, Howard and Charlesworth555
Table 13. Choice of antimuscarinic medication for secretions at end of life
Note: The dose of the syringe driver should be increased if two or more ‘when required’ doses are needed. CSCI = continuous subcutaneous infusion
Bleeding, airway and wound management
Key recommendations for bleeding, airway and wound management in the palliative context:
• Rule out any reversible cause, e.g. infection or contributing medications, and consider whether oncological or surgical intervention remains appropriate (good practice point (G))
• Be proactive in initiating advance care planning discussions, especially where risks of acute events could arise (G)
• In an acute, potentially life-threatening event, it is vital that the patient is never left alone (G)
• Manage distress and associated symptoms with both pharmacological and non-pharmacological measures (G)
Generally, it is important to identify individuals with head and neck cancer who are at risk of acute, potentially life-threatening events such as a bleeding episode or airway difficulties. These types of events are estimated to occur for 6–11 per cent of head and neck cancer patients.Reference Mayland, Ingarfield, Rogers, Dey, Thomas and Waylen581 Identification of the risk enables timely discussions to occur within the MDT about whether further investigation and interventions may be indicated, such as embolisation or radiotherapy.
Planning ahead
It is important to proactively clarify whether specific anti-cancer treatments are options, and to conduct conversations surrounding the focus and limitations of further care, and the likelihood of an acute and terminal event. These discussions need to be had with the patient (where possible), and their family or caregivers. Airway difficulties and bleeding are frightening situations, and being told there is ‘nothing that can be done’ does not alleviate anxiety. Rather, an explanation of symptom management as an active way of providing care should be undertaken. The aim is to understand the patient's priorities and care needs, and to develop an individualised care plan in the case of a significant event. In situations involving bleeding, it is important to acknowledge that not all patients with disease near major arteries will suffer a terminal haemorrhage. This topic should be discussed sensitively, emphasising that any plans are in case of a bleed, rather than this being a certainty.
Patients and carers may be traumatised by an acute episode of airway compromise or bleeding. If the patient survives, fear of further episodes may impact future care plans, including place of care and death. If the event was precipitated, for example by infection, thick secretions or poor tracheostomy care, addressing these issues may help prevent another episode. It is essential that relevant healthcare professionals are made aware of the likelihood of a further event, and know what actions to take in this situation. For those patients returning to their own home, discussions with community services, such as their general practitioner and district nurses, are imperative.
Witnessing an acute death from airway compromise or terminal haemorrhage can be very distressing for family members and healthcare professionals alike, and it is important to offer a debrief and ongoing support.
Bleeding
The identified risk factors for minor and potentially terminal bleeds are shown in Table 14.
Table 14. Risk factors for bleeding
Note: there may be no warning at all. CT = computed tomography
Minor bleeding
A minor or ‘herald’ bleed resolves spontaneously or with minimal intervention. They often, however, signal a risk of ongoing acute events, including terminal bleeding, and should trigger advance care planning discussions. Several factors and treatments should be considered to reduce the risk of bleeding and taken into account when treating minor bleeds (Table 15).
Table 15. Prevention and treatment of minor and major bleeding
NSAIDs = non-steroidal anti-inflammatory drugs; IV = intravenous; IM = intramuscular
Terminal haemorrhage
Three to four per cent of head and neck cancer patients experience a terminal haemorrhage, where the volume of blood loss causes circulatory collapse.Reference Harris and Noble582 In the majority of terminal haemorrhages, the patient will lose consciousness before any medication can be administered or have time to take effect. Remaining with the patient for psychological support is the most important treatment (Table 15).Reference Harris and Noble582
Airway management and breathlessness
Airway management in a palliative situation, where a definitive treatment or procedure is not planned, involves measures to control symptoms. These include tracheostomy and laryngectomy care (if already in place), and the use of adrenaline nebulisers and steroids (despite a lack of supporting evidence),Reference Hardy, Haywood, Rickett, Sallnow and Good583 and are usually guided by local policies. Additionally, end-of-life guidance on managing breathlessness may have local variation, but is likely to include the use of opioids and benzodiazepines (Table 16).
Table 16. Management of breathlessness and distress
CSCI = continuous subcutaneous infusion; max = maximum; IM = intramuscular; IV = intravenous
If the patient survives an acute event but remains terminal, the use of continuous benzodiazepines and opioids (e.g. via a syringe driver) will control ongoing symptoms. If the patient survives an event and is able to take oral medication, consider benzodiazepines and/or opioids.
Complex wound management
Head and neck cancers often result in wounds that are unusually shaped, in a prominent position, are painful, result in fistulas or gross disfigurement, can bleed, become infected, or are malodorous. Once nutritional status and pressure area care are optimised, and palliative oncological or surgical intervention have been considered, a stepwise management approach can be taken:
• Exclude or confirm infection by microbiology swabbing and treat appropriately: for systemic or local infections, see local antibiotic policies; for malodour, use metronidazole, orally or topically). Infection can change a wound – make it more painful, induce colour change, cause healing to remain static or deteriorate, and worsen pain, odour or exudate. Superadded infection, particularly of a fungating wound, is an important reversible cause for minor bleeding.
• Treat the pain (as per previous subsection); consider topical opioid on a dressing and providing pain relief prior to anticipated dressing changes.
• Appropriate dressings – general principles are outlined in Table 17; refer to local tissue viability formularies.
• Psychological support for the patient or carers if disfigured. This could be a simple conversation whilst changing dressings, or more formal psychological support if available.
Table 17. Key considerations in wound management
Advance care planning and end-of-life care
Key recommendations for advance care planning and end-of-life care:
• Consider advance care planning whenever there is a ‘trigger’ or transition in care (good practice point (G))
• Prepare for end-of-life care at the earliest opportunity and discuss preferences ahead of time (G)
• Plan ahead for anticipated symptoms or events (G)
• Recognition that a person might be in the last days of life is critical to providing good end-of-life care (G)
Advance care planning in head and neck cancer patients can be challenging (Table 18).Reference Rietjens, Sudore, Connolly, van Delden, Drickamer and Droger584 Communication about illness trajectories and outcomes is critical to the initiation of advance care planning discussions. Studies of head and neck cancer patients demonstrate differences in the understanding and interpretation of information between healthcare professionals, patients and carers, alongside differences in preferences regarding openness to prognostic discussions.Reference Vukkadala, Fardeen, Ramchandran and Divi585 Whilst acknowledging that there will be differences in the desire to discuss the future, the opportunity for such a discussion may affect patients’ access to palliative care and hospice care at the end of life.Reference Paladino, Koritsanszky, Nisotel, Neville, Miller and Sanders586
Table 18. European Association of Palliative Care definition of advance care planning
ACP = advance care planning
There are many different models that support discussions relating to advance care planning; these can be helpful frameworks to support clinicians when initiating conversations. Models such as the Serious Illness Conversation GuideReference Paladino, Koritsanszky, Nisotel, Neville, Miller and Sanders586 and mnemonic frameworks (e.g. ‘REMAP’, which stands for Reframe, Expect emotion, Map out patient goals, Align with goals, and Propose a planReference Childers, Back, Tulsky and Arnold587) focus on eliciting patients’ values and preferences for care, alongside their goals of care. Advance care planning should be considered to support options for care, including the continuation of further lines of therapy, alongside less interventional approaches and end-of-life care planning.Reference Vukkadala, Fardeen, Ramchandran and Divi585 Importantly, advance care planning may not reduce hospital attendance for patients with head and neck cancer, as has been anticipated in patients with other conditions; rather, the difficulty of managing symptoms such as bleeding, airway obstruction and complex wounds in the home setting frequently necessitates hospital-based care. Crucial to introducing advance care planning into practice is the recognition of key ‘triggers’ that prompt the clinician to consider and raise conversations regarding such planning (Table 19).
Table 19. Examples of triggers for introducing conversations about advance care planning
Resuscitation
Patients with head and neck cancer often have a treatment course characterised by frequent interventions and a high level of tertiary based care.Reference Mayland, Ho, Doughty, Rogers, Peddinti and Chada381 There is some evidence to suggest these patients may be less likely than others to accept ‘Do not attempt cardiopulmonary resuscitation’ forms.Reference Parsons, de la Cruz, Zhukovsky, Hui, Delgado-Guay and Akitoye588 The integration of resuscitation discussions into structured advance care planning discussions is recommended, and may increase the readiness of patients to understand and accept the limitations of medical care.Reference Field, Fritz, Baker, Grove and Perkins589 Thus, discussions about resuscitation should be prompted by transitions in the patient journey; these may be similar triggers to those used for prompting advance care planning conversations. For those at risk of catastrophic events, this is critical, with the recognition that should such an event take place, cardiopulmonary resuscitation would be ineffective.
End-of-life care
It can be difficult to be certain that a person is in the last days of life (Table 20). Anticipating this, however, is critical to allow a person to communicate with loved ones, have the potential to be in a place of their choosing, complete any important rituals or attend to spiritual care needs. Care in the last days of life should follow National Institute for Health and Care Excellence (NICE) recommendations (Table 21).590
Table 20. Indicators that an individual is entering last days of life
Table 21. Care in last days of life
MDT = multidisciplinary team
For those approaching the last days of life with established artificial nutrition (enteral or parenteral), similar challenges to those with clinically assisted hydration can arise. It is important for patients and/or those individuals close to them to understand that there is a change in the goal of the nutritional support, from maintaining a good nutritional status, to providing comfort and symptom control. The manner in which this happens should be tailored to the individual, and their preferences and needs. For many, stopping the feed when in the last days of life allows the best balance between the burden of the feeding and the clinical benefit (which is minimal at this stage of life).Reference Orrevall591 Given the emotional and sometimes spiritual importance of feeding, however, it might be important to discuss a gradual reduction in feeding rather than an abrupt stopping, depending on the person's wishes.
Medication at end of life
‘Anticipatory’ medications should be prescribed to ensure there is no delay in the administration of a required drug, especially for a symptom or situation that can be anticipated (Table 22).590
Table 22. Summary of first-line options for anticipated symptoms at end of life*
* Refer also to previous relevant sections about symptom control
‘One Chance to Get it Right’ details the principles for caring for dying people, and highlights five key priorities for care (Table 23).592 This advocates the use of an individualised care plan in the last days of life and supports all members of the patient's care team to adopt an appropriate focus of care for the patient, alongside attention to supporting carers. Detailed plans, such as those described above, for anticipated emergencies, are considered essential.
Table 23. The five priorities for caring for a dying person
Current and future research
Over the last decade, many trials have been conducted assessing the merits of palliative care.Reference Haun, Estel, Rucker, Friederich, Villalobos and Thomas593,Reference Davis, Temel, Balboni and Glare594 Generally, it is advocated that palliative care be integrated into standard oncological care early in the illness trajectory, often alongside active anti-cancer treatments.Reference Ferrell, Temel, Temin, Alesi, Balboni and Basch547 Within an Indian healthcare context, early routine specialist palliative care input for those with head and neck cancer did not improve quality of life at three months.Reference Patil, Singhai, Noronha, Bhattacharjee, Deodhar and Salins595 Models of care, however, need to be targeted to individual populations. Questions therefore remain about the optimum model of care needed to improve outcomes for this complex population. Addressing variability in terms of clinical outcomes is essential. Geographical and socio-economic inequalities persist, and ultimately affect place of death.596
The limited evidence base for symptom control highlights the need for further work and dedicated funding. Within cancer research, less than 0.3 per cent of the £500 million is allocated to palliative care.Reference Higginson597 Clinical trials need to extend inclusion beyond those patients with good performance status, and be combined with novel methodologies to help overcome practical and ethical challenges.
Improving the general palliative care skills and knowledge of the head and neck cancer MDT, and integrating a needs-driven mechanism for access to specialist palliative care services, seem essential.Reference Mayland and Payne598 Models of integrated care, which include community providers, are pertinent to avoid ‘silo’ working,Reference Ullgren, Kirkpatrick, Kilpelainen and Sharp548 and to enhance decision-making and advance care planning, encourage mutual learning, and optimise quality of life for patients and their family carers.
Introduction
This chapter aims to describe the investigations and management for some of the specific common effects of treatment for patients treated for head and neck cancer.
Chyle leak following head and neck cancer surgery
Recommendations
• Surgery to the supraclavicular region on both sides of the neck should necessitate careful inspection for a potential chyle leak at completion (evidence-based recommendation (R))
• If an intra-operative chyle leak is identified, every effort should be made to repair or close the damaged thoracic duct at the time of surgery (R)
• Biochemical testing to be undertaken to confirm a clinically suspected post-operative chyle leak (good practice point (G))
• Monitor fluid balance, electrolytes, protein levels and white cell count (R)
• Consider non-operative measures that can be implemented to reduce chyle flow (R)
• Operative interventions will be influenced by local experience and available resources (G)
• Consideration should be given to involving interventional radiology and thoracic teams, even if not available locally for persistent leaks (G)
Introduction
Lymph returns proteins and excess interstitial fluid to the bloodstream. It also transports fats from the digestive system via chylomicrons, and is rich in electrolytes, fat-soluble vitamins, trace elements, glucose and white blood cells. Flow rates in the thoracic duct are such that damage to it can result in significant fluid shifts.
The thoracic duct originates from the cisterna chyli, at the level of the second lumbar spinal vertebra (L2), and drains lymph from the left side of the body and the right side below the diaphragm. The right lymphatic duct receives lymph from the right chest, arm, and head and neck region. The thoracic duct ascends through the thoracic cavity to enter the route of the neck, with the duct extending up to 5 cm above the clavicle before terminating into the venous circulation.Reference Delaney, Shi, Shokrani and Sinha599 Whilst this is most commonly into the left internal jugular vein (IJV), it can also drain into the confluence of the IJV and subclavian vein, the subclavian vein, or even into the brachiocephalic vein, external jugular vein or vertebral vein, or, even more rarely, to the right-sided vessels. Furthermore, whilst a single duct is most common, there may also be multiple drainage channels.Reference Brennan, Blythe, Herd, Habib and Anand600,Reference de Gier, Balm, Bruning, Gregor and Hilgers601
Iatrogenic damage to the thoracic ductal drainage system during head and neck surgery resulting in a chyle leak occurs in 0.5–1.4 per cent of thyroidectomies and in 2–8 per cent of neck dissections.Reference Delaney, Shi, Shokrani and Sinha599 The variable anatomy and fragile composition of the thoracic duct render it prone to inadvertent injury. The head and neck surgeon may also encounter a chyle leak as a result of penetrating neck trauma, cervical rib resection, or following sentinel or cervical node biopsy.Reference Brennan, Blythe, Herd, Habib and Anand600
A chyle leak into the neck can have serious consequences, in terms of additional interventions, delayed discharge and wound healing. Systemic effects of a large volume of chyle leak can result in hypovolaemia, electrolyte imbalances (hyponatraemia, hypochloraemia and hypoproteinaemia), malnutrition and immunosuppression. Local effects of a chylous collection include delayed wound healing, infection, wound breakdown and fistula formation. The presence of chyle within a contained space may decrease tissue perfusion, resulting in skin or flap necrosis. It is also possible that chyle may penetrate the chest, forming a chylothorax.Reference Delaney, Shi, Shokrani and Sinha599,Reference Brennan, Blythe, Herd, Habib and Anand600
Intra-operative diagnosis and treatment
Surgery to the supraclavicular region on both sides of the neck should necessitate careful inspection for a potential chyle leak at completion. Identifying a chyle leak can be aided by manoeuvres that increase the intrabdominal and/or intrathoracic pressure. These include Trendelenburg positioning, a Valsalva manoeuvre or abdominal compression. Meticulous haemostasis, careful drying and observation of the area can aid the detection of even small chyle leaks.
If an intra-operative chyle leak is identified, every effort should be made to repair or close the damaged thoracic duct at the time of surgery. If clearly visible, ligation of the thoracic duct can be readily achieved with clips or by over-sewing the duct. Additionally, the use of locoregional muscle flaps to cover the area should be considered. A number of these flaps have been proposed, including the clavicular head of the sternocleidomastoid, the scalenus medius and the omohyoid; one or more of these can be raised and rotated into the space (usually posteriorly to the IJV). Alternatively, a regional pedicled flap such as a pectoralis major flap can be used to fill the space. A number of glues, such as fibrin- or cyanoacrylate-based products, have been proposed to help.Reference Delaney, Shi, Shokrani and Sinha599,Reference Brennan, Blythe, Herd, Habib and Anand600 If using such products, care should be taken to isolate nerves (such as the vagus) traversing the region. Figure 1 shows a flow chart summarising the identification and management of a chyle leak following surgery for head and neck cancer.
Figure 1. Flow chart summarising the identification and management of chyle leak following surgery for head and neck cancer.
Post-operative diagnosis and treatment
Table 1 summarises the clinical and biochemical features of a chyle leak. Whilst the classical creamy or milky appearance of the drain fluid is highly indicative of a chyle leak, this may not be evident for a few days, especially if there is serosanguinous discharge or a delay in enteral feeding.Reference Erisen, Coskun and Basut602,Reference Rodgers, Johnson, Petruzzelli, Warty and Wagner603
Table 1. Clinical and biochemical diagnosis of chyle leak
The volume of chyle produced in a 24-hour period is one of the main determinants of treatment. Low output chyle leaks (less than 500 ml/day) are initially managed with conservative non-operative treatment, whilst persistent high output chyle leaks (more than 500 ml/day) will usually require operative treatment. However, volume alone is not the only determinant of treatment. Management will also be influenced by patient factors, the duration of the chyle leak and the impact of losing a large volume of electrolyte-rich fluid. Further factors to consider include the response of the drain output to treatment measures, individual surgeon preference and the range of local services available.Reference Delaney, Shi, Shokrani and Sinha599,Reference Brennan, Blythe, Herd, Habib and Anand600,Reference Campisi, Boccardo, Piazza and Campisi604 It is important to note that large fluid shifts can occur, so fluid balance and electrolytes should be monitored regularly, along with white cell counts, blood glucose and albumin levels.
Non-operative interventions
Physical activity increases chyle flow, so bed rest, with elevation of the head of the bed, along with measures to reduce straining such as the use of stool softeners, are helpful.
Dietary modifications with a fat-free, low-fat or medium-chain fatty acid diet should be instigated. Medium-chain fatty acids are largely water-soluble and are absorbed via the portal venous circulation, reducing lymphatic flow. This diet can be administered enterally or parenterally.Reference Talwar, Donnelly, Skelly and Donaldson605 For high-volume chyle leaks (more than 500 ml), parenteral nutrition can be considered as nutritional intervention.
For output of less than 500 ml, enterally fed patients should receive either a fat-free nutritional product or a medium-chain triglyceride-based feed.
For patients who are able to eat and drink a low-fat diet, supplements that are fat-free or where the fat is made up of medium-chain triglycerides should be used to meet nutritional requirements. A low-fat diet may be defined as providing less than 5 kcal per serving of food. Compliance with a low-fat diet can be low.
Pressure dressings, suction drainage and negative-pressure therapy have all been described to reduce the available space for chyle to accumulate. These measures need to be balanced against the risks of impaired flap perfusion and inducing an exudate with suction.Reference Delaney, Shi, Shokrani and Sinha599
Systemic drug therapy with somatostatin or a somatostatin analogue such as octreotide has been shown to slow chyle production and is effective in a chyle leak following a neck dissection. The timing of starting octreotide does vary, as does the treatment dose, although 100 μg subcutaneously every 8 hours is a common regimen.Reference Delaney, Shi, Shokrani and Sinha599,Reference Brennan, Blythe, Herd, Habib and Anand600,Reference Campisi, Boccardo, Piazza and Campisi604,Reference Magoo, Bhate, Santhosh Kumar, Kakodkar, Gajul and Mastud606
Sclerosing agents introduced via drainage tubes have been described. However, these should be used with caution given the potential problems if re-operation is required.
Surgical interventions
The decision to surgically explore a chylous fistula requires careful planning, and will be influenced by surgeon choice and locally available resources. There is no high-level evidence for a particular intervention.
The aim of surgical exploration of the neck is to facilitate surgical repair to stop or reduce a chyle leak, or to ligate the thoracic duct. However, as a result of the small size of lymphatic vessels, it is quite often difficult to identify specific leakage points, and there may be multiple sites responsible. As described above, identification of the leak site can be helped with manoeuvres that increase the intrabdominal or intrathoracic pressure, including: Trendelenburg positioning, Valsalva manoeuvre or abdominal compression.Reference Delaney, Shi, Shokrani and Sinha599
Closure of the supraclavicular space can be aided by locoregional flaps. The use of sclerosing agents, adhesive agents or mesh can also help.Reference Brennan, Blythe, Herd, Habib and Anand600 Once again, care is required to protect other vital structures (vagus and phrenic nerves, and the brachial plexus) when using these agents.
Interventional radiology and cardiothoracic interventions
If local neck control cannot be achieved, percutaneous lymphangiography guided cannulation of the thoracic duct and its embolisation has been reported to have a high success rate. Thoracic duct ligation can be achieved by a number of techniques, including video-assisted thoracoscopic surgery, and has been found to have a high success rate. Several other procedures, including thoracotomy, pleurodesis and decortication, pericardial ‘window’, and pleura-venous and pleura-peritoneal shunts, have been described.Reference Delaney, Shi, Shokrani and Sinha599,Reference Wilkerson, Haque, Pitkin and Soon607
Osteoradionecrosis following head and neck cancer treatment
Recommendations
• For early osteoradionecrosis, pragmatic conservative and symptomatic management involves debridement of sharp and exposed necrotic bone, analgesia, antiseptic mouthwash, and antibiotics (good practice point (G))
• Minor bone spicules may heal with entirely conservative management (evidence-based recommendation (R))
• The role of rim resection and saucerisation, i.e. resection of all necrotic bone in cases of Notani stage 1 and 2 osteoradionecrosis, remains uncertain (G)
• The use of hyperbaric oxygen in the management or prevention of osteoradionecrosis is not supported by recent randomised, controlled clinical trials (R)
• The efficacy of the pharmacological therapy with pentoxifylline, vitamin E and clodronate (‘PENTOCLO’) protocol requires further evaluation in controlled trials (R)
• In patients with Notani stage 3 (advanced) osteoradionecrosis or in those with disease refractory to medical or conservative management, definitive surgical resection and reconstruction is likely to be necessary. This may result in acceptable healing outcomes, but with significant complications and compromised function (R)
Introduction
Osteoradionecrosis has been defined as ‘exposed irradiated bone that fails to heal over a period of three months in the absence of local tumour’.Reference Harris608 This is a much-feared complication of (chemo-)radiotherapy occurring in 5–10 per cent of irradiated patients.Reference Frankart, Frankart, Cervenka, Tang, Krishnan and Takiar609 This is associated with pain, infection, fistulation, haemorrhage and progressive disfigurement. In the most common site, the mandible, osteoradionecrosis is often associated with trismus, leading to malnutrition and fracture. Osteoradionecrosis may also occur in the midface, skull and even cervical spine. A parallel condition of the cartilage (chondroradionecrosis) may affect the larynx. Osteoradionecrosis may co-exist with late radiation toxicities of adjacent tissues, such as xerostomia, atrophic mucosae, fibrosis, skin telangiectasia, trismus and neuropathic pain, any of which may complicate treatment and compromise outcomes. Osteoradionecrosis is associated with significant mortality in its own right, but also is occasionally coincident with the presentation of local tumour recurrence.
Classification
Classifications of osteoradionecrosis, in particular for the mandible, have been used to categorise and manage osteoradionecrosis in clinical practice and in the context of prospective trials. Some prior classifications have depended on progression, or responses to a particular therapy, or presumed indications for one treatment over another. After a recent review of various classifications,Reference Shaw, Tesfaye, Bickerstaff, Silcocks and Butterworth610 Notani and colleagues’ classification of osteoradionecrosis has shown many pragmatic advantages.Reference Notani, Yamazaki, Kitada, Sakakibara, Fukuda and Omori611 They classify mandibular osteoradionecrosis by: cases limited to the dentoalveolar process (Notani stage 1), cases involving the body of the mandible above the inferior dental canal (Notani stage 2), and cases with full thickness, fracture or extra-oral fistulation (Notani stage 3). One recent additional category by virtue of its favourable prognosis is minor bone spicules (less than 20 mmReference Brennan, Blythe, Herd, Habib and Anand600 exposed bone),Reference Shaw, Tesfaye, Bickerstaff, Silcocks and Butterworth610 which often heal without intervention.Reference Shaw, Butterworth, Silcocks, Tesfaye, Bickerstaff and Jackson498 There are no commonly used classifications of osteoradionecrosis outside the mandible, but it is recommended that the extent and anatomy of exposed bone and involved necrotic bone are described.
Management
Conservative and supportive care
Conservative measures include: oral hygiene improvement, irrigation, antibiotic therapy, analgesia and nutritional status optimisation.Reference Rice, Polyzois, Ekanayake, Omer and Stassen612 There is a paucity of data on the natural history and course of osteoradionecrosis treated with conservative management. In advanced osteoradionecrosis, conservative approaches are neither cost-effective nor effective, and have been criticised given the lack of evidence.
Combined pharmacological therapy with pentoxifylline, vitamin E and clodronate
A better understanding of the pathophysiology of osteoradionecrosis, especially the radiation-induced fibro-atrophic theory, has resulted in diverse treatments, including attempts at medical management. Delanian et al.Reference Delanian, Chatel, Porcher, Depondt and Lefaix613 presented details of the medical management of osteoradionecrosis with a combination of pentoxifylline, and vitamin E with clodronate. Each patient was given a daily combination of twice-daily 400 mg pentoxifylline (800 mg/day) plus 500 IU vitamin E (1000 IU/day) and once-daily 1600 mg/day clodronate from Monday to Friday (5 days per week), alternated with 20 mg prednisone plus 1000 mg ciprofloxacin on the weekend (2 days per week). It was concluded that long-term pharmacological therapy with pentoxifylline, vitamin E and clodronate (‘PENTOCLO’) treatment is curative for refractory osteoradionecrosis, and induces mucosal and bone healing with significant symptom improvement. That was a phase II trial, with most patients having a defect of less than 2 cm (Notani stage 1), with the median time for healing of six months. Limited studies to date have demonstrated the effective pharmacological use of ‘PENTOCLO’ in treating osteoradionecrosis,Reference Kolokythas, Rasmussen, Reardon and Feng614 and controlled trials are essential in order to confirm these results.
Hyperbaric oxygen
Prior enthusiasm for the use of hyperbaric oxygen in the management of osteoradionecrosis arose from uncontrolled case series and a single-centre randomised, controlled trial of prevention following dental extractions.Reference Marx, Johnson and Kline615 Since then, two multi-centre randomised, controlled trials have shown no benefit of hyperbaric oxygen in the management of established osteoradionecrosis, either used aloneReference Annane, Depondt, Aubert, Villart, Gehanno and Gajdos616 or as an adjunct to surgical excision.617 A recent randomised, controlled trial of hyperbaric oxygen to prevent osteoradionecrosis following high-risk surgical procedures to the mandible showed no benefit, with the rate in both study arms of around 6 per cent, lower than had been previously reported in trials.Reference Shaw, Butterworth, Silcocks, Tesfaye, Bickerstaff and Jackson498
Surgery
Surgical interventions should be appropriate for the stage of osteoradionecrosis. Surgical procedures include sequestrectomy, marginal or rim resection (‘saucerisation’), and segmental resection, usually implying oromandibular reconstruction with a composite free flap. The goal of surgery is usually the removal of bone with compromised perfusion, whilst retaining viable bone in order to promote healing. There is no convincing evidence that either sequestrectomy or rim resection have superior outcomes to conservative or medical management. A concern following surgery is that vascularised soft tissue coverage of the resultant bony defect is often compromised. One relatively novel and untested approach worthy of further investigation is the use of marginal resection of Notani stage 1 and 2 mandibular osteoradionecrosis and subsequent revascularisation with a free flap bearing periosteum.Reference Bettoni, Olivetto, Duisit, Caula, Bitar and Lengele618 Segmental reconstruction with free flap reconstruction has acceptable microvascular patency, with a recent meta-analysis reporting patency in over 90 per cent of cases.Reference Lee, Chin, Eslick, Sritharan and Paramaesvaran619 However, healing and functional outcomes are understandably compromised in a surgical field affected by late radiation toxicity, and are associated with significant complications (40 per cent) including fistula, infection and hardware exposure.Reference Lee, Chin, Eslick, Sritharan and Paramaesvaran619 For many Notani stage 3 cases, there is sometimes no acceptable alternative than to undertake major resection and reconstruction. The outcomes of surgery can be improved by virtual planning, but will ultimately be informed by access to suitable vessels, co-morbidities and the need for occlusal rehabilitation.
Role of palliative care
In a small group of patients, mandibular osteoradionecrosis may be present in combination with: affected sites such as the base of skull, temporomandibular joint (TMJ) distraction with extradural or subdural collections, and middle-ear complications. In addition, patients with a history of extensive treatment for head and neck cancer often present with significant co-morbidities. Patients with mandibular osteoradionecrosis demonstrate reconstructive challenges associated with poor healing, soft tissue defects and vessel-depleted necks. Surgical complications such as wound infection, skin necrosis, salivary fistulae and partial flap loss, and in rare instances carotid blow-out, have also been reported. In a patient with significant co-morbidities, extensive craniofacial osteoradionecrosis may be beyond any surgical intervention, and best supportive care may be the only pragmatic option; these patients have symptoms that are very difficult to manage and they often succumb to these complications.
Overall prognosis
Patients with early osteoradionecrosis (minor bone spicules and Notani stage 1), can be managed successfully through conservative management with occasional support in an out-patient clinic. Advanced osteoradionecrosis is a debilitating condition with significant adverse effects to a patient's health-related quality of life. Early surgical intervention can contribute to a successful outcome.Reference Lee, Chin, Eslick, Sritharan and Paramaesvaran619
Trismus following head and neck cancer treatment
Recommendations
• The primary focus should be directed towards prevention by identifying high-risk patients before treatment of head and neck cancer (evidence-based recommendation (R))
• The mainstay of treatment is physiotherapy; jaw-stretching exercises with a trismus device should be commenced as soon as feasible following surgery and/or radiotherapy (RT) (R)/(good practice point (G))
• Every effort should be made to establish the cause of trismus for targeted investigations and treatment (G)
• Mouth-opening distance should be measured and recorded at each follow-up consultation, and a realistic goal for mouth opening agreed with the patient (G)
Introduction
Trismus is a common but often overlooked sequelae of head and neck cancer treatment involving surgery and/or RT that involves the masticatory system. It can have a significant impact on the physical, psychological and social well-being of patients, and is generally defined as a reduced jaw opening of 35 mm or less.Reference Dijkstra, Huisman and Roodenburg620,Reference Scott, Butterworth, Lowe and Rogers621
Trismus is associated with several potential complications that lead to a poor overall quality of life for patients. It may result in poor nutrition due to difficulty chewing and poor dentition, and periodontal disease as a result of limited access for both oral hygiene and dental treatment. Chewing and speech difficulties may cause patients to avoid eating in public, leading to social isolation, and even anxiety and depression. Restricted mouth opening may compromise cancer surveillance, and may complicate airway management during general anaesthesia for procedures that may be required down the line. Impaired clearance of oropharyngeal secretions may also lead to a risk of aspiration.Reference Scott, Butterworth, Lowe and Rogers621–Reference Kamstra, van Leeuwen, Roodenburg and Dijkstra623 It is therefore important that trismus is actively addressed and dealt with by clinicians as soon as possible (Table 2).
Table 2. Summary of management of trismus post head and neck cancer treatment
CT = computed tomography; MRI = magnetic resonance imaging; PET = positron emission tomography
Causes
There are several possible causes of trismus in patients treated for head and neck cancer. Disease progression or tumour recurrence must be considered in the differential diagnosis and excluded with utmost urgency.
More commonly, trismus may be a result of surgical resection causing scarring and tissue contracture in the masticatory apparatus, or occur following damage to the neural innervation to masticatory muscles. Surgical resection extending into the masticator space, thereby affecting muscles of mastication such as the masseter, temporalis, medial pterygoid and superior head of lateral pterygoid muscles, are particularly inclined to cause trismus. Similarly, tumour resections involving the infratemporal fossa, parapharyngeal space and TMJ may also result in trismus.
The prevalence of trismus in head and neck cancer patients treated with RT ranges from 30 to 50 per cent,Reference de Gier, Balm, Bruning, Gregor and Hilgers601 and concomitant chemoradiotherapy may be associated with a higher prevalence of trismus.Reference Abboud, Hassin-Baer, Alon, Gluck, Dobriyan and Amit622 Trismus may begin towards the end of radiation treatment, and the adverse effect of progressive fibrosis may continue for years.
It is important to consider odontogenic causes of restricted mouth opening. Odontogenic infection originating from dental (apical) abscesses or periodontal disease with bacterial spread into the masticator space may cause inflammation and pain, resulting in trismus. Head and neck cancer patients may struggle with maintaining oral hygiene, and may suffer from xerostomia, which can lead to the rapid deterioration of dentition and subsequent orofacial infections. Infected or failed reconstructive fixation metal plates and screws can cause pain and restricted mouth opening. Dislocated condyles or stress-induced TMJ dysfunction syndrome may also be the cause of restricted mouth opening.
Investigations
Thorough clinical examination may not be possible because of restricted access. Flexible nasoendoscopy is useful not only to examine the nasal cavity and rest of the upper aerodigestive tract, but also to visualise the oral cavity and oropharynx by passing the scope tip through a small gap between the incisor teeth (if present).
Any suspicion of residual disease or recurrence should be excluded with imaging modalities such as computed tomography (CT), magnetic resonance imaging or positron emission tomography-CT. Interpretation of these scans may be difficult because of post-surgical and radiation changes, which may warrant multiple imaging modalities for the accurate interpretation of radiological findings.
Examination under general anaesthesia may be required in some patients, and focused histological sampling should be considered, guided by clinical examination and radiological findings.
Management
Management of trismus requires a multidisciplinary approach with the primary focus aimed at prevention and early intervention.
Prevention
Preventative measures are important because, once the trismus is established, many patients will not return to a pre-treatment level of mouth opening.Reference Kamstra, van Leeuwen, Roodenburg and Dijkstra623 Identifying high-risk patients may help facilitate early preventative measures and intervention. Patients with tumours affecting, or in close proximity to the masticatory apparatus, should be prepared for possible restricted mouth opening following treatment. Every attempt must be made to reduce the impact of surgical resection and radiation exposure to adjacent structures, whilst balancing the importance of oncologically safe treatment. Decreasing the radiation dose delivered to the surrounding structures involved in mastication, using intensity-modulated RT, reduces the incidence rate and severity of radiation-induced trismus after RT.Reference Bensadoun, Riesenbeck, Lockhart, Elting, Spijkervet and Brennan624
Physiotherapy should begin as early as possible after surgery and/or RT, to maintain or improve jaw mobilisation. Those patients with reconstructive fixation plates should be approached judiciously. Early intervention is likely to result in better long-term outcomes.Reference Abboud, Hassin-Baer, Alon, Gluck, Dobriyan and Amit622,Reference Bensadoun, Riesenbeck, Lockhart, Elting, Spijkervet and Brennan624
Primary reconstruction of surgical defects at the time of tumour ablation is advocated, to prevent excessive scarring and contraction. Prophylactic removal of the coronoid process (where the temporalis muscle inserts) is recommended for surgery involving resection of the posterior maxilla and mandible. The temporalis muscle is at high risk of contracture and fibrosis post-surgery and/or RT, which can result in jaw tightening and closure. It may be beneficial to detach the insertions of masseter and medial pterygoid muscles as a part of the surgical resection if appropriate.
Comprehensive dental examination and restorative care, and extraction of teeth with guarded prognosis, at the time of resection surgery or prior to RT commencement, reduces the risk of odontogenic infections. Strict oral hygiene measures, a low-sugar diet and fluoride treatment are recommended to prevent dental caries and the rapid deterioration of dentition, especially in such patients with radiation-induced xerostomia. Artificial saliva, frequent sips of water or sialagogues to encourage saliva flow may be considered.
Treatment
It is important to determine the cause of trismus so that the treatment can be guided appropriately. Mouth opening should be measured and recorded regularly at follow-up consultations, and a realistic target agreed with patients.
For trismus related to surgery and/or RT, the first line of treatment is jaw mobilisation exercises using various jaw-opening devices. Jaw exercises with a vertical and horizontal range of movements should commence as soon as possible, and should be undertaken several times a day. Exercises may have to be continued for one to two years to maintain the results.Reference Abboud, Hassin-Baer, Alon, Gluck, Dobriyan and Amit622 Input from both physiotherapy and speech and language therapy teams will be invaluable in instigating this treatment.
Trismus appliances that are commonly used in dentate or partially dentate patients include wooden spatulas (tongue depressors), the TheraBite® Jaw Motion Rehabilitation System™, the Dynasplint® Trismus System and corkscrews. Readily available tongue depressors are easy to use and can be employed early in patients with restricted mouth opening. Care should be taken not to cause excessive discomfort or pain whilst using these devices, to avoid damage to dentition or soft tissue. In a randomised feasibility study of 71 patients, there was improved mouth opening in both groups of patients using either wooden spatulas or the TheraBite system at six months, with no significant difference between the two groups.Reference Lee, Yeo, Rogers, Caress, Molassiotis and Ryder625
A systematic review found large variations in the stretching techniques, duration of stretching and frequency of exercises. In the studies analysed, both preventative and therapeutic measures were found to increase mouth opening after exercise therapy. There was a considerable range in jaw-opening changes, and a no stretching technique was superior to others. Multiple studies in the systematic review reported post-stretching mouth opening, which remained at less than 35 mm, indicating that trismus persisted in many patients despite treatment.Reference Erisen, Coskun and Basut602 Other adjunctive measures, such as heat therapy, low-level laser therapy, pentoxiphylline and botulinum A toxin injection, have been reported in the literature, but overall evidence for their effectiveness is lacking.
There may be scope for surgical intervention in some patients, but the risk of complications is high in this group of patients. Consideration may be given to the excision of scar tissue and fibrotic bands (e.g. buccal mucosa bands). If mandibular coronoidectomy was not performed at the time of primary surgery, this may be considered, particularly if temporalis muscle contracture is suspected. Manipulation of the jaw under anaesthesia, followed by the maintenance of mouth opening with physiotherapy, may also be considered.
Aspiration and dysphagia following head and neck cancer treatment
Recommendations
• Functional and radiological swallowing assessment should be considered in all patients with post-treatment dysphagia and aspiration (evidence-based recommendation (R))
• Endoscopic dilatation should be offered to patients with a symptomatic pharyngeal stricture (R)
• Multidisciplinary team (MDT) input and shared decision-making with the patient must support consideration of a functional laryngectomy (R)
• Gastrostomy and tracheostomy insertion are reasonable interventions to manage laryngeal and pharyngeal dysfunction following head and neck cancer treatment (good practice point (G))
• Cross-sectional imaging should be performed prior to major surgical intervention, to exclude recurrent disease (G)
Introduction
Radiotherapy and surgery induced fibrosis can affect many patients who are cured following their head and neck cancer treatment. Laryngeal and pharyngeal dysfunction due to fibrosis leads to unpredictable levels of dysphagia, aspiration risk and airway compromise. The mainstay of managing post-treatment dysphagia and/or aspiration should be speech and language therapy and dietetic input (Chapters 9 and 10). This chapter will concentrate on potential surgical interventions to treat these treatment-related sequelae.
Tracheostomy
Tracheostomy may be required to relieve airway compromise. Conservative methods to improve the airway, such as transoral laser cordectomy or arytenoidectomy, are unlikely to be appropriate in this particular patient group as the aspiration risk may be exacerbated. A tracheostomy may worsen dysphagia,Reference Skoretz, Anger, Wellman, Takai and Empey626 causing further impairment of laryngeal elevation, and is unlikely to reduce aspiration risk, although it may be used to help manage symptomatic aspiration.
Gastrostomy
Gastrostomy is an option for patients at risk of aspiration and who struggle to meet their oral calorie intake. Reducing the need for oral intake may allow a more targeted oral diet with a lower aspiration risk. Whilst a gastrostomy may help to reduce the risk of aspiration and pneumonia for some patients with severe long-term fibrosis effects, a gastrostomy may not reduce this risk.Reference Hutcheson, Lewin, Barringer, Lisec, Gunn and Moore339
Dilatation
Dilatation is frequently performed for pharyngeal stenosis. Oropharyngeal and supraglottic stenoses are less common, and are more challenging to manage. Endoscopic dilatation can be performed under general anaesthesia via direct sight with serial bougies or balloon dilatations. The main risk is perforation. Endoscopic dilatation can be performed under local anaesthetic with or without sedation. More challenging stenoses can be dilated using radiological guidance with bougies or balloons inserted over a guide wire, to reduce the risk of creating a false passage and perforation risk. Complete pharyngeal stenoses can be approached in a retrograde fashion via a gastrostomy. A narrow gastroscope can be directed superiorly within the oesophagus to the pharyngeal stricture, or a guide wire can be advanced superiorly in the oesophagus with radiological confirmation. The stricture can then be opened under direct transoral visualisation towards the light of the endoscope or by blunt dissection towards the guide wire – the so called ‘rendezvous’ procedure.Reference Bertolini, Meyenberger, Putora, Albrecht, Broglie and Stoeckli627
Case series show that pharyngeal dilatation is successful in improving swallowing in up to 75 per cent of patients.Reference Moss, Pang, Orosco, Weissbrod, Brumund and Weisman628 Poor swallowing outcomes may be expected for patients with fibrotic strictures who require repeat procedures.Reference Chapuy, Annino, Tishler, Haddad, Snavely and Goguen629
Functional laryngectomy
‘Functional laryngectomy’ is an option for patients with a dysfunctional larynx leading to aspiration and dysphagia, with or without airway compromise. It is infrequently performed, and therefore very little best practice evidence exists. The long-term results of the Radiation Therapy Oncology Group ‘RTOG 91-11’ trial showed that 9 out of 148 patients undergoing laryngectomy did so for functional reasons.Reference Forastiere, Zhang, Weber, Maor, Goepfert and Pajak630
Patients must be managed in an MDT approach, with the opportunity to discuss the option with speech and language therapists and dietetic specialists, and ideally with a similar patient. The main aims of surgery are often to prevent aspiration and pneumonia. The secondary aim is to improve swallowing. Many patients achieve their aim of swallowing, with a proportion requiring ongoing enteral feeding.Reference Hutcheson, Alvarez, Barringer, Kupferman, Lapine and Lewin631 Speech rehabilitation with a tracheoesophageal puncture voice is likely to be more variable than following primary laryngectomy, but can be achieved in the majority of carefully selected patients.Reference Hutcheson, Alvarez, Barringer, Kupferman, Lapine and Lewin631
Patient selection is key to potentially successful functional outcomes. An assessment of the extent of fibrosis is required prior to surgery, ideally under general anaesthesia to assess for hypopharyngeal stenosis. Cross-sectional imaging is recommended to exclude recurrence or distant metastases. Speech and language therapy input should include a functional and radiological evaluation of swallowing, to counsel on potential outcomes. The most favourable scenario is that of a dysfunctional larynx with an otherwise healthy pharynx. In this situation, surgical dissection can be kept to a minimum, with the possible preservation of the strap muscles and hyoid bone (a narrow field laryngectomy).
If pharyngeal fibrosis and structuring is present, a pharyngectomy to excise the scarred mucosa and reconstruction should be considered, to augment the pharyngeal diameter. These cases are far more challenging than the straightforward functional laryngectomy. The fibrotic process may continue following surgery, affecting the reconstruction and leading to further stricture formation.
Reconstruction after pharyngolaryngectomy is discussed in Chapter 7.
Other surgical interventions
Other surgical interventions have been described to manage post-treatment dysphagia and aspiration that may be considered prior to laryngectomy or for patients refusing laryngectomy who wish to preserve their laryngeal voice.
Cricopharyngeal myotomy has been described for this group of patients.Reference Dawe, Patterson, Hamilton and Hartley632 Hyoid suspension is another technique, generally preserved for patients with neurological laryngeal dysfunction. This tends to have poor results in the irradiated patient because of immobility of the larynx. Many patients will have global pharyngeal weakness and fibrosis causing their dysphagia following RT. The potential improvements from these specific interventions require careful consideration.
Tubed supraglottic closure has been recently described, with encouraging results.Reference Ku, Vlantis, Cho, Yeung, Ho and Hui633 Laryngeal voice in the presence of a tracheostomy can be achieved whilst minimising the aspiration risk by reducing the supraglottic aperture.
Radiotherapy-induced xerostomia
Recommendations
• All patients should be asked about dry mouth symptoms (xerostomia) as part of their post-treatment clinical reviews (evidence-based recommendation (R))
• Salivary substitutes and oral lubricants in a variety of forms (sprays, gel, mouthwash, slow-release adhesive discs) may be recommended for the pragmatic temporary reduction of dry mouth symptoms. When residual saliva gland function is present, tactile stimulation of the salivary reflex with topical sialagogues (sugar-free gums, pastilles and lozenges) may temporarily increase natural salivation and reduce dry mouth symptoms. Evidence behind both interventions is not robust, and the perceived benefits can vary widely (good practice point (G))
• When residual salivary gland function is present, and topical sialagogues do not offer satisfactory relief from persistent xerostomia, systemic sialagogues (e.g. the parasympathomimetic cholinergic agonist pilocarpine hydrochloride) should be considered if there are no absolute contraindications in the patient's medical history. The efficacy of pilocarpine in increasing salivary flow and reducing xerostomia is supported by robust evidence. Adverse effects are common but usually mild, and should be discussed with the patient (R)
• If the above strategies are not effective, clinicians may consider acupuncture or neuro-electrostimulation (intra-oral or extra-oral devices). The evidence supporting these strategies is not robust, and availability may vary widely (G)
Introduction
Xerostomia is defined as the subjective sensation of oral dryness, whereas hyposalivation reflects an objective, measurable decrease in salivary flow.Reference Epstein, Thariat, Bensadoun, Barasch, Murphy and Kolnick634
Radiation exposure of salivary glands located within the treatment portal results in a dramatic loss of gland function within the first week of treatment, with a subsequent permanent decrease in salivary flow rate in the vast majority of patients.Reference Burlage, Coppes, Meertens, Stokman and Vissink635,Reference Nagler636 The critical dose limit for parotid and submandibular salivary gland tissue is less than 30 Gy; exceeding this limit typically leads to the acute and eventually progressive, irreversible loss of saliva-producing acinar cells, impaired parasympathetic innervation, and injury to glandular vascular structures.Reference Konings, Coppes and Vissink637 Hyposalivation and xerostomia can also be associated with chemotherapy, although function loss and associated symptoms tend to be less severe and often transient.Reference Jensen, Mouridsen, Reibel, Brünner and Nauntofte638
More than 60 per cent of individuals with head and neck cancer receiving radiation as a monotherapy or in combination with chemotherapy develop irreversible hyposalivation and experience permanent xerostomia.Reference Wijers, Levendag, Braaksma, Boonzaaijer, Visch and Schmitz639 Intra-oral discomfort, difficulties with speech, chewing and swallowing, as well as increased risks of secondary oral infection (e.g. candidiasis and suppurative sialadenitis) and dental disease are commonly seen as consequences of hyposalivation in head and neck cancer survivors.Reference Porter, Fedele and Habbab640 Dental disease in turn may increase the risk of jawbone osteoradionecrosis. Ultimately, a restriction in daily activities, reduced quality of life, poorer general health, social disability and malnutrition may be observed.Reference van der Laan, Van den Bosch, Schuit, Steenbakkers, van der Schaaf and Langendijk641,Reference Langendijk, Doornaert, Verdonck-de Leeuw, Leemans, Aaronson and Slotman247
This chapter proposes a working management plan for hyposalivation and xerostomia (Figure 2).
Figure 2. Flow chart summarising a working management plan for hyposalivation and xerostomia.
Oral mucosal lubricants and saliva substitutes
Various oral mucosal lubricants and saliva substitutes with constituents resembling the chemical-physical properties of saliva have been developed, and are commercially available in the form of moisturising gels, mouthwashes, sprays or adhesive discs. As individual preferences and perceived responses to different types of salivary substitutes can vary widely, it is difficult to suggest the superiority of one formulation over another. Evidence supporting the use of oral mucosal lubricants and saliva substitutes is limited, and a clear understanding of the magnitude of their effect is lacking. Realistically, only mild and short-lived xerostomia alleviation may be expected.Reference Mercadante, Al Hamad, Lodi, Porter and Fedele642 Nevertheless, oral mucosal lubricants and saliva substitutes have a very safe profile, with very few, if any, adverse effects.
Topical sialagogues
The residual secretory capacity of the salivary glands, if any, may allow increased secretion of natural saliva, which arguably can provide the best protection for the oral tissues and relief from uncomfortable dry mouth symptoms. This can be attained by means of mechanical and gustatory stimuli, with sugar-free chewing gums, and pastilles or lozenges being able to trigger the salivary reflex, increase whole saliva secretion and lessen xerostomia.Reference Senahayake, Piggott and Hamilton-Miller643,Reference Jensdottir, Nauntofte, Buchwald, Hansen and Bardow644 Topical sialagogues have a very safe profile; however, the evidence supporting their efficacy is limited. Clinical experience suggests that the beneficial effects of topical sialagogues may be short-lived.
Systemic sialagogues
Randomised, placebo-controlled trials have provided convincing evidence that patients with RT-related xerostomia can experience increased salivation and reduced dry mouth symptoms with regular use of pilocarpine 5 mg tablets, one tablet three times a day.Reference Jensen, Vissink, Limesand and Reyland645,Reference LeVeque, Montgomery, Potter, Zimmer, Rieke and Steiger646 Adverse effects are common but not severe, and include sweating, increased urinary frequency, dizziness, gastrointestinal discomfort and nausea, palpitations, and asthenia.Reference Cheng, Xu, Liu, Wang, Liu and Li647 Contraindications include uncontrolled asthma, cardio-renal disease and chronic obstructive pulmonary disease.Reference Cheng, Xu, Liu, Wang, Liu and Li647
Acupuncture
Acupuncture may be offered to patients with radiation-induced xerostomia, especially where the above topical treatments have proven unsatisfactory, and systemic pilocarpine is not effective, is contraindicated or is poorly tolerated. Stimulation of residual salivary functional capacity by acupuncture may increase whole salivary flow rates and alleviate xerostomia up to six months following RT completion.Reference Simcock, Fallowfield, Monson, Solis-Trapala, Parlour and Langridge648,Reference Blom and Lundeberg649 The evidence is, however, limited by the lack of true controls in clinical studies and by the considerable heterogeneity in acupuncture protocols.Reference Li, Tian and He650,Reference Assy and Brand651 In the available studies, no serious adverse events were attributed to acupuncture, although somnolence, tiredness, and minor bruising or bleeding at the puncture site have been reported.Reference Ni, Tian, Chen, Liu, Li and Li652 Furthermore, acupuncture services are not widely available.
Neuro-electrostimulation of salivary glands
Non-pharmacological, non-invasive electrostimulating devices, such as transcutaneous electrical nerve stimulation, have been developed and tested, with the aims of conveying pulsed electrical currents across the intact surface of the skin or oral mucosa, and stimulating the underlying nerves that modulate salivary gland function. Clinical trials of transcutaneous electrical nerve stimulation devices and intra-oral electrostimulators have reported increased saliva secretion and possible reductions in dry mouth symptoms, although the evidence is limited by differences and inconsistencies in terms of the duration and type of electrical stimulation used, as well as the study design.Reference Salimi, Saavedra, Andrews, FitzGerald and Winter653 No serious adverse events have been reported with the use of available salivary neuro-electrostimulating devices.
Unilateral vocal fold paralysis following treatment for head and neck cancer
Recommendations
• All patients with a unilateral vocal fold palsy should be assessed by a speech and language therapist (evidence-based recommendation (R))
• Transection of the recurrent laryngeal nerve at the time of surgery should be followed by primary repair where skills allow for this (good practice point (G))
• Early medialisation should be considered in all patients (G)
Introduction and presentation
Unilateral vocal fold paralysis is not uncommon in patients undergoing treatment for head and neck cancer, particularly following surgery for thyroid cancer. Symptoms of unilateral vocal fold paralysis include a breathy and quiet voice. Many patients will also instinctively tension the vocal folds, resulting in a high-pitched voice.
The loss of adequate glottic closure affects swallow function, with aspiration of solids and liquids, but also saliva. Spontaneous coughing on the patient's own secretions is not uncommon. The loss of glottic competence also means that the strength of the cough is diminished (characteristically described as ‘bovine’). It is now recognised that silent aspiration results in excess morbidity and mortality.Reference Nouraei, Allen, Kaddour, Middleton, Aylin and Darzi654 Figure 3 shows a suggested algorithm for the management of unilateral vocal fold paralysis.
Figure 3. Suggested algorithm for the management of unilateral vocal fold (VF) paralysis. EMG = electromyography
Features of unilateral vocal fold paralysis:
• Breathy, weak voice
• Altered pitch (often high-pitched)
• Vocal fatigue
• Hyperventilation
• Ineffective cough
• Aspiration (particularly liquids)
• Pneumonia
The degree of breathy dysphonia is variable and depends on the position of the paralysed vocal fold; a vocal fold sitting in a median or paramedian position results in better glottic closure, and hence a stronger voice. Conversely, a vocal fold in a lateral position will result in a very breathy voice. Most treatments for unilateral vocal fold paralysis are aimed at pushing the paralysed vocal fold into a more medial position (vocal fold augmentation or medialisation). Some patients will still aspirate despite good compensation, and this may be related to sensory deficits from the injury.
Management
The management of unilateral vocal fold paralysis has changed significantly in recent years; in the past, a period of clinical observation (‘watchful waiting’) was advocated. However, with the advent of newer injectable and absorbable materials, combined with better out-patient endoscopic systems, medialisation procedures are now relatively straightforward to undertake in the clinic in an awake, unsedated patient.
Patients with unilateral vocal fold paralysis should be seen by the speech and language therapy team without delay. If there is evidence of aspiration, appropriate action is necessary (see section on aspiration, above).
Procedures to medialise or augment the paralysed vocal fold aim to provide bulk to the paraglottic space, pushing the medial vibratory edge of the vocal fold to the midline so that glottic closure can be improved. The medialisation material may either be injected directly into the vocal fold (vocal fold medialisation injection (injection laryngoplasty)) or may be placed into the paraglottic space via a window in the thyroid cartilage (Isshiki type 1 thyroplasty (medialisation laryngoplasty)).
Medialisation (augmentation) injection
Vocal fold medialisation injection (injection laryngoplasty) is relatively easy to perform under local anaesthesia in the clinic setting. Equipment requirements are minimal: a distal chip endoscopic system, local anaesthetic and the injection material itself.
Early medialisation following the onset of unilateral vocal fold paralysis improves long-term outcomes. A series of studies has demonstrated that the longer the delay in performing a medialisation procedure, the more likely it is that the patient will require a thyroplasty in the future.Reference Alghonaim, Roskies, Kost and Young655–Reference Snyder, Angelos, Carty, Doherty, Howe and Lee658 There is therefore little justification for ‘watchful waiting’, and this is radically altering clinical practice amongst laryngologists.
Different materials can be used, the most common of which are:
• Calcium hydroxylapatite (Prolaryn™ Plus, Renú® Voice), which is easy to handle and requires no specific preparation. Its typical duration of action is around 12–18 months.
• Hyaluronic acid (various proprietary preparations, including Restylane®). This typically lasts around four months. It is therefore ideally suited to those patients in whom resolution of the unilateral vocal fold paralysis is anticipated, restoring the voice for the intervening recovery period.
• Dissolvable gels (Renú® Gel, Radiesse™ Voice Gel, Prolaryn™ Gel) are synthetic products that have a similar short duration of action (typically a few months) to hyaluronic acid, and these are used in similar situations to hyaluronic acid.
Injection medialisation can be performed percutaneously or per-orally. All the local anaesthetic techniques described below take just a few minutes to perform, and the patient will typically leave the clinic a few minutes after it is completed. The options are:
• Percutaneous injection under endoscopic guidance (transhyoid, transcricothyroid, transthyroid cartilage)
• Transoral injection under endoscopic guidance
Isshiki type 1 thyroplasty (medialisation laryngoplasty)
Thyroplasty has the advantage of being a ‘definitive’ procedure, with an implant that (in theory) will not resorb or move. In long-standing cases of unilateral vocal fold paralysis, particularly when previous injections have been performed, a thyroplasty is often the most appropriate choice of procedure.
This procedure is performed in the operating theatre under local anaesthesia. A window in the thyroid cartilage is made at the level of the paralysed vocal fold; an implant material (which may be Silastic™, Gore-Tex® ribbon, metal or other) is placed through the window into the paraglottic space.
Laryngeal reinnervation
If the recurrent laryngeal nerve is transected at the time of neck surgery, it can be primarily repaired, with or without the use of a nerve interposition graft.
Non-selective reinnervation procedures aim to restore tone and bulk to the paralysed vocal fold, but do not achieve normal laryngeal movement. A branch of the ansa cervicalis can be anastomosed to the distal stump of the recurrent laryngeal nerve. Early studies have shown promising results,Reference Fancello, Nouraei and Heathcote659 but improvements in voice often take several months to be seen, so the reinnervation is often combined with a temporising medialisation injection with (for example) hyaluronic acid.
Studies are planned to compare thyroplasty with laryngeal reinnervation, and a feasibility study is underway.Reference Blackshaw, Carding, Jepson, Mat Baki, Ambler and Schilder660
Thyroid testing and dysfunction following head and neck cancer treatment
Introduction
Radiation damages the thyroid gland and can cause, in patients treated for head and neck cancers, mainly two disorders: short-term thyroiditis and long-term hypothyroidism.Reference Lin, Yang, He, Wang, Gao and Tam661 Less frequently, central hypothyroidism (pituitary irradiation), Graves’ disease (hyperthyroidism), including Graves’ ophthalmopathy (through the release of thyroid antigens) and benign nodularity, and radiation-induced thyroid carcinoma may occur.Reference Jereczek-Fossa, Alterio, Jassem, Gibelli, Tradati and Orecchia662
Radiation-induced hypothyroidism
The importance of recognising hypothyroidism cannot be understated. Manifestations of hypothyroidism include slowed mentation, depression, skin dryness, pleural and pericardial effusions, decreased gastrointestinal tract motility, weight gain, and cold intolerance.Reference Sinard, Tobin, Mazzaferri, Hodgson, Young and Kunz169
Radiation-induced hypothyroidism is the most common post-RT thyroid complication. The incidence of post-treatment hypothyroidism among patients with head and neck cancer is 10–40 per cent.Reference Lin, Yang, He, Wang, Gao and Tam661,Reference Feen Ronjom663 It can develop at any time after RT, as long as 10 years post treatment, but with peak occurrence at 1–3 years after RT.Reference Lin, Yang, He, Wang, Gao and Tam661
Patients treated with total laryngectomy and RT are at the greatest risk for developing radiation-induced hypothyroidism.Reference Sinard, Tobin, Mazzaferri, Hodgson, Young and Kunz169
Primary hypothyroidism can be divided into clinical and sub-clinical states. Clinical hypothyroidism is characterised by an elevated thyrotropin (thyroid-stimulating hormone (TSH)) level with a decreased thyroxine (T4) level, or TSH of 10 mIU/l or more, regardless of symptoms. In these cases, hormone replacement with levothyroxine is recommended.Reference Jereczek-Fossa, Alterio, Jassem, Gibelli, Tradati and Orecchia662 Patients with sub-clinical hypothyroidism have an elevated TSH level with normal free T4 levels.Reference Sinard, Tobin, Mazzaferri, Hodgson, Young and Kunz169 If the patient has symptoms of hypothyroidism and TSH is higher than 5 mIU/l, it is reasonable to start a six-month trial of levothyroxine monotherapy to see whether symptoms improve.664
Thyroid function studies should be considered before beginning treatment, with further thyroid function tests at 3–6 months after the completion of therapy, and then at 12-month intervals for 2 years followed by annual evaluation, being considered reasonable. Other radiation-induced thyroid disorders (thyroiditis, Graves’ disease, thyroid cancer) are rarer, and treatment is similar to that employed in spontaneously occurring conditions.Reference Jereczek-Fossa, Alterio, Jassem, Gibelli, Tradati and Orecchia662
Clinical management of radiation-induced hypothyroidism
• All patients with overt hypothyroidism (TSH level over 10 mIU/l) should be treated with levothyroxine.Reference Khandelwal and Tandon665
• Initial treatment should entail thyroxine in lower doses in: older adult patients, patients with coronary artery disease and patients with long-standing, severe hypothyroidism.Reference Khandelwal and Tandon665
• In primary hypothyroidism, treatment is monitored in terms of serum TSH levels, with the target level being less than 2.5 mIU/l.
Sub-clinical radiation-induced hypothyroidism
• Consider treatment with levothyroxine for: patients in whom serum TSH is between 5 mIU/l and 10 mIU/l along with symptoms of hypothyroidism, patients with infertility, and patients with goitre or positive anti-thyroid peroxidase antibodies.71,72
• When treatment is indicated, a three to six month therapeutic trial is justified. If the patient feels improved by therapy, it is reasonable to continue treatment.667
• In patients with central hypothyroidism, treatment is tailored according to free or total T4 levels, which should be maintained in the upper half of the normal range for age.Reference Khandelwal and Tandon665
Treatment with levothyroxine
• Treatment with levothyroxine is lifelong.
• The levothyroxine starting dose should be 25–100 μg.
• In older adults with a history of ischaemic heart disease, the levothyroxine dose can be raised by 25 μg increments at six-week intervals until the TSH goal is attained.Reference Wass and Owen666
• Thyroid-stimulating hormone should be checked only after six weeks after any dose change; once stabilised, TSH should be checked on an annual basis.Reference Wass and Owen666
• Levothyroxine is best taken in the morning, with water, on an empty stomach, at least half an hour before eating and drinking anything.667
• In patients with persistently elevated TSH despite an apparently adequate replacement dose of T4, consider poor compliance, coeliac disease, malabsorption and the presence of drug interactions.Reference Khandelwal and Tandon665
• Over-replacement is common; it is associated with an increased risk of atrial fibrillation and osteoporosis, and hence should be avoided.Reference Khandelwal and Tandon665
Hearing loss and tinnitus following head and neck cancer treatment
Recommendations and key points
• Where surgical resection (most commonly temporal bone resection) is thought to disrupt any anatomy integral to hearing, it is vital that these patients undergo thorough audiological testing and counselling pre-operatively (evidence-based recommendation (R))
• Total or near-total unilateral hearing loss: consider referral for a bone conduction hearing implant or a bilateral contralateral routing of signals (‘BiCROS’) aid (good practice point (G))
• Numerous therapeutic agents have been proposed as potentially protective from the ototoxic effects of cisplatin; however, none have clinically significant evidence to support their use. There are less ototoxic alternatives to cisplatin; however, they lack the survival benefit that is evidenced with cisplatin (R)
• The management of otitis media with effusion (OME), commonly encountered in the context of post-RT nasopharyngeal carcinoma (NPC), remains controversial. Conservative measures to manage nasopharyngeal and middle-ear inflammation, as well as hearing aids to rehabilitate hearing, should be considered prior to offering more invasive treatments (R)
• Identifying patients who are at risk of developing ototoxicity pre-treatment can help tailor treatment in the chemoradiotherapy setting, to try and minimise the impact on their quality of life (G)
Introduction
Hearing loss and tinnitus are potential post-treatment sequelae of all modalities used to treat head and neck cancer, namely:
• Cochlear toxicity from chemotherapy
• Cochlear toxicity from RT
• Direct effects of surgery on the external ear canal and/or middle ear
• Effects of surgery and/or RT on Eustachian tube function, causing middle-ear effusion
Hearing loss and tinnitus can have profound effects on a patient's quality of life, particularly in older adults.Reference Dalton, Cruickshanks, Klein, Klein, Wiley and Nondahl668 It is therefore essential to not only help patients in their rehabilitation after treatment, but also to explore how to limit the ototoxic effects of treatment whilst maintaining excellent oncological outcomes. Hearing loss and tinnitus are both included in the National Cancer Institute Common Terminology Criteria for Adverse Events.Reference Basch, Reeve, Mitchell, Clauser, Minasian and Dueck669
Surgery
Temporal bone resection can result in total or near-total conductive hearing loss, and, if extended to the inner ear, sensorineural hearing loss (SNHL) and balance impairment.
Pre-operative audiological assessment of the ipsilateral ear is necessary, and audiological involvement is recommended. Both conductive hearing loss and SNHL can be rehabilitated through a bone conduction hearing implant or a bilateral contralateral routing of signals aid. A bone conduction hearing implant may be contraindicated within any post-operative RT field however.
Chemotherapy
Cisplatin is the most commonly used chemotherapeutic agent in head and neck oncology. Cisplatin-induced ototoxicity results in high frequency SNHL, which may be permanent and is often associated with tinnitus.Reference Frisina, Wheeler, Fossa, Kerns, Fung and Sesso670 Fifty per cent of patients receiving more than 200 mg/mReference Brennan, Blythe, Herd, Habib and Anand600 have a significant reduction in their hearing, with a severe to profound loss in both ears.Reference Frisina, Wheeler, Fossa, Kerns, Fung and Sesso670–Reference Moroso and Blair672 It is therefore part of pre-treatment counselling to make patients aware of these potential side effects and to screen them for pre-existing hearing loss. It is also important to advise patients to contact the chemotherapy hotline or other appropriate service should they develop hearing loss or tinnitus during their treatment. This allows clinicians to then consider means of mitigating ototoxicity.
Fractionating the cisplatin regimen to weekly doses reduces the likelihood of ototoxicity occurring.673 There are alternative chemotherapeutic agents with potentially more favourable ototoxicity profiles. Carboplatin is a second-generation platinum-based drug that has a similar mode of action; however, it is associated with less ototoxicity, and less nephrotoxicity and peripheral neuropathy.Reference Lokich and Anderson674 Carboplatin, however, has failed to demonstrate the same survival benefit as cisplatin.Reference Guan, Li, Zhang, Xiao, Chen and Zhang675
If there was a scenario where patient-specific factors such as occupation or hobbies, or pre-existing profound hearing loss, were very significant and chemotherapy was indicated, then the above alternatives can be considered. It would obviously be important to counsel the patient thoroughly on the potential reduction in the efficacy of their treatment and to involve the wider MDT in the decision-making process.
There have been attempts to see whether any steps can be taken to protect the inner ear from the ototoxic effects of cisplatin. A wide variety of therapeutics, both systemic and intratympanic, have been proposed to try and ameliorate the ototoxic effects of cisplatin. Unfortunately, to date, none have demonstrated any convincing protection.
Radiotherapy
Radiotherapy causes ototoxicity by a number of proposed mechanisms, including effects on outer hair cells, atrophy of the vestibulocochlear nerveReference Moretti676 and endarteritis affecting the cochlear blood vessels.Reference Young, Cheng and Ko677 It has been established that the effects of RT on hearing are dose-related, and combination chemoradiotherapy exacerbates ototoxicity. Radiotherapy ototoxicity generally causes a high frequency SNHL (4–8 kHz) that may be transient.Reference Anteunis, Wanders, Hendriks, Langendijk, Manni and de Jong678
The advent of intensity-modulated RT and subsequently volumetric-modulated arc therapy has allowed a reduction of the radiation dose received by the cochlea. The critical cochlear dose to cause an SNHL is unclear, and is quoted as anywhere between 24.2 Gy and 60 Gy in different patient populations.Reference Chen, Liao, Tsai, Yeh, Wang and Tang679,Reference van der Putten, de Bree, Plukker, Langendijk, Smits and Burlage680 Even with these modern advancements in RT techniques, in the context of NPC, it is hard to limit the radiation dose to the cochlea to less than 15 Gy.Reference Lamaj, Vu, van Timmeren, Leonardi, Marc and Pytko681 As a consequence, some authors have advocated pre-treatment risk stratification to try and identify patients who are at high risk of developing ototoxicity, so patients can be appropriately counselled and RT planning tailored.Reference Theunissen, Zuur, Jóźwiak, Lopez-Yurda, Hauptmann and Rasch682,Reference Zuur, Simis, Lamers, Hart, Dreschler and Balm683 Increasing age, cochlear dose, the addition of cisplatin and poor pre-existing hearing have all been identified as risk factors for developing clinically significant ototoxicity following RT.Reference Mujica–Mota, Waissbluth and Daniel684
Otitis media with effusion
Specific attention should be given to OME in relation to patients with NPC. Interestingly, over 40 per cent of patients with NPC will present with OME due to tumour occlusion or Eustachian tube dysfunction.Reference Sham, Wei, Lau, Yau and Choy685 Radiotherapy with or without chemotherapy is the mainstay of treatment for the majority of these patients (see above). The incidence of post-radiation OME ranges between 7.1 per cent and 53 per cent.Reference Wei, Engzell, Lam and Lau686–Reference Tsang, Kwong, Ho, To, Ho and Wei688 This wide range is likely dose-related; however, the advent of intensity-modulated RT has not significantly reduced the incidence.Reference Hsin, Chen, Young and Liu689,Reference Hsin, Tseng, Lin and Chen690 Pre-treatment OME is understandably an adverse risk factor for it remaining following treatment.
The management options for patients with OME with associated hearing loss are the same as in other settings. Many authors have historically advocated a hearing aid, as this rehabilitates both conductive and SNHL and is non-invasive.Reference Skinner and Van Hasselt691,Reference Chen, Young, Hsu and Hsu692 Myringotomy and grommet insertion remain controversial, and there is conflicting evidence in the literature; some authors go as far as to say grommets are contraindicated, quoting persistent otorrhoea rates of 68 per cent and highlighting almost certain recurrence in a condition that is known to last for as long as 10 years following treatment.Reference Skinner and Van Hasselt691 A more contemporary (2017) randomised, controlled trial comparing observation versus grommet insertion demonstrated more favourable outcomes in patients with grommets, finding that only 10 per cent of patients suffered from persistent otorrhoea.Reference Charusripan and Khattiyawittayakun693
It is reasonable to suggest that a stepwise approach is sensible in this patient cohort. Conservative measures such as a hearing aid, managing nasopharyngeal inflammation with topical nasal steroids, and salt-water nasal douches should be instigated first. If these fail to rehabilitate symptoms effectively, more invasive measures such as myringotomy and grommets can be discussed. Informing a patient that this treatment is unlikely to be definitive and may result in a chronically discharging ear is vital in the joint decision-making process.
Tinnitus
Tinnitus is a notoriously difficult condition to manage, even in a non-cancer setting. Problematic tinnitus (as recorded by the Tinnitus Handicap Inventory)Reference Newman, Jacobson and Spitzer694 is more prevalent in patients undergoing chemoradiotherapy rather than RT alone.Reference Niemensivu, Saarilahti, Ylikoski, Aarnisalo and Mäkitie695 The management of tinnitus post treatment should be carried out as in other settings, with patient counselling, noise distraction and cognitive behavioural therapy as the mainstays of treatment. Some patients with Eustachian tube dysfunction may experience autophony or tinnitus associated with patulous Eustachian tube dysfunction.Reference Young, Cheng and Ko677 There is no evidence-based treatment for patulous Eustachian tube dysfunction in the setting of post-cancer treatment.
Introduction
Cancers of the lip and oral cavity are common, with over 377 000 incident cases annually worldwide.Reference Sung, Ferlay, Siegel, Laversanne, Soerjomataram and Jemal199 In the UK, there are over 3500 new diagnoses each year. The incidence is increasingReference Louie, Mehanna and Sasieni210 and represents currently approximately 1 per cent of cancer incidence.696 Cancers of the lip are the most common group of malignant tumours affecting the head and neck region.
Within the oral cavity, the tongue and floor of the mouth are the most common subsites affected. Cancers of the lip require separate consideration as their natural history may differ from oral cavity disease.
The overwhelming majority of oral cavity cancers are squamous cell carcinomas (SCCs). Those not of squamous origin are pre-dominantly derived from salivary tissues and are discussed elsewhere in these guidelines. Surgical resection is the primary treatment modality for the vast majority of oral and lip cancers.
There are significant functional and cosmetic sequelae of the management of oral cavity tumours, as well as frequent medical co-morbidities and social issues in this patient group. Hence, multidisciplinary team (MDT) management is particularly important.
Pathology
Oral cavity squamous cell carcinoma and dysplasia
Oral cavity SCC may develop de novo, or from a pre-malignant dysplastic lesion that appears clinically as leukoplakia, erythroplakia or a combination of the two. In both instances, chronic exposure to carcinogens such as tobacco and/or alcohol is thought to be important. Dental trauma can also be a risk factor, especially in the lateral tongue.
Malignant and pre-malignant lesions of the lip and oral cavity present as a spectrum of disease with varying degrees of cellular atypia, ranging from mild dysplasia to widely invasive carcinoma.Reference El-Naggar, Chan, Grandis, Takata and Slootweg65 Malignant transformation is reported to occur in approximately 12 per cent of dysplastic oral lesions.Reference Mehanna, Rattay, Smith and McConkey697
Oral dysplasia should be reported according to the World Health Organization (WHO) classification, i.e. a three-tier system of mild, moderate and severe dysplasia, with carcinoma in situ being synonymous with severe dysplasia (see Chapter 3, on pathology).
Histological subtypes of oral cavity SCC have prognostic relevance. For example, verrucous carcinoma has a better prognosis compared to spindle cell carcinoma which generally has a poor outcome. Oncogenic human papillomavirus (HPV) infection is detected in only a small proportion (5 per cent) of oral cavity SCC cases, and there is some evidence to support differential outcomes on the basis of HPV status,Reference Burr, Harari, Ko, Chen, Yu and Baschnagel698,Reference Ko, Harari, Sacotte, Chen, Wieland and Yu699 but not to the extent seen in the oropharynx.Reference Chung, Zhang, Kong, Harris, Fertig and Harari700,Reference Lingen, Xiao, Schmitt, Jiang, Pickard and Kreinbrink701
Depth of invasion is particularly important in oral cavity SCC, hence its incorporation into the most recent staging, as per AJCC Cancer Staging Manual (eighth edition)Reference Amin, Edge, Greene, Byrd, Brookland and Washington87 (see below). Oral tongue SCC of greater than 4 mm tumour thickness is considered to represent a risk of occult cervical lymph node metastasis of greater than 20 per cent.Reference Huang, Hwang, Lockwood, Goldstein and O'Sullivan702
Tumours with a non-cohesive invasive front, lymphovascular and/or vascular, with lymphatic or perineural invasion, are associated with an increased risk of locoregional relapse. These pathological factors therefore supplement the tumour–node–metastasis (TNM) classification and are now incorporated in histopathology reporting datasets.
Lip
Squamous cell carcinoma is the commonest histological tumour type in lip cancers, followed by basal cell carcinoma. The clinical behaviour of lip SCC is similar to that of skin cancer (see also Chapter 27, on non-melanoma skin cancer). Aetiological factors for lip cancer include solar radiation, tobacco smoking and viruses.
The most common non-mucosal form of lip cancer arises from minor salivary glands, which (in contrast to SCC of the lip) occurs in the upper lip more commonly than the lower.
Clinical presentation and diagnosis
Presentation
Most oral cavity SCCs (over 95 per cent) present as ulcers or masses. Early lesions can be subtle, and appear as flat, discoloured areas (leukoplakia or erythroplakiaReference Rethman, Carpenter, Cohen, Epstein, Evans and Flaitz703). A non-healing ulcer is the most common presentation. Advanced tumours may present with invasion of neighbouring structures, causing tooth mobility, trismus, sensory change, referred otalgia and extraoral masses.
The clinical presentation of cancer of the lip is usually that of an exophytic, crusted lesion with variable invasion into underlying muscle (related to the size of the primary tumour). The adjacent lip often shows features of actinic sun damage such as colour change, mucosal thinning, and various associated areas of leukoplakia.Reference Wolff, Follmann and Nast704 About 90 per cent of tumours arise in the lower lip, with 7 per cent occurring in the upper lip and 3 per cent at the oral commissure.
Diagnosis
A systematic approach of examination must be adopted to include the primary site and neck, with an assessment of the index tumour size as well as any potential invasion of local structures.
Diagnosis is confirmed histologically by biopsy for any lesion suspected to be either dysplastic (providing grade of dysplasia) or malignant, and is typically performed in the out-patient setting under local anaesthesia for accessible lesions. The use of flexible nasendoscopy facilitates both the assessment of primary tumours posteriorly positioned in the oral cavity and the assessment of adjacent mucosal structures at risk of synchronous primary malignancies. Examination under anaesthesia (EUA) might be necessary for more posterior lesions, for mapping biopsies, and/or to aid with staging and operation planning.
Imaging
Imaging should ideally be conducted before biopsy, but not at the expense of diagnostic delay. In practice, this means that biopsy in clinic is typically performed before imaging, but imaging should be carried out before EUA. It should be borne in mind that inflammation caused by biopsy might alter the radiological size of some smaller oral cancers and regional lymph nodes.
Oral cavity SCC should be staged with cross-sectional imaging as routine, by either computed tomography (CT) or magnetic resonance imaging (MRI). The chest should be imaged to exclude synchronous primary lung cancer and/or distant metastases.Reference Duprez, Berwouts, De Neve, Bonte, Boterberg and Deron705 This may also demonstrate other simultaneous pulmonary parenchymal disease. National Institute for Health and Care Excellence (NICE) guidance (NG36) recommended cessation in systemic staging for T1/2N0 disease unless indicated otherwise.706 However, most patients continue to have systemic staging, similarly to laryngeal cancer, factoring the importance of detecting synchronous lung primary cancers.
Imaging of the primary site in early-stage tumours of the lip is usually not indicated. However, advanced tumours, particularly if they are adherent to the adjacent mandible, require CT or MRI to allow complete staging and treatment planning with regard to resection margins which may include adjacent bone. Where cross-sectional imaging is not indicated, ultrasound assessment of the clinically N0 neck should be considered, to adequately stage the neck (bilaterally).
Table 1 shows the recommendations for routine imaging modalities in the staging of oral cavity SCC.706
Table 1. Recommendations for routine imaging modalities in oral cavity SCC staging
*Magnetic resonance imaging (MRI) is superior for assessing soft tissue involvement (e.g. tongue); computed tomography (CT) is superior in the assessment of mandible erosion. †An orthopantomogram (OPG) should be taken to assess the adjacent dentition, and may support the determination of bone invasion alongside clinical assessment and cross-sectional imaging. ‡National Institute for Health and Care Excellence guidance (NG36) recommends cessation in systemic staging for T1/2N0 disease unless indicated otherwise;706 positron emission tomography (PET)-CT is utilised for N3 disease. SCC = squamous cell carcinoma
Staging
Staging of primary cancer of the lip and oral cavity (according to the AJCC Cancer Staging Manual, eighth edition) is summarised in Tables 2–4.Reference Amin, Edge, Greene, Byrd, Brookland and Washington87 The main change in the eighth edition, specific to the oral cavity, is the inclusion of the impact of depth of invasion. The presence of bone invasion carries a negative influence on disease-specific survival (approximately halving survival).Reference Montero, Yu, Palmer, Patel, Ganly and Shah707 This is reflected in the TNM staging classification, whereby the presence of bone invasion (beyond merely superficial cortical erosion) upstages tumours to T4.
Table 2. Tumour (T) staging for oral cavity and lip cancerReference Amin, Edge, Greene, Byrd, Brookland and Washington87*
* As per AJCC Cancer Staging Manual (eighth edition). †‘DOI’ is depth of invasion; tumour thickness is not necessarily synonymous with depth of invasion. ‡Note: superficial erosion of bone or tooth socket (alone) by a gingival primary is not sufficient to classify a tumour as T4a. ICA = internal carotid artery
Table 3. Node (N) staging for oral cavity and lip cancerReference Amin, Edge, Greene, Byrd, Brookland and Washington87*
* As per AJCC Cancer Staging Manual (eighth edition).
Table 4. Group staging for oral cavity and lip cancerReference Amin, Edge, Greene, Byrd, Brookland and Washington87*
* As per AJCC Cancer Staging Manual (eighth edition).
Cancers of the lip vermilion are staged as oral cavity and lip; those arising from outside of the vermilion are staged as skin cancers.
Management – oral cavity cancers
Recommendations
• Surgery is the mainstay of management for oral cavity tumours (evidence-based recommendation (R))
• Offer surgical excision of resectable lesions with a high risk of malignant transformation (R)
• Tumour resection should be performed with a clinical clearance of 1 cm, vital structures permitting (good practice point (G))
• Proactive or elective neck treatment should be offered for all oral cavity tumours (R)
• Elective neck dissection for clinically node-negative disease should include levels I–III (R)
• Therapeutic neck dissection for clinically node-positive disease should include at least levels Ia, Ib, IIa, IIb and III (R)
• Post-operative radiotherapy (RT) should be considered for locally advanced disease (R)
• Adjuvant chemoradiotherapy in the presence of advanced neck disease (extracapsular spread) or positive margins improves control rates (R)
General principles
Whilst there are no data from randomised, control trials exclusively comparing the different treatment modalities available in the management of oral cavity cancer, it is generally accepted that the primary treatment modality in suitable patients is surgery.
Two-year crude survival rates are around 85 per cent for stage I disease, 70 per cent for stage II disease,Reference Ganly, Goldstein, Carlson, Patel, O'Sullivan and Lee708 50 per cent for stage III disease and 40 per cent for stage IV (non-metastatic) disease.Reference Zhang, Dziegielewski, Biron, Szudek, Al-Qahatani and O'Connell709
Management of oral dysplasia
Management of oral pre-malignant lesions remains controversial and lacks level I evidence to support practice. The overall malignant transformation rate in a recent systematic review was 27 per cent.Reference Yan, Reddy, Nguyen, Chi, Neville and Day710
Intervention should be guided by risk stratification for malignant transformation. Grade (either by WHO grading or by two-tier classification) is only one factor that predicts transformation. Risk assessment is an evolving field. In addition to the grade of dysplasia, site (floor of mouth and lateral tongue), size, appearance (erosive appearance or erythroplakia), carcinogen exposure and past oral cancer history influence risk.Reference Ranganathan and Kavitha711
Whilst decision-making will vary, low-risk cases may be observed clinically with an appropriate review regime and interval clinical photography instigated. It should be ascertained that the biopsy sample was representative of the clinically affected area. Patients with resectable lesions that have a high risk of transformation should be offered surgery, but some lesions are pan-oral or so extensive that surveillance is the only pragmatic option. Where surgical intervention is elected, excision should be undertaken to facilitate histological assessment of the specimen.
Surgery – primary cancer
Curative surgery for cancer of the oral cavity involves resection of the tumour with an appropriate uninvolved margin. Primary reconstruction in order to restore functional integrity should be offered when required. The size and location of the primary tumour determines the need for adjuncts such as access procedures and/or temporary tracheostomy. A full description of surgical techniques is beyond the scope of these guidelines; however, important principles are set out below:
• The primary aim of surgery in oral cavity cancer is tumour resection with a clinical clearance of ideally 1 cm (vital structures permitting) to achieve a histopathological margin of at least 5 mm.
• Most tumours in the anterior aspect of the oral cavity can be accessed via the transoral route. This is ideal, as in so doing the circumferential muscular sphincter is maintained and scars avoided.
• As tumours increase in volume and/or are positioned more posteriorly in the oral cavity, a controlled resection may be facilitated by a lingual release or lip-split mandibulotomy.
• The method of ablation, be it by scalpel, laser, diathermy or Coblation®, is a matter of surgeon preference.
• The use of intra-operative frozen sections to assist marginal clearance is resource-intensive and its benefits remain controversial.Reference Gerber, Gengler, Gratz and Kruse712 Although specificity is good, there is suboptimal sensitivity, which can give a false sense of security and invariably prolongs operative time.
• Attempts to reduce the incidence of dysplastic or in situ disease at the margins with topical adjuncts, such as Lugol's iodine, to guide assessment have been investigated in clinical trials, although definitive outcomes are awaited.Reference McCaul, Cymerman, Hislop, McConkey, McMahon and Mehanna713
• Where bone resection is required, the assessment is based upon both clinical inspection and radiological findings. Intra-operative techniques such as periosteal stripping may guide resection margins.
Bone invasion
The extent to which bone invasion influences survival depends on both the depth of tumour invasion and overall size.Reference Ebrahimi, Murali, Gao, Elliott and Clark714,Reference Fried, Mullins, Weissler, Shores, Zanation and Hackman715 Where bone involvement is confined to just cortical erosion, no adverse impact on survival is apparent. In situations where the tumour abuts but does not invade the bone, it is not clear whether incorporating uninvolved bone to help obtain a negative resection margin confers an oncological advantage. However, it remains an important consideration to avoid close or involved margins at the bony interface, thus influencing pathological clearance assessment (no residual tumour (R0) vs microscopic residual tumour (R1)) and hence adjuvant therapy with consequent morbidity. Assessment of periosteum can also address these issues when tumour abuts bone.
By contrast, medullary bone involvement confers a poor prognosis, with a significantly increased risk of cancer-specific death even after adjustment for tumour size and other covariates. Medullary bone invasion is an independent predictor of distant metastatic disease.Reference Ebrahimi, Murali, Gao, Elliott and Clark714,Reference Fried, Mullins, Weissler, Shores, Zanation and Hackman715
A 2018 systematic review and meta-analysis of 15 retrospective cohort studies, including over 1600 individuals,Reference Gou, Yang, Qiao, Ye, Yan and Li716 provided evidence for improved local disease control for individuals undergoing segmental mandibulectomy. There was no statistically significant difference in terms of survival when marginal mandibulectomy was compared to segmental mandibulectomy, although there was a weak trend towards improved overall survival for the latter. In cases where medullary invasion of the mandible occurred, segmental mandibulectomy provided better disease-free survival.
Surgery – reconstruction
Oral cavity cancer frequently requires the reconstruction of ablated tissues to provide restoration of form and functional integrity of the oral cavity and its adjacent bony structures. The importance of appropriate reconstruction in this vital area, including the mandible and maxilla, cannot be overstated. There is a plethora of retrospective series reporting technique and outcome of a wide range of reconstructive techniques for the repair of defects following ablation for oral cavity tumours. The literature suffers from a wide range of heterogeneous factors introducing bias, including tumour sites, stages, patient variables, institutional preferences, surgical techniques, study designs, small numbers, lack of clarity for treatment intention and the reporting of different outcome measures.
Reconstructive options, including local flaps, regional pedicled flaps, and, more frequently, soft tissue and composite microvascular free-tissue transfer, are discussed elsewhere in the guidelines (Chapter 7).
Management of neck lymph node metastasis
Clinically node-negative neck
Occult nodal metastases are present in up to 30 per cent of patients with oral cavity SCC. Randomised, controlled trial evidence supports the provision of an elective neck dissection in the clinically N0 neck, as this confers improved overall and disease-free survival compared with initial surveillance.Reference D'Cruz, Vaish, Kapre, Dandekar, Gupta and Hawaldar717,Reference Hutchison, Ridout, Cheung, Shah, Hardee and Surwald718 The relevance of using tumour thickness as a determinant for elective neck management is still debated. However, the NICE (NG36) guidance recommends that all patients with T1–2 oral SCC be offered surgical management of the neck.719 In a significant proportion of patients, the neck will be accessed for microvascular reconstruction, and hence the issue of elective neck dissection is relatively minor, with little additional surgical morbidity when the neck is accessed and dissected for vessel preparation.
When undertaking an elective neck dissection for oral cavity SCC, levels I–III should be included. Evidence for the exclusion of level IIb for tumours other than those arising from the oral tongue remains insufficient to recommend a practice change.Reference Lea, Bachar, Sawka, Lakra, Gilbert and Irish720–Reference Pandey, Karthikeyan, Joshi, Kumar and Shukla722
Sentinel lymph node biopsy has been demonstrated to be a safe oncological technique for staging the clinically N0 neck in patients with T1–2 oral SCC. In addition to defining the presence of metastasis in the sentinel node, the technique presents potential benefits through the staging of both sides of the neck (12.4 per cent of well lateralised oral tumours demonstrate contralateral drainageReference Schilling, Stoeckli, Haerle, Broglie, Huber and Sorensen723), in addition to the avoidance of selective neck dissection for those staged pathologically N0. The sensitivity of sentinel lymph node biopsy is approximately 85 per cent and the negative predictive value is about 95 per cent.
Sentinel lymph node biopsy may be justifiable in terms of clinical utility and cost effectiveness, and NICE guidance (NG36) recommends that a sentinel lymph node biopsy be offered to patients with T1–2 oral cancer.719 However, debate regarding the optimal approach for the clinically N0 neck in early oral cancer remains.Reference McGurk, Chegini, Schilling and Lai724,Reference Vassiliou, Acero, Gulati, Holzle, Hutchison and Prabhu725 There is a lack of evidence for a reduction in morbidity when sentinel lymph node biopsy is compared to elective neck dissection. The Sentinel European Node Trial (‘SENT’) reported a false-negative rate of 14 per cent of cases,Reference Schilling, Stoeckli, Haerle, Broglie, Huber and Sorensen723 and, as with any nodal recurrence, clinical outcomes following salvage neck dissection were inferior compared with primary, elective neck dissection. These data should be viewed in the context of the neck recurrence rate for elective neck dissection. A recent systematic review and meta-analysis of evidence reported regional recurrence in T1/2 patients with a pathologically N0 neck following elective neck dissection in 7.5 per cent of cases (range, 1.5–14 per cent).Reference Chegini, Schilling, Walgama, Yu, Thankappan and Iyer726 Randomised, controlled trials of sentinel lymph node dissection versus elective neck dissection powered to assess comparative survival outcomes are ongoing and capable of guiding future practice.Reference Schilling, Stoeckli, Vigili, de Bree, Lai and Alvarez727
Where primary tumours abut the midline, consideration should be made for elective surgical treatment of the contralateral neck given the potential for contralateral neck drainage.
For patients who decline elective surgical management of the neck, regular ultrasound surveillance can be considered.
Clinically node-positive neck
Neck dissection at the time of surgery is indicated when there is clinico-radiological evidence of neck metastasis. The extent of neck dissection is determined by the levels and overall neck disease burden. However, at least levels Ia, Ib, IIa, IIb and III should be dissected. In the largest prospective study, of 583 neck dissections, 91 per cent of nodal metastases were at level I–III.Reference Pantvaidya, Pal, Vaidya, Pai and D'Cruz728 Metastasis at nodal station IIb, IV and V was reported at 3.8 per cent, 4.8 per cent and 3.3 per cent, respectively, with no skip metastases at level IV in the absence of metastasis at levels I–III.Reference Pantvaidya, Pal, Vaidya, Pai and D'Cruz728 A systematic review and meta-analysis of retrospective studies comparing selective neck dissection (I–III) with comprehensive neck dissection in oral SCC patients with a clinically node-positive neck suggested comparable oncological outcomes.Reference Liang, Zhang, Kong, Liang and Liao729 (See also Chapter 26, on the management of neck metastases.)
Radiotherapy in oral cavity squamous cell cancer
The role of RT in oral SCC is essentially as post-operative adjuvant therapy with or without synchronous chemotherapy. It is not a standard of care for definitive treatment, but may be considered for patients unfit or unwilling to undergo surgery. Radiotherapy can be delivered by external beam RT (Table 5) or brachytherapy.Reference Langendijk, Ferlito, Takes, Rodrigo, Suarez and Strojan730–Reference Nutting, Morden, Harrington, Urbano, Bhide and Clark106 The latter requires specialist expertise not widely available in the UK, and is therefore not discussed further. Intensity-modulated RT is the accepted standard of care for patients undergoing primary and adjuvant external beam RT.
Table 5. External beam radiotherapy dosage
* The dosage of 65–66 Gy in 30 fractions over six weeks (with elective dose of 54 Gy in 30 fractions) has been adopted as a primary treatment or adjuvant treatment for high-risk patients in most UK centres or trials.Reference Nutting, Morden, Harrington, Urbano, Bhide and Clark106
Post-operative radiotherapy and chemoradiotherapy
Recommendations
• Assess suitability for radical surgery and post-operative RT or chemoradiotherapy for patients with locally advanced disease before surgery (good practice point (G))
• Post-operative RT with concurrent chemotherapy (chemoradiotherapy) should be offered to eligible patients with involved positive resection margins (≤1 mm) and/or extra-nodal extension (evidence-based recommendation (R))
Combined modality treatment with surgery followed by post-operative RT or chemoradiotherapy should be considered in all patients with locally advanced or high-risk disease. This is discussed in more detail in Chapter 4. The suitability of patients with locally advanced disease to undergo multimodality treatment should be evaluated at the outset, as well as the likelihood of disease control balanced with the functional impact of treatment. This is particularly important in this patient group who often have significant co-morbidities and social issues. Alternatives to radical treatment should be discussed with the patient as part of the informed consent and decision-making process.
The need for post-operative treatment should be confirmed in the MDT meeting after definitive pathology reporting. The presence of high-risk features (extra-nodal extension and/or positive resection margins) are definite indications for RT,Reference Huang, Johnson, Schmidt-Ullrich and Grimes732 with concurrent chemotherapy (chemoradiotherapy) in those eligible.Reference Bernier, Cooper, Pajak, van Glabbeke, Bourhis and Forastiere733,Reference Cooper, Zhang, Pajak, Forastiere, Jacobs and Saxman734 Other adverse features for considering adjuvant RT include close margins (1–5 mm), pathologically T3/4, node-positive disease, perineural invasion and lymphovascular invasion, and tumours with a non-cohesive invasive front.147
Pre-operative imaging, examination reports, intra-operative findings, and the final pathology result should be available to inform treatment volume delineation.
The clinical target volume should include the primary and nodal tumour bed, with a suitable margin to account for microscopic spread,Reference Metcalfe, Aspin, Speight, Ermis, Ramasamy and Cardale735 including all pathologically involved nodal levels. The elective clinical target volume should include at-risk uninvolved nodal levels; this will vary according to primary tumour and nodal factors.
Inclusion of the contralateral (undissected) but clinically or radiologically node-negative neck is controversial. Whilst unilateral RT may allow toxicity reduction with sparing of the contralateral mucosa and parotid, it has been shown that recurrences are unlikely to be successfully salvaged in oral cavity SCC.Reference Kowalski737 This, along with patient fitness, must be considered when assessing the risk of treating or omitting the contralateral neck. Radiotherapy to the contralateral neck is recommended in cases following surgery to the primary site and ipsilateral neck when any of the following apply: pathological T3/4 tumour stage, a primary tumour within 10 mm (or less) of the midline, and ipsilateral nodal metastasis (with extra-nodal extension).147,Reference Metcalfe, Aspin, Speight, Ermis, Ramasamy and Cardale735,Reference Chan, Huang, Le, Yu, Dawson and Kim738–Reference Kao and Hsu739
Post-operative RT should start within six to seven weeks post-operatively. Delayed adjuvant RT and prolonged duration of the treatment package is associated with reduced locoregional control and overall survival.Reference Rosenthal, Liu, Lee, Vapiwala, Chalian and Weinstein736
Primary radiotherapy in oral squamous cell carcinoma
External beam RT is not recommended as the primary curative treatment in oral cavity SCC.Reference Ellis, Graboyes, Wahlquist, Neskey, Kaczmar and Schopper740–Reference Liu, Liu, Kao, Qin, Lin and Fang743 In selected patients, usually those unwilling to undergo surgery, it may be carefully considered. This may be as a single modality for early-stage disease or utilised with concurrent platinum or cetuximab for locally advanced disease.Reference Bonner, Harari, Giralt, Azarnia, Shin and Cohen744–Reference Pignon, le Maitre, Maillard and Bourhis745 Whilst surgical and non-surgical treatment have not been compared prospectively, retrospective data suggest that disease control is likely inferior with RT.Reference Ellis, Graboyes, Wahlquist, Neskey, Kaczmar and Schopper740–Reference Liu, Liu, Kao, Qin, Lin and Fang742
Morbidity is significant and primary RT is therefore not an alternative to surgery, irrespective of patient performance status.
Treatment – lip cancer
Recommendations
• Early-stage lip cancer can be treated by surgical resection or RT (evidence-based recommendation (R))
• The standard of care for advanced lip cancer is primary surgery (R)
• There is no evidence to support elective treatment for cervical lymph nodes when there is no indication of lymph node metastases (R)
• In the absence of clear margins, further surgical excision to achieve this may be preferable to adjuvant RT and should be considered (good practice point (G))
Early-stage lip cancer can be treated by surgery or RT. Prognosis is generally excellent, as patients tend to present early. Locally advanced disease is best treated by surgery. Lymph node metastases are relatively uncommon.
The five-year crude survival rates for surgical treatment are about 85–95 per cent for T1 to T2 tumours, dropping to 40–70 per cent for T3 and T4 tumours.Reference Ozturk, Gode, Erdogan, Akyildiz and Apaydin746,Reference Biasoli, Valente, Mantovan, Collado, Neto and Sundefeld747 The local recurrence rate is low because of the relative ease of surgical excision and accurate margin assessment. Re-excision following local failure retains a salvage rate of 75–80 per cent.Reference Zitsch, Park, Renner and Rea748
Early-stage lip cancer (T1/2)
Although there is a paucity of comparative data, it is accepted that early-stage cancers can be treated equally well by surgery or RT. However, surgery represents the commoner modality and the simplest treatment pathway, with small lesions being managed in a single stage by simple surgical excision and primary closure. When used, external beam RT using electrons or orthovoltage photons may be used to treat the full lip thickness whilst minimising the dose (and therefore the toxicity) to the oral cavity. The Royal College of Radiologists dose fractionation guidance recommends a variety of doses, including 35 Gy in 5 fractions, 45 Gy in 10 fractions, 50 Gy in 15–20 fractions, 55 Gy in 20 fractions and 60 Gy in 30 fractions.147 Decisions regarding radiation method and dose depend on the size and depth of the area treated, the radiation tolerance of the tissue and patient fitness, and so are based largely on clinical judgement. For most cancers, 50 Gy in 15 fractions over three weeks using a single anterior field with orthovoltage may be suitable.
Topical and tissue destructive methods of treatment are reserved for non-invasive lesions, and are not recommended for the treatment of invasive carcinomas. Superficial field change lesions affecting the external vermilion of the lip, such as leukoplakia or actinic keratosis, may be managed with a range of techniques. These include carbon dioxide laser ablation and cryotherapy. Larger confluent lesions may be suitable for lip shave and mucosal advancement surgery.
Advanced stage lip cancer (T3/4)
It is generally accepted that the standard of care for advanced lip cancer is primary surgery. As is the case for larger T2 cancers, advanced stage lip cancer requires either local flaps to reconstruct, or, infrequently, free-tissue transfer to restore a circumferential oral seal and adjacent tissue loss. Adjuvant treatment is determined on the basis of histopathological stage and adverse features, as is the case for oral cavity SCC.
Principles of surgery
There is little in the way of agreed consensus as to what surgical margins of clearance are required for lip SCC. This relates to the nature of the lip resting between the oral cavity proper, in which a 5 mm pathological margin is regarded as the minimum for a clear surgical margin, and the surrounding skin (cutaneous SCC), in which, although the aim is 4–6 mm, a 1 mm pathological margin is regarded as adequate. Other relevant factors include the size of the tumour and the fact that lip function can be compromised with larger excisions. In general, for cancers of the wet vermilion that verge into the oral cavity proper, margins as for mucosal SCC should be achieved (5 mm). For other areas of the lip vermilion, a margin of 3 mm is probably adequate, but practice varies.Reference de Visscher, Gooris, Vermey and Roodenburg749
Small lesions are managed by simple surgical excision and primary closure. There are reports of using Moh's surgery, as there are for small cutaneous SCC elsewhere in the head and neck, but with little data specifically on lip SCC.Reference Hunt, Earp, Brown, Veitch and Wernham750
Small lower lip lesions are managed by simple surgical excision and primary closure (such as pentagonal wedge excision or ‘W’-plasty resection techniques). Small upper lip lesions can be treated in a similar manner to small lower lip defects, but issues of symmetry can affect the aesthetic outcome, especially in younger patients.
Surgery for larger lip lesions requires greater consideration of the functional outcomes of lip reconstruction (including sensation and muscle function). Whatever technique is chosen, the repair should provide sufficient mucosa contiguous to the commissure, to avoid contracture and microstomia. Full thickness flaps (skin, muscle and mucosa) used in tissue advancement or lip-sharing techniques (unilaterally or bilaterally) are useful in this setting. Various eponymous techniques are described. Ultimately, if full thickness repair including innervated orbicularis oris muscle covered with skin and mucosa can be achieved, this typically gives the best outcome. Balanced microstomia can be addressed with lip stretching exercises once healing is complete, and this often gives acceptable results. Extensive defects of the lip may require remote tissue to be imported in order to achieve healing. Cheek flaps or free flaps can provide adequate tissue, but often functional and aesthetic outcomes are poor because of a lack of innervated orbicularis muscle, poor sensory recovery, and/or differences in skin texture and colour.
Management of the neck in lip cancer
Most large series in the literature show that the majority of patients have small lesions without occult cervical metastases.
The primary lymphatic drainage of the lips is to the submental and submandibular cervical lymph nodes. Elective neck dissection is not performed routinely for lip cancers because of the low rates of occult metastasis. However high-risk tumours that are thought to significantly involve the oral mucosa can be treated as higher-risk oral cavity tumours. Tumours that involve the dry vermilion and skin external to this without mucosal involvement may be treated in the way that cutaneous SCCs are (see Chapter 27, on non-melanoma skin cancer); in such cases, elective neck dissection would not typically be undertaken.
The role of sentinel lymph node biopsy in lip cancer is not clearly defined and is not routine treatment at this stage.
The presence of regional metastases at presentation is a poor prognostic indicator.
Clinically node-positive neck
Management of the node-positive neck in lip cancer is broadly consistent with treatment of anterior oral cavity tumours with nodal metastasis (see above). The extent of neck dissection is governed by the size and location of both the primary tumour and secondary lymph node mass(es). Resection of appropriate levels of the neck may be considered on a case by case basis, but central tumours may often require a bilateral neck dissection. There is conflicting evidence as to need for comprehensive neck dissection in the setting of upper anterior neck lymph node metastasis.Reference Veness751,Reference Ebrahimi, Moncrieff, Clark, Shannon, Gao and Milross752 A more selective neck dissection, omitting levels IIb and V in particular, may be reasonable in certain situations.
Post-operative radiotherapy and chemoradiotherapy
Indications for adjuvant treatment are comparable to the remainder of oral cavity SCC. In the absence of clear margins, further surgical excision to achieve this may be preferable to adjuvant RT and should be considered.
Recurrent oral squamous cell carcinoma
Patients with locally recurrent disease should be fully restaged and assessed for consideration of curative treatment. This can include salvage surgery and/or RT. Careful patient selection is essential. (See also Chapter 5.)
Palliative treatment
The population who are eligible or might benefit from palliative treatment is heterogeneous, and as such there is no firm evidence base to recommend a specific regimen. Given the limited survival often reported in this patient group, treatment should be of the shortest possible duration, whilst ensuring effective palliation and minimal side effects.Reference Shahid Iqbal, Kelly, Kovarik, Goranov, Shaikh and Morgan753
Potential benefit from treatment must be balanced with toxicity and possible alternatives (e.g. pharmacological interventions) in the context of the patient's anticipated life expectancy.
Patients with adequate performance status who have inoperable, recurrent or metastatic oral cancer may be considered for palliative systemic anticancer treatment (see Chapter 4). Any role of debulking surgery is minimal.
Pending research questions
• Chemotherapy prevention studies – can novel agents or repurposed drugs support the prevention of high-risk lesions (dysplasia) undergoing malignant transformation?
• Combination immunotherapy strategies – how and when should immunotherapy be integrated into curative treatment pathways for locally advanced oral cavity cancer?
• Does the omission of the pathologically node-negative neck from post-operative RT fields reduce toxicity without compromising survival?
• Are smaller surgical margins safe for smaller non-metastatic cancers?Reference Brennan, Dylgjeri, Coletta, Arakeri and Goodson754
Studies due to report
A direct comparison of elective neck dissection with sentinel lymph node biopsy in early-stage oral cavity cancer (NCT04333537) is being assessed currently in a phase III clinical trial seeking evidence for the equivalence (or otherwise) of the two treatment strategies in terms of survival.
Introduction
Internationally, the incidence of oropharyngeal squamous cell carcinoma (SCC) has been increasing. The most recent UK figures from the Office of National Statistics show that the annual incidence increased from 1029 to 2997 cases between 2000 and 2016, with the rate almost doubling over the last 10 years. Human papillomavirus (HPV) is the proposed driver of the increase in global rates, with a 20.6 per cent rise in worldwide prevalence of HPV-positive oropharyngeal SCC.Reference Stein, Saha, Kraninger, Swick, Yu and Lambert755 Human papillomavirus was associated with over 70 per cent of oropharyngeal SCCs by 2009 in both the UK and USA, with little change in the prevalence in non-HPV-related oropharyngeal SCCs.Reference Mehanna, Beech, Nicholson, El-Hariry, McConkey and Paleri204
Presentation and diagnosis
Recommendations
• Record the site, size and fixity of the oropharyngeal primary tumour, and any restrictions in transoral access (before biopsy or tonsillectomy) (good practice point (G))
Patients with oropharyngeal cancers may present with a variety of symptoms. These include: dysphagia, odynophagia, pain in the throat, tongue or referred otalgia, weight loss, and altered speech. Speech changes can include problems with articulation, a ‘hot potato’ type voice, or, rarely, hoarseness. While some patients may present with symptoms, others will present with an otherwise asymptomatic neck lump and be investigated via the carcinoma of unknown primary pathway (see Chapter 27), with an oropharyngeal primary being discovered during diagnostic investigation.
Oropharyngeal cancer primary sites may be biopsied under local anaesthesia if practical and safe, or under general anaesthesia. If general anaesthesia assessment is performed, the transoral access for the primary site, the fixity of the tumour and the feasibility for resection should be documented. Cervical lymphadenopathy when present should be assessed, and a core or fine needle biopsy performed for diagnosis. Performing this under ultrasound guidance is preferable.756
Support from allied health members of the multidisciplinary team (MDT) should be offered to all patients, given the physical and psychosocial effects of cancer, and, in particular, the potential impact on diet and swallowing.Reference Schache, Kerawala, Ahmed, Brennan, Cook and Garrett71
Imaging
Recommendations
Recommendations for pre-treatment imaging in oropharyngeal SCC are shown in Table 1.
Table 1. Pre-treatment imaging in oropharyngeal SCC
SCC = squamous cell carcinoma; MRI = magnetic resonance imaging; CT = computed tomography; PET = positron emission tomography; N = nodal stage
Magnetic resonance imaging (MRI) with contrast is optimal for primary tumour staging; it is preferable to computed tomography (CT) as it provides improved soft tissue contrast resolution between tumour, muscle and mucosa. It also allows for the evaluation of: early bone marrow involvement, pterygopalatine fossa extension, prevertebral muscle involvement,Reference Hsu, Loevner, Karpati, Ahmed, Mong and Battineni757 retropharyngeal lymph node involvement, the relationship to the internal carotid artery and the perineural spread of the tumour.Reference Kato, Kanematsu, Watanabe, Mizuta and Aoki758,Reference Baulch, Gandhi, Sommerville and Panizza759 Magnetic resonance imaging can stage the primary site and neck nodes at the same time. Magnetic resonance imaging is also much less degraded by artefacts from dental amalgam when compared to contrast-enhanced CT.
Computed tomography may be required to evaluate invasion of bony structures (e.g. mandible and skull base), or when MRI is contraindicated (e.g. pacemaker).
Ultrasound with or without needle sampling (fine needle aspirate biopsy or core biopsy as appropriate) can provide further information as to the status of indeterminate cervical nodes.Reference Lewis-Jones, Colley and Gibson760,Reference Olliff, Richards, Connor, Wong, Beale, Madani and Nicholson761
Systemic staging is recommended for all patients. Although National Institute for Health and Care Excellence (NICE) guidance states that, in tumour–node stage T1–2N0 disease, systemic staging may be omitted,756 this is unusual practice and most patients undergo CT of the thorax.
Fluoro-deoxy-glucose (FDG) positron emission tomography (PET)-CT has been shown to provide no significant benefit in the pre-treatment evaluation of staging oropharyngeal SCC, with only a 50 per cent specificity in the node-negative (N0) neck and similar sensitivity to conventional modalities.Reference Kyzas, Evangelou, Denaxa-Kyza and Ioannidis762 It may have some utility in treatment planning, but this has not yet been defined. Current NICE guidelines recommend FDG PET-CT for patients with N3 nodal stage.756
Staging
All tumours should be staged using the TNM Classification of Malignant Tumours (eighth edition).Reference Brierley, Gospodarowicz and Wittekind763 There are now separate classifications for the clinical and pathological staging of neck lymph node metastases. The eighth edition also introduced a new classification for p16-positive oropharyngeal cancers (Tables 2–6). P16 immunohistochemistry overexpression is a surrogate marker for HPV infection. As p16 (HPV) positive oropharyngeal SCCs have a favourable prognosis,Reference El-Naggar, Chan, Takata, Grandis and Slootweg764 the staging for these two diseases is distinct.Reference Brierley, Gospodarowicz and Wittekind763
Table 2. Primary tumour (T) staging for oropharyngeal SCC (p16-negative and p16-positive)*
* According to the TNM Classification of Malignant Tumours (eighth edition).Reference Brierley, Gospodarowicz and Wittekind763 SCC = squamous cell carcinoma
Table 3. Nodal (N) staging for p16-negative oropharyngeal SCC*
* According to the TNM Classification of Malignant Tumours (eighth edition).Reference Brierley, Gospodarowicz and Wittekind763 SCC = squamous cell carcinoma
Table 4. Nodal (N) staging for p16-positive oropharyngeal SCC*
* According to the TNM Classification of Malignant Tumours (eighth edition).Reference Brierley, Gospodarowicz and Wittekind763 SCC = squamous cell carcinoma; N/A = not applicable
Table 5. Group staging for p16-negative oropharyngeal SCC*
* According to the TNM Classification of Malignant Tumours (eighth edition).Reference Brierley, Gospodarowicz and Wittekind763 SCC = squamous cell carcinoma
Table 6. Group staging for p16-positive oropharyngeal SCC*
* According to the TNM Classification of Malignant Tumours (eighth edition).Reference Brierley, Gospodarowicz and Wittekind763 SCC = squamous cell carcinoma
Pathology
Recommendations
• Testing for HPV should be carried out for all oropharyngeal SCCs (evidence-based recommendation (R))
• Testing for HPV-related disease should be performed by an appropriately accredited laboratory (good practice point (G))
• Surgical specimens should be orientated clearly and details communicated to the pathologist (R)
Surgical specimens
Regarding surgical specimens, see also Chapter 3, on pathology. When surgery is performed for diagnosis or treatment, evaluation of the specimens can be challenging for the pathologist, especially if a mosaic resection has been performed. Good communication with the surgical team is required to establish the orientation of the specimen and to identify critical margins.
Orientation can be challenging for transoral specimens. Specimens should be orientated, and labelled using a cork board, or sutured to a foam pad or acetate sheet and annotated with labels.Reference Winter, Corbridge, Shah, Millard and Cox765 The pathologist can then document the specimen with photography and ink the excision margins appropriately.
Intra-operative frozen sections can be used to guide surgery, but this requires appropriate planning and resources.
The definition of close and positive margins is controversial and confounded by difficulties in accurately assessing composite resection specimens. Ultimately, the margin status influences decisions around the provision of further surgery and the use of non-surgical adjuvant treatment, and is best formulated in the context of a head and neck cancer MDT meeting or within a trial protocol.
Human papillomavirus testing
There are numerous methods to test for HPV-related oropharyngeal SCC. P16 immunohistochemistry is a simple, thoroughly validated surrogate marker for HPV-related oropharyngeal SCC and has a prognostic impact. The most commonly used cut-off for a p16-positive result is strong and diffuse nuclear and cytoplasmic staining in 70 per cent or more of the malignant cells. Human papillomavirus specific tests are directed at the detection of high-risk HPV DNA or RNA by in situ hybridisation or polymerase chain reaction methods.Reference Lewis, Beadle, Bishop, Chernock, Colasacco and Lacchetti766 The World Health Organization, American Joint Committee on Cancer, Union for International Cancer Control, International Collaboration on Cancer Reporting and College of American Pathologists recommended p16 immunohistochemistry to infer HPV status for disease classification.Reference Brierley, Gospodarowicz and Wittekind763,Reference Lewis, Beadle, Bishop, Chernock, Colasacco and Lacchetti766–Reference De Virgilio, Costantino, Mercante, Pellini, Ferreli and Malvezzi768 Nevertheless, there is emerging evidence from studies in Europe that p16 testing alone may not be sufficient for accurate prognosis. Specifically, patients with p16-positive, HPV DNA or RNA negative oropharyngeal SCC have a similar poor prognosis to patients with p16-negative disease.Reference Nauta, Rietbergen, van Bokhoven, Bloemena, Lissenberg-Witte and Heideman769 An established algorithm using p16 immunohistochemistry followed by HPV-specific testing of p16-positive cases is required for accurate prognostication and for recruitment of patients to interventional clinical trials.Reference Craig, Anderson, Schache, Moran, Graham and Currie770–Reference Craig, Anderson, Moran, Graham, Currie and Rooney772 The HPV testing should be performed in a quality assured laboratory with appropriate accreditation (e.g. International Organization for Standardization accreditation ISO15189:2012).
Human papillomavirus status and impact on management
Recommendations
• Use HPV status to stage and counsel patients regarding prognosis, but not to modify treatment outside of the trial setting (good practice point (G))
The TNM Classification of Malignant Tumours (eighth edition) incorporated a different staging system for p16-positive and p16-negative tumours (utilising this as a surrogate marker for HPV) because of their different prognostic outcomes. However, there is currently no evidence to offer treatments of reduced intensity based on HPV status.756 For example, there should be no difference in management between a patient with p16-positive oropharyngeal SCC with two lymph node metastases (N1) and a patient with the same, but p16-negative (N2b).
Management of early oropharyngeal cancer (T1–2N0–1)
Recommendations
• Offer patients with early oropharyngeal SCC information regarding primary surgical and non-surgical approaches for curative treatment (evidence-based recommendation (R))
• Every MDT should have the facility and expertise to offer radical radiotherapy (RT) (with concurrent chemotherapy) and transoral surgery (R)
• Consider primary non-surgical treatment if adjuvant radiation following transoral surgery is likely to include concurrent chemotherapy. This avoids triple modality treatment. This is particularly the case in patients with HPV-related oropharyngeal SCC who have an excellent prognosis (good practice point (G))
• Open primary surgery is not recommended for early oropharyngeal SCC (G)
There remains considerable ongoing debate regarding the management of early oropharyngeal SCC. Open primary surgery is not recommended; the essential options for treatment are either:
(1) Primary RT (with chemotherapy, if suitable, for tumours with more than one lymph node or a lymph node sized 3 cm or larger (i.e. N2 using p16-negative classification for all oropharyngeal SCC)); or
(2) Primary transoral surgery and neck dissection with or without post-operative RT or chemoradiotherapy.
In many cases, it is the lymph node status that informs the choice, i.e. whether concurrent chemotherapy would be indicated as part of treatment. For oropharyngeal SCC with no lymph node metastases or limited to a single lymph node sized less than 3 cm, the choice is between single modality treatment with RT or surgery (but possibly with post-operative RT). One key issue with surgery in such circumstances is the determination of a margin status indicating R0 resection (no residual tumour) and no need for adjuvant treatment. There is growing consensus that 2–3 mm (and possibly less), rather than 5 mm, is appropriate for p16-positive oropharyngeal SCC.Reference Warner, O'Hara, Lin, Oozeer, Fox and Meikle773
The NICE concluded that transoral surgical resection or primary RT should be offered for T1–2N0 tumours of the oropharynx, with post-operative RT (with or without chemotherapy) if adverse risk factors are identified.756
Where patients are suitable for either primary RT or transoral surgery, this should be discussed in an MDT and the current equipoise discussed with the patient. As part of the discussion, the patient should be aware of the possibility of requiring post-operative RT or chemoradiotherapy.
Evidence to date has shown that both primary surgery and RT offer broadly comparable and excellent survival and functional outcomes for patients with early oropharyngeal SCC. A number of trials have been performed (but findings are as yet unreported) or are underway to address the clinical question of how best to treat these tumours, both in terms of oncological effectiveness and treatment effects (particularly swallowing). These trials include direct comparisons between treatment modalities and de-escalation strategies, and are discussed at the end of this chapter.
Management of early oropharyngeal cancer – surgery
Recommendations
• Consider transoral surgery as the first treatment option for patients with a high likelihood of achieving R0 resection (evidence-based recommendation (R))
• Offer ipsilateral neck dissection (levels II–IV) for patients with well lateralised cancer and clinically N0 or clinically N1 neck disease, when the primary tumour is being managed by transoral surgery (R)
• Consider contralateral selective neck dissection for patients with clinical and radiological absence of disease, when the non-lateralised primary tumour is being managed by transoral surgery and ipsilateral neck dissection (good practice point (G))
• Offer ipsilateral feeder vessel ligation for patients undergoing transoral resection of oropharyngeal cancer (R)
Primary site
T1 and T2 tumours and selected T3 tumours may be considered for surgical resection.Reference Holsinger, McWhorter, Ménard, Garcia and Laccourreye774 The aim of surgery is to obtain an R0 resection (see above regarding margins). Important considerations include access, fixity, palatal extent, laterality and vessel location.
Surgical techniques include transoral robotic surgery, transoral laser microsurgical resection or other endoscopic resections (e.g. monopolar diathermy with endoscope visualisation). No evidence exists to suggest the superiority of any of these techniques. Transoral surgery may involve en bloc resections or sectional (mosaic resections), with or without separate margins.
Post-operative haemorrhage following transoral resection is well recognised, with major or severe haemorrhage reported in 6.7 per cent and 2.6 per cent of cases respectively; 60 per cent of these patients will need to go to the operating theatre for haemostasis.Reference Sharbel, Abkemeier, Sullivan, Zimmerman, Albergotti and Duvvuri775 In order to help decrease the incidence of life-threatening bleeding, ligation of the external carotid artery branches (lingual and facial branches) is recommended at the time of neck dissection.Reference Stokes, Ramadan, Lawson, Ferris, Holsinger and Turner776
Neck
Neck treatment should cover levels II–IV on the ipsilateral side.Reference Stanford-Moore, Ochoa, Larson, Han, Hoppe and Ryan777 Between 10 and 31 per cent of patients who are clinically staged as T1–2N0 will have occult nodal disease.Reference Smith, Gallitto, Lehrer, Wasserman, Gupta and Sharma778 As occult level IB metastasis risk is low, routine dissection of this level is not indicated in clinically N0 disease. Level I disease is associated with worse outcomes, especially with a higher T or N stage.Reference Lee, Kelly, Park, Yarbrough, Burtness and Husain779,Reference Xiao, Ward, Yang, Adelstein, Koyfman and Prendes780
Contralateral nodal involvement is generally low (up to 4 per cent), with the highest risk when disease is within 1 cm of the midline.Reference Smith, Gallitto, Lehrer, Wasserman, Gupta and Sharma778 Therefore, it may be appropriate to consider contralateral super-selective neck dissection for at-risk tumours if this will alter the overall treatment plan.Reference Stanford-Moore, Ochoa, Larson, Han, Hoppe and Ryan777
Primary and neck surgery can be completed at the same sitting or separated by up to two weeks, with the neck surgery preceding the transoral resection.
Post-operative adjuvant treatment
Recommendations
• Offer patients post-operative RT in the presence of adverse pathological features (close margins, multiple nodes) (evidence-based recommendation (R))
• Offer patients aged under 70 years with positive margins and extra-nodal extension post-operative RT with concurrent chemotherapy (R)
There are no recommendations for adjuvant RT treatment that is specific to oropharyngeal SCC. See Chapter 4 for general considerations. The recommended adjuvant dose is 60 Gy in 30 fractions, with a dose of up to 66 Gy in 33 fractions to high-risk sub-volumes.
Patients with extra-nodal extension or positive resection margins (less than 1 mm), who are aged less than 70 years, should be offered post-operative RT with concurrent chemotherapy, unless part of a trial.Reference Bernier, Cooper, Pajak, van Glabbeke, Bourhis and Forastiere733 No evidence supports altering the adjuvant regimen based on HPV status, and this should only be done within a trial envelope. Preparation for adjuvant treatment (e.g. dental extractions, percutaneous endoscopic gastrostomy placement) can be completed at the time of primary surgery.
Management of early oropharyngeal cancer – non-surgical treatments
Recommendations
• Intensity-modulated RT should be used with the ‘5 + 5’ technique (good practice point (G))
• The dose should be equivalent to 70 Gy (evidence-based recommendation (R))
• When the tumour is not lateralised, the contralateral neck should be treated (G)
• Assessment of the nodal burden should be considered in lateralised tumours when considering omitting contralateral RT (R)
Radical radiotherapy
Recommendations for the treatment of oropharyngeal SCC are provided per the Royal College of Radiologists’ 2021 head and neck consensus statement and the European Organisation for Research and Treatment of Cancer 2017 consensus guidelines.781,Reference Gregoire, Evans, Le, Bourhis, Budach and Chen782 Intensity-modulated RT should be used ideally with a ‘5 + 5’ technique, as per protocol, but consider larger margins if there is uncertainty regarding the gross tumour volume. Organs at risk for these fields include the spinal cord, brainstem and parotid glands. It is suggested that a dose equivalent to 70 Gy in 35 fractions (typically 65–66 Gy in 30 fractions) is used.783
With the aim of reducing toxicity, the high level II lymph nodes (i.e. cranial border of level II defined as where the internal jugular vein crosses the posterior belly of the digastric muscle) should be omitted from the elective target volume in an uninvolved contralateral neck.Reference Spencer, Gay, Haughey, Nussenbaum, Adkins and Wildes784
It is also possible to consider omitting the contralateral retropharyngeal lymph nodes from the elective target volume and when delivering radical RT, as long as there are no ipsilateral involved retropharyngeal lymph nodes and the gross tumour volume of the primary does not involve the soft palate or posterior pharyngeal wall.Reference Biau, Lapeyre, Troussier, Budach, Giralt and Grau146
Neck radiotherapy
The contralateral neck can be omitted for well lateralised T1–2 SCC of the tonsil with a N0 neck or with one involved ipsilateral neck node.
‘Well lateralised’ is defined as a tumour confined to the palatine tonsil, tonsillar fossa or lateral pharyngeal wall, with greater than 10 mm clearance from midline, not involving the base of the tongue or posterior pharyngeal wall, and extending on to the adjacent soft palate by less than 10 mm. Omission can be considered in well lateralised T1–2 of the tonsil with ipsilateral nodes but a low nodal burden (i.e. fewer than three nodes, less than 3 cm, only levels II–III).Reference Sher, Adelstein, Bajaj, Brizel, Cohen and Halthore785,Reference Al-Mamgani, van Werkhoven, Navran, Karakullukcu, Hamming-Vrieze and Machiels786
Management of advanced oropharyngeal squamous cell carcinoma (T3–4 or N2+)
Recommendations
• Offer RT with concurrent chemotherapy with platinum-based chemotherapy to suitable patients with advanced staged disease (good practice point (G))
• Consider tri-modality treatment in select cases, following full discussion of the benefits and drawbacks with patients (G)
• Consider early, rapid nutritional intervention (evidence-based recommendation (R))
Concurrent chemoradiotherapy for advanced disease when indicated can improve survival, and should be the standard of care for patients aged under 70 years with no contraindications. Evidence suggests an overall survival benefit of 4–8 per cent for patients aged under 70 years who are receiving concurrent chemotherapy with radiation.Reference Blanchard, Baujat, Holostenco, Bourredjem, Baey and Bourhis787 Weekly cisplatin may be considered in patients not suitable for three-weekly cisplatin.Reference Bauml, Vinnakota, Anna Park, Bates, Fojo and Aggarwal788 Cetuximab as an alternative concurrent chemotherapeutic agent is inferior to cisplatin.Reference Mehanna, Robinson, Hartley, Kong, Foran and Fulton-Lieuw136,Reference Gillison, Trotti, Harris, Eisbruch, Harari and Adelstein789
Whilst chemoradiotherapy remains the standard of care for advanced oropharyngeal SCC, triple modality treatment may provide a survival benefit in a subset of patients with advanced tonsil cancer.Reference Roden, Schreiber and Givi790 Surgery may have a role in cases when concurrent chemotherapy is contraindicated (i.e. as preferred to RT alone).
Where surgery is offered, this should be with the aim of achieving complete resection at the primary site. Open surgery can be considered for T3/4 disease, especially in the presence of bone invasion.
In the presence of advanced neck disease, it is important to assess the likelihood of achieving macroscopic clearance when performing a neck dissection. This is less likely in N3 disease or where skin resection may be required. The radiological likelihood of extra-nodal extension, especially in patients with HPV-negative disease, should be assessed at the MDT, as this will upstage tumours and will necessitate adjuvant chemoradiotherapy (triple modality treatment) if there is pathological evidence of extra-nodal extension and positive surgical margins.Reference Bernier, Cooper, Pajak, van Glabbeke, Bourhis and Forastiere733
Tumours in the oropharynx may affect the airway at presentation, and large tumours and subsequent swelling with RT may compromise airway safety. If the risk of airway swelling due to treatment is felt to be significant, it is advisable to offer prophylactic tracheostomy in order to avoid breaks in treatment. An increased duration of nutritional support is likely with advanced tumours, and this should be factored into treatment planning; the prophylactic placement of a feeding tube or a low threshold for triggering reactive nasogastric tube feeding should be considered in this patient group. Patient choice and local availability will influence this decision.
Post-treatment imaging and follow up
Recommendations
• Offer all patients PET-CT at a minimum interval of 12 weeks after primary non-surgical treatment (evidence-based recommendation (R))
• Offer neck dissection to patients with an incomplete lymph node response on PET-CT imaging after primary non-surgical treatment (good practice point (G))
• Consider a neck dissection or further PET-CT scan at 10–12 weeks in patients treated for HPV-positive oropharyngeal SCC with primary non-surgical treatment who have an equivocal response on initial PET-CT (G)
Following non-surgical treatment, FDG PET-CT at a minimum of 12 weeks has been shown to accurately evaluate the response to treatment, and, in the absence of residual structural or metabolic activity, no surgical treatment is needed.Reference Mehanna, Beech, Nicholson, El-Hariry, McConkey and Paleri204,756 For HPV-positive patients with equivocal FDG PET-CT findings at 12 weeks, a repeat scan following a further 10–12 weeks can be considered as an alternative to a neck dissection.Reference Zhou, Rulach, Hendry, Stobo, James and Dempsey791–Reference Liu, Milne, Lock, Panizza, Bernard and Foote793 Equivocal FDG PET-CT findings in patients with HPV-negative oropharyngeal SCC are associated with a high specificity for residual disease and will need surgical treatment.Reference Bird, Barrington, Thavaraj, Jeannon, Lyons and Oakley794,Reference de Ridder, Gouw, Navran, Hamming-Vrieze, Jasperse and van den Brekel795 Post-treatment neck dissection should include at least the nodal level(s) demonstrated on the FDG PET-CT and an adjacent level if possible. Evidence for the extent of neck dissection is absent.
In those patients treated with primary surgery with or without adjuvant therapy, a post-treatment baseline MRI scan at 12 weeks is helpful as a future method of comparing changes.
Ultrasound evaluation with needle sampling can be helpful in the decision-making process for patients showing equivocal residual nodal disease on FDG PET-CT. Diffusion-weighted MRI sequences can help to distinguish recurrent disease from radiation-induced soft tissue changes.Reference Vandecaveye, De Keyzer, Nuyts, Deraedt, Dirix and Hamaekers796 Fluoro-deoxy-glucose PET-CT may be a useful adjunct to CT or MRI in inconclusive cases and to rule out distant metastatic disease, particularly if further treatment is planned.
For those patients undergoing non-surgical treatment, there is no evidence that further routine surveillance imaging, besides regular clinical follow up, results in better survival rates for patients who remain asymptomatic after being treated with curative intent and who show a full metabolic and structural response on the post-treatment FDG PET.Reference Wang797
Recurrent disease
Recommendations
• All recurrent cases of oropharyngeal SCC should be re-staged (evidence-based recommendation (R))
• The MDT should have pathways to offer all potential treatments for recurrent cases of oropharyngeal SCC (R)
See also Chapter 5, on follow up, surveillance and recurrent disease. Recurrent disease following previous irradiation is challenging. The local stage and extent of recurrence, coupled with the type of treatment given previously, will be important determinants of the treatment options. Complete re-staging of the primary site and systemic body imaging are recommended, as well as assessment of nutritional status, speech and swallowing function.
Surgical resection in those patients previously treated with RT remains an option in a selected group of patients. Surgery is complicated and should be undertaken by an experienced team. Open surgery is more frequently offered for larger tumours, when transoral access is compromised or when vascularised tissue reconstruction is required. Transoral resection can be offered in select patient groups, including those with vascularised tissue reconstruction, with good oncological and functional outcomes.Reference Warner, O'Hara, Lin, Oozeer, Fox and Meikle773
Other management options apply as for other recurrent cancers, discussed in Chapters 4 and 5, and include: RT for patients previously treated with surgery only; re-irradiation (only in select patients); surgery for isolated lymph node recurrence (Chapter 26); systemic palliative therapy with chemotherapy and/or immunotherapy; and best supportive care.
Ongoing research
It remains unknown whether there is functional benefit from either non-surgical primary treatment or transoral surgery.Reference Dawe, Patterson and O'Hara798,Reference Nichols, Theurer, Prisman, Read, Berthelet and Tran799 This is especially important in the context of de-intensified radiation regimens offering comparable outcomes.Reference Mehanna, Wong, McConkey, Rahman, Robinson and Hartley51
A number of trials on oropharyngeal cancer are currently actively recruiting in the UK, which will report in the next five years. These include the ‘Post-Operative Treatment for HPV-positive Tumours’ (‘PATHOS’) trial, which will report on the impact of lower-dose adjuvant RT in intermediate risk oropharyngeal SCC and the omission of chemotherapy in high-risk oropharyngeal SCC following primary transoral resection. Furthermore, a direct comparison of surgery versus RT for early-stage oropharyngeal SCC will come from the ‘Study Assessing The “Best of” Radiotherapy vs the “Best of” Surgery in Patients With Oropharyngeal Carcinoma (Best Of)’. ‘A trial of proton beam RT for oropharyngeal cancer’ (‘TORPEdO’) is evaluating whether proton beam therapy can reduce the side effects of non-surgical treatments, whilst the ‘Phase III Randomised Controlled Trial Comparing Alternative Regimens for Escalating Treatment of Intermediate and High-risk Oropharyngeal Cancer’ (‘ComPARE’) is investigating standard chemoradiotherapy against dose-escalated chemoradiotherapy or the addition of the programmed death-ligand 1 (PD-L1) inhibitor durvalumab.
Important research questions to be answered
(1) Is p16 immunohistochemistry alone sufficient for prognostication?
(2) Can treatment or surveillance protocols be stratified based on HPV status?
(3) What are oncologically safe transoral surgical margins?
(4) What is the functional and survival profile of transoral surgery and risk-stratified adjuvant treatment versus standard non-surgical treatment?
(5) Does immunotherapy in combination with other agents confer a survival benefit in oropharyngeal SCC?
(6) Would proton beam therapy provide functional or survival benefits over intensity-modulated RT?
Introduction
Nasopharyngeal carcinoma (NPC) has a unique pattern of endemic distribution, with the highest incidences in Southeast and East Asia. Nasopharyngeal carcinoma is non-endemic and rare in Western populations.Reference Chua, Wee, Hui and Chan801,Reference Chen, Chan, Le, Blanchard, Sun and Ma802
In non-endemic populations, there is a bimodal peak, with an initial peak in adolescents/young adults and a second peak after 65 years of age.Reference Chua, Wee, Hui and Chan801,Reference Bray, Ferlay, Soerjomataram, Siegel, Torre and Jemal803
Management of NPC is particularly challenging in non-endemic areas such as the UK. The validity of basing the management of NPC in non-endemic areas upon evidence derived from large endemic trials remains uncertain.
Presentation and diagnosis
Summary recommendations
• Diagnosis should be made by an endoscopic guided biopsy of the nasopharynx (good practice point (G))
• Epstein–Barr virus (EBV) status of lymph node biopsies should be assessed in the investigation of an unknown primary cancer of the head and neck region (evidence-based recommendation (R)). In this scenario, 18F-fluoro-deoxy-glucose (FDG) positron emission tomography (PET) imaging can guide biopsy of the nasopharynx.
Clinical presentation of nasopharyngeal carcinoma (NPC) is usually related either to the extent of primary disease or regional neck lymph node metastases. Because of its highly infiltrative nature, NPC spreads easily in a stepwise fashion via pathways of lower resistance, and via neural pathways and foramina.Reference Dubrulle, Souillard and Hermans804,Reference Lee, Ng, Pan, Poh, Ahn and AlHussain805 Involvement of the nasal cavity, paranasal sinuses, skull base, cavernous sinus, brain parenchyma or orbit can cause a variety of symptoms. These include unilateral nasal obstruction, epistaxis, deafness (typically unilateral) and cranial nerve palsies (with cranial nerves III, IV, Va,b, VI and XII being most commonly affected). Around three-quarters of patients will have regional lymph node metastases at presentation. The retropharyngeal and level II lymph nodes are commonly involved, and skipped lymph node metastases are uncommon.Reference Chua, Wee, Hui and Chan801 An endoscopic guided biopsy of the primary should be performed to confirm the diagnosis.
Initial presentation may be with lymphadenopathy, with no evidence of a primary site clinically or radiologically (i.e. an unknown primary; see Chapter 27). Imaging (magnetic resonance imaging (MRI) of the neck or PET-computed tomography (CT)) and/or in situ hybridisation determining EBV status may suggest a nasopharyngeal primary, and endoscopic biopsies of the nasopharyngeal mucosa should then be performed.
Imaging
Table 1 shows a summary of imaging recommendations for nasopharyngeal carcinoma.806,Reference Chen, Ismaila, Chua, Colevas, Haddad and Huang807
Table 1. Summary of imaging recommendations for nasopharyngeal carcinoma
MRI = magnetic resonance imaging; FNA = fine needle aspiration; CT = computed tomography; FDG = fluoro-deoxy-glucose; PET = positron emission tomography; CRT = chemoradiotherapy; NICE = National Institute for Health and Care Excellence
Summary recommendations
• Magnetic resonance imaging of the skull base and neck is the imaging modality of choice for local staging (good practice point (G))
• Whole-body CT or 18FDG-PET-CT is required for staging (G)
• Magnetic resonance imaging co-registration is recommended for radiotherapy (RT) planning (G)
• 18F-fluoro-deoxy-glucose-PET-CT is the most accurate modality for response assessment (evidence-based recommendation (R))
Accurate mapping of the entirety of perineural spread, the extent of skull base infiltration and the proximity to organs at risk (e.g. optic chiasm) is crucial for RT planning. A close collaborative approach between the radiologist and clinical oncologist is therefore recommended. A rigid co-registration of an MRI acquired within two to three weeks of the planning CT scan improves accuracy of both target volume and organ-at-risk delineation.Reference Lee, Ng, Pan, Poh, Ahn and AlHussain805 Although not currently available in many UK centres, the method of ensuring maximum accuracy of co-registration is for the MRI to be performed in the RT treatment position with the immobilisation devices used for treatment.Reference Lee, Ng, Pan, Poh, Ahn and AlHussain805,Reference Prestwich, Sykes, Carey, Sen, Dyker and Scarsbrook808
Staging and prognostic markers
Staging
Nasopharyngeal carcinoma has its own nodal (N) staging. This is unchanged in the TNM Classification of Malignant Tumours (eighth edition) (Tables 2 and 3).Reference Brierley, Gospodarowicz and Wittekind809
Table 2. Tumour (T) and nodal (N) staging for nasopharyngeal carcinoma*
* According to the TNM Classification of Malignant Tumours (eighth edition).Reference Brierley, Gospodarowicz and Wittekind809
Table 3. Group staging for nasopharyngeal carcinoma*
* According to the TNM Classification of Malignant Tumours (eighth edition).Reference Brierley, Gospodarowicz and Wittekind809
Other prognostic markers for survival risk stratification
In addition to the tumour–node–metastasis (TNM) stage, other independent prognostic markers are important. Plasma Epstein–Barr virus DNA is associated with non-keratinising nasopharyngeal carcinoma, with higher pre-treatment levels having an inferior prognosis.Reference Chen, Chan, Le, Blanchard, Sun and Ma802
Pathology
Based upon the World Health Organization criteria, nasopharyngeal carcinoma is categorised into three subtypes: keratinising; non-keratinising, which includes differentiated and undifferentiated; and a third group of basaloid squamous cell carcinoma. Non-keratinising cancer comprises more than 95 per cent of endemic cases, and there is a strong association with Epstein–Barr virus infection in endemic regions.Reference Chua, Wee, Hui and Chan801,Reference Chen, Chan, Le, Blanchard, Sun and Ma802 Keratinising cancer is more common in non-endemic regions.Reference Chua, Wee, Hui and Chan801 Another potential viral aetiological factor is human papillomavirus (HPV).Reference Chua, Wee, Hui and Chan801,Reference Chan, Lo, Lau and Lam810 In non-endemic areas, HPV appears more frequently in keratinising cancer cases, although is rarely described with an uncertain impact upon prognosis.Reference Huang, Chan and Liu811 Epstein–Barr virus and HPV infections are nearly always mutually exclusive.Reference Chua, Wee, Hui and Chan801
Treatment of non-metastatic nasopharyngeal carcinoma
Table 4 shows a summary of treatment approaches for non-metastatic (M0) nasopharyngeal carcinoma (NPC).
Table 4. Summary of treatment approaches for non-metastatic (M0) nasopharyngeal carcinoma
* See main text for comment on induction chemotherapy plus chemoradiotherapy (CRT) and chemoradiotherapy alone for disease stages T3N1 and T4N0. TNM = tumour–node–metastasis; RT = radiotherapy; EBV = Epstein–Barr virus; GP = gemcitabine, cisplatin; TPF = docetaxel, cisplatin, 5-fluorouracil; PF = cisplatin, 5-fluorouracil
Summary recommendations
• Intensity-modulated RT is standard of care (evidence-based recommendation (R))
• Chemoradiotherapy is recommended for stage II (with the exception of T2N0), stage III and stage IVa disease (R)
• Induction chemotherapy should be considered for all patients with high-risk locoregionally advanced NPC (R)
• In patients who cannot receive induction chemotherapy, adjuvant chemotherapy should be considered (R)
Radiotherapy
Intensity-modulated RT is the current standard treatment.Reference Chen, Ismaila, Chua, Colevas, Haddad and Huang807 Randomised, controlled trials have shown that intensity-modulated RT improves locoregional control and overall survival, and reduces late toxicity including xerostomia, temporal lobe damage and trismus.Reference Zhang, Mo, Du, Wang, Liu and Wei105,Reference Lee, Ma, Ng and Chan812 Local control following intensity-modulated RT for T3 disease or lower is more than 90 per cent, but is lower for T4 disease, with reported local control rates of 74–80 per cent.Reference Lee, Ma, Ng and Chan812 Conventional dose fractionation remains standard, with no established benefit for altered fractionation schedules including modest acceleration,Reference Chua, Wee, Hui and Chan801,Reference Chen, Ismaila, Chua, Colevas, Haddad and Huang807 although a moderately hypofractionated dose of 68.2 Gy in 30 fractions (2.27 Gy per fraction) can be used for small tumours (T1–2N0–1).Reference Chen, Ismaila, Chua, Colevas, Haddad and Huang807,Reference Miao, Di, Chen, Wang, Cao and Xiao813 A commonly agreed standard total dose is in the order of 70 Gy in 33–35 fractions or equivalent (2–2.12 Gy per fraction), with 54–60 Gy for at-risk areas.Reference Chen, Ismaila, Chua, Colevas, Haddad and Huang807,Reference Lee, Ma, Ng and Chan812 Use of a dose of 65 Gy in 30 fractions has also been reported within the UK.Reference Miah, Bhide, Del Rosario, Matthews, Nicol and Tanay814 There is a clear dose–response relationship; under-dosing of the tumour target (65 Gy or lower) is associated with inferior local control and survival.Reference Ng, Lee, Chang, Chan, Chan and Cheung815 Significant anatomical changes are common during treatment because of shrinkage of the primary tumour and lymph node disease, along with changes in body contour due to weight loss.Reference Lee, Ma, Ng and Chan812 Correction of set-up errors and a low threshold for re-planning associated with volume changes during RT are key aspects of treatment delivery.
Primary tumour target volume delineation is a complex process; recent international guidelines are available.Reference Lee, Ng, Pan, Poh, Ahn and AlHussain805 Magnetic resonance imaging co-registration is recommended.Reference Chen, Ismaila, Chua, Colevas, Haddad and Huang807
Lymph node involvement typically occurs in an orderly fashion, with retropharyngeal and level II being the most commonly involved levels; skip metastases are unusual.Reference Ho, Tham, Earnest, Lee and Lu816 Bilateral neck treatment is standard, and traditionally has involved the inclusion of the entire lymph node draining basin including bilateral retropharyngeal lymph nodes down to the lower neck.Reference Ho, Tham, Earnest, Lee and Lu816 There is interest in a more selective approach to neck treatment with the possibility of omission of the lower neck volume in the uninvolved side,Reference Chen, Ismaila, Chua, Colevas, Haddad and Huang807 based on a meta-analysis of nine studies.Reference Huang, Xu, Zhang, Zhou, Mao and Liu817 Elective level Ib irradiation can be omitted in the absence of involvement of the anterior half of the nasal cavity, or if there are level II lymph nodes that are more than 2 cm in size or with evidence of extracapsular spread or bilateral involvement.Reference Lee, Ng, Pan, Poh, Ahn and AlHussain805,Reference Chen, Ismaila, Chua, Colevas, Haddad and Huang807,Reference Ou, Miao, Wang, Ding, He and Hu818 In approximately 25 per cent of NPC patients, involved level II lymph nodes are located above the caudal edge of the lateral process of C1 (which is defined as the superior boundary of level II).Reference Gregoire, Ang, Budach, Grau, Hamoir and Langendijk819 Lymph node mapping studiesReference Lee, Ng, Pan, Poh, Ahn and AlHussain805 have highlighted the need to extend the cranial border to the skull base in order to extend coverage of retropharyngeal lymph nodes, to include the medial retropharyngeal lymph nodes to the caudal edge of C2, and to cover the full posterior triangle including transverse vessels of the neck.
Following induction chemotherapy, international guidelines recommend contouring to include the pre-chemotherapy disease extent,Reference Lee, Ng, Pan, Poh, Ahn and AlHussain805,Reference Chen, Ismaila, Chua, Colevas, Haddad and Huang807 whilst respecting tolerances of organs at risk. Recent studies suggest the potential for an intermediate dose to ‘resolved’ pre-induction induction chemotherapy areas of disease, whilst delivering the full dose to post-chemotherapy residuum.Reference Zhao, Miao, Hua, Wang, Han and Lu820,Reference Yang, Chen, Lin, Rong, Yang and Wen821 These data are useful for some cases of advanced disease following a response to induction chemotherapy, in which it would not be possible to keep organs at risk such as optic pathways within tolerance if treating the pre-chemotherapy extent to full prescription doses.
Proton therapy offers steep dose fall offs beyond target volumes. Dosimetric studies suggest that proton therapy for NPC would deliver lower doses to critical organs at risk, whilst early clinical studies show high rates of locoregional control.Reference Leeman, Romesser, Zhou, McBride, Riaz and Sherman822 It is plausible that proton therapy can result in the reduction of acute toxicities compared to conventional intensity-modulated RT. Proton therapy can be considered for patients aged less than 25 years via the NHS England Proton Beam National Clinical Reference panel.
Role of concurrent chemoradiotherapy
Locally advanced disease (stages III–IVa/b)
Multiple studies in patients with locally advanced (stage III and IVA) NPC have established the use of concurrent chemotherapy as a standard of care,Reference Chua, Wee, Hui and Chan801 supported by an updated Meta-Analysis of Chemo-therapy in Nasopharynx Carcinoma (‘MAC-NPC’).Reference Blanchard, Lee, Marguet, Leclercq, Ng and Ma823 The Meta-Analysis does not show that the benefit of concurrent chemotherapy on overall survival is lost with age, although the hazard ratio for progression-free survival is non-significant for patients aged over 60 years.Reference Blanchard, Lee, Marguet, Leclercq, Ng and Ma823 Concurrent cisplatin, either 100 mg/mReference Chen, Chan, Le, Blanchard, Sun and Ma802 three weekly or 40 mg/mReference Chen, Chan, Le, Blanchard, Sun and Ma802 weekly, are widely used and a recent randomised phase 3 trial confirms that both regimens harbour comparable efficacy, albeit haematological toxicities and late-onset auditory loss were increased with the weekly regimen.Reference Xia, Lv, Liang, Liu, Sun and Zeng824 Achieving a cumulative dose of 200 mg/mReference Chen, Chan, Le, Blanchard, Sun and Ma802 of concurrent cisplatin appears optimal for definitive chemoradiotherapy.Reference Chen, Chan, Le, Blanchard, Sun and Ma802 For patients with a contraindication to cisplatin, carboplatin area under the curve based dosing of 5–6 mg/ml/minute may be a suitable alternative.Reference Chen, Ismaila, Chua, Colevas, Haddad and Huang807,Reference Huang, Cao, Guo, Mo, Guo and Xiang825,Reference Chitapanarux, Lorvidhaya, Kamnerdsupaphon, Sumitsawan, Tharavichitkul and Sukthomya826
Stage II disease
The role of concurrent chemotherapy for stage II NPC is controversial in the intensity-modulated RT era. Stage II disease represents a heterogeneous group, encompassing T2N0 and T1/2N1 disease. Local control is high with RT alone, although a pre-intensity-modulated RT randomised trial has demonstrated a survival advantage for concurrent cisplatin with the benefit being mediated via improved distant metastasis-free survival.Reference Chen, Wen, Guo, Liu, Huang and Mo827 Meta-analyses based on mainly retrospective studies in the endemic populationReference Liu, Jin, Liu, Xiang, Yan and Yang828,Reference Xu, Zhang, Chen, Liu, Zhou and Lin829 do not suggest a clear benefit of concurrent chemotherapy for stage II disease in the intensity-modulated RT era. Recent consensus guidelines recommend the use of concurrent chemotherapy for T1–2N1, reflecting the higher risk of developing distant metastatic disease in stage 2 disease with nodal involvement.Reference Chen, Ismaila, Chua, Colevas, Haddad and Huang807
Role of induction or adjuvant chemotherapy
Given that distant metastases development is a major cause of treatment failure, multiple trials have examined the intensification of systemic therapy with induction chemotherapy or adjuvant chemotherapy.
Induction chemotherapy is an appealing strategy, with high rates of compliance and the potential to eradicate occult metastases early in treatment.Reference Zhang, Huang, Hong, Yang, Yu and Jia830 Results of recent phase III trials have established induction chemotherapy as a standard of care for locoregionally advanced NPC, with reduced rates of distant failure and improved survival outcomesReference Yang, Cao, Guo, Hua, Huang and Zhang831–Reference Sun, Li, Chen, Zhang, Hu and Xie833 following no demonstrated benefit in a prior study.Reference Tan, Lim, Fong, Cheah, Soong and Ang834 It should be noted that these trials excluded older patients, and/or patients with T3–4N0 and T3N1 NPC, in two of the trials.Reference Yang, Cao, Guo, Hua, Huang and Zhang831–Reference Sun, Li, Chen, Zhang, Hu and Xie833 Patients with T3N0 disease are lower risk, and the addition of induction chemotherapy would be based on individual discussions of benefits and risks.Reference Chen, Ismaila, Chua, Colevas, Haddad and Huang807 For patients with T4N0 and T3N1 disease, induction chemotherapy should be offered, although chemoradiotherapy alone can be considered in selected cases, e.g. a non-bulky and/or single-involved neck node (less than 4 cm), and/or a low Epstein–Barr virus (EBV) DNA copy number titre (where available). Whether these results can be extrapolated to the non-endemic population is unclear. Chemoradiotherapy should start 21–28 days from the 1st day of the final cycle of induction chemotherapy.Reference Chen, Ismaila, Chua, Colevas, Haddad and Huang807
The early chemoradiotherapy trials for locoregionally advanced NPC have often included adjuvant chemotherapy with chemoradiotherapy. Whilst the addition of systemic therapy proved to be superior over RT alone, the issue with adjuvant chemotherapy had always been the fact that it is poorly tolerated after an intensive course of chemoradiotherapy.Reference Chua, Wee, Hui and Chan801 A phase III trial failed to show a benefit of chemoradiotherapy-adjuvant chemotherapy over chemoradiotherapy alone in NPC patients who are at high risk of distant metastatic relapse (T3–4N+ or N2–3).Reference Chen, Hu, Chen, Hu, Cheng and Sun835 Adjuvant chemotherapy is an approach that has rarely been used in non-teenage/young adult populations in the UK.Reference Slevin, Pan, Mistry, Sen, Foran and Slevin836,Reference Petkar, Bhide, Newbold, Harrington and Nutting837 However, the updated network Meta-Analysis of Chemo-therapy in Nasopharynx Carcinoma, which comprised mostly endemic NPC cases, showed that concurrent chemoradiotherapy followed by adjuvant chemotherapy was ranked above chemoradiotherapy alone for the failure-free and overall survival.Reference Petit, Lee, Carmel, Ng, Ma and Chan838 Cisplatin and 5-fluorouracil is the standard doublet adjuvant chemotherapy regimen; however, two recent randomised, controlled phase III trials showed that adjuvant full dose or metronomic dose capecitabine is an efficacious adjuvant chemotherapy.Reference Chen, Liu, Zhou, Yang, Jin and Zhu839,Reference Miao, Wang, Tan, Li, Yi and Zhang840
Teenage and young adult population
Based on randomised trial data, the European Cooperative Study Group for Pediatric Rare Tumors (‘EXPeRT’) / Paediatric Rare Tumours Network – European Registry (‘PARTNER’) consensus group recommends treatment of NPC in children and adolescents using: doublet induction chemotherapy with cisplatin and 5-fluoro-uracil, RT doses of 54 Gy in good responders, and the consideration of adjuvant interferon therapy.Reference Ben-Ami, Kontny, Surun, Brecht, Almaraz and Dragomir841 It is important to note that the paediatric trials overwhelmingly involved non-keratinising, undifferentiated EBV-positive patients. This wide differentiation in practice versus adult management presents a challenge for treatment selection in young adults aged over 16 years. Limited prospective and retrospective data suggest that a dose-attenuated approach is effective in EBV-related disease in the young adult population; however, larger prospective studies are needed in this patient group.Reference Claude, Jouglar, Duverge and Orbach842,Reference Uezono, Indelicato, Rotondo, Sandler, Katzenstein and Dagan843
Management of recurrent or distant metastatic disease
Table 5 shows the different approaches to the management of recurrent or metastatic disease.
Table 5. Approaches to management of recurrent or metastatic disease
CRT = chemoradiotherapy
Summary recommendations
• Treatment approaches are individualised for the different recurrence scenarios (good practice point (G))
• Standard of care for first-line palliative systemic therapy is cisplatin plus gemcitabine, with recent trials showing an advantage for this combination with immune checkpoint blockade (evidence-based recommendation (R))
• For patients with de novo metastatic disease and a demonstrated complete or partial response to first-line systemic therapy, consider consolidation RT to the primary and regional sites (G)
• For patients with locally recurrent disease, consider salvage surgery for selected superficial T1–3 tumours; in unresectable cases, re-irradiation with or without chemotherapy can be considered, if deemed at low risk of severe complications with re-treatment (G)
• Isolated regionally recurrent disease can be managed with neck dissection (G)
• For patients who relapse with oligometastases, consider metastasis-directed therapy with or without systemic chemotherapy (G)
Nasopharyngeal carcinoma presenting with synchronous distant metastases
A national cancer database analysis of 718 patients demonstrated superior survival outcomes for patients treated with chemotherapy and RT to the primary site versus those receiving chemotherapy alone, with the benefit maintained in models of single versus multiple organ distant metastases.Reference Rusthoven, Lanning, Jones, Amini, Koshy and Sher844 The overall survival benefits were seen amongst patients receiving 50 Gy or more. This is supported by a phase III trial that closed early after recruiting 126 patients because of an imbalance of deaths, with a median overall survival for the chemotherapy and RT arm of 40.2 months versus 24.5 months in the chemotherapy alone arm.Reference You, Liu, Huang, Zou, Sun and He845
Nasopharyngeal carcinoma with oligometastatic disease
Nasopharyngeal carcinoma patients with oligometastatic disease have a considerably better prognosis than those with widespread distant metastases.Reference Tian, Zou, Xiao, Zeng, Yuan and Bai846 Patients with lung-only metastasis have a more favourable outlook, whilst liver metastases are associated with inferior survival outcomes.Reference Tian, Zou, Xiao, Zeng, Yuan and Bai846 For patients presenting with synchronous oligometastatic disease, aggressive treatment with ablative treatment (e.g. stereotactic RT or metastasectomy) of oligometastases, systemic therapy and consolidation RT to the primary tumour is advocated.Reference Chua, Wee, Hui and Chan801,Reference Tian, Zou, Xiao, Zeng, Yuan and Bai846 Longer-term survival is possible for selected patients with metachronous oligometastases following ablative therapy.Reference Bates, De Leo, Morris, Amdur and Dagan847,Reference Pan, Wu, Yu, Li, Huang and Wang848
Palliative systemic therapy for distant metastatic disease
The seminal phase III trial by Zhang et al. established cisplatin with gemcitabine as the standard of care.Reference Zhang, Chen, Hu, Zhang, Zhu and Yang832 Two other randomised, controlled phase III trials investigated the role of combining the anti-programmed cell death-1 (anti-PD1) antibody; recent results confirmed a further improvement in progression-free survival when compared to cisplatin plus gemcitabine (the ‘JUPITER-02’ trial, investigating the efficacy and safety of toripalimab injection combined with chemotherapy for nasopharyngeal cancer, reported median progression-free survival of 11.7 months vs 8.0 months;Reference Xu, Mai, Chen, Chen, Hu and Yang849 the ‘CAPTAIN-1st’ trial, a phase III study of camrelizumab in combination with chemotherapy in recurrent or metastatic nasopharyngeal carcinoma, reported median progression-free survival of 9.7 months vs 6.9 monthsReference Yang, Qu, Li, Hu, Xu and Li850). Another recent clinical trial from ChinaReference Liu, Li, Wang, Liang, Lv and Ye851 also reported prolonged progression-free survival with maintenance capecitabine after cisplatin plus gemcitabine (median progression-free survival of 35.2 months vs 9.1 months), albeit the results were immature for overall survival. Single-agent chemotherapy with taxanes or capecitabine are appropriate second-line therapies, with response rates of 30–40 per cent.Reference Chua, Wee, Hui and Chan801
Local recurrence
Long-term disease control is achievable, and so aggressive treatment is often appropriate, especially for limited local recurrences, although the risk of complications is high. Options include surgery or re-irradiation with or without chemotherapy. Prior treatment including doses to organs at risk, latency from treatment, late toxicity from prior treatment, age, co-morbidity, performance status and extent of recurrence are all key factors.
Open or endoscopic nasopharyngectomy should be considered if it is considered likely that clear margins can be achieved;Reference Ng, Soong, Chan Ahn, AlHussain, Choi and Corry852 one series reported five-year post-recurrence survival of 56 per cent following surgery for early stage recurrence.Reference Ng, Wong, Cheung, Chow, Poon and Lai853 The availability of surgical expertise is critical, and re-irradiation is an alternative.Reference Ng, Soong, Chan Ahn, AlHussain, Choi and Corry852 Complications of open surgical approaches are high (over 40 per cent, with risks of trismus, fistula and facial numbness), and endoscopic approaches have been developed.Reference Chen, Wang, Zhu, Shen, Qiu and Luo854,Reference Zou, Han, Ma, Deng, Jiang and Guo855
Recent international guidelines for re-irradiation have been published.Reference Ng, Soong, Chan Ahn, AlHussain, Choi and Corry852 Re-irradiation is an effective treatment, with a five-year failure-free survival rate of 72 per cent and an overall survival rate of 41 per cent reported in a meta-analysis, although risks of late complications are high, with one-third experiencing fatal toxicity.Reference Leong, Soon, Lee, Wong, Tham and Ho856 Re-irradiation is not appropriate for disease that recurs less than one year from treatment indicating radio-resistance. Re-irradiation after surgery is advisable for positive margins and can be considered for close margins.Reference Ng, Soong, Chan Ahn, AlHussain, Choi and Corry852 Nonetheless, it must be cautioned that re-irradiation following surgery, especially for recurrent tumours, is linked to substantially increased rates of severe morbidities.Reference Maghami, Koyfman and Weiss857 Re-irradiation should also be considered for unresectable local recurrence. Based on an extrapolation from the de novo trials, consideration can be given to integrating induction chemotherapy with or without concurrent chemotherapy with re-irradiation. The total dose for re-irradiation is recommended to be in the order of 60–66 Gy EQD2 (equivalent dose in 2 Gy fractions);Reference Ng, Soong, Chan Ahn, AlHussain, Choi and Corry852 hyper-fractionation may improve the therapeutic window (ongoing trial NCT02456506).Reference Ng, Soong, Chan Ahn, AlHussain, Choi and Corry852 Elective lymph node treatment is not recommended in this scenario.Reference Ng, Soong, Chan Ahn, AlHussain, Choi and Corry852 Using a survival risk prediction model, Li and colleagues were able to stratify patients into low- and high-risk groups; the latter was at risk of re-irradiation mortality.Reference Li, Tian, Tan, Liu, Kusumawidjaja and Ong858 For this group, chemotherapy with or without anti-PD1 antibody could be appropriate.Reference Chen, Chan, Le, Blanchard, Sun and Ma802
Isolated regional recurrence
Isolated regional lymph node recurrences can be managed with a neck dissection. Selective and comprehensive neck dissections have similar outcomes, with a three-year overall survival rate of 67 per cent; negative prognostic markers include extracapsular spread, recurrent N stage and positive margins.Reference Liu, Li, You, Li, Liu and Yang859
Follow up
The recommendations for follow up are set out in Chapter 5.
Recommendations
• Response assessment should take place at least 12 weeks post treatmentReference Chen, Chan, Le, Blanchard, Sun and Ma802 (evidence-based recommendation (R)), including clinical examination, nasoendoscopy and imaging (PET-CT/MRI), with or without Epstein–Barr virus (EBV) DNA titres (if detectable pre-treatment) (good practice point (G))
• Ongoing follow up should include clinical examination with nasendoscopy (G)
• Surveillance imaging is not routinely recommended (G)
• Monitoring for late complications is a key component of follow up (G)
In endemic areas, persistent detection or re-emergence of plasma EBV DNA post treatment has been shown to correspond to an increased likelihood of relapse and a precursor of clinical disease; monitoring of this biomarker may thus have a role in surveillance.Reference Chua, Wee, Hui and Chan801,Reference Zhou, Wu, Deng, Gao, Lv and Lin860
Monitoring for late complications of treatment is a key component of follow up (see Chapter 16). Late effects include sensorineural and conductive hearing loss, visual impairment, endocrine (pituitary and thyroid) dysfunction, xerostomia, dysphagia, soft tissue fibrosis, osteoradionecrosis, and neurological complications (e.g. temporal lobe injury, cranial neuropathy, brachial plexus neuropathy).Reference Chua, Wee, Hui and Chan801,Reference Chen, Chan, Le, Blanchard, Sun and Ma802
Studies due to report
There is an ongoing phase 3 trial (NCT026330202) comparing concurrent chemoradiotherapy versus RT alone for stage II and T3N0M0 disease, as well as an ongoing NRGHN001 trial (NCT02135042) exploring the use of post-RT plasma Epstein–Barr virus DNA to identify subgroups who may benefit from adjuvant chemotherapy.
Research questions
• Establishing the place for proton therapy.
• Incorporating immunotherapy into current treatment paradigms.
• Selecting patients for induction chemotherapy, and determining the optimal induction chemotherapy regimen.
Introduction
Cancers originating in the hypopharynx are relatively rare, accounting for approximately 3 per cent of head and neck cancers in the UK.861 They are more prevalent in men, and the prognosis is generally worse than cancer in other head and neck sites, with an overall five-year survival rate of around 25 per cent in the UK,861 increasing to 40 per cent for patients suitable for curative treatment.Reference Newman, Connolly, Illing, Kilgore, Locher and Carroll862 This is partly due to advanced stage disease at presentation. However, early-stage tumours (T1–2) treated with curative intent have an overall five-year survival of around 60 per cent or more.Reference Kim, Kim, Koh, Chung, Son and Oh863
The hypopharynx is subdivided into the pyriform sinuses, the posterior pharyngeal wall and the post-cricoid areas, with the majority of cases being located in the pyriform sinuses (60–85 per cent).Reference Newman, Connolly, Illing, Kilgore, Locher and Carroll862 Lymphatic drainage from the hypopharynx is rich and drains to the jugular as well as retropharyngeal lymph nodes. Over 50 per cent of patients present with metastatic lymphadenopathy.
Many patients have significant co-morbidities. Baseline performance status is an independent prognostic factor for overall survival, and this should be borne in mind when determining management options.
Diagnosis
Recommendations
• Perform endoscopic examination under general anaesthesia for biopsy and to determine the extent of primary tumour (if suitable for anaesthesia) (evidence-based recommendation (R))
• Narrow-band imaging increases the diagnostic potential of endoscopy in the early detection of hypopharyngeal malignant lesions (good practice point (G))
• Transnasal oesophagoscopy can facilitate early evaluation and detection, and allow for biopsies in the clinic (G)
Presenting symptoms and physical examination
Symptoms from hypopharyngeal cancer (HPC) are determined by the size and location of the tumour. These include sore throat, odynophagia, dysphagia, referred otalgia, blood-stained phlegm, airway obstruction and aspiration. However, the most common presentation of HPC is a neck mass, and the incidence of regional metastasis at presentation is 70 per cent, especially in cancers that originate in the pyriform sinuses. This incidence is higher than in any other head and neck location.Reference Spector864
Assessment of a patient with HPC should comprise a full head and neck examination, including oral cavity examination, neck palpation and fibre-optic examination of the upper aerodigestive track. Suggestive signs include the presence of a mucosal ulceration, erythematous lesions, vocal fold immobility or pooling of secretions.
Narrow-band imaging increases the diagnostic benefit of endoscopy in the detection of early malignant lesions within the hypopharynx.Reference Waclawek, Milonski and Olszewski865 Laryngopharyngeal evaluation with transnasal oesophagoscopy, including hypopharyngeal biopsies under topical anaesthesia, can be used to help with early evaluation and biopsy, therefore shortening the diagnostic pathway.Reference Schimberg, Wellenstein, van den Broek, Honings, van den Hoogen and Marres866
Endoscopy and tissue sampling
Examination under general anaesthesia enables tissue biopsy, assessment of the extent of the tumour and determination of the presence of synchronous tumours, which can be found in 10–15 per cent of the patients.Reference Hoffman, Karnell, Shah, Ariyan, Brown and Fee867 Again, narrow-band imaging may help. Photographic documentation of the clinical lesion as well as documentation of transoral access should be included, using the appropriate grading system.
Imaging
Table 1 shows recommendations for pre-treatment imaging in cancers originating in the hypopharynx.
Table 1. Recommendations for pre-treatment imaging in cancers originating in the hypopharynx
MRI = magnetic resonance imaging; CT = computed tomography; PET = positron emission tomography; T = tumour stage; N = nodal stage
Recommendations
• Positron emission tomography (PET-) computed tomography (CT) is advised in recurrent HPC cases, in advanced disease cases and to determine the inferior extent of the tumour (good practice point (G))
Imaging should be conducted prior to a biopsy wherever possible, to avoid a post-biopsy artefact caused by oedema and subsequent over-staging.
Assessment of the primary tumour extent and size should be performed with magnetic resonance imaging (MRI) pre- and post-gadolinium enhancement and/or with multi-slice intravenous contrast-enhanced CT. Magnetic resonance imaging is recommended for the evaluation of soft-tissue extension, and in specific clinical scenarios such as possible early cartilage invasion. A combination of axial, coronal and sagittal T1-weighted and T2-weighted sequences may be required. A slice thickness of 3 mm is recommended for the assessment of early laryngeal cartilage invasion. A CT scan can be performed at the same time as the CT scan of the chest, to enable full staging of the tumour.
The PET scan is useful to confirm the extent of the primary site, specifically the inferior extent in post-cricoid cancers, in order to detect distant metastatic spread in cases of occult disease and in recurrent tumours.Reference Simo, Rovira and Townley868 Current National Institute for Health and Care Excellence (NICE) guidelines indicate the use of PET-CT for T4 cancers of the hypopharynx and nodal stage N3 cancers, which are most likely to have distant metastases.869
As for other upper aerodigestive tract cancers, the NICE guidelines do not mandate systemic imaging for T1–2N0 cancers. However, it is standard practice to stage the thorax via a chest CT (if PET-CT is not indicated) in all patients, as there is a relatively high incidence of second primaries or metastasis at presentation.
Staging
The Union for International Cancer Control tumour–node-metastasis (TNM) classification (eighth edition) for HPC is described in Tables 2–5.Reference Brierley, Gospodarowicz and Wittekind763
Table 2. Staging of primary tumours in hypopharyngeal cancer
*According to the TNM Classification of Malignant Tumours (eighth edition).Reference Brierley, Gospodarowicz and Wittekind763 †Central compartment soft tissue includes pre-laryngeal strap muscles and subcutaneous fat.
Table 3. Clinical staging of regional lymph nodes in hypopharyngeal cancer*
* According to the TNM Classification of Malignant Tumours (eighth edition).Reference Brierley, Gospodarowicz and Wittekind763
Table 4. Pathological staging of regional lymph nodes in hypopharyngeal cancer*
* According to the TNM Classification of Malignant Tumours (eighth edition).Reference Brierley, Gospodarowicz and Wittekind763 †Indicates upstaging from clinical stage when extra-nodal extension is confirmed.
Table 5. Group staging of hypopharyngeal cancer*
* According to the TNM Classification of Malignant Tumours (eighth edition).Reference Brierley, Gospodarowicz and Wittekind763
Pathology
The vast majority (95 per cent) of HPCs are squamous cell carcinomas (SCCs), with two-thirds of these being keratinising SCC. Other tumour types include lymphoma, sarcoma and adenocarcinoma; these are typically poorly differentiated and aggressive.Reference Hoffman, Karnell, Shah, Ariyan, Brown and Fee867
Only a small proportion of HPC cases have been associated with human papillomavirus infection,Reference Eckel and Bradley870 and there is no proven impact on HPC treatment or prognosis as yet.
Pre- and post-treatment speech and swallow rehabilitation
All patients presenting with head and neck cancer should undergo speech and swallowing and dietetic pre-treatment assessment, as discussed in other chapters.
Management of hypopharyngeal cancer
Table 6 shows the treatment options for hypopharyngeal cancer.
Table 6. Treatment of hypopharyngeal cancer
* Applies only to radiotherapy. TNM = tumour–node–metastasis; RT = radiotherapy; TLM = transoral laser microsurgery; CRT = chemoradiotherapy
Recommendations
• Treat early-stage cancers with single-modality treatment and offer patients all viable options (evidence-based recommendation (R))
• Offer primary concomitant chemoradiotherapy/radiotherapy (RT) or surgery plus (chemo)radiotherapy to patients with locally advanced disease as options for treatment with curative intent (R)
• Consider post-operative RT in patients with adverse pathological risk factors (good practice point (G))
• The addition of chemotherapy (in suitable patients) combined with post-operative RT is recommended in patients with positive margins or extracapsular extension (R)
Early-stage hypopharyngeal cancer T1–2N0
Early-stage HPC should be treated as much as possible with a single-modality treatment (surgery or RT).Reference Machiels, Rene Leemans, Golusinski, Grau, Licitra and Gregoire871,Reference Garneau, Bakst and Miles872 Most patients are treated with RT. There are insufficient data to ascertain the functional and oncological results from primary surgery, and to determine how often this is possible as a single-modality treatment option. However, reported outcomes suggest equipoise with RT in selected patients.Reference Kim, Kim, Koh, Chung, Son and Oh863
Treatment with RT should consist of intensity-modulated RT or volumetric-modulated arc therapy.Reference Machiels, Rene Leemans, Golusinski, Grau, Licitra and Gregoire871 The standard primary radical RT fractionation used for head and neck cancer patients in the UK currently varies mainly between 70 Gy in 35 fractions over seven weeks and 65 Gy in 30 fractions over six weeks.Reference Vasiliadou, Noble, Hartley, Moleron, Sanghera and Urbano873 Radiotherapy should include neck levels IIa, III and IV. Prophylactic irradiation of retropharyngeal nodes also provides effective regional control.Reference Nguyen, Vock, Vinh-Hung, Almeida, Ewell and Betz874
Surgical management includes open surgical procedures and transoral approaches (transoral laser surgery and transoral robotic surgery). This should be combined with selective neck dissection, bilateral for midline tumours. Transoral approaches are preferred, having less morbidity compared to open surgery, and with a reported laryngeal preservation rate of 70 per cent.Reference Imanishi, Ozawa, Sakamoto, Fujii, Shigetomi and Habu875,Reference Wei876 Complete excision with margins of 5 mm or greater should be the mainstay of treatment. Positive margins should be considered for re-resection or post-operative chemoradiotherapy. Other adverse pathological features, including on neck dissection, may also mandate adjuvant therapy.Reference Lacas, Carmel, Landais, Wong, Licitra and Tobias109
Locally advanced cancer (stage III–IV hypopharyngeal cancer)
The options for curative treatment are laryngeal preservation with radical RT with concurrent chemotherapy, or primary surgery with pharyngolaryngectomy.
As for other upper aerodigestive tract cancers, concomitant chemoradiotherapy has greater locoregional control and improved overall survival in comparison with RT alone, but its effect decreases with age over 70 years; RT alone is generally used for patients older than 70 years. Induction chemotherapy is not recommended.
Concomitant chemoradiotherapy has the obvious advantage of organ preservation, with surgery reserved for locally recurrent disease, if appropriate (albeit with increased rates of post-operative morbidity). However, functional status before treatment is a key determinant of functional status after chemoradiotherapy. Pre-treatment extensive invasion of surrounding structures (T4a tumours) and poor laryngeal function are associated with poor functional outcomes. According to Chen et al., in patients with T4a disease, hyoid bone invasion significantly increases the severe complication rate in organ preservation treatment, with overall pharyngeal dysfunction, laryngeal dysfunction and aspiration rates of 36 per cent, 27 per cent and 25 per cent, respectively.Reference Chen, Huang, Tsou, Bau and Tsai877
Hence, it is generally recommended that patients with T4a tumours be treated with pharyngolaryngectomy.Reference Chen, Huang, Tsou, Bau and Tsai877 Reconstruction after pharyngolaryngectomy is discussed in Chapter 7.
In select smaller stage III tumours, partial laryngectomy with partial pharyngectomy might be considered.Reference Laccourreye, Ishoo, de Mones, Garcia, Kania and Hans878
Preservation of part or all of the thyroid gland should be the aim in patients undergoing surgery for cancers originating in the hypopharynx, as thyroid invasion is uncommon.Reference Elliott, Odell, Tysome, Connor, Siddiqui and Jeannon879
Standard indications for considering post-operative RT apply after surgery, as for other SCCs of the upper aerodigestive tract (see Chapter 4). These include pathologically T3–4 tumours, positive margins (≤1 mm), close margins (1–5 mm), perineural or lymphovascular spread, more than one involved lymph node, or extracapsular extension. The radiotherapy dose should be up to 60–66 Gy in 30–33 daily fractions. Concomitant chemoradiotherapy with platinum-based chemotherapy is indicated in the presence of extracapsular extension or positive margins. Post-operative RT should be started within six to seven weeks after surgery (Table 6).Reference Machiels, Rene Leemans, Golusinski, Grau, Licitra and Gregoire871
Management of the neck
See also Chapter 26.
Recommendations
• The neck should be treated bilaterally for tumours involving the midline, post-cricoid area, medial wall of the pyriform sinus or posterior pharyngeal wall (evidence-based recommendation (R))
• Clinically node-negative (N0) neck – levels II to IV should be treated electively (R)
• Clinically node-positive (N+) neck – levels II(a/b) to V should be treated (R)
For patients with no evidence of cervical lymph node metastases (clinically N0), elective treatment by selective neck dissection or irradiation of cervical lymph node levels II–IV is recommended.Reference Schmitz, Machiels, Weynand, Gregoire and Hamoir880 Treatment should be bilateral for tumours involving the midline, post-cricoid area, the medial wall of the pyriform sinus, the posterior pharyngeal wall, the posterior annular region or tumours across the midline.Reference Viñals, Rodrigues, Lopez, Payro, Porté and Sildenikova236,Reference Johnson, Bacon, Myers and Wagner881
There is no clear evidence regarding the extent of neck dissection for patients undergoing primary surgery with N+ disease. In addition to levels IIA–IV, neck dissection of levels IIB and V would be recommended, accepting the increased risk of accessory nerve dysfunction. Level I can be spared unless involved radiologically.
Follow up
Locally advanced head and neck cancer carries a high risk of local recurrence (15–40 per cent depending on staging and location) and distant metastasis. Follow up is described in Chapter 5. In patients who have had treatment for HPC, surveillance should include flexible laryngoscopy (if pharyngolaryngectomy has not been performed). Persistent or recurrent pain should be considered as a serious warning sign of recurrent HPC, even if there is no endoscopic evidence of persistent and/or recurrent disease.Reference Simo, Rovira and Townley868,Reference Scharpf, Karnell, Christensen and Funk882 Assessment under general anaesthesia and biopsies are indicated in cases of suspected recurrence after performing a PET-CT. Of patients treated with RT to the larynx, 6–53 per cent will develop hypothyroidism.Reference Brunet, Tornari, Ezebuiro, Kennedy, Connor and Oakley883 Therefore, thyroid-stimulating hormone levels should be monitored every 6–12 months. Total laryngectomy alone is also associated with hypothyroidism in 13–48 per cent of cases.Reference Brunet, Tornari, Ezebuiro, Kennedy, Connor and Oakley883
Post-cricoid hypopharyngeal stenosis can occur after organ-preservation strategies for HPC. Its pathogenesis is not well established, although it could be the consequence of chemoradiation-induced mucositis which results in ulceration of the mucosal surface of the post-cricoid region, with subsequent circumferential fibrosis formation which leads to post-cricoid stenosis.Reference Sullivan, Jaklitsch, Haddad, Goguen, Gagne and Wirth884 Dilatation for this complication or for stenosis following pharyngolaryngectomy may be required in some patients.
Recurrent tumours
The management of recurrent head and neck cancer in general is discussed in Chapter 5. Locoregional recurrence cancers should be restaged with endoscopy under general anaesthesia and MRI (of the neck); for distant metastases and second primary disease, CT of the thorax or (preferably) PET-CT should be conducted.Reference Isles, McConkey and Mehanna885,Reference Ul-Hassan, Simo, Guerrero-Urbano, Oakley, Jeannon and Cook886 The management of isolated recurrence in cervical lymph nodes is discussed in Chapter 26.
With regard to HPC, a small proportion of patients treated initially with chemoradiotherapy or RT may be suitable for salvage surgery. For such individuals, the outcomes are often poor. For example, one series reported a 40 per cent rate of major post-operative complications, with a median overall survival of 17 months and a 5-year survival rate of 30 per cent.Reference Zhang, Huang, Xu, Liu, Li and An887
Other options, discussed in Chapter 4, include palliative chemotherapy, immunotherapy, (rarely) re-irradiation and best supportive care.
The complications and toxicities from further surgery, RT and systemic treatments can significantly impact the quality of life of patients with a very low likelihood of cure, even for those undergoing treatment with curative intent. Hence, multidisciplinary team discussion about treatment options, as well as detailed discussion with patients regarding the potential benefits and the risks of treatments, are crucial.
Studies due to report
There are no National Cancer Research Institute phase III/III clinical trials specific to hypopharyngeal cancer. Global trials can be found in: https://clinicaltrials.gov.
Important research questions to be answered
• Refining the choice between primary surgery and organ-preservation RT/chemoradiotherapy in advanced HPC.
• Defining the place of transoral surgery in early HPC.
• Predicting the outcome after salvage pharyngolaryngectomy: identifying patients in whom salvage surgery will not improve survival and who might benefit from immunotherapy despite having potentially resectable disease.
Introduction
This chapter concentrates on squamous cell carcinoma (SCC), which accounts for over 90 per cent of laryngeal cancer cases. Laryngeal SCC is primarily caused by cigarette smoking, although approximately 20 per cent of patients are non-smokers.Reference Shoffel-Havakuk, O'Dell, Johns, Reder, Popova and Halperin888
For most patients with laryngeal cancer, primary surgery or primary radiotherapy (RT) (with or without chemotherapy) are appropriate treatment options. There is a lack of clinical trial data comparing different treatment modalities for laryngeal SCC, and there have been no significant published clinical trials since the last version of these guidelines. In situations where surgery and RT-based treatments are thought to be equally effective, the decision of which treatment to proceed with is based on the values and perspectives of the individual patient.
Presentation and diagnosis
Laryngeal SCC usually presents with a hoarse voice. Supraglottic laryngeal SCC may present with pain. Symptoms suggestive of more advanced laryngeal SCC include: odynophagia, dysphagia, referred otalgia, palpable lymphadenopathy and stridor.
Out-patient clinical examination must include flexible laryngoscopy. Photographic documentation is good practice for diagnosis, referrals, follow-up surveillance and training.Reference Schache, Kerawala, Ahmed, Brennan, Cook and Garrett71 Vocal fold mobility must be documented.
Patients with clinically concerning laryngeal lesions should undergo laryngoscopy under general anaesthesia, which is more frequently performed with an array of angled endoscopes than with the microscope. Narrow-band imaging and similar technologies may improve diagnosis.Reference Stanikova, Walderova, Jancatova, Formanek, Zelenik and Kominek889 In addition to an appropriate biopsy, photographic documentation of the lesion should be made in the operating theatre. There should be sufficient anatomical detail, particularly details on transoral access, to allow the treating multidisciplinary team (MDT) to establish appropriate treatment options, especially the option of transoral microsurgery. In patients not fit for general anaesthetic assessment, channelled laryngoscopy may permit biopsy under local anaesthesia.
Recommendations
• Photographic documentation should be available, and be used as a record of out-patient and operative laryngoscopy findings (good practice point (G))
• Vocal fold mobility must be documented at out-patient laryngoscopy (evidence-based recommendation (R))
• Any restriction in transoral access should be documented when performing diagnostic laryngoscopy (R)
Imaging
Table 1 shows recommendations for pre-treatment imaging (note that points 1–3 below correspond with the same numbers in Table 1).Reference Shoffel-Havakuk, O'Dell, Johns, Reder, Popova and Halperin888–Reference Stanikova, Walderova, Jancatova, Formanek, Zelenik and Kominek889
Table 1. Recommendations for pre-treatment imaging
Note that the numbers in parentheses correspond with the same numbers in the main text when referring to Table 1. MRI = magnetic resonance imaging; CT = computed tomography; PET = positron emission tomography; N = nodal stage
(1) Imaging is not necessary for superficial tumour stage T1 cancers not involving the anterior commissure, unless there is concern regarding lateral extension and paraglottic involvement.
(2) Whilst the probability of distant metastasis is very low so as not to require systemic imaging, there is some uncertainty about the use of thorax computed tomography (CT) for screening for synchronous primary bronchial malignancy, given the similar risk factors and relatively high instance of pulmonary nodules and lung cancer in patients with laryngeal carcinoma.Reference Piersiala, Akst, Hillel and Best50 Despite National Institute for Health and Care Excellence guidance to the contrary,890 mainstream practice follows a pragmatic approach of routinely performing CT of the thorax in patients with laryngeal carcinoma, irrespective of primary stage, with a recent study showing a 9 per cent rate of second primary cancers or distant metastases in patients with T1/2 laryngeal SCC who are smokers.Reference Chew, Cosway and Hamilton891
(3) Positron emission tomography (PET)-CT is preferred for nodal stage N3 disease.890
Computed tomography imaging benefits from a short scanning time, wide availability and excellent anatomical resolution. The superior contrast resolution of magnetic resonance imaging (MRI) enables better evaluation of the thyroid cartilage.Reference Becker, Zbären, Casselman, Kohler, Dulguerov and Becker892 Magnetic resonance imaging is also superior for assessing tongue base and pre-epiglottic fat invasion.Reference Olliff, Richards, Connor, Wong, Beale, Madani and Nicholson893
After RT with or without chemotherapy for advanced disease, post-treatment baseline imaging is recommended, with PET-CT following RT or chemoradiotherapy for stage III/IV disease.
Staging
Laryngeal cancer staging relies heavily on clinical examination findings, including the site of origin, involvement of compartments and determination of vocal fold mobility to assess the tumour. Radiological or pathological staging of early laryngeal cancers often requires clinical information for accurate staging.
The eighth edition of the American Joint Committee on Cancer/Union for International Cancer Control tumour–node–metastasis (TNM) stagingReference Brierley, Gospodarowicz and Wittekind894 made no changes to primary site staging for laryngeal carcinoma (Tables 2–4).
Table 2. Primary tumour staging for laryngeal cancer*
* According to the TNM Classification of Malignant Tumours, eighth edition.Reference Brierley, Gospodarowicz and Wittekind894
Table 3. Nodal staging for laryngeal cancer*
* According to the TNM Classification of Malignant Tumours, eighth edition.Reference Brierley, Gospodarowicz and Wittekind894 †The presence of skin involvement or soft tissue invasion with deep fixation or tethering to underlying muscle or adjacent structures, or clinical signs of nerve involvement, is classified as extra-nodal extension. Midline nodes are considered ipsilateral nodes.
Table 4. Group staging for laryngeal cancer*
* According to the TNM Classification of Malignant Tumours, eighth edition.Reference Brierley, Gospodarowicz and Wittekind894
Pathology
Biopsy samples of the larynx are often fragmented and poorly orientated. The use of devices producing thermal artefacts to take purely diagnostic samples is not recommended given the unacceptable level of tissue distortion often associated with small samples. Biopsy material should be entirely processed and examined at multiple levels. Minimum criteria should include confirmation of the type of carcinoma, grade, if SCC, and presence and grade of epithelial dysplasia.
Squamous cell carcinoma is the most common cancer of the larynx (more than 90 per cent of all cases).
A number of morphological variants of SCC have been noted to arise within the larynx: verrucous SCC, basaloid SCC, papillary SCC, spindle cell SCC, adenosquamous carcinoma and lymphoepithelial carcinoma.
These variants are associated with varying difficulties in pathological diagnosis, with reported differences in prognosis, metastatic potential and response rates to different treatment modalities. There is no good evidence for a different treatment approach for these variants and, stage for stage, treatment should be the same as for classical SCC.
Other histological types of malignant disease arising within the larynx include: neuroendocrine carcinoma, minor salivary gland carcinoma (see also Chapter 22), melanoma (see also Chapter 27), and soft tissue tumours such as liposarcoma, granular cell tumour, inflammatory myofibroblastic tumour, chondroma and chondrosarcoma. These are relatively rare. Neuroendocrine and soft tissue tumours should be co-managed with referral to specialised neuroendocrine tumour and sarcoma MDTs respectively.
Speech and language therapy
Laryngeal cancer and its treatment can profoundly affect voice and swallowing. Hence, the involvement of speech and language therapy in patient management is pivotal. Speech and language therapy in head and neck cancer, including in relation to total laryngectomy, is discussed in detail in Chapter 10. Any consequent effects on nutrition are discussed in Chapter 9.
Recommendations
• All patients with laryngeal SCC should be assessed by a speech therapist before treatment and during rehabilitation (good practice point (G))
• The impact of treatment options on voice quality, swallowing and other areas of function should be discussed with all patients (evidence-based recommendation (R))
• All patients undergoing total laryngectomy should be considered for speech valve rehabilitation (R)
Management of laryngeal dysplasia
Laryngeal dysplasia can be associated with any abnormal mucosal appearance, most commonly leukoplakia or erythroleukoplakia, but also hyperkeratosis, mucosal reddening or simply thickening.
The World Health Organization (WHO) classification system divides laryngeal dysplasia into two categories: low-grade and high-grade dysplasia. Carcinoma in situ is included within high-grade dysplasia, and the WHO 2017 guidelines allow the use of this term if preferred.Reference El-Naggar, Chan, Grandis, Takata and Slootweg65 The risk of malignant transformation is approximately 10 per cent for low-grade dysplasia and 30 per cent for high-grade dysplasia, with a mean time to malignancy of six years.Reference Weller, Nankivell, McConkey, Paleri and Mehanna895 A useful flowchart describing the management of laryngeal dysplasia has been written by Cosway and Paleri.Reference Cosway and Paleri896
Recommendations
• Describe laryngeal dysplasia as low or high grade. Avoid the use of other descriptors (e.g. moderate) (evidence-based recommendation (R))
• Treatment of low-grade dysplasia – excisional biopsy, with a minimum of six months’ surveillance post-excision biopsy. Patients with continued mucosal abnormalities should remain under surveillance, especially if they continue to smokeReference Brierley, Gospodarowicz and Wittekind894 (good practice point (G))
• Treatment of high-grade dysplasia (including carcinoma in situ) – definitive treatment required (R). Transoral laser microsurgery is preferred. Follow up as per T1a laryngeal SCC recommendations
• Patients should be offered support in stopping smoking (R)
• Photographic documentation should be used for follow up and surveillance (G)
Narrow-band imaging may assist in follow up and surveillance. Radiotherapy can be used to treat high-grade dysplasia in patients with multiple recurrences, a high anaesthetic risk, wide-field multiple lesions, or poor endoscopic access preventing transoral laser microsurgery, and is associated with roughly equivalently high rates of local control as compared to transoral laser microsurgery.Reference Eckel, Simo, Quer, Odell, Paleri and Klussmann897
Early laryngeal cancer (T1–2N0M0)
Key recommendations
• The options of transoral laser microsurgery or RT should be discussed with every patient (evidence-based recommendation (R))
• Transoral laser microsurgery for T1a glottic tumours is a cost-effective treatment and recommended in National Institute for Health and Care Excellence (NICE) guidance (good practice point (G))
• Single-modality treatment should be the principal aim (R)
• Transoral surgical techniques should not be considered if complete surgical excision of the tumour is in doubt (R)
Table 5 summarises the recommendations for the primary treatment of early-stage cancers.
Table 5. Summary recommendations for primary treatment of early-stage cancers
* Transoral laser microsurgery, transoral robotic surgery or transoral endoscopic techniques. T = tumour stage; N = nodal stage; TLM = transoral laser microsurgery; RT = radiotherapy
Early laryngeal SCC comprises T1a glottic laryngeal SCC in 50–60 per cent of cases; 20 per cent of cases are T2 glottic laryngeal SCC and 12 per cent are supraglottic laryngeal SCC.898
Laryngeal SCC staged T1–2N0 should be treated ideally with single-modality therapy using transoral surgical techniques or RT. Transoral surgery requires appropriate expertise and patient selection to achieve clear margins and minimise the requirement for post-operative RT. There are no suitable randomised, controlled trials (RCTs) assessing oncological or functional outcomes between the two treatment modalities.
Glottis
Stage T1a disease
Local and overall survival are similar for both RT and transoral laser microsurgery. For example, a pooled case series showed very similar three-year local control rates for both treatments, of 89 per cent (88.9 per cent for transoral laser surgery and 89.3 per cent for RT).Reference O'Hara, Markey and Homer899 Whilst treatment with initial transoral laser microsurgery may offer superior laryngeal preservation rates,Reference Abdurehim, Hua, Yasin, Xukurhan, Imam and Yuqin900 voice outcomes appear broadly similar, with the only RCT published showing a less breathy voice after RT.Reference Aaltonen, Rautiainen, Sellman, Saarilahti, Mäkitie and Rihkanen366
Transoral laser microsurgery is a cost-effective strategy to treat T1a glottic laryngeal SCC.890,Reference Prettyjohns, Winter, Kerawala, Paleri, Robinson and Bhide901 Difficult transoral access and posterior glottic tumours around the arytenoid preclude straightforward transoral laser microsurgery resection. Radiotherapy is recommended for these tumours.
Stage T1b–2N0 disease
There is much less data for non-T1a glottic laryngeal SCC. Large database analysis and pooled data suggest approximate equivalence in local control and overall survival, with approximate local control rates of around 75 per cent at five years for T2 tumours. The outcomes for T1b laryngeal SCC are somewhere between this and those for T1a tumours, with little data for this particular subgroup of patients.Reference O'Hara, Markey and Homer899,Reference Zhan, Yang, Song and Yan902–Reference Warner, Lee and Homer904
The evidence for a difference in voice outcome is poor but, intuitively, given that transoral laser microsurgery must involve resection of the anterior commissure and/or into the paraglottic space, voice outcomes are expected to be worse for transoral laser microsurgery.
It should be noted that the previous subdivisions of T2a and T2b (reflecting impairment of vocal fold mobility) are not used in the eighth edition of the TNM Classification of Malignant Tumours.Reference Brierley, Gospodarowicz and Wittekind894 Outcomes for T2b laryngeal SCC treated with both transoral laser microsurgery or RT are worse than for T2a disease.Reference Canis, Martin, Ihler, Wolff, Kron and Matthias905,Reference Dixon, Douglas, Shaukat, Garcez, Lee and Sykes906 With improved radiology, many T2 tumours with impaired mobility may be upstaged to T3 on the basis of paraglottic space invasion. The addition of chemotherapy to primary RT may be considered for these deeply invasive T2 tumours with impaired vocal fold mobility.
Pathological margins after transoral laser microsurgery
Incompletely excised tumours require further treatment, either additional surgical resection or RT. Patients with involved margins have a significant rate of residual disease at second-look surgery (about 25 per cent).Reference To, Qureishi, Mortimore and De907 A UK consensus groupReference Bradley, Mackenzie, Wight, Pracy and Paleri908 recommended considering 1 mm clearance to require no further treatment following transoral laser microsurgery, and advocated second-look procedures in instances of confident per-operative microsurgical perception of clearance but with tumour at the histological margin. The use of separate marginal biopsies following the main tumour resection is another popular strategy, in which negative marginal biopsies may constitute adequate surgical clearance.Reference Wilkie, Lightbody, Lythgoe, Tandon, Lancaster and Jones909
Recommendations
• Clear margins can be assumed with histological clearance of 1 mm (good practice point (G))
• Further surgery (second-look or wider clearance) is necessary for assumed involved margins (G)
Supraglottis
Transoral techniques and the use of RT have significantly reduced the indications for open partial supraglottic laryngectomy. No high-quality data exist to demonstrate any differences in oncological outcomes between RT and transoral surgical techniques, nor in terms of functional outcomes. Individual patient and tumour characteristics should be considered when deciding between the two modalities. Large database analyses assessing outcomes between RT and surgery include only small numbers of patients treated surgically, very likely highly selected on their tumour characteristics.Reference Megwalu and Panossian910 Elective, bilateral neck treatment is often recommended for supraglottic disease, but with little evidence base for T1 tumours, although these are unusual.Reference Arshad, Jayaprakash, Gupta, Cohan, Ambujakshan and Rigual911 Well lateralised low-volume supraglottic tumours not involving the epiglottis or pre-epiglottic space may be considered for ipsilateral neck treatment alone.
Open partial laryngectomy for early laryngeal squamous cell carcinoma
Open partial laryngectomy for early-stage disease is an alternative treatment to RT or transoral laser microsurgery in cases of difficult transoral access. Open partial laryngectomy may be more suitable for intermediate-stage tumours or those that involve tissues posterior to the vocal process of the arytenoid,Reference Succo, Crosetti, Bertolin, Piazza, Molteni and Cirillo912 as an alternative to RT with or without chemotherapy. Pooled case series data including over 5000 patients showed that open partial laryngectomy can offer local control rates of nearly 90 per cent at two years.Reference Thomas, Drinnan, Natesh, Mehanna, Jones and Paleri913 However, the functional results are worse in almost all respects, and rehabilitation is challenging. Open partial laryngectomy should only be considered for patients with good pre-treatment swallowing function and appropriate lung function, to allow successful functional rehabilitation, and where there is a good reason not to favour transoral surgery or RT.
Advanced laryngeal cancer
Summary recommendations
• Pre-treatment laryngeal function must be assessed to guide treatment options (evidence-based recommendation (R))
• Patients with T3 and carefully selected early T4a tumours with preserved pre-treatment laryngeal function should be offered primary RT with or without chemotherapy (R)
• Patients with obstructive tumours and/or impaired laryngeal function, and most T4a cases, should be offered primary total laryngectomy (R)
• Open partial laryngectomy may be appropriate for carefully selected patients (good practice point (G))
• All patients treated with primary surgery should be offered adjuvant RT with or without chemotherapy, other than carefully selected patients with pathologically staged T3N0 disease (G)
• Patients with unresectable disease (T4b) may be offered radical RT with or without chemotherapy, a palliative RT regimen, or alternative systemic anti-cancer therapies (G)
The Veterans Affairs trial,Reference Wolf, Fisher, Hong, Hillman and Spaulding914 and subsequent Radiation Therapy Oncology Group ‘RTOG 91-11’ trials,Reference Forastiere, Goepfert, Maor, Pajak, Weber and Morrison915 showed that advanced laryngeal SCC is a radiosensitive disease, with concurrent chemoradiotherapy leading to high laryngeal preservation and locoregional control than RT alone or with induction chemotherapy followed by RT. No randomised trials have assessed surgery versus RT with or without chemotherapy. Since these trials, clinicians have had to collate experience and observational evidence in order to attempt to define recommendations that improve shared decision-making with individual patients.Reference Forastiere, Ismaila, Lewin, Nathan, Adelstein and Eisbruch916
Consistent subsequent data have shown that disease-specific and overall survival for T3 laryngeal SCC are similar when comparing RT or chemoradiotherapy and total laryngectomy.Reference Forastiere, Ismaila, Lewin, Nathan, Adelstein and Eisbruch916,Reference Patel, Qureshi, Dyer, Jalisi, Grillone and Truong917
Where there is threatened or actual airway obstruction, a tracheostomy would be necessary before chemoradiotherapy, and there is no functional advantage in preserving the larynx. When there are other forms of laryngeal dysfunction (e.g. aspiration), laryngeal function is similarly already compromised. Pre-treatment chronic aspiration will likely persist after oncologically successful chemoradiotherapy.Reference Eskander, Blakaj and Dziegielewski918 Hence, patients with T3 laryngeal SCC with airway obstruction or laryngeal dysfunction should be offered primary surgery with total laryngectomy.
Whilst there is consistent evidence of superior survival for patients with T4a laryngeal SCC treated with surgery compared to chemoradiotherapy, there is limited evidence to suggest that the survival disadvantage associated with primary chemoradiotherapy is either non-existent or at least more limited when the primary disease is low volume with limited cartilage involvement. Hence, in carefully selected patients with low-volume T4a tumours, and with limited cartilage invasion, together with well-preserved function, primary chemoradiotherapy should be considered as a means of laryngeal preservation, with little compromise in local control or survival.Reference Forastiere, Ismaila, Lewin, Nathan, Adelstein and Eisbruch916
However, for other patients with T4a laryngeal SCC, the standard of care is primary surgery, with consistently reported improved locoregional disease control in particular, and, to a lesser extent, improved overall survival.
Whilst the evidence offers the basis for broad guidelines as above, any guideline cannot capture the nuanced nature of decision-making bearing in mind the impact of laryngectomy, which varies from patient to patient. For example, in a patient who values survival above all else, and who has a large-volume T3 tumour with lymph node metastases, but is not suitable for chemotherapy, then total laryngectomy with neck dissection and post-operative RT, rather than RT only, may be the preferred option for them. The equipoise in survival between RT or chemoradiotherapy and total laryngectomy for T3 laryngeal SCC in general may not be applicable to this subgroup.Reference Dziegielewski, Reschly, Morris, DeJesus, Silver and Boyce919 Or, for a patient who might struggle to come to terms with laryngectomy, with a T4a tumour with high-volume or multiple lymphadenopathy, who is fit for chemoradiotherapy, primary chemoradiotherapy may be preferred, as overall survival is poor whatever the initial treatment, as long as pre-treatment function is not overtly compromised.
Table 6 summarises the recommendations for the primary treatment of advanced cancers.
Table 6. Summary recommendations for primary treatment of advanced cancers
T = tumour stage; N = nodal stage; RT = radiotherapy; M = metastasis stage
Chemoradiotherapy should be offered to patients who are suitable (see Chapter 4). Total laryngectomy might need to include partial or circumferential pharyngectomy and tissue transfer for reconstruction (see Chapter 7). Adequate pharyngectomy with adequate tumour margins is essential. The overall incidence of thyroid gland invasion is low, and, therefore, thyroidectomy should be reserved for cases considered to be at risk, as opposed to being a routine measure for all total laryngectomies. Advanced laryngeal and hypopharyngeal carcinomas involving the subglottis carry a significantly elevated risk of thyroid gland invasion compared with those that spare this subsite.Reference Kumar, Drinnan, Robinson, Meikle, Stafford and Welch920
Post-operative radiotherapy with or without chemotherapy
All patients should be offered post-operative RT for the recognised pathological indications (see Chapter 4). This will include all patients with T4 disease. Oncological outcomes may not be improved with post-operative RT for pathologically staged T3N0 disease, when considering large database analyses.Reference Cervenka, Rao, Farwell and Bewley921
Neck metastases in advanced laryngeal squamous cell carcinoma
Patients with T3–4N0 disease should have levels IIA–IV treated electively.Reference O'Hara, Simo, McQueen, Andi, Lester and Giddings922 For disease involving the supraglottis, and any tumour involving or crossing the midline, bilateral neck treatment is generally recommended. There is no clear evidence regarding the extent of neck dissection for patients undergoing primary surgery with N+ disease. In addition to levels IIA–IV, neck dissection of levels IIB and V would be recommended, accepting the increased risk of accessory nerve dysfunction. Level I can be spared unless involved radiologically. (See also Chapter 26.)
Open partial laryngectomy for locally advanced laryngeal squamous cell carcinoma
Open partial laryngectomy can be considered for carefully selected T3 tumours, especially those anteriorly situated in the larynx.Reference Succo, Crosetti, Bertolin, Piazza, Molteni and Cirillo912 Functional outcomes are superior if post-operative RT can be avoided – i.e. with complete tumour resection and by selecting patients with N0 disease.Reference Forastiere, Ismaila, Lewin, Nathan, Adelstein and Eisbruch916
Follow up
The recommendations for follow up are set out in Chapter 5. Routine imaging is not required after primary surgery or RT for early-stage (T1–2N0) disease, unless there is clinical concern.
Recommendations
• All patients with a preserved larynx should undergo flexible laryngoscopy at follow-up appointments (evidence-based recommendation (R)), ideally with photographic documentation (good practice point (G))
• Thyroid function should be assessed at least every six months for any patients undergoing neck RT or open laryngeal surgery (R)
Management of complications in laryngeal cancer treatment
Complications of both RT and surgery are not uncommon, especially for advanced laryngeal SCC. These are discussed in Chapter 16.
Recurrent disease
Summary recommendations
• Consider both primary and recurrent disease stage and extent (evidence-based recommendation (R))
• Consider organ preservation for early-stage laryngeal SCC recurrences (R)
• Consider vascularised tissue transfer in salvage total laryngectomy, to reduce risks of wound breakdown and fistula (R)
All recurrent tumours must be staged clinically and radiologically, for locoregional and distant disease. Both CT and MRI may be required to stage the primary disease for complex salvage surgical decisions. The management of recurrence in general is discussed in Chapter 5, and non-surgical management options, including systemic therapy, are discussed in Chapter 4.
Patients with early-stage recurrences following previous transoral laser microsurgery may be suitable for further transoral laser microsurgery, RT or open partial laryngectomy, assuming laryngeal function is preserved.
Patients with early-stage disease following previous RT may be suitable for laryngeal preservation surgery (transoral surgery or open partial laryngectomy). Open partial laryngectomy may be preferred for recurrent disease after RT. Transoral laser microsurgery may be appropriate for selected small, accessible, well-defined glottic lesions, but, overall, it is associated with a lower local control rate than open partial laryngectomy.Reference Ramakrishnan, Drinnan, Kwong, Grant, Mehanna and Jones923,Reference Paleri, Thomas, Basavaiah, Drinnan, Mehanna and Jones924 Open partial laryngectomy requires suitable pre-treatment swallowing and lung function to aid rehabilitation. Total laryngectomy should be considered if transoral surgery or open partial laryngectomy are not possible.
Advanced stage disease following previous RT should be managed with total laryngectomy if the disease is resectable with no distant metastasis. Very selected patients with recurrent T3 disease may be suitable for open partial laryngectomy, accepting prolonged rehabilitation.
Vascularised tissue transfer should be considered in salvage total laryngectomy, to reduce risks of wound breakdown and fistula.Reference Paleri, Drinnan, van den Brekel, Hinni, Bradley and Wolf925 This may be myofascial as a second layer to the pharyngeal primary repair, or with skin if required in order to allow wider extra-laryngeal surgical clearance margins without causing stricture.
For patients who have had full treatment (i.e. total laryngectomy and RT or chemoradiotherapy, in any order), most will have a very low chance of survival and will be best treated for palliation (see Chapters 4 and 15). Patients with unresectable disease, or those who are unfit to undergo major surgery, may be offered immunotherapy, chemotherapy or re-irradiation (see Chapter 4).
Very occasionally, a patient with, for example, peri-stomal recurrence may be suitable for salvage surgery, usually involving manubriumectomy, mediastinal dissection, pharyngectomy and challenging reconstruction, but this is exceptional in a patient group with very poor overall survival prospects.
The management of lymph nodes in the setting of salvage laryngectomy for local recurrence and management of isolated lymph node recurrence is discussed in Chapter 26.
Studies due to report
‘Laryngeal Cancer Cohort’ (‘LARCH’) – an enhanced cohort study of patients with newly diagnosed laryngeal cancer (UK, Health Research Authority).
‘Vocal-cord vs. Complete Laryngeal Radiotherapy for Early Glottic Cancer’ (‘VOCAL’); ClinicalTrials.gov identifier: NCT03759431. This is a phase II trial assessing the non-inferiority of local control achieved with vocal-fold-only RT compared to complete larynx RT in T1N0 glottic laryngeal SCC.
There are no National Cancer Research Institute phase III/III clinical trials specific to laryngeal cancer. Other global trials can be found in: https://clinicaltrials.gov.
Important research questions to be answered
(1) Methods to define patient selection for primary treatment (e.g. total laryngectomy or chemoradiotherapy): chemo-selection protocols and improved radiogenomics prognostication.
(2) More evidence for local control, comparing transoral laser microsurgery and RT for early-stage laryngeal SCC, particularly for T1b and T2 laryngeal SCC.
(3) More evidence for voice outcomes following transoral laser microsurgery and RT for early-stage laryngeal SCC.
Introduction
The scope of this chapter pertains to cancers involving the bony lateral skull base and its associated structures. These include primary temporal bone malignancies which usually arise from the external auditory canal or middle ear, and locally advanced skin or parotid malignancies which involve the lateral skull base though direct extension. If treated surgically, similar principles apply to these groups.
Primary temporal bone malignancies are uncommon, and account for less than 1 per cent of all head and neck cancers. The majority of these are squamous cell carcinoma (SCC), accounting for approximately 60–80 per cent of cases.Reference Gidley and DeMonte926 Locally advanced skin cancers can arise from the pinna or peri-auricular skin, and are usually SCC, basal cell carcinoma (BCC), melanomas or skin adnexal tumours. Parotid malignancies with lateral skull base involvement are often high-grade cancers of salivary gland origin, and can also include metastatic SCC to the parotid. Other sites of tumour origin include the infratemporal fossa and temporomandibular joint (TMJ), where a variety of pathologies, including sarcomas, can arise.
This area is therefore highly heterogeneous. In addition, the largest subgroup, primary temporal bone cancer, is a rare cancer, and hence there is relatively poor evidence to inform best practice.
Presentation, investigation and diagnosis
Recommendations
• Consider malignancy in cases of progressive otalgia, otorrhoea, bleeding, hearing loss and/or facial nerve palsy when there are abnormal clinical signs (good practice point (G))
• Skull base osteomyelitis can mimic malignancy;Reference Masterson, Rouhani, Donnelly, Tysome, Patel and Jefferies927 biopsy is essential for diagnosis (G)
• Audiology should be performed in primary temporal bone cancers and for all cases to be treated by temporal bone surgery (G)
• Imaging for temporal bone cancers should include both magnetic resonance imaging (MRI) and computed tomography (CT) scanning (G)
Presentation
Clinical presentation will depend on the site of origin and stage of tumour.
Primary temporal bone malignancies of the external auditory canal or middle ear often present with unilateral symptoms of pain, otorrhoea, bleeding and hearing loss,Reference Masterson, Rouhani, Donnelly, Tysome, Patel and Jefferies927 and with facial nerve palsy in advanced cases. Examination may reveal a lesion which can be polypoidal or exophytic. External auditory canal and middle-ear tumours can also arise as a result of chronic inflammation or infection.Reference Allanson, Low, Clark and Gupta928 Both the symptoms and signs of this rare malignancy can therefore be similar to chronic external and middle-ear inflammation, often resulting in a late diagnosis. In many cases, it is the degree or persistence of pain that raises the possibility of malignancy.
Skin cancers will usually be visible as irregular, ulcerated lesions, with advanced cases demonstrating the fixation and invasion of underlying structures. Advanced parotid cancers are often palpable, with a parotid gland mass, which may involve the skin or TMJ. Deep lobe origin or extension is common, with the possible involvement of pterygoid musculature and the mandibular ramus. Facial palsy can be a presenting sign, but in many cases nerve function may be preserved till late, even in cases of tumour encasement of the nerve.
Tumours of the infratemporal fossa may be asymptomatic until significant tumour progression, at which stage symptoms can include (hemifacial) pain, trismus and lower cranial nerve dysfunction. Clinical examination may show fullness in the zygomatic region and loss of sensation in the V3 (mandibular nerve) distribution.
Diagnosis
Diagnosis is usually made through histological assessment of a biopsy specimen. This is generally straightforward to undertake in visible skin or external auditory canal tumours. Middle-ear cancers may require transtympanic or transmastoid biopsy. Image-guided fine needle aspiration cytology or core biopsy is possible in most parotid or infratemporal fossa tumours.
Histological or cytological confirmation of malignancy is important, as other benign or inflammatory conditions, such as skull base osteomyelitis, may be part of the differential diagnosis.
Audiology
Pure tone audiometry may show conductive hearing loss with or without sensorineural loss in primary temporal bone cancers. It will also provide information to predict post-operative hearing problems and rehabilitation options.
Imaging
Table 1 summarises the pre-treatment imaging recommendations.
Table 1. Pre-treatment imaging recommendations
* Positron emission tomography computed tomography (PET-CT) is preferred in high-grade non-squamous cell carcinoma primary and/or nodal stage N3 neck disease. CT = computed tomography; MRI = magnetic resonance imaging
Cross-sectional imaging provides valuable information on tumour stage, extent and surgical resectability.
Both CT and MRI with contrast are used for assessment of the primary tumour. High-resolution CT is essential for bone assessment of temporal bone erosion, skull base foramina, middle-ear ossicles, mandible and TMJ. Magnetic resonance imaging is helpful for soft tissue, including the dura and brain, masticator space and parotid gland, and in differentiating fluid from the tumour, particularly within the mastoid air cells.
In cases where primary surgery is planned, and there is close proximity of the tumour to the internal carotid artery, which may require sacrifice, then a pre-operative balloon occlusion test with or without embolisation can be considered.
Staging and pathology
Staging
Recommendations
• Use modified Pittsburgh staging for primary temporal bone cancersReference Nabuurs, Kievit, Labbé, Leemans, Smit and van den Brekel930 (good practice point (G))
• Use the applicable site of origin (based on Union for International Cancer Control (‘UICC’) TNM Classification of Malignant Tumours, eighth editionReference Brierley, Gospodarowicz and Wittekind763) for non-primary temporal bone lateral skull base malignancies (G)
The American Joint Committee on Cancer (‘AJCC’)/Union for International Cancer Control tumour–node–metastasis (TNM) staging does not include a specific cancer staging system for lateral skull base or temporal bone malignancies. However, certain sites of origin can fit into the TNM Classification of Malignant Tumours, eighth edition.Reference Brierley, Gospodarowicz and Wittekind763 These sites are: skin carcinoma of the head and neck (for external auditory canal, pinna and peri-auricular skin), the major salivary gland for parotid cancers, tumours of bone and soft tissue for sarcomas, and head and neck malignant melanoma. These are described in the relevant chapters of these guidelines.
The modified Pittsburgh staging system,Reference Moody, Hirsch and Myers929 described for cancer arising from the external auditory canal, is often used for other primary temporal bone cancers and can be used for cutaneous cancers as well. It provides an intuitive staging system and good prognostication, and is also helpful for clinical decision-making and surgical planning.Reference Nabuurs, Kievit, Labbé, Leemans, Smit and van den Brekel930 This is therefore the recommended staging system for primary temporal bone malignancies.
Recent studies have shown a number of additional factors that are important in prognostication; these include: dural involvement, histological grade of tumour and route of spread. With regard to the latter, tumour stage T4 cancers on the basis of anterior spread have a much better prognosis than T4 on the basis of spread in other directions.Reference Zanoletti, Franz, Cazzador, Franchella, Calvanese and Nicolai931,Reference Zanoletti, Marioni, Stritoni, Lionello, Giacomelli and Martini932 There is an argument for revising the T4 classification. Clinicians should be aware of this, as the former group of T4 patients have a better prognosis.
Neck and distant metastatic disease are assessed using the standard American Joint Committee on Cancer/Union for International Cancer Control TNM Classification of Malignant Tumours (eighth edition) classification for head and neck cancers (Tables 2–4).Reference Brierley, Gospodarowicz and Wittekind763
Table 2. Modified Pittsburg primary tumour stagingReference Moody, Hirsch and Myers929
EAC = external auditory canal; TMJ = temporomandibular joint
Table 3. Nodal staging*
* According to the TNM Classification of Malignant Tumours (eighth edition).Reference Brierley, Gospodarowicz and Wittekind763
Table 4. Metastasis staging*
* According to the TNM Classification of Malignant Tumours (eighth edition).Reference Brierley, Gospodarowicz and Wittekind763
Pathology
The main pathologies can be considered by site of origin. Table 5 shows the main pathologies affecting the lateral skull base.
Table 5. Main pathologies affecting lateral skull base
EAC = external auditory canal; SCC = squamous cell carcinoma; BCC = basal cell carcinoma; TMJ = temporomandibular joint
Treatment
Recommendations
• Primary surgery is generally accepted as the standard of care in suitable patients (evidence-based recommendation (R))
• Surgery will usually entail at least a lateral temporal bone resection, with consideration of adjacent structures (good practice point (G))
• The parotid and neck regional lymph nodes should be treated electively in most cases of cutaneous SCC and T3–4 primary temporal bone SCC (R)
• Post-operatively radiotherapy (RT) as intensity-modulated RT is the standard of care in T2–4 cancers (G)
• Surgery should be avoided in advanced cases with dural and/or carotid artery involvement, as the prognosis is very poor and the likelihood of surgical morbidity is high (G)
General principles
Primary surgery is the mainstay of treatment for most lateral skull base cancers.
The decision for surgery is dependent on the extent of tumour and the ability to achieve an appropriate oncological resection. Patient factors and co-morbidities must also be weighed into the decision-making process. Cases should be referred for management to designated supra-regional head and neck or skull base oncology multidisciplinary teams (MDTs) with expertise in managing such cases.
Primary RT or chemoradiotherapy may be used as an approach in selected cases where surgery is not appropriate or acceptable. Palliative RT and/or chemotherapy may also have a role for symptom control or in cases with distant metastatic disease.
Surgical management and considerations are described below. Similar approaches for surgery, post-operative management and rehabilitation may be required for primary temporal bone malignancies and advanced parotid or skin cancers.
Primary temporal bone cancers
It is generally accepted that primary temporal bone cancers, arising from the external auditory canal or middle ear, are best treated with surgery followed by adjuvant post-operative RT.Reference Masterson, Rouhani, Donnelly, Tysome, Patel and Jefferies927,Reference Zanoletti, Marioni, Stritoni, Lionello, Giacomelli and Martini932 However, there is very little evidence comparing different modalities of treatment.
Survival in early-stage T1–2 disease is good with primary surgery and post-operative RT if necessary. Data show a five-year disease-free survival rate of 67–100 per cent and a disease-specific survival rate of 92–100 per cent. Most surgery entails a lateral temporal bone resection with little resultant major morbidity.Reference Morita, Homma, Nakamaru, Sakashita, Hatakeyama and Kano933
Locally advanced tumours, treated with primary surgery and post-operative RT, have lower survival with T3–4 disease, with a disease-free survival rate of 41–59 per cent and a disease-specific survival rate of 48–65 per cent. However, rates have been improving with time, indicating that advances in treatment have impacted positively in this group.Reference Nabuurs, Kievit, Labbé, Leemans, Smit and van den Brekel930,Reference Bacciu, Clemente, Piccirillo, Ferrari and Sanna934,Reference Lovin and Gidley935 Most data, for patients treated with curative intent, arise from patients who have been treated with primary surgery. However, more limited data from patients with T3–4 cancers treated with primary chemoradiotherapy show near equivalent survival.Reference Morita, Homma, Nakamaru, Sakashita, Hatakeyama and Kano933,Reference Shinomiya, Hasegawa, Yamashita, Ejima, Kenji and Otsuki936,Reference Takenaka, Cho, Nakahara, Yamamoto, Yasui and Inohara937
Involvement of the dura confers a worse prognosis. One group reported five-year disease-specific survival rate of 37 per cent in 14 patients with brain involvement.Reference Masterson, Rouhani, Donnelly, Tysome, Patel and Jefferies927 However, this has not been reported by others, e.g. a 0 per cent disease-specific survival rate with dural involvement in nine patients was reported by one groupReference Zanoletti, Franz, Cazzador, Franchella, Calvanese and Nicolai931 and 0 per cent rate in nine patients by another.Reference Bacciu, Clemente, Piccirillo, Ferrari and Sanna934
Encasement of the petrous internal carotid artery (ICA) is a poor prognostic sign, and surgery for such cases will almost always cause highly morbid neurological functional loss with a very poor chance of survival.Reference Masterson, Rouhani, Donnelly, Tysome, Patel and Jefferies927 Ferrari et al. reported a small series of 10 patients with ICA resection, with a 10 per cent peri-operative mortality rate and a 25 per cent major neuromorbidity rate of the survivors, with a mean overall survival of 27 months.Reference Ferrari, Zanoletti, Taboni, Cazzador, Tealdo and Schreiber938
There is varying evidence as to the prognostic impact of facial nerve involvement, with some series showing worse outcomes and others showing no significance.Reference Cristalli, Manciocco, Pichi, Marucci, Arcangeli and Telera939–Reference Xie, Zhang and Dai941
As in other head and neck cancers, the presence of nodal disease has a significant impact on disease-specific and overall survival (by about 40–50 per cent),Reference Nabuurs, Kievit, Labbé, Leemans, Smit and van den Brekel930,Reference Zanoletti, Marioni, Stritoni, Lionello, Giacomelli and Martini932 as do positive surgical margins.Reference Bacciu, Clemente, Piccirillo, Ferrari and Sanna934
The route of spread for advanced cancer is highly prognostic, with a much better prognosis for advanced cancers that are T4 on the basis of anterior spread (to the parotid gland and/or pre-auricular tissue or skin) than T4 on the basis of spread in other directions (55 per cent vs 16 per cent five-year disease-specific survival rate in 21 patients).Reference Zanoletti, Franz, Cazzador, Franchella, Calvanese and Nicolai931
Surgery will usually require at least a lateral temporal bone resection, with advanced-stage tumours requiring an extended temporal bone resection. Adequate surgery is required to achieve optimal prognosis; this should be performed en bloc when possible, to maximise the chance of margin control. Surgery, other than for palliation, should not be performed without a high chance of complete microscopic resection.
Stage T1 tumours
In selected superficial cases, a sleeve resection may be considered. This may be appropriate, for example, for an early-stage superficial BCC involving the cartilaginous external auditory canal. Otherwise, a lateral temporal bone resection is required to achieve adequate margins.
Stage T2 tumours
A lateral temporal bone resection is recommended as a minimum for all T2 cases. The specific details of surgery are outside the scope of this chapter, but it involves en bloc resection of the external auditory canal and tympanic membrane lateral to the facial nerve and stapes.
Stage T3 tumours
For T3 tumours not extending medial to the tympanic membrane, a lateral temporal bone resection may be sufficient. Evidence from pathological specimens shows that tumours eroding the anterior bony external auditory canal should also undergo TMJ resection. If there is middle-ear involvement, an extended temporal bone resection should be performed.
Stage T4 tumours – extended temporal bone resection
The term extended temporal bone resection encompasses various types of temporal bone resection, with described terminology including subtotal, near total and total temporal bone resection.
The essential elements of extended temporal bone resection are: (1) facial nerve sacrifice; (2) posterior and middle craniotomy; (3) labyrinthectomy; (4) transection of the internal auditory canal; (5) resection of the petrous tip; (6) exposure of the intra-petrous portion of the carotid; and (7) total parotidectomy.
Additional options include: craniectomy (squamous temporal bone; sphenoid wing, posterior fossa); mandibulectomy; parapharyngeal or infratemporal fossa resection; extension to the jugular foramen; lower cranial nerve sacrifice; ICA; dura; and brain.
Table 6 summarises the surgical management of primary temporal bone cancer.
Table 6. Surgical management of primary temporal bone cancer
* Modified Pittsburgh classification.Reference Moody, Hirsch and Myers929 EAM = external auditory meatus
Advanced skin and parotid gland malignancy
Locally advanced cutaneous or parotid malignancy may also necessitate lateral temporal bone resection as part of the resection, in cases where the tumour abuts or locally erodes into the temporal bone. Tumour extension up to or through the stylomastoid foramen may also occur, with or without facial palsy.
In this setting, lateral temporal bone resection can enable complete tumour resection, and allow improved access to gain better margins and to allow proximal dissection of the facial neve in the temporal bone for grafting. This approach has shown improved locoregional control in cases of locally advanced parotid and cutaneous malignancies.Reference Mehra, Morris, Shah, Bilsky, Selesnick and Kraus942–Reference Shao, Wong, McIvor, Mylnarek, Chaplin and Izzard945
Management of additional structures
Facial nerve
In cases of pre-existing facial palsy, the involved portion of the intra-temporal and extra-temporal facial nerve will require resection, ideally with frozen section control.
Facial nerve sacrifice may also be required when the nerve is functioning normally pre-operatively, if there is tumour encasement of the nerve, extension through the stylomastoid foramen, or where preservation would compromise successful tumour extirpation.
Temporomandibular joint
The TMJ is in close proximity to the external auditory canal and parotid, and consideration should be given to it as part of operative planning. Resection of the TMJ is not required routinely in every lateral temporal bone resection and does contribute to post-operative morbidity.Reference Masterson, Rouhani, Donnelly, Tysome, Patel and Jefferies927,Reference de Casso, Kwhaja, Davies, Al-Ani, Saeed and Homer946 Where there is anterior external auditory canal bony canal erosion, microscopic tumour involvement of the TMJ is significantly more likely.Reference Hosokawa, Mizuta, Takahashi, Okamura, Takizawa and Hosokawa947 A selective approach is therefore advised, with TMJ resection more likely to be required in advanced tumours and those with anterior extension.
Dura or brain
Cases with dura involvement can be operated upon, in highly select cases, with dura resection. Gross brain parenchymal invasion has a low likelihood of curative outcomes, and, in this setting, a primary surgical approach is not recommended.
Mandible
Adjacent bony structures include the mandibular ramus and zygomatic arch. Partial or complete resection of these structures may be necessary to achieve appropriate margins or access.
Parotid gland
The parotid gland may also be involved through direct tumour invasion. In cases of parotid involvement through direct invasion, parotidectomy should be undertaken, guided by the degree of local extension. Parotidectomy may also be carried out as part of surgical access for a complete en bloc resection, or for possible micrometastatic lymph node disease (see below).
Internal carotid artery
Limited disease at the ICA in an otherwise resectable tumour could, in select cases, warrant consideration of ICA sacrifice. Pre-operative balloon occlusion testing should be undertaken and, if there is a high likelihood of sacrifice, occlusion of the artery with coiling can be performed.
Management of regional lymph nodes
The parotid gland and cervical lymph nodes are the regional nodal basins in primary temporal bone and cutaneous malignancies.
Clinically staged node-negative (N0) disease
For primary temporal bone cancers, routine elective treatment of the parotid and neck in T1–2 SCC is not necessary, with occult metastasis rates of 0 and 7 per cent respectively.Reference Borsetto, Vijendren, Franchin, Donnelly, Axon and Smith948 However, most patients with T1–2 primary temporal bone SCC will receive surgical treatment, and elective treatment of these nodal basins may be undertaken for surgical access requirements or can be considered part of adjuvant post-operative RT, factoring in histopathology. Elective treatment of the parotid and neck is recommended in T3–4 primary temporal bone SCC, with occult metastasis rates of around 20 per cent,Reference Borsetto, Vijendren, Franchin, Donnelly, Axon and Smith948 in keeping with generally accepted practice.
Elective treatment should be undertaken in the majority of cutaneous SCC cases requiring lateral temporal bone resection.Reference Wilkie, Lancaster, Roland and Jones949
If elective treatment of the node-negative neck and parotid is planned, this would usually be undertaken as part of the primary surgery, with dissection of nodal levels 2(a/b) and 3 as well as a superficial parotidectomy.
Clinically staged node-positive (N+) metastatic disease
In the setting of clinical-radiological evidence of nodal metastatic disease to the parotid or neck, both parotidectomyReference Xie, Wang, Zhang and Liu950 and neck dissection should be performed, the extent of which is dependent on the burden of disease. This should at least involve a superficial parotidectomy, and dissection of nodal levels to typically include 1b, 2, 3, 4 and 5a. Dissection of the deep lobe of the parotid is more controversial in the context of facial nerve preservation and the absence of deep lobe involvement.
Radiotherapy and chemoradiotherapy
Radiotherapy is usually given in the adjuvant post-operative setting. Most T2–4 primary temporal bone cancers and locally advanced malignant parotid tumours show improved survival rates with adjuvant RT. Selected T1 tumours with clear resection margins may be managed with surgery alone.Reference Kunst, Lavieille and Marres951
Highly conformed intensity-modulated RT is the standard of care to reduce the dose to normal structures, including the adjacent inner ear and oral cavity, and to reduce the volume of bone treated to a high dose. Post-operative doses used for head and neck cancer are 60 Gy in 30 fractions for moderate-risk disease and 66 Gy in 33 fractions for high-risk disease.
Synchronous post-operative treatment with cisplatin can be considered for high-risk SCC, e.g. involved surgical margins (see Chapter 4).
Primary RT or chemoradiotherapy may be used as an approach in selected cases as an alternative to surgery, with similar outcome data for advanced disease.Reference Morita, Homma, Nakamaru, Sakashita, Hatakeyama and Kano933,Reference Shinomiya, Hasegawa, Yamashita, Ejima, Kenji and Otsuki936,Reference Takenaka, Cho, Nakahara, Yamamoto, Yasui and Inohara937 The use of induction pre-operative chemoradiation therapy has also been reported to help obtain surgical tumour-free margins in advanced disease,Reference Takenaka, Cho, Nakahara, Yamamoto, Yasui and Inohara937 but is not in common practice.
Inoperable tumours
In cases where surgery is not undertaken, definitive or palliative treatment can be given. The choice is dependent on the rationale for not undertaking surgery, the extent of tumour, and patient age and co-morbidities.
Advanced disease at presentation may make the tumour either technically inoperable or render the chance of cure very low with factors such as gross brain invasion, cavernous sinus involvement or carotid encasement. It may therefore be deemed that the benefit of surgery is outweighed by the poor prognosis and attendant morbidity. In an otherwise fit patient, definitive RT or chemoradiotherapy using intensity-modulated RT may be considered. Long-term durable results, albeit with lower success, can be seen with this approach.Reference Takenaka, Cho, Nakahara, Yamamoto, Yasui and Inohara937
Palliative RT can be given for the control of local symptoms such as pain and fungating disease. Palliative systemic treatment can also be considered for reasons of local control and to manage distant metastatic disease.
Reconstruction and rehabilitation following surgery
Recommendations
• The anterolateral thigh free flap offers excellent reconstruction for lateral skull base defects (good practice point (G))
• Facial nerve rehabilitation should be initiated at primary surgery (G)
• Osseo-integrated bone-anchored hearing aids (BAHAs) can be considered for hearing rehabilitation (G)
• Selected condylar defects may be left unreconstructed with minor occlusal disturbance (G)
There are a number of important considerations in reconstruction and rehabilitation following lateral skull base surgery. One or more of the following areas may require addressing:
• Management of any dural defect
• Filling of the volume defect and coverage of skin defects
• Management of pinna defects
• Facial nerve management and rehabilitation
• Hearing and balance rehabilitation
• Management of cranial nerve or neurological deficits
• Temporomandibular joint loss or dysfunction
Dura, tissue volume and skin
Dural defects are normally repaired with non-vascularised tissue such as autologous fascia lata grafts, pericardial xenografts, or synthetic materials.
Volume and skin defects are determined by the extent of surgery. For smaller skin defects without much volume loss, options include a radial forearm free flap, cervicofacial rotation flap, temporalis flap and supraclavicular artery island flap. These can be used to reconstruct small skin or auricle defects with modest volume loss.Reference Rosenthal, King, McGrew, Carroll, Magnuson and Wax952,Reference Hanasono, Silva, Yu, Skoracki, Sturgis and Gidley953
For most defects after temporal bone resection, the anterolateral thigh free flap offers optimal reconstruction, providing bulk (variable by the inclusion of the vastus lateralis) and enough skin for most defects (which can be reduced by de-epithelialisation if the pinna is not resected).Reference Rosenthal, King, McGrew, Carroll, Magnuson and Wax952,Reference Hanasono, Silva, Yu, Skoracki, Sturgis and Gidley953 It is reliable, has the requisite tissue and is associated with minimal donor site morbidity. It allows vascularised fascia lata to be used for static facial suspension or the lateral cutaneous femoral nerve for either sensory innervation of the flap or an interpositional facial nerve graft. Also, the accessible donor site allows for concomitant flap harvest and tumour ablation. Alternative flaps include the latissimus dorsi, rectus abdominis or deep inferior epigastric artery perforator, radial forearm, medial sural artery, and lateral arm flaps.
In a vessel-depleted neck or in a patient unsuitable for microvascular surgery, a lower trapezius muscle island flap or supraclavicular artery island flap (if the transverse cervical vessels are intact), or a superior trapezius flap (when a radical neck dissection has been performed), can be used. The use of a pectoralis major or deltopectoral flap is suboptimal as the lateral skull base is at or beyond the limits of rotation in many cases.
Facial function
In cases in which facial nerve sacrifice is necessary, one or more of the steps described below should be considered. It should be borne in mind that the best time to perform many of these interventions is at the time of tumour resection, as virtually every patient in this group will go on to have post-operative RT. Patients will require detailed pre- and post-operative counselling for the functional and cosmetic sequelae resulting from a facial nerve palsy.
For cases with pre-operative facial function, a cable graft to intra-parotid branches can be performed if: (a) there is enough proven tumour-free proximal facial nerve (otherwise a facial-hypoglossal anastomosis can be considered); and (b) the peripheral branches can be identified (this may be difficult when a radical en bloc parotidectomy with overlying skin is performed). Useful donor nerves include the greater auricular nerve, sural nerve or lateral cutaneous nerve of the thigh (easily available if harvesting an anterolateral thigh free flap). Free flaps give the option of skin, muscle, and fat or fascia for dead space filling, as well as vascularised nerves to bridge any facial nerve defect. Commonly used chimeric free flap options include the anterolateral thigh or latissimus dorsi free flaps. Such vascularised nerve conduits may take time to function. Buddying the buccal branch or zygomatic branches onto the nerve to the masseter provides more rapid reanimation and neuromuscular end plate function. This allows time for the vascularised nerve conduit to work.
Other options, and for cases with pre-operative facial palsy, include static procedures such as a fascia lata sling for oral commissure or cheek suspension, or dynamic procedures such as lengthened temporalis myoplasty (e.g. Labbé type I or II), if the deep temporal nerve and artery are preserved. Oculoplastic interventions (e.g. upper lid weight, canthoplasty) can be performed at the time of tumour resection or later on. Immediate on-table options include a lateral tarsal strip procedure or lateral tarsorrhaphy.
Hearing
Ipsilateral total or total conductive hearing deficit is an inevitable outcome of temporal bone resection. Pre-operative audiological assessment of the contralateral ear will identify patients with a pre-existing deficit. This may be corrected or improved with appropriate aiding in either the pre- or post-operative period. Total conductive hearing loss can be rehabilitated through an osseo-integrated BAHA. Total hearing loss can be rehabilitated through either a BAHA or a bilateral contralateral routing of signals aid.
Balance
Post-operative vertigo is expected if there is resection of a functioning labyrinth. If vestibular compensation is protracted and incomplete, referral for vestibular rehabilitation services should be considered.
Lower cranial nerves
In addition to VIIth cranial nerve (facial nerve) issues, all lower cranial nerves essential for swallowing and voice (IXth, Xth and XIIth cranial nerves) are at risk of injury or sacrifice during surgery for advanced tumours. Care of the patient in this situation must include close involvement of speech and language therapists. Interventions include: either pre- or post-operative percutaneous gastrostomy; a nasogastric tube; or a tracheostomy if aspirating on saliva. Later interventions include vocal fold medialisation and cricopharyngeal myotomy.
Temporomandibular joint
It is feasible to leave selected condylar resections unreconstructed, accepting minor dental occlusal disturbance. Where mandibular reconstruction is required, a composite microvascular flap such as a chimeric thoracodorsal artery perforator, scapula osteomusculocutaneous or deep superior epigastric artery with costochondral junction flap can restore a large mandibular and lateral skull defect.
Follow up
Given the numerous considerations of rehabilitation, follow up should ideally be in an MDT clinic setting. Post-treatment imaging and surveillance will include an initial post-treatment scan approximately 12 weeks after the completion of definitive treatment, which will act as a baseline scan. Magnetic resonance imaging after surgery, with or without positron emission tomography (PET)-CT, is most useful in this regard. As most tumour recurrences are not detectable by clinical examination, especially if a free flap has been used, six-monthly MRI scans for the first year and annual MRI scans until year five are recommended.
Recurrent local disease
When only limited primary surgery (less than lateral temporal bone resection) has been previously performed, surgery based on the principles described above is appropriate.
Local recurrence after comprehensive treatment may be aggressive and is associated with a poor prognosis. In the majority of cases, recurrences in this setting are inoperable. Longer survival can be achieved through palliative treatment than through surgery.Reference Zanoletti, Marioni, Franchella, Lovato, Giacomelli and Martini954 Consideration may be given to re-irradiation if there is a sufficient time interval from prior RT. Palliative options can also be considered (see Chapter 4).
Other lateral skull base cancers
There is large variability in both the histology and anatomical origin of these tumours. They may arise from the contents of the infratemporal fossa, temporal bone or adjacent structures. They include sarcomas, malignant neurogenic tumours and paragangliomas. This heterogeneity makes it difficult to be generalisable in terms of guidelines. Similar principles of management as described in this chapter will usually however apply, with surgery being the mainstay of treatment, involving oncological resection of the primary tumour, followed by adjuvant RT.
The skillset required to deal with such cases is highly specialised, requiring surgeons who have expertise in various forms of surgical access and resection of the infratemporal fossa, with joint neurosurgery input when necessary. Referrals should be made to designated supra-regional head and neck or skull base oncology MDTs with experience in managing these cases.
Studies due to report
Current evidence is based on retrospective case series and expert opinion. No randomised, controlled trials exist or are likely to, given the low incidence of cases.
Cemiplimab, an anti-programmed cell death 1 immune checkpoint inhibitor, has shown promise in the treatment of non-resectable advanced cutaneous SCC.Reference Migden, Rischin, Schmults, Guminski, Hauschild and Lewis955 It is currently being investigated in a phase 2 clinical trial in a neo-adjuvant setting prior to surgery.Reference Migden, Rischin, Schmults, Guminski, Hauschild and Lewis955
Vismodegib is an inhibitor of the hedgehog signalling pathway, and is under investigation in locally advanced BCCs, where given pre-operatively it may allow for downstaging of the extent of surgery.Reference Bertrand, Guerreschi, Basset-Seguin, Saiag, Dupuy and Dalac-Rat956
Important research questions
The role of proton therapy in a pre- or post-operative adjuvant setting is also of interest given the close anatomical proximity of the lateral skull base to critical functionally significant structures.
Evidence comparing different modalities of primary treatment would be desirable, but may be unachievable in this rare tumour group.
Introduction
Paranasal sinus and skull base malignancies are a rare and heterogeneous group of diseases representing less than 5 per cent of all head and neck cancers. The incidence of these cancers is just less than 1 in 100 000 people per year, peaking between the ages of 50 and 60 years.Reference Dutta, Dubal, Svider, Liu, Baredes and Eloy957 Arising within a relatively small anatomical area, these rare tumours comprise a variety of histological types. The typical presentation is with locally advanced stage disease, and subsequent treatment is complicated by the proximity of critical anatomical structures. Clinical practice is based on low-quality evidence, with wide variation across the UK.Reference Iyizoba-Ebozue, Fleming, Prestwich and Thomson958
Presentation and diagnosis
The propensity for sinonasal malignancy to present at a locally advanced stage reflects both the innocuous symptoms of early-stage disease and the difficulty of endonasal examination in the community. Unilateral nasal symptoms should raise suspicion. Localised disease may present with epistaxis, nasal obstruction and hyposmia. Advanced disease, with adjacent structural involvement, may result in corresponding functional symptoms affecting the: orbit (diplopia, proptosis); pre-masseteric space, pterygopalatine or infratemporal fossa (trismus, facial pain, facial sensory change); anterior cranial fossa (headaches, anosmia); palate (intraoral mass, loose dentition, malocclusion); and lacrimal apparatus (epiphora).
Occupational risk factors include exposure to wood and leather dust for intestinal-type adenocarcinoma. Nickel or chromium, organic solvents, and construction and textile site environments are mainly risk factors for squamous cell carcinoma (SCC).Reference Binazzi, Ferrante and Marinaccio959 A full ENT, head and neck, and cranial nerve examination must be supplemented by nasendoscopy. Biopsy can be performed in the clinic or in an operating theatre setting.Reference Iyizoba-Ebozue, Fleming, Prestwich and Thomson958
Imaging prior to biopsy should be carried out before general anaesthetic biopsy. However, a small out-patient biopsy from a non-vascular mass will not interfere with subsequent imaging and can expedite diagnosis. Biopsy, which is usually performed by a referring surgeon, should not take the form of a debulking, as this can make surgical planning more difficult, and may lead to an underappreciation of tumour extent (even if full pre-biopsy imaging is performed).
After diagnosis, in addition to standard head and neck multidisciplinary team (MDT) discussion with its core members, input from rhinology and skull base, neurosurgery, and oculoplastic surgery departments may be required. Additional MDT input will be required for sarcomas and may be necessary for neuroendocrine carcinomas.
Recommendations
• Sinonasal malignancy is a rare but complex disease that should be managed by a specialist MDT that includes anterior skull base expertise and oculoplastic input as necessary (evidence-based recommendation (R))
• History and examination should focus not only on the nasal cavity and paranasal sinuses, but also any potentially involved adjacent structures (good practice point (G))
• Biopsy should not take the form of macroscopic tumour debulking prior to referral to the appropriate MDT (G)
Imaging
The role of imaging is to define staging, help plan the surgical approach and enable planning of subsequent radiotherapy (RT) treatment volumes.
Table 1 shows the recommendations for different imaging modalities.Reference Gilain, Houette, Montalban, Mom and Saroul960
Table 1. Recommendations for different imaging modalities
*A combination of computed tomography (CT) and magnetic resonance imaging (MRI) of the sinuses and neck including the skull base is generally required for complete staging before treatment. Computed tomography is superior for delineation of bony and skull base erosion. Magnetic resonance imaging is superior for assessing soft tissue involvement (orbit, dura and brain parenchyma), perineural tumour spread, and differentiation of tumour from inflammation and secretions. †Positron emission tomography (PET)-CT can be considered for high-grade disease, or when there is significant nodal disease. Melanoma staging requires whole-body CT (or PET-CT) and brain imaging to identify occult tumours.Reference Gilain, Houette, Montalban, Mom and Saroul960 OPG = orthopantomogram
Pathology
Sinonasal tumours are histologically very diverse, and arise from surface epithelium or underlying seromucinous glands. Tumour types should be categorised according to the World Health Organization classification of head and neck tumours (fifth edition, 2022), which has several changes for sinonasal tract cancers.Reference Thompson and Bishop961 The commonest types are shown in Table 2.Reference Gilain, Houette, Montalban, Mom and Saroul960,Reference Sharma, Tang, Takiar, Wise-Draper and Langevin962
Table 2. Pathological summary of sinonasal tumours
SCC = squamous cell carcinoma; H&N = head and neck; HPV = human papillomavirus; SW1/SNF = switch/sucrose non-fermentable; EBV = Epstein–Barr virus
Correct diagnosis of these lesions is critical, but often poses significant pathological difficulties because of overlapping features, with up to 30 per cent of alterations reported from second opinions at centres of excellence.Reference Bell, Hanna, Weber, DeMonte, Triantafyllou and Lewis963 This is further complicated by a number of new disease entities being recognised on genetic and phenotypic testing, including NUT carcinomas, SW1/SNF (switch/sucrose non-fermentable) complex deficient sinonasal carcinoma and human papillomavirus (HPV)-related multi-phenotypic sinonasal carcinoma. To date, these have largely been of histopathological interest. Whilst any differences in the clinical behaviours are now being recognised in these rare subtypes, the principles of treatment remain the same as for other sinonasal carcinomas, which are fairly uniform, with the exception of increasingly used induction chemotherapy for sinonasal undifferentiated carcinoma.
Staging
AJCC/UICC tumour–node–metastasis staging
The eighth edition of the American Joint Committee on Cancer (‘AJCC’) / Union for International Cancer Control (‘UICC’) TNM Classification of Malignant Tumours made no changes to primary site staging for paranasal sinus tumours (Tables 3–5).Reference Brierley, Gospodarowicz and Wittekind964 Pathology is a major determinant for both disease behaviour and prognosis, and, as such, the tumour–node–metastasis (TNM) system is often a less useful clinical guide for management in this tumour site. Furthermore, primary sites of the frontal and sphenoid sinus are not covered. Historical classifications such as Ohngren's line,Reference Ohngren965 predicting a poor prognosis for tumours located postero-superiorly to a line connecting the angle of the mandible to the medial canthus, retain clinical relevance.
Table 3. Primary tumour staging*
* According to the TNM Classification of Malignant Tumours (eighth edition).Reference Brierley, Gospodarowicz and Wittekind964
Table 4. Nodal staging*
* According to the TNM Classification of Malignant Tumours (eighth edition).Reference Brierley, Gospodarowicz and Wittekind964
Table 5. Group staging*
* According to the TNM Classification of Malignant Tumours (eighth edition).Reference Brierley, Gospodarowicz and Wittekind964
The final TNM stage is used to categorise patients into prognostic disease stage groups, with five-year overall survival rates for different stages as follows: stage I (63 per cent), stage II (61 per cent), stage III (50 per cent) and stage IV (35 per cent).Reference Quan, Yan, Zhang, Zou, Yuan and Wang966 However, histopathological subtype specific data are likely to be more clinically relevant than staging.
A variety of different staging systems are used in practice for specific tumour types.
Olfactory neuroblastoma
Staging classification of olfactory neuroblastoma was first proposed by Kadish et al. in 1976,Reference Kadish, Goodman and Wang967 and modified by Morita and colleagues in 1993Reference Morita, Ebersold, Olsen, Foote, Lewis and Quast968 to include four stages (Table 6). This staging is also sometimes used for other tumour histological types.
Table 6. Kadish et al. staging of esthesioneuroblastomaReference Kadish, Goodman and Wang967, Reference Morita, Ebersold, Olsen, Foote, Lewis and Quast968
In variable clinical use, this and other complimentary alternative systems help guide treatment and surgical approaches, but are poor predictors of survival. Disease-grade stratification with pathological Hyams grading is complementary to radiological staging and independently associated with prognosis.Reference Thompson and Bishop961
Sinonasal malignant melanoma
The American Joint Committee on Cancer / Union for International Cancer Control staging system for mucosal melanoma of the head and neck is the most commonly used in clinical practice and is described in Chapter 29.
Nasal vestibule cancer
Often amalgamated within skin classifications, the nasal vestibule is a rare subsite and hence there is a lack of consensus about which staging system is most appropriate to use. However, Wang's classification (Table 7) is popular as it contains the most clinically relevant criteria.Reference Wang969
Table 7. Wang's staging of nasal vestibule carcinomaReference Wang969
Treatment
General principles
Recommendations
• The heterogeneity of tumour histological types, and the extent and nature of invasion, together with a poor evidence base means that MDT discussion is nuanced to the individual patient (good practice point (G))
• Sinonasal cancer management should take place in high-volume centres with appropriate histopathological, oncological, surgical, oculoplastic and reconstructive expertise (G)
• Most operable tumours are treated by primary surgery with post-operative RT or chemoradiotherapy (G)
• Tumours classified as ‘T1’ nasal vestibule cancers according to WangReference Wang969 should be treated by single-modality treatment, either RT or surgery (evidence-based recommendation (R))
Although many studies have shown improved survival over the last two to three decades, the overall prognosis remains poor, with five-year overall survival rates of around 40–50 per cent.Reference Slevin, Pan, Mistry, Denholm, Shor and Oong970 Achieving local tumour control is challenging because of the close proximity of critical anatomical structures. Local recurrence dominates failures (and hence a low threshold for post-operative RT), and isolated distant metastases are only seen in about 5 per cent of cases.Reference Dulguerov, Jacobsen, Allal, Lehmann and Calcaterra971
As for many rare cancer sites, there are no phase III trial data to direct therapy. Surgery is generally accepted to be the mainstay of treatment for operable tumours in suitable patients, with most cases also having post-operative RT, with or without chemotherapy. Single-modality surgical treatment is possible in selected T1 ethmoidal, or T1 and limited T2 (infrastructure or middle meatal extension) maxillary sinus cancers, if clear pathological margins can be confidently demonstrated in the absence of unfavourable histopathological features. Primary (chemo)radiotherapy can be considered for inoperable disease. Induction chemotherapy is variably but increasingly used for sinonasal undifferentiated carcinoma (discussed below). Cross-specialty involvement is crucial in these complex cases; where possible, sinonasal cancer management should take place in high-volume centres, with appropriate histopathological, oncological, surgical and reconstructive expertise.
The input and possible refinement of chemotherapy (neo-adjuvant, adjuvant or palliative) by a neuroendocrine MDT may be necessary for sinonasal neuroendocrine carcinomas of epithelial origin. Sarcomas should be discussed within a sarcoma MDT in conjunction with the head and neck MDT. Most patients with chondrosarcomas have primary surgery followed by adjuvant therapy. Other sinonasal sarcoma treatment is more nuanced in terms of the exact sequence and nature of multimodality treatment.
Nasal vestibule cancers are generally SCC, relatively rare, and behave more like cutaneous SCC. Hence, the evidence base for best treatment is poor. Generally, there is equipoise between surgery or RT for disease classified as ‘T1’ by Wang,Reference Wang969 and possibly for limited ‘T2’ disease.Reference Agger, von Buchwald, Madsen, Yde, Lesnikova and Christensen972 The choice may depend on the confidence of surgical resection achieving clear margins and the prospects of satisfactory reconstruction. For other larger tumours, multimodality treatment may be recommended, but that may depend on the radicality of surgery (e.g. where total rhinectomy is performed for T2/3 tumours with clear margins and no other adverse features, nasal vestibule cancer could be treated using surgery alone).
Inverted papilloma is a benign tumour with a significant risk of malignant transformation. It is most commonly staged based on local extent using the Krouse staging system.Reference Krouse973 Contrary to previous opinion, these tumours are generally not HPV driven with no evidence of high-risk HPV transcripts. It is locally aggressive, with recurrence rate estimates of up to 14 per cent and potential malignancy transformation risk of up to 8 per cent.Reference Long, Jabarin, Harvey, Ham, Javer and Janjua974 Treatment usually involves endoscopic resection, with subperiosteal clearance at the tumour insertion point of particular importance in order to minimise the risk of recurrence.Reference Lisan, Laccourreye and Bonfils975 When SCC is found histologically, management should be appropriate for this malignant diagnosis.
Surgery
A full description of surgical techniques is out of the scope of these guidelines; however, it is important that clinicians managing this condition are aware of the different options to best approach this disease. A number of different algorithms have been published to help inform surgical decision-making. We present one example, adapted from Naunheim et al.,Reference Naunheim, Goyal, Dedmon, Chambers, Sedaghat and Bleier976 as an example guide for approaching this disease (Figure 1).
Figure 1. Flowchart for surgical approach to sinonasal or anterior skull base malignancy, adapted from Naunheim et al.Reference Naunheim, Goyal, Dedmon, Chambers, Sedaghat and Bleier976 *Endoscopic assistance can be utilised in open approaches to: assess disease intra-operatively, perform targeted incisions (septum, lamina papyracea and skull base, pterygopalatine fossa) and facilitate repair. EEA = endoscopic endonasal approach
Craniofacial resection has long been held as the historical ‘gold standard’ method for ensuring adequate removal of anterior skull base tumours. However, endoscopic resection, facilitated by technological advances, has been shown to have equivalent long-term outcomes with less morbidity.Reference Nicolai, Castelnuovo, Lombardi, Battaglia, Bignami and Pianta977 This has addressed potential criticisms of endoscopic piecemeal resections. Even in open surgery, a true en bloc resection is rarely achieved because of tumour and anatomical factors.
This body of evidence now supports a treatment paradigm that focuses on negative surgical margins rather than the method of resection or choice of surgical approach.Reference Torabi, Spock, Cardoso, Chao, Morse and Manes978 The surgical options available, tailored to a patient's needs, can be viewed as different combinations of facial and cranial access (Table 8), together with additional access approaches (e.g. lateral).
Table 8. Summary of combinations of surgical approaches to paranasal sinuses and adjacent structures
Post-operative histopathology should be reported to include at least core items, as for other head and neck cancers, and should be discussed within the MDT meeting. Surgical–pathological correlation is vital to understand resection margins. This is challenging for sinonasal cancers because of the anatomical complexity and the typical piecemeal nature of resection. Reliance on multiple margin specimens or biopsies is common, but there needs to be a high degree of confidence if a decision to not offer post-operative RT is based on the assumption of adequately clear margins.
Recommendations
• Involvement of an MDT with the correct cross-specialty skill mix is critical to devise an appropriate surgical plan, including, where necessary, input from restorative dentistry and prosthetics, neurosurgery, rhinology, and oculoplastic surgery (evidence-based recommendation (R))
• The surgical approach selected should maximise the chance of achieving an R0 microscopically margin-negative resection (no residual tumour) (R)
• Surgical–pathological correlation discussed within an MDT meeting is vital to understand the completeness of excision (R)
Management of the eye
It is accepted that intraconal invasion, demonstrated clinically by visual loss, restriction of ocular mobility or globe infiltration, or by cross-sectional imaging, mandates exenteration. However, eye preservation in the presence of periorbita or even focal extraconal fat involvement remains controversial. Case series have reported high rates of orbital preservation and function.Reference Iannetti, Valentini, Rinna, Ventucci and Marianetti979,Reference Muscatello, Fortunato, Seccia, Marchetti and Lenzi980 However, published data are considerably heterogeneous,Reference Reyes, Mason, Solares, Bush and Carrau981 and a nuanced approach based on experience and individual patient circumstances is required. It should also be considered that most patients who have surgery near to or within the orbit will undergo post-operative RT, and hence will experience the combined effects of both treatments. When there is orbital involvement and eye preservation is aimed for, oculoplastic surgery input is ideal.
Detailed long-term data on radiation-induced orbital effects are lacking. Up to 83 per cent of patients suffer post-treatment complications at five years.Reference Kamrava, Lamb, McCannel, Singh and Damato982 De-intensification strategies related to eye preservation should therefore be adopted, ideally within a prospective clinical trial context.
Recommendations
• Orbital exenteration should be performed for intraconal invasion of sinonasal malignancy (evidence-based recommendation (R))
• Further prospective research is required for a histologically driven approach to eye conservation surgery (good practice point (G))
Radiotherapy and chemotherapy
Nasal vestibule cancers staged as T1 can be treated with RT as a single-modality treatment.Reference Jeannon, Riddle, Irish, O'Sullivan, Brown and Gullane983,Reference Filtenborg, Lilja-Fischer, Sharma, Primdahl, Kjems and Plaschke984
For other sinonasal cancers, definitive (chemo)radiotherapy can be considered for inoperable disease (approximately 20 per cent of patients) or in those who choose to avoid surgery.
There is very little data comparing definitive RT with surgery. The studies that have been published suggest comparable outcomes for SCC,Reference Duru Birgi, Teo, Dyker, Sen and Prestwich985 despite treatment selection bias. Salvage surgery is possible after definitive RT for residual disease (similar to laryngectomy after initial RT for T3 laryngeal cancer, for example), although this is an uncommon treatment strategy, with little outcome data. For sinonasal undifferentiated carcinoma, definitive RT may be given after a significant response to induction chemotherapy avoiding surgery.Reference Amit, Abdelmeguid, Watcherporn, Takahashi, Tam and Bell986
Radiotherapy dose–volume coverage is complex because of the proximity of critical structures including the eye, optic pathway, base of skull, brainstem and brain. There is potential for severe late radiation toxicities, including ocular, auditory, olfactory, endocrine, sinus or neurocognitive dysfunction.Reference Patel, Chitguppi, Vimawala, Epps, Fastenberg and Evans987 The use of intensity-modulated RT is the standard of care, whilst observational data from non-comparative studies suggest proton beam therapy may improve survival and late toxicity outcomes.Reference Patel, Wang, Wong, Murad, Buckey and Mohammed124,Reference Dagan, Bryant, Li, Yeung, Justice and Dzieglewiski988
For post-operative RT, doses of 60–66 Gy in 30–33 fractions of 2.0 Gy, or equivalent, are used. For definitive RT, a dose of 70 Gy in 35 fractions of 2.0 Gy, or equivalent, is used. For post-operative volumes, the gross tumour volume refers to the macroscopic tumour prior to surgery (and, where used, the pre-chemotherapy tumour volume), based on pre-treatment magnetic resonance imaging (MRI) and computed tomography (CT) scans. The high-dose clinical target volume includes a 1 cm expansion from gross tumour volume edited for natural fascial, air and bone boundaries, and the lower-dose clinical target volume includes the surgical bed or resection cavity, surgical clips, and the ipsilateral sinonasal compartment of the involved sinuses. For definitive RT volumes, the gross tumour volume refers to the macroscopic tumour (where used, the pre-chemotherapy tumour volume). The high-dose clinical target volume includes a 0.5 cm expansion from the gross tumour volume, edited for natural fascial, air and bone boundaries, and the lower-dose clinical target volume includes a 1 cm expansion from the gross tumour volume and the ipsilateral sinonasal compartment of the involved sinus(es).
The basis for the use of concurrent chemotherapy is extrapolated from head and neck pharyngeal SCCs.Reference Bernier, Cooper, Pajak, van Glabbeke, Bourhis and Forastiere733,Reference Pignon, le Maitre, Maillard and Bourhis989 For sinonasal SCCs (and select other high-grade pathologies), concurrent cisplatin chemotherapy may be used in the definitive setting, and post-operatively, where there are less than 1 mm surgical margins or pathological extra-capsular extension of the nodes.
The role of cisplatin-based induction chemotherapy for locally advanced inoperable sinonasal SCC or high-grade pathologies remains to be established. It may be used where the tumour is in close proximity to critical organs, predominantly for a down-staging effect prior to definitive surgery or chemoradiotherapy.Reference Khoury, Jang, Carrau, Ready, Barak and Hachem990 For sinonasal undifferentiated carcinomas, there is often a significant response to induction chemotherapy, and treatment selection based on this response has been published in large series, with chemotherapy responders receiving RT without surgery.Reference Amit, Abdelmeguid, Watcherporn, Takahashi, Tam and Bell986 A histology-driven approach of induction chemotherapy followed by response-based treatment selection is an important area of current research.
Recommendations
• Most patients treated by primary surgery undergo post-operative RT (good practice point (G))
• The indications for concurrent chemotherapy are extrapolated from head and neck SCC in general (G)
• Induction chemotherapy should be considered for sinonasal undifferentiated carcinoma (G)
Management of the neck
Lymphatics directed to the upper jugular, perifacial and retropharyngeal nodes comprise the main drainage pathways, with levels I and II being the most commonly involved regions. Overall, less than 10 per cent of patients present with regional disease,Reference Sangal, Lee, Brady, Patel, Eloy and Baredes991 although nodal recurrence – which is highly dependent on histology and site – can occur in up to 33 per cent of patients. Table 9 indicates the major risk factors, with increasing cumulative incidence of nodal disease.Reference Sangal, Lee, Brady, Patel, Eloy and Baredes991
Table 9. Risk factors for positive nodal disease
SCC = squamous cell carcinoma
Clinico-radiological nodal metastases are treated as for other head and neck cancers, by neck dissection for those having primary surgery, with or without post-operative (chemo)radiotherapy.
Elective treatment of the clinically node-negative neck, however, should be carefully considered depending on the risk factors involved. When an open surgical approach with free flap reconstruction is planned, elective neck dissection should be performed in conjunction with vessel exposure. Otherwise, elective neck irradiation for the above high-risk features should be considered, with the retropharyngeal and lower jugular (level IV) nodes included in the clinical target volume.Reference Biau, Lapeyre, Troussier, Budach, Giralt and Grau146 This approach has been shown to decrease the regional relapse rate to 5–10 per cent.Reference Paré, Blanchard, Rosellini, Aupérin, Gorphe and Casiraghi992,Reference Jeremic, Nguyen-Tan and Bamberg993
The highest level of evidence for elective neck management in sinonasal undifferentiated carcinoma comes from a recent meta-analysis with a positive nodal rate at presentation of 16 per cent, with 26.9 per cent regional failure at two years,Reference Morand, Anderegg, Vital, Ikenberg, Huber and Soyka994 advocating elective neck treatment with irradiation or neck dissection in locally advanced disease.
The role of elective neck treatment in other histological subtypes is more uncertain, but would appear to be unnecessary in low-grade pathologies.Reference de Gabory, Maunoury, Maurice-Tison, Merza Abdulkhaleq, Darrouzet and Bébéar995–Reference Shay, Ganti, Raman, Kuhar, Auger and Eggerstedt997
Recommendations
• Therapeutic neck treatment when there are clinical-radiological nodal metastases follows that for head and neck cancer in general, namely surgery (with or without post-operative (chemo)radiotherapy) or primary (chemo)radiotherapy (evidence-based recommendation (R))
• Elective neck treatment (generally bilateral) for the clinically staged node-negative (N0) neck with irradiation or neck dissection is recommended for T3/4 SCC and sinonasal undifferentiated carcinoma (R)
• There is insufficient evidence to support recommendations in other sinonasal subtypes, although elective neck treatment in high-grade tumours should be considered to reduce regional failure (good practice point (G))
Reconstruction and rehabilitation
The aims of reconstruction serve two primary purposes, namely functional (separation of the intracranial cavity and sinonasal airspace; separation of the mouth, and nose or sinuses; creation of a barrier between sinonasal airspace and skin; orbital support), and aesthetic (addressing skin loss, facial contour deformity).
Although there is the potential for refinements later in the treatment pathway, reconstruction should be addressed at primary surgical planning and prior to adjuvant RT.
There are further details on maxilla and mid-facial reconstruction in Chapter 7 (on reconstructive considerations) and Chapter 13 (on restorative dentistry), which discuss extra-oral and oral prosthetics. Anterior skull base defects for reconstruction can be divided into four broad categories.
Skull base only (bone with or without dural defect)
There are a variety of choices for closure, ranging from non-vascularised fat or a fascial graft (inlay and/or onlay), often in conjunction with fibrin glue, or a multilayer closure with local nasoseptal flaps (after limited endoscopic resections) or a pericranial flap (after a bicoronal flap) if available.
Skull base plus facial defects and/or orbital exenteration
Local skin flap options are limited in volume to close the dead space, and local muscle options (e.g. temporoparietal flaps) are limited in reach to the central area. These can, however, be combined with osseo-integration for facial prostheses in patients not suitable for microvascular free flaps. Otherwise, the optimal approach is microvascular free tissue transfer, providing well-vascularised tissue with a suitable volume for dead space filling as well as skin for cutaneous reconstruction.
Skull base plus lower maxillary or palate defects
For large volume defects that result in communication from the oral cavity to the skull base (see Chapter 7), composite free tissue transfer is generally required. There should be enough tissue transfer to safely close off any skull base defect. Soft-tissue-only reconstructions will obturate the operative site, but tend to descend and retract with gravity, causing mid-facial ptosis, speech and swallowing issues as well as aesthetic concerns.
Frontal sinus and frontal bone reconstruction
If the inner table is resected, the sinus can be cranialised. If an open sub-cranial approach is used, the outer table is re-plated with the upper nasal bones. This can occasionally break down after RT (usually as a late effect). If ablation of both tables is necessary, the resulting through and through defect requires reconstruction. Soft tissue only at this site retracts and potentially descends. Osseous free flaps can be used for such defects with associated soft tissue, to protect pedicles from aerodigestive tract secretions, close dural defects and fill any local dead space. Autogenous calvarial grafts or synthetic cranioplasty materials (titanium, polyetheretherketone (‘PEEK’) and so on) can be employed, but their non-vascularised nature may be complicated within irradiated fields.
Recommendations
• Early involvement of reconstructive and prosthetic specialists is vital to achieve optimal functional and cosmetic outcomes (evidence-based recommendation (R))
Follow up
Follow up should be guided by local and standard head and neck practice (Chapter 5). Support for those patients with functional and aesthetic defects is essential. Some patients will have on-going prosthetic intervention (intra-oral and extra-oral). In patients with orbital preservation, treatment effects are common, which can include epiphora, the effects of changes in globe position, and radiation-induced retinal and optic nerve damage and cataracts. Hence, ophthalmology input is required for this patient group.
Imaging by MRI of the primary site three to four months after treatment completion is standard practice for establishing a baseline for future comparison, as anatomy will usually be altered through treatment.Reference Byrd, Clair and El-Sayed998 The positron emission tomography (PET)-CT will often be performed within three to four months of definitive radiation or systemic therapy, for assessment of treatment response and to identify any residual tumour.
It is worth noting that hypermetabolism following sinonasal malignancy treatment can be prolonged.Reference Schwartz, Brooks, Stubbs, Ghosh, Tajudeen and Khalili999 Additional re-imaging should be considered depending on baseline imaging findings, concerning signs or symptoms, risk of recurrence, and ease of clinical surveillance (e.g. after reconstruction when endoscopy is not reliable or possible). In most cases, imaging is an important component of surveillance.
Recommendations
• Baseline imaging by MRI of the primary site should be conducted around three to four months post-treatment, to allow for future comparison (evidence-based recommendation (R))
• Tumour surveillance is both clinical and radiological (R)
Recurrent tumours
Recommendations
• Many locally recurrent tumours are incurable. Careful selection for salvage surgery is recommended on the basis of prognosis and morbidity of salvage surgery (good practice point (G))
Most recurrences of sinonasal cancer after treatment are local. Local recurrences are prognostic for poor overall survival.Reference Dagan, Bryant, Li, Yeung, Justice and Dzieglewiski988 Surgery usually provides the only feasible treatment option. Recurrences should be assessed on a case-by-case basis. Tumours of high-risk histological subtype or grade (most types except lower grade salivary cancers and low-grade olfactory neuroblastomas), and/or with orbital and skull base involvement, have a very poor prognosis. Occasionally, limited additional surgery is appropriate. Local recurrences limited to the ethmoid sinus region have a better prognosis.Reference Kaplan, Kim, Wang and Snyderman1000 Local recurrences of intestinal-type adenocarcinoma tumours are frequently localised, which may in part reflect the multifocality of this tumour, and many are suitable for endoscopic resection.Reference Camp, Van Gerven, Poorten, Nuyts, Hermans and Hauben1001 Otherwise, radical salvage surgery involves significant morbidity, and clear margins are difficult to achieve.
There is a lack of evidence for immune-based targeted therapies in sinonasal malignancy. Despite programmed death-ligand 1 (PD-L1) positivity not appearing to have prognostic value, its use is likely to be extrapolated from other head and neck subsite protocols.
Studies due to report
‘Docetaxel, Cisplatin and Fluorouracil in Treating Patients With Previously Untreated Stage II-IV Nasal Cavity and Paranasal Sinus Cancer’ – a phase II trial of induction chemotherapy (docetaxel, cisplatin and fluorouracil) in patients with previously untreated stage II–IV nasal cavity and/or paranasal sinus cancer. Location: MD Anderson Cancer Center, Texas, USA.
‘Prospective Sinonasal Cancer Multi-institution Study’ – a multi-institution prospective study of patients with sinonasal malignancies, to study disease course, treatment outcomes and patient quality of life. Location: Mayo Clinic in Rochester, New York, USA.
‘Intensity-Modulated or Proton Radiation Therapy for Sinonasal Malignancy’ – a prospective, non-randomised, phase II study comparing local control and the toxicity profile of proton beam therapy and conventional intensity-modulated RT in advanced sinonasal malignancy. Location: Massachusetts General Hospital, Boston, USA.
Research questions
Is there a role for induction chemotherapy for specific sinonasal malignancies in addition to sinonasal undifferentiated carcinoma, e.g. for chemo-selection or as a pre-surgical organ preservation approach?
Is there a genomic or transcriptomic profile that will further help risk-stratify patients or predict treatment response? This will require a collaborative national or international sample collection given the disease rarity.
What is the role of charged particle therapy as a primary and adjuvant treatment modality compared to standard intensity-modulated RT in a phase III, prospective randomised, controlled trial?
Introduction
Malignant salivary gland tumours contribute to about 5 per cent of head and neck cancers, with an incidence of 8–13 per one million per year.Reference McGurk and Renehan1002 Their relative rarity, diverse histopathology and heterogeneity of clinical behaviour mean that there are limited trial data from which to develop evidence-based management recommendations.
Tumour grade is an important factor to consider, as it relates to prognosis and treatment. As a general principle, surgery forms the mainstay of treatment. Adjuvant radiotherapy (RT) should also be considered for the majority of malignant salivary tumours (except small, low-grade tumours).
Presentation and diagnosis
Presentation
In general, malignant salivary gland tumours present either as painless, mobile lumps, or with an enlarging, painful, fixed mass, with or without facial palsy and regional lymph node spread. It is notable that most salivary gland malignancies present without signs of malignancy, i.e. without skin involvement, facial nerve weakness, short history, rapid growth, pain and/or paraesthesia, regional lymphadenopathy. These clinical features occur only in 25–35 per cent of parotid carcinomas.Reference Wong1003 Therefore, all salivary gland masses must be carefully investigated to establish whether they are benign or malignant.
Table 1 describes data from three large UK series.Reference Jones, Craig, Speight and Franklin1004–Reference Barnes, Eveson, Reichart and Sidransky1006 Although, overall, tumours are more common in the parotid; those that are malignant have a higher proportion in the submandibular, sublingual and minor salivary glands.Reference Spiro1007,Reference Alsanie, Rajab, Cottom, Adegun, Agarwal and Jay1008
Table 1. Site and malignancy rate in salivary gland tumours
Diagnosis
Recommendations
• Ultrasound-guided fine needle aspiration cytology (FNAC) or core biopsy are recommended for all salivary gland tumours (evidence-based recommendation (R))
• Both FNAC and core biopsy should be available (R)
• Magnetic resonance imaging (MRI) is the preferred axial imaging modality (good practice point (G))
Patients should be assessed in a rapid access clinic by appropriate clinicians with access to ultrasound and cytopathology (see Chapter 2). Every effort should be made to determine whether the tumour is benign or malignant before proceeding to surgical treatment. However, definitive exclusion of malignancy may only be possible following histological evaluation of the entirely excised tumour in some cases.
Fine needle aspiration cytology and core biopsy
Fine needle aspiration cytology and/or core biopsy should be considered in all salivary gland tumours. Fine needle aspiration cytology is less invasive, and can provide a diagnosis of neoplasia and may differentiate benign from malignant. The large number of salivary gland tumour entities, the morphological complexity and heterogeneity within a single neoplasm, the metaplastic changes and the absence of tumour architecture in cytology specimens limit the accuracy of FNAC. Fine needle aspiration cytology has a sensitivity of 67–98 per cent, a specificity of 82–100 per cent and accuracy of 81–99 per cent.Reference Jeong, Park, Cha, Sung, Kim and Ahn1009–Reference Feinstein, Alonso, Yang and St John1011 The current and most widely accepted reporting system for salivary gland cytopathology is the Milan system (Table 2).Reference Rossi, Baloch, Pusztaszeri and Faquin1012
Table 2. Milan system for reporting salivary gland cytopathology
FNA = fine needle aspiration
Core biopsy in general is more helpful to assess tumour morphology and establish the type of tumour. A meta-analysis of ultrasound-guided core biopsies, mostly using 18–20 gauge spring-loaded needles, reported a pooled sensitivity of 94 per cent and a specificity of 98 per cent.Reference Kim and Kim1013
Occasionally, excision may be required for definitive tumour characterisation, but only if all efforts to establish a diagnosis fail. Similarly, open biopsy should be avoided if possible, given the risk of tumour seeding and spillage, but it has a role in fungating tumours.
Minor salivary gland tumour biopsies may be taken by skin punch or incision biopsy.
Frozen section
The indications of frozen sections for intra-operative diagnosis are highly limited and its routine use is discouraged. Occasionally, however, intra-operative frozen section may be considered in cases where attempts at cytological diagnosis have failed on at least two occasions, and may help to stratify the risk of malignancy and extent of surgery in such cases.Reference Olsen, Moore and Lewis1014,Reference Choy, Bundele, Fu, Li, Gan and Rao1015
Imaging
Ultrasound imaging is a key part of the initial assessment of major salivary gland tumours, assisting with staging and tissue sampling. Ultrasound should be used to guide needle aspiration or core biopsy in nearly all major salivary gland tumours, and may assist with the assessment of regional adenopathy.
When malignancy is suspected or confirmed, contrast MRI is the preferred axial imaging modality, although computed tomography (CT) is an alternative and can characterise bone invasion in locally advanced tumours. For tumours that are thought to be malignant, CT scanning of the thorax, or positron emission tomography (PET)-CT, are used to assess for distant metastases.
Carcinomas of minor salivary glands are imaged as tumours of the lip and oral cavity.
Recommendations
Recommendations for different imaging modalities:
• Primary major salivary gland – MRI with or without CT
• Regional neck – MRI or CT
• Distant – CT or PET-CT (the latter can be considered for high-grade disease or when there is significant nodal disease)
Staging
The eighth edition of the TNM Classification of Malignant Tumours staging system for salivary gland primary tumours is unchanged from the seventh edition (Tables 3–5).Reference Brierley, Gospodarowicz and Wittekind1016 Carcinomas of minor salivary glands are staged according to tumours of the lip and oral cavity.
Table 3. Tumour staging for major salivary gland tumours*
*According to the TNM Classification of Malignant Tumours (eighth edition).Reference Brierley, Gospodarowicz and Wittekind1016 †Extra-parenchymal extension refers to clinical or macroscopic evidence of soft tissue invasion; microscopic evidence alone does not constitute extra-parenchymal extension for classification purposes.
Table 4. Nodal staging for major salivary gland tumours*
* According to the TNM Classification of Malignant Tumours (eighth edition).Reference Brierley, Gospodarowicz and Wittekind1016
Table 5. Group staging for major salivary gland tumours*
* According to the TNM Classification of Malignant Tumours (eighth edition).Reference Brierley, Gospodarowicz and Wittekind1016
Pathology, classification and tumour biology
Recommendations
• All treatment centres should have access to diagnostic molecular testing (good practice point (G))
• Consider pathology review when initial biopsy or surgery conducted outside of the treatment centre (G)
Salivary gland tumours are a heterogeneous group. Detailed description of salivary gland tumour pathology is beyond the scope of this guideline. New tumour subtypes have been identified, largely based on molecular alterations, rearrangements and fusions, leading to the existence of a new ‘molecular gold standard’. The World Health Organization (WHO) 2022 classification of head and neck tumours incorporates some new entities (Table 6).Reference Skálová, Hyrcza and Leivo1017 The classification contains a category ‘Salivary carcinoma not otherwise specified (NOS) and emerging entities’, which includes some poorly differentiated carcinomas and oncocytic carcinomas.
Table 6. WHO classification of salivary gland tumours (2022)Reference Skálová, Hyrcza and Leivo1017
WHO = World Health Organization
Histological grading of salivary gland carcinomas can serve as an independent predictor of biological behaviour, prognosis and outcome, which may help optimise therapy. A positive correlation between histological grade and clinical stage has been consistently identified.Reference Seethala1018,Reference Seethala1019
Examples of salivary gland neoplasms that are typically low-grade include acinic cell carcinoma, basal cell adenocarcinoma, epithelial-myoepithelial carcinoma, secretory carcinoma and clear cell carcinoma. Myoepithelial carcinoma is typically of intermediate grade. Salivary duct carcinoma, lymphoepithelial carcinoma, small cell and large cell neuroendocrine carcinoma, large cell undifferentiated carcinoma, primary squamous cell carcinoma, and carcinosarcoma are typically high-grade. Some tumours, such as mucoepidermoid carcinoma, adenoid cystic carcinoma, intraductal carcinoma, carcinoma ex pleomorphic adenoma and adenocarcinoma not otherwise specified, can show variable grades.
A solid and highly infiltrative growth pattern can also be a feature of high-grade transformation in otherwise typically low-grade salivary gland malignancies.
Treatment
Recommendations
• Wide local excision achieving clear margins is the principle surgical operation for most submandibular gland cancers (evidence-based recommendation (R))
• Total parotidectomy with preservation of facial nerve function is the principle of treatment for most parotid cancers (R)
• In general, if facial nerve function is normal pre-operatively, and is not involved or encased by the tumour, the facial nerve function should preserved (good practice point (G))
• Resection of facial nerve branches should be undertaken where there is pre-operative and/or intra-operative evidence of nerve invasion by the cancer (R)
• Clinically staged node-positive (N+) neck – neck dissection should be performed (R)
• Clinically staged node-negative (N0) neck – high or intermediate or indeterminate grade disease – selective neck dissection should be performed as elective treatment (R)
• Clinically staged N0 neck – low-grade tumours – elective selective neck dissection should be considered (G)
• The standard indications for post-operative RT (see Chapter 4) apply to salivary gland cancers (R)
• Additional considerations specific to salivary gland cancers include all stage 3/4 tumours, high-grade cancers and adenoid cystic carcinoma (G)
Surgery
Surgery is the treatment of choice for resectable salivary gland cancers. As with all cancers, it is important to achieve clear resection margins.Reference Spiro1007,Reference Morse, Fujiwara, Judson, Prasad and Mehra1020 Whilst head and neck malignancies are typically resected with a margin of up to 1 cm in order to achieve a 5 mm pathological margin (where anatomical and functional considerations permit), in salivary gland cancers, the optimal resection margin has not been established in a randomised study, and close margins may not necessarily adversely affect the oncological outcome.Reference Malik, Devabalan, Bernstein, Awad, Gujral and Partridge1021 Attaining an uninvolved closest resection margin is balanced against the need for the preservation of function and cosmesis and preservation of the facial nerve. A close resection margin, therefore, is not uncommon.
Submandibular gland
Wide excision is appropriate for tumours confined to the gland, combined with neck dissection. Some argue in favour of a wider resection for adenoid cystic carcinoma, but even with radical surgery it is frequently difficult to obtain adequate surgical margins.Reference Garden, Weber, Morrison, Ang and Peters1022 It should be noted that the advice for radical surgery in submandibular gland malignancy is at variance with recommendations for the preservation of the uninvolved facial nerve in parotid malignancy. However, it is generally accepted that high-grade tumours should be treated aggressively, involving excision of the gland with wider resection margins and with the resection of involved nerves if necessary. Large infiltrative tumours with bony involvement are treated with composite resection of the tumour, adjacent soft tissue cuff and segmental mandibulectomy.
Parotid gland
Surgery is the treatment of choice for malignant parotid tumours, with the extent of surgery tailored to the individual tumour. In general, a conservative total parotidectomy with preservation of facial nerve function is the principle of treatment for most lesions, except for small, low-grade malignancies which may be treated by a partial or superficial parotidectomy.
The facial nerve is a critical structure that limits the margin of resection in parotid malignancies. If there is a facial nerve palsy pre-operatively, the facial nerve should be resected, with frozen section evaluation of surgical margins to ensure adequate clearance.
If there is a plane of dissection between the functioning nerve and tumour, the surgeon should undertake facial nerve preservation. If it is evident that the facial nerve is invaded or encased by a confirmed cancer, the involved trunk or branches should be resected to optimise local tumour control. Where facial nerve resection forms part of a planned resection of parotid carcinoma, the ablative procedure should be accompanied by planned reconstructive procedures including nerve grafting and/or slings. The eye should be protected and assessed for any procedures at a later date.
It should be noted that resection of the uninvolved facial nerve in adenoid cystic carcinoma has been shown not to affect local control.Reference Iseli, Karnell, Preston, Graham, Funk and Buatti1023
Inadvertent injury to the facial nerve should be repaired as soon as possible, ideally at the time of tumour resection. Direct microsurgical repair without tension, or a cable nerve graft, offer the best chance of recovery. Facial re-animation techniques may be considered if specialist input is available.
More extensive parotidectomy should be undertaken in locally advanced cancers invading extra-parotid structures (e.g. lateral skull base – see Chapter 25).
Minor salivary glands
Tumours of the minor salivary glands have been reported to have a worse oncological outcome compared to parotid and submandibular gland malignancies.Reference Santos ES, Rodrigues-Fernandes, Speight, Khurram, Alsanie and Costa Normando1024 The outcome is worse for occult sites such as the nasal cavity, nasopharynx, larynx and trachea.
Most cases are treated in a similar way to squamous cell carcinoma, with en bloc resection, with the depth of excision compatible to treatment of squamous cell carcinoma, to ensure adequate resection margins. Significant defects are reconstructed as appropriate with local or free flaps.
Neck metastases
Clinically node-positive (N+) disease
Patients with confirmed neck metastases should have a therapeutic neck dissection. The levels of neck dissection will be influenced by the primary tumour site and stage, histological subtype and grade.
For a parotid primary tumour, the neck dissection should include at least levels IIA, III and IV,Reference Ali, Palmer, DiLorenzo, Shah, Patel and Ganly1025 and consideration should be given to the dissection of adjacent uninvolved levels.
For submandibular gland primary tumours, the neck dissection should include at least levels Ia, IIa and III. Again, consideration should be given to the dissection of adjacent uninvolved levels.
Clinically node-negative (N0) disease
The evidence base for the management of a clinically N0 neck is poor. The frequent lack of knowledge of the histological subtype and grade pre-operatively is a complicating factor in deciding on the appropriate management.Reference Erovic, Shah, Bruch, Johnston, Kim and O'Sullivan1026 Clinical staging underestimates the rates of regional lymph node metastases, with the rate of occult lymph node spread being 23 per cent in elective neck dissections in one pooled analysisReference Valstar, van den Brekel and Smeele1027,Reference Xiao, Zhan, White-Gilbertson and Day1028 and 34 per cent for high-grade cancers in another.Reference Westergaard-Nielsen, Rosenberg, Gerke, Dyrvig, Godballe and Bjorndal1029 Elective neck dissection should be performed in the presence of adverse features, a high-grade and an advanced stage.Reference Jinnin, Kawata, Higashino, Nishikawa, Terada and Haginomori1030
Neck dissection should include levels Ia, IIa, IIb and III for parotid primary tumours, and levels I, IIa and III for submandibular gland primary tumours.Reference Ali, Palmer, DiLorenzo, Shah, Patel and Ganly1025
In low-grade tumours, with clinically staged N0 necks, the same elective neck dissection should be considered in most cases, but the decision-making may be more nuanced, e.g. with more selective (less extensive) elective neck dissections.
When a malignant diagnosis is not known or suspected before or at surgery, the clinically staged N0 neck may be addressed by RT and/or re-operative neck dissection, but any elective reoperation should not delay RT.
Post-operative radiotherapy
Adjuvant RT improves survival in patients with early- and advanced-stage salivary gland cancers, according to an analysis of 8580 patients from the US National Cancer Database, whilst concurrent chemotherapy did not improve outcome.Reference Cheraghlou, Kuo, Mehra, Agogo, Bhatia and Husain1031
The decision for RT should be made after discussion at the specialist multidisciplinary team (MDT) meeting. The standard indications for post-operative RT (see Chapter 4) apply to salivary gland cancers. Additional considerations specific to salivary gland cancers include all stage 3/4 tumours, high-grade cancers and adenoid cystic carcinoma.
Hence, adjuvant RT is recommended in cases including:
• High-grade tumours
• Advanced-stage tumours (stage 3/4 tumours)
• Involved resection margins (R1–2, microscopic or macroscopic residual tumour)
• Perineural invasion
• Lymphovascular invasion
• Extra-nodal extension
• Following revision resection
• Adenoid cystic carcinoma
Radiotherapy is not required for small, low-grade tumours that have been completely excised.
Recurrent malignant parotid gland tumours
This situation requires careful evaluation of the patient with repeat imaging and a review of the histology from the initial excision. Treatment will usually require more radical surgery, with sacrifice of the facial nerve and overlying skin if there is any suspicion of involvement by the tumour. Radical resections of the skull base have not to date shown convincing evidence of improved survival. Where an R0 resection (no residual tumour) is considered not achievable, surgery should not be attempted. Chemotherapy, RT and drug trials should be considered for palliation.
Post-treatment surveillance
Patients should be followed up for at least five years, in keeping with MDT protocols. In salivary gland malignancies with adverse features, the survival curves tend to begin to plateau towards 10 years rather than 5 years, and follow up should be considered for longer. Ultrasound may be a useful supplement to clinical follow up. Baseline CT, MRI or PET-CT imaging at three months may be useful for comparison if there is a suspicion of recurrent disease at a later stage.
There is no consensus regarding chest imaging as part of follow up for cancers that have a predilection for distant metastases to the lungs (e.g. adenoid cystic carcinoma). There are ongoing trial therapies for metastatic salivary tumours and, whilst their efficacy is as yet not proven, detection of distant metastases.
Benign tumours
When treatment is considered, surgery is the treatment of choice. For benign tumours of the submandibular gland, the gland should be excised in a supracapsular plane. A wide dissection of local tissues is not required.
For benign parotid tumours, resection of the tumour should be performed without breach of the tumour capsule and with preservation of the facial nerve. There is no consensus within the literature as to the extent of resection. Procedures include: formal superficial parotidectomy, partial parotidectomy and extra-capsular dissection. The extent of surgery is largely determined by the location and size of the tumour, and by the preference of the operating surgeon. There is a trend towards reducing the extent of the resection with the proviso of preserving the tumour capsule and facial nerve function. The facial nerve is often in close proximity to the tumour, thereby not infrequently leading to close margins on final histology assessment. It is important to avoid tumour rupture, by careful dissection.
Tumour spillage carries with it an increased rate of tumour recurrence over a prolonged period.Reference Park, Cho, Kang, Roh, Choi and Kim1032 Tumour spillage may be frank and noted at the time of surgery, or more commonly noted on the final histology report. If tumour spillage has been noted, consideration should be given to long-term follow up, and the case should be discussed in the MDT meeting as there is a potential role for RT in a select few cases. Radiotherapy is generally not indicated for the majority of cases of tumour spillage and is generally not recommended in younger patients because of the risk of radiation-induced tumours.
Recurrent benign tumours require careful investigation, including ultrasound with fine needle aspiration (FNA) and MRI imaging. Treatment is usually surgical excision of the recurrent tumour with preservation of facial nerve function, to include resection of all recurrent nodules. In some cases, surgical excision of isolated tumour nodules may suffice. Recurrence is often multifocal, and re-recurrence is therefore common given the challenges of excising all of the recurrent nodules. Therefore, adjuvant RT should be considered for such cases in the MDT meeting.
Additional considerations regarding specific tumour histological types
Commoner histological types
Mucoepidermoid carcinoma
• Most common major salivary gland tumour (4–9 per cent) (including in children), with over 90 per cent occurring in the parotid gland.
• These tumours have a variable grade.
• Histopathological division into low, intermediate and high grades correlates somewhat with prognosis; although low-grade tumours can on occasion metastasise to cervical lymph nodes. Five-year survival rates vary between 86 per cent for low-grade and 22 per cent for high-grade tumours.
• Treatment guidelines vary according to grade.
Adenoid cystic carcinoma
• Accounts for 25 per cent of malignant salivary gland tumours – all major glands and minor glands.
• Slow, pervasive growth and a high incidence of perineural infiltration which can cause pain.
• Predilection for distant pulmonary metastases.
• Variable histological appearance. It can be difficult to correlate with clinical behaviour, although, in general, tubular and cribriform patterns are associated with a better prognosis than solid pattern tumours. Tumours with NOTCH1 mutation have a poorer prognosis than tumours with MYB or MYB1 and/or NFIB translocations.Reference Miller, Au, Mokhtari, Goss, Faden and Varvares1033
• Szanto et al.Reference Szanto, Luna, Tortoledo and White1034 and Perzin et al.Reference Perzin, Gullane and Clairmont1035 described a commonly used grading scheme: grade 1 – predominantly tubular, no solid component; grade 2 – predominantly cribriform, less than 30 per cent solid component; and grade 3 – solid component of more than 30 per cent.
• Only 20 per cent of patients with pulmonary metastases survive more than five years.
• Stage I and II cancers may be cured, although the survival curve never flattens, even after 20 years. Long-term follow up is recommended because there is a high late recurrence rate. Stage III and IV disease is associated with a poor prognosis, with low survival rates at 10 years.
• There are insufficient data to clarify whether pulmonary metastectomy (and therefore chest imaging surveillance) is beneficial in some patients.
Acinic cell carcinoma
• Around 3 per cent of parotid tumours
• Usually low grade
• Can be multifocal in origin and occasionally bilateral
• Survival rates of 90 per cent at 5 years and 55 per cent at 20 years
• Lymph node metastases occur in approximately 10 per cent of cases
• Treatment is in keeping with grade and size of tumour
Adenocarcinoma not otherwise specified
• This uncommon tumour is most frequently (90 per cent) found in the parotid gland.
• It has a variable grade; histological appearance varies between low-grade, well-differentiated papillary or mucinous patterns, to high-grade, undifferentiated lesions.
• The incidence of distant metastases is 40 per cent for high-grade tumours, and is directly related to survival rates (five-year survival rate of 75 per cent for low-grade tumours and 19 per cent for high-grade tumours).
• Treatment is usually as for high-grade salivary cancer, unless proven low-grade and small.
Carcinoma ex pleomorphic adenoma
• Carcinoma ex pleomorphic adenoma is a broad category of carcinomas of the salivary glands that often pose a diagnostic challenge to clinicians and pathologists.
• Carcinoma ex pleomorphic adenoma are carcinomas with histological, molecular or historic evidence of arising in or from a primary or recurrent pleomorphic adenoma.
• These tumours are classified as in situ carcinomas, non-invasive or intracapsular, and minimally invasive (less than 4–6 mm) or invasive. The extent of invasion is an important prognostic factor; the clinical behaviour of non-invasive tumours is similar to that of pleomorphic adenoma with a good prognosis, whilst tumours extending more than 6 mm have been shown to have a poor prognosis.
• Different patterns of malignant change can occur in carcinoma ex pleomorphic adenoma, with the other types being a carcinosarcoma and metastasising pleomorphic adenoma. The malignant component is most often salivary duct carcinoma, but can also be adenocarcinoma not otherwise specified, adenoid cystic or mucoepidermoid carcinoma.
• The remainder are probably not a homogeneous group of tumours, and may occur de novo rather than following a malignant generation of pleomorphic adenoma.
• Locoregional recurrence is considered to be an important prognostic factor with a poor prognosis after detection of recurrence. The reported survival rate at 5 years is 30–40 per cent, dropping to 20 per cent at 15 years.
Squamous cell carcinoma
• Squamous cell carcinoma affecting the parotid gland is a rare primary tumour, and is usually one or more regional lymph node metastases from a current or previous cutaneous squamous cell carcinoma of the scalp or face including the nasal cavity and ear canal.
• It tends to occur in older adults (seventh and eighth decades of life).
• It is associated with poor prognosis and should be treated in a similar fashion to high-grade cancers.
• The five-year disease-specific survival rate is about 25 per cent, with high rates of distant metastasis.
New malignant tumour entities and recent modification of terminology
Secretory carcinoma
Formerly known as mammary analogue secretory carcinoma, this salivary gland adenocarcinoma is characterised by ETV6 gene rearrangement and is composed of intercalated duct-type cells; 70 per cent of the tumours have been described in the parotid gland, followed by 20 per cent in the minor salivary glands of the oral cavity and 8 per cent in the submandibular gland.
Most cases of secretory carcinoma are low-grade, and present as a slow-growing, painless, well-demarcated mass with a possible cystic component. Up to 25 per cent of cases can develop lymph node metastasis. Lymphovascular and perineural invasion are uncommon. Architectural patterns within the tumour can vary and are similar to acinic cell carcinomas. Distant metastases and tumour-related deaths are rare, except in rare cases with high-grade transformation.
Microsecretory adenocarcinoma
These tumours were previously reported as a low-grade variant of adenocarcinoma not otherwise specified. Given the small number of cases identified as microsecretory adenocarcinoma, data on prognosis and overall survival are currently limited.
Sclerosing microcystic adenocarcinoma
Sclerosing microcystic adenocarcinoma has also been described. It resembles microcystic adnexal carcinoma of the skin, showing an infiltrative duct forming tumour and syringoma-like formations. Most cases occur in the oral cavity. This is classed as a low-grade variant of adenocarcinoma not otherwise specified spectrum.
Intraductal carcinoma
This refers to a malignant salivary gland tumour where all salivary proliferations are entirely or predominantly intraductal in location. This entity was previously known as low-grade salivary duct carcinoma and low-grade cribriform cystadenocarcinoma. It is predominantly seen in the parotid gland. Subtypes include: intercalated duct-type and oncocytic, both of which are rarely associated with invasion; apocrine type, which can show low- or high-grade cytomorphology; and mixed intercalated duct-apocrine tumour.
Intraductal carcinoma without invasion has an excellent prognosis following complete surgical excision, requiring no adjuvant therapy. Apocrine intraductal carcinoma with invasion has a poorer prognosis. A diagnosis of intraductal carcinoma on a core biopsy means that the tumour needs to be excised with caution, and the clinician should be made aware of the possibility that an invasive component could be present in the excision specimen.
Polymorphous adenocarcinoma
Previously named polymorphous low-grade adenocarcinoma, this is a malignant tumour. It is usually low-grade, but some – known as cribriform adenocarcinoma of the salivary gland – have behaved more like high-grade tumours. Cribriform adenocarcinoma may be considered an emerging new tumour entity. It demonstrates a predilection for the tongue base, and has an up to 65 per cent risk of nodal metastases.
Future research
Clinical trials in the management of salivary carcinoma are a particular challenge because these cancers are rare. Working towards meaningful trials in this area, the establishment of tumour registries on a prospective basis would provide a foundation for future trial design. This would provide accurate observational data and allow outcome comparison, and hence facilitate the development of research questions and trial design for this rare and diverse tumour group. Widespread collaboration – an increasing trend in head and neck surgical research in the UK – will be necessary to provide these processes and data.
Introduction
Thyroid cancer is an uncommon malignancy, with an incidence in the UK of approximately 8 per 100 000 women and 3 per 100 000 men.Reference Li, Dal Maso and Vaccarella1036 Evidence suggests an increasing incidence due to increased rates of detection; however, the survival rates remain static. Thyroid nodules are common, with 50–70 per cent of the general population showing ultrasound evidence of nodules.
The long-term prognosis for differentiated thyroid cancer, which includes papillary and follicular subtypes, is excellent, with a survival rate for adults of over 90 per cent at 10 years’ follow up.Reference Craig, Smart, Fielding, Ramsay and Krukowski1037 However, up to 20 per cent of patients develop locoregional recurrence requiring further treatment, and 5 per cent go on to develop distant metastases. Recent improvements in the understanding of biology have led to a move away from aggressive treatment of differentiated cancers, based on appropriate initial and subsequent dynamic risk stratification associated with the response to initial therapy.
Medullary and anaplastic thyroid cancers are rare and more aggressive. A structured approach to the assessment and management of these conditions is essential to optimise outcomes for these patients. The main types of thyroid cancer are summarised in Table 1.Reference Scoazec, Couvelard and Reseau1038
Table 1. Pathological characteristics of thyroid cancerReference Scoazec, Couvelard and Reseau1038
The primary treatment of most thyroid carcinomas is surgical, with radio-iodine therapy for many patients after total thyroidectomy. Advancements in targeted therapies now offers hope to even those with recurrent, radio-iodine resistant disease, and anaplastic cancers.
There have been several detailed guidelines published on the diagnosis and management of thyroid cancer.Reference Mitchell, Gandhi, Scott-Coombes and Perros1039 Three key ones are the guidelines for thyroid cancer management (2014) by the British Thyroid Association,Reference Perros, Boelaert, Colley, Evans, Evans and Gerrard Ba1040 the revised American Thyroid Association Management Guidelines (2016)Reference Haugen, Alexander, Bible, Doherty, Mandel and Nikiforov1041 and, most recently, the National Institute for Health and Care Excellence (NICE) Guideline NG230 (2022).1042 Given differences in presentation, pathophysiology and outcomes, separate guidelines exist for children with differentiated thyroid cancer.Reference Francis, Waguespack, Bauer, Angelos, Benvenga and Cerutti1043
Clinical presentation and referral
Patients generally present with a thyroid lump or mass, or lymph node metastasis. Many cancers are also detected as incidental findings on imaging performed for unrelated reasons.
Clinical features associated with an increased risk of malignancy in individuals with a thyroid nodule (and which should be referred urgently as suspected cancer) include:
• Stridor
• Presentation at the extremes of age (less than 18 years or 70 years or more)
• Rapid growth over a period of weeks
• Fixation to adjacent structures
• Voice change due to vocal fold paralysis
• Associated lymphadenopathy
• History of neck or upper body irradiation
• Strong family history of thyroid cancer
As the majority of thyroid nodules are benign, in the absence of the above ‘red flags’, urgent (suspected cancer) referral is not indicated.
Assessment
Recommendations
• Ultrasound scanning of the thyroid and cervical lymph nodes is mandated in the investigation of thyroid nodules and masses (evidence-based recommendation (R))
• Thyroid nodules should be classified using a recognised system such as the European Thyroid Association Thyroid Imaging Reporting And Data Systems (EU-TIRADS) or by ultrasound ‘U’ grading (R)
• Fine needle aspiration cytology should be carried out under ultrasound guidance (R)
• Fine needle aspiration cytology should be performed for all nodules with potentially suspicious ultrasound features (EU-TIRADS score of 4 or more; or ‘U’ grades of U3–U5) and for patients with risk factors for malignancy (R)
• Cytological analysis and categorisation should be reported according to the current Royal College of Pathologists guidance (R)
• Ultrasound assessment of cervical nodes should be performed in fine needle aspiration cytology (FNAC)-proven cancer (R)
• Contrast-enhanced cross-sectional imaging should be carried out in suspected cases of retrosternal extension, when there are bulky lateral nodal metastases or when there is suspicion of locally invasive disease (good practice point (G))
• For patients with Thy 3f or Thy 4 FNAC, a diagnostic hemithyroidectomy at least is recommended (R)
The key elements of investigation are as follows, not all of which may apply to any given patient:
• Palpation of the neck including assessment of potential retrosternal extension.
• Fibre-optic laryngoscopy if there is airway compromise or voice change (and pre-operative laryngoscopy if surgery is indicated – best practice).
• Blood tests: thyroid-stimulating hormone (TSH); auto-antibody status if thyroiditis is suspected and/or to assist with indeterminate FNAC; calcitonin – only in suspected cases of medullary thyroid cancer (e.g. family history, or after imaging or cytology); and bone profile including vitamin D (if surgery is contemplated).
• Ultrasound of the neck.
• Fine needle aspiration cytology under ultrasound guidance depending on ultrasound findings (see below).
• Computed tomography (CT) scan of the neck and thorax if retrosternal extension is suspected, or for selected cases of confirmed or suspected cancer (see below).
Ultrasound and fine needle aspiration cytology of thyroid nodules
The crucial diagnostic investigations for a thyroid nodule or mass are ultrasound and, when indicated, FNAC, under ultrasound guidance for optimal accuracy.Reference Yeh, Bauer, Bernet, Ferris, Loevner and Mandel1044 The 2014 British Thyroid Association guidelines for thyroid cancer management recommend the U grading system.Reference Perros, Boelaert, Colley, Evans, Evans and Gerrard Ba1040 However, the 2022 NICE guidance simply recommends using an established system for grading ultrasound appearance.1042 Ultrasound imaging should include assessment of cervical lymph nodes.
A description of the ‘U’ grading system can be found in the 2014 British Thyroid Association guidelines.Reference Perros, Boelaert, Colley, Evans, Evans and Gerrard Ba1040 The EU-TIRADS classification system is described by Russ et al.Reference Russ, Bonnema, Erdogan, Durante, Ngu and Leenhardt1045 Nodules graded as U2 or with an EU-TIRADS score of less than 4 can be considered benign and do not require FNAC in the absence of risk factors (i.e. family history, radiation exposure, fluoro-deoxy-glucose (FDG) avidity for incidental nodules, symptoms (e.g. growth) or signs (e.g. vocal fold palsy)). Such patients do not require further management for suspected or possible thyroid cancer.
Further management is then determined largely by cytopathology, which should be reported using the Royal College of Pathologists’ thyroid classification (Table 2).Reference Poller, Bongiovanni and Trimboli1046 When a repeat sample is required, the 2022 NICE guidelines advocate core needle biopsy as an alternative to repeat FNAC, particularly for Thy 3a lesions.1042 These guidelines also advocate considering further investigation of Thy 2 nodules (by repeat ultrasound scan, with or without core needle biopsy or FNAC) before discharge, although this is not widely adopted presently.
Table 2. Cytopathology classification of thyroid nodules
CNB = core needle biopsy; FNAC = fine needle aspiration cytology; MDT = multidisciplinary team
Fine needle aspiration cytology should also be carried out on any abnormal lymph nodes.
Thy 1 and Thy 2 can be further classified into Thy 1c and Thy 2c for cystic lesions. In a pure or simple cyst on ultrasound, a Thy 1c cytology may be considered definitive.
Immunohistochemistry on FNAC material and/or core biopsy may be helpful for confirmation of the type of suspected malignancy in the thyroid or in a lateral neck lymph node, particularly if metastasis to the thyroid is considered, or if there is suspicion of de-differentiated disease, anaplastic thyroid carcinoma or lymphoma. In some very limited circumstances, BRAF V600E molecular testing of FNAC or core biopsy material may be useful, to confirm a diagnosis of carcinoma on pre-operative FNAC or needle core biopsy. More extensive molecular testing of cytologically indeterminate thyroid nodules (e.g. Afirma®, Veracyte® or ThyroSeq®), although utilised overseas, has not become standard practice in the UK. Other proposed strategies for risk stratification include utilising the presence of nuclear atypia or ultrasound characteristics.
Computed tomography / axial imaging
Computed tomography of the neck and thorax should be carried out in cases with suspected retrosternal extension, tracheal compression or invasion. It can be considered for complete systemic staging and the evaluation of mediastinal disease for T3–4 tumours, for tumours with aggressive features, and in cases with bulky or multiple cervical lymph node metastases, with NICE guidance stating it can be considered for N1 disease.
Borderline neoplasms of the thyroid gland
A non-invasive follicular thyroid neoplasm with papillary-like nuclei, a follicular tumour of uncertain malignant potential and a well-differentiated tumour of uncertain malignant potentialReference Nikiforov, Seethala, Tallini, Baloch, Basolo and Thompson1047 cannot be reliably diagnosed pre-operatively, although a non-invasive follicular thyroid neoplasm with papillary-like nuclei may be suspected by ultrasound or FNAC findings. These lesions require meticulous pathological assessment, typically after diagnostic hemithyroidectomy. BRAF V600E mutation assessment may be helpful, as a non-invasive follicular thyroid neoplasm with papillary-like nuclei does not show BRAF V600E mutation. The UK incidence of a non-invasive follicular thyroid neoplasm with papillary-like nuclei is low (estimated as below 5 per cent of all newly diagnosed malignant thyroid lesions).Reference Johnson, Stephenson and Poller1048 These lesions are not regarded as malignant and need no further management after hemithyroidectomy. The need for and extent of follow up or surveillance varies and is not evidence-based.
Staging
The tumour, nodes and metastases (AJCC Cancer Staging Manual, eighth edition) staging systemReference Amin, Edge, Greene, Byrd, Brookland and Washington87 (Tables 3–6)Reference Tuttle, Haugen and Perrier1049,Reference Yang, Zhong, Huang, Lian and Fang1050 is used to stage thyroid cancers; this should be used in all cases prior to treatment, and may change based on intra-operative findings or the post-operative pathology. Post-operatively, ‘R’ is used to describe the adequacy of surgical margins (R1, microscopic residual disease; R2, macroscopic residual disease), although interpretation of this requires communication between pathologist and surgeon. The main changes in the eighth edition of the AJCC Cancer Staging Manual centre on downstaging more patients into lower stages than reflect their better prognosis. This includes the selection of age cut-off criterion to 55 years rather than 45 years.
Table 3. Tumour staging for differentiated and medullary thyroid cancer with relationship to overall disease stage*
* According to the AJCC Cancer Staging Manual (eighth edition).Reference Amin, Edge, Greene, Byrd, Brookland and Washington87 †All may be solitary (‘s’) or multifocal (‘m’) – the largest tumour determines the classification
Table 4. Nodal staging for differentiated and medullary thyroid cancer with relationship to overall disease stage*
* According to the AJCC Cancer Staging Manual (eighth edition).Reference Amin, Edge, Greene, Byrd, Brookland and Washington87
Table 5. Metastasis staging for differentiated and medullary thyroid cancer with relationship to overall disease stage*
* According to the AJCC Cancer Staging Manual (eighth edition).Reference Amin, Edge, Greene, Byrd, Brookland and Washington87
Table 6. Group staging for differentiated and medullary thyroid cancer with relationship to overall disease stage*
* According to the AJCC Cancer Staging Manual (eighth edition).Reference Amin, Edge, Greene, Byrd, Brookland and Washington87 ETE = extra-thyroidal extension; DSS = disease-specific survival rate; CSS = cancer-specific survival rate
Thyroid surgery
Surgeons performing operations for confirmed or suspected thyroid cancer should be core members of the thyroid cancer multidisciplinary team (MDT) and should perform a minimum of 20 thyroidectomies per year.Reference Chadwick, Kinsman and Walton1051 Complex surgery and lymph node surgery should be undertaken by nominated surgeons in the team with specific training in, and experience of, thyroid oncology.
Management of differentiated thyroid carcinoma
Microcarcinomas (T1a)
Microcarcinomas are differentiated thyroid carcinomas sized less than 10 mm in maximum dimension and are predominantly papillary carcinomas. When a microcarcinoma is found after thyroid lobectomy, if it is a solitary focus, and there are no high-risk features (e.g. multifocal disease, aggressive histological variant, family history), no further treatment is required. No follow up is required. When there are higher-risk features (or abnormalities in the contralateral lobe on ultrasonography), completion thyroidectomy can be considered. Management after completion thyroidectomy is the same as for papillary thyroid carcinoma, as described below.
For cases that are diagnosed through FNAC, active surveillance of these extremely low-risk tumours may be appropriate, with groups in JapanReference Ito, Miyauchi, Inoue, Fukushima, Kihara and Higashiyama1052 and the USAReference Tuttle, Fagin, Minkowitz, Wong, Roman and Patel1053 showing that this is safe in selected patients, with only around one-third of patients demonstrating progression during 10–20 years of follow up. At present, this has not been widely adopted in the UK, except for in high-risk surgical patients.
Initial thyroid surgery for papillary thyroid cancer
A strategy for the surgical treatment of papillary thyroid cancer is detailed in Table 7. All cases should be discussed pre-operatively at the thyroid cancer MDT. Previous recommendations considered all patients suitable for total thyroidectomy to render them able to receive radioactive iodine. However, a recognition that only a select group of patients with aggressive disease will benefit from radioactive iodine has resulted in an increased percentage of patients being considered for less than total thyroidectomy (i.e. thyroid lobectomy).
Table 7. Surgical options for papillary thyroid cancer in patients fit for treatment
TNM = tumour–node–metastasis; MDT = multidisciplinary team
Although rates of lymph node occult disease are relatively high, few patients ever develop clinically evident metastases. Prophylactic lateral neck dissection for a clinically staged node-negative (N0) neck is not indicated. Most clinicians now only consider a prophylactic central neck dissection if there is extensive local disease (tumour stage T3/4) or lateral nodal disease (N1b).
Radical locoregional surgery should be considered to provide long-term locoregional disease control in the metastasis stage M1 setting and to facilitate radio-iodine ablation of distant metastatic disease.
Initial surgery for follicular thyroid cancer and Hürthle cell cancer
The majority of patients undergoing surgery for follicular thyroid cancer will be undiagnosed at the time of the initial surgery (typically Thy 3f cytology). Frozen section histology cannot reliably differentiate benign follicular lesions from follicular thyroid cancer, and is not recommended. After diagnostic hemithyroidectomy, MDT discussion will inform whether completion thyroidectomy is indicated to facilitate radioactive iodine ablation.
Prophylactic level VI dissection is not indicated, with low rates of occult disease in follicular thyroid cancer.
An operative strategy for surgical treatment of follicular cancer is outlined in Table 8.
Table 8. Surgical options for follicular thyroid cancer in patients fit for treatment
TNM = tumour–node–metastasis; MDT = multidisciplinary team
Although data are conflicting, Hürthle cell cancers (follicular oncocytic) can be more aggressive, and many centres would generally treat these by total (or completion) thyroidectomy and radioactive iodine, although hemithyroidectomy can be considered in selected cases, along the same principles outlined in Table 8. Patients with tumours that exceed 4 cm but who show no signs of aggressive disease (pathologically staged T3a) have traditionally been offered total or completion thyroidectomy and radioactive iodine, but, if histologically non-aggressive, selected cases may be managed without the need for completion thyroidectomy and routine ablation.
Completion thyroidectomy
Many cases of differentiated thyroid cancer are diagnosed after initial hemithyroidectomy; when total thyroidectomy is recommended, as described above, completion surgery is necessary. In patients at higher risk of morbidity, e.g. with recurrent laryngeal nerve palsy before or after initial surgery, the benefits of completion surgery need to be weighed up against the risks.
Management of lymph nodes in differentiated thyroid cancer
Prophylactic level VI lymph node dissection is associated with a higher incidence of recurrent laryngeal nerve damage and long-term permanent hypoparathyroidism.Reference Salem, Bergenfelz, Nordenstrom and Almquist1054 It is therefore not routinely recommended, but it may be indicated in individuals with high-risk tumours.
Ipsilateral prophylactic central neck dissection is often recommended in patients with known involved lateral nodes. Therapeutic central neck dissection is recommended when the presence of lymph node metastasis is confirmed (usually as a per-operative finding).
Clinically involved lateral cervical lymph nodes should be managed by selective neck dissection (of at least levels IIa–Vb). Involvement of level I and Va nodes is rare in differentiated thyroid cancer and should only be dissected if involved. Prophylactic lateral compartment neck dissection for node-negative patients is not recommended.
Recommendations
• Total thyroidectomy is recommended for most patients with tumours greater than 4 cm in diameter, or tumours of any size with any of the following characteristics: multifocal disease, bilateral disease, extra-thyroidal spread (pathologically staged T3b and T4a disease), familial disease, and those with clinically or radiologically involved nodes and/or distant metastases (evidence-based recommendation (R))
• Subtotal thyroidectomy should not be used in the management of thyroid cancer (good practice point (G))
• Central compartment neck dissection is not recommended for patients without clinical or radiological evidence of locoregionally advanced disease (R)
• Patients with metastases in the lateral compartment should undergo therapeutic lateral compartment neck dissection, usually with ipsilateral central compartment neck dissection (R)
Locally advanced disease with extra-thyroidal spread
Where possible, locally advanced disease should be resected. Preservation of recurrent laryngeal nerves should be attempted in almost all cases, particularly where pre-operative function is preserved.Reference Shindo, Caruana, Kandil, McCaffrey, Orloff and Porterfield1055,Reference Fundakowski, Hales, Agrawal, Barczynski, Camacho and Hartl1056 For minimal involvement, organs should be preserved with ‘shave procedures’ where appropriate. Extensive resection of the trachea, larynx and oesophagus should be considered if potentially curative.Reference Wang, Nixon, Patel, Palmer, Tuttle and Shaha1057 Where disease is unresectable, or after R2 resection (macroscopic residual disease) of such cases, external beam radiotherapy, radioactive iodine, and systemic therapies such as tyrosine kinase inhibitors should be considered.
Post-operative management
After total thyroidectomy for differentiated thyroid cancer, patients should be commenced on TSH-suppressive (aiming for TSH of less than 0.1) doses of levothyroxine (approximately 2 μg/kg). Radio-iodine ablation is not required for low-risk patients; in such patients, TSH should be maintained in the low normal range.
Calcium and parathyroid hormone (PTH) levels should be routinely checked within 24 hours, and hypocalcaemia should be treated and monitored appropriately according to local guidelines.
Based on the British Association of Endocrine and Thyroid Surgeons’ position statement from 2010,Reference Palazzo1058 post-operative laryngoscopy is required as a key quality performance indicator and should be mandatory. As an alternative, a normal signal from intra-operative nerve monitoring at the end of the operation can be used as a surrogate marker of intact recurrent laryngeal nerve function in patients without post-operative voice changes.
Persistent voice dysfunction should be investigated, with referral to a specialised practitioner for assessment and speech therapy. Patients with long-term hypocalcaemia (hypoparathyroidism) should have their calcium, vitamin D and PTH levels regularly monitored, either in association with an endocrinologist or with their general practitioner.
Thyroglobulin levels should be checked no earlier than six weeks after surgery. Achieving unmeasurable thyroglobulin levels after total thyroidectomy for early disease could become a quality control parameter, but this is yet to be introduced into routine clinical practice.
Post-operative radio-iodine (iodine-131) ablation and external beam radiotherapy in differentiated thyroid cancer
All patients with thyroid cancer should be clinically re-staged in the MDT meeting and scored using one of the clinicopathological scoring systems (e.g. American Thyroid Association Management GuidelinesReference Haugen, Alexander, Bible, Doherty, Mandel and Nikiforov1041). This will aid planned follow up, and help identify high-risk patients and those who would benefit from radio-iodine therapy (Table 9).
Table 9. Indications for radioactive iodine ablation following total thyroidectomy or differentiated thyroid cancer
* Consider 3.7 GBq (rather than 1.1 GBq) for tumour–node–metastasis (TNM) stage T4N1bM1 or aggressive subtypes.
Randomised trials have shown that in low- and intermediate-risk differentiated thyroid cancer ablation patients, success and five-year recurrence rates with a lower dose of radioactive iodine ablation (1.1 GBq) are non-inferior to the previous standard dose of radioactive iodine ablation (3.7 GBq).Reference Dehbi, Mallick, Wadsley, Newbold, Harmer and Hackshaw1059
Patients should be prepared for iodine-131 (I131) by having a low-iodine diet for one to two weeks prior to treatment. Recombinant human TSH therapy prior to I131 is preferable to thyroid hormone withdrawal, and is preferred by patients as it avoids periods of hypothyroidism.
Pregnancy and breast feeding should be excluded prior to giving I131. A post-ablation whole body scan, should be performed after I131 when residual activity levels permit satisfactory imaging. Practically, this is generally 1–10 days following treatment. Following I131, TSH should be suppressed to less than 0.1 mIU/l (approximately 2 μg/kg).
Adjuvant external beam radiotherapy should only be considered in individual cases where there remains gross disease following surgery and I131.
Between 9 and 12 months following I131 for low-risk differentiated thyroid cancer, patients should undergo dynamic risk stratification (Table 10). Patients are then categorised as having either an excellent response, an indeterminate response or an incomplete biochemical or structural response.Reference Tuttle, Tala, Shah, Leboeuf, Ghossein and Gonen1060 The extent of TSH suppression is then accordingly modified.
Table 10. Post-treatment dynamic risk stratification and TSH suppression targetsReference Haugen, Alexander, Bible, Doherty, Mandel and Nikiforov1041
*After 12 months of treatment for differentiated thyroid cancer. †Thyroid-stimulating hormone (TSH) suppression targets for patients treated with complete resection, total thyroidectomy and radioactive iodine ablation. ‡Assumes the absence of interference in the thyroglobulin assay. Tg = thyroglobulin
Post-treatment follow up and surveillance
Thyroglobulin monitoring is most effective following total thyroidectomy with or without I131, and is an important modality in detecting persistent or recurrent disease. Physicians should be aware that thyroglobulin estimations vary according to the assay method. The presence of anti-thyroglobulin antibodies (in up to 30 per cent of cases) renders thyroglobulin levels unreliable. In this situation, serial anti-thyroglobulin antibody levels can be used cautiously as a surrogate tumour marker, along with ultrasound or other imaging.
Patients should have their thyroglobulin levels checked at 6–12 monthly intervals. Thyroglobulin evaluation is most effective following TSH stimulation, either by direct recombinant human TSH stimulation or by withdrawal of thyroid hormone replacement. However, super-sensitive modern assays may avoid the need for stimulation. There is no clear consensus as to when, how often and in which patients these more sensitive methods of thyroglobulin monitoring are required (i.e. when standard non-supressed thyroglobulin is normal).
A rise in thyroglobulin on serial monitoring may be suggestive of recurrent or residual disease, and needs further evaluation with imaging in the first instance.
A risk-adapted approach to follow up is recommended. For patients treated with surgery only without radioactive iodine ablation, those with T1a tumours can be offered discharge. For other surgically treated low-risk patients, ultrasound at 6–12 months and then annual follow up for 5 years can be offered (ultrasound and thyroglobulin in patients who have had total thyroidectomy).
For other patients who have had post-operative radioactive iodine ablation, there is no formalised standard of care to determine thyroid cancer follow up in terms of length of follow up and intensity (e.g. the repetition of dynamic risk stratification). A suggested guide appears in the 2022 NICE guidelines (Table 11).1042
Table 11. Risk-stratified follow up following total thyroidectomy with radioactive iodine ablation*
* Adapted from National Institute for Health and Care Excellence guidelines.1042 DRS = dynamic risk stratification; Tg = thyroglobulin
Recommendations
• A risk-adapted approach to follow up for differentiated thyroid cancer is recommended (good practice point (G))
• Follow up should be based on serum thyroglobulin, thyroglobulin antibody and TSH levels, with an ultrasound scan where indicated (evidence-based recommendation (R))
Management of recurrent differentiated thyroid cancer
Recurrence in differentiated thyroid cancer is a complex subject; it can be local, regional, distant or solely biochemical (thyroglobulin). A rise in thyroglobulin is now the most common mode of presentation. When this occurs, neck ultrasound is the first investigation, with or without FNAC of any identifiable recurrence. If this shows recurrence, axial neck imaging may be considered for lesions in proximity to central neck structures.
If there is no evidence of a recurrence in the neck, CT of the thorax or FDG positron emission tomography (PET)-CT can be considered.
The aim of treatment for locoregional recurrence is to surgically clear all macroscopic disease, if possible weighing up the benefits and risks. Small-volume recurrent lymph nodes that are distant from vital structures may be monitored in the first instance. In the setting of residual disease after surgery for recurrence, for unresectable locoregional disease or for distant metastases, further radio-iodine therapy can be given, and the prognosis can still be favourable as long as the tumour remains radioactive iodine avid.
A small percentage of recurrent disease will become de-differentiated and refractory to radioactive iodine, at which point the 10-year survival can drop significantly. In these cases, local treatment with external beam radiotherapy, including newer forms (e.g. stereotactic radiotherapy, surgery for bone disease, embolisation, radio frequency ablation, alcohol injection, vertebroplasty and so on), may be considered for local control and palliation of symptoms.
In specialist units, selected patients with progressive symptomatic metastatic disease that is refractory to radioactive iodine are treated with kinase inhibitors such sorafenib, lenvatinib, larotrectinib and entrectinib within clinical trials if appropriate. These drugs can offer benefit in progression-free survival not overall survival, but patients need monitoring for side effects and deterioration in quality of life.Reference Urken, Milas, Randolph, Tufano, Bergman and Bernet1061,Reference Scharpf, Tuttle, Wong, Ridge, Smith and Hartl1062
Recommendations
• Patients with iodine-refractory disease should be referred to a centre with experience of managing these patients within clinical trials (good practice point (G))
Medullary thyroid cancer
Diagnosis and assessment
Medullary thyroid cancer is rare (approximately 1–3 per cent of all thyroid cancer cases).1063 The same investigations as for any patient with a thyroid mass are required. Fine needle aspiration cytology may be diagnostic or suspicious for medullary thyroid cancer, and ultrasound imaging may also show features associated with medullary thyroid cancer. When medullary thyroid cancer is suspected or confirmed, serum calcitonin and carcinoembryonic antigen should be measured. This can assist with diagnosis when medullary thyroid cancer is suspected and measurements show a pre-treatment baseline level. Calcitonin levels are also predictive of the disease load or extent, and prognosis. In addition to an ultrasound scan of the neck, CT or magnetic resonance imaging (MRI) may be indicated, to guide the extent of surgical resection at initial surgery. Calcitonin gene product can cause systemic symptoms. Enquiry about symptoms such as flushing, diarrhoea or irritable bowel symptoms should be made.
All cases should be referred for surgical treatment to a regional designated cancer centre of the Thyroid Cancer Network. Twenty-five per cent of medullary thyroid cancer cases are familial (MEN2A, MEN2B and familial non-MEN medullary thyroid cancer). A family history must be ascertained, as should urinary or serum metanephrine levels to exclude phaeochromocytoma, and PTH and calcium levels to exclude hyperparathyroidism. If a phaeochromocytoma is diagnosed, adrenalectomy should be performed prior to offering thyroid surgery for medullary thyroid cancer.
Staging
The tumour–node–metastasis (TNM) staging for medullary thyroid cancer follows the same criteria as for differentiated thyroid cancer (Table 12); however, the stage grouping differs.
Table 12. Disease staging for medullary thyroid carcinoma
Treatment
All patients with medullary thyroid cancer should undergo:
• Total thyroidectomy and central compartment node clearance (level VI). This should be performed even in the presence of disseminated metastases to aim for disease control in the central neck.
Lateral neck lymph nodes (clinically staged node-positive (N+)):
• Patients with clinically involved lateral compartment nodes should have a therapeutic lateral neck dissection (levels IIa–Vb) to eradicate regional disease on that side.
Lateral neck lymph nodes (clinically staged N0):
• Prophylactic ipsilateral lateral neck dissection (levels IIa–Vb) should be considered in patients with involved ipsilateral central neck nodes and in those with high calcitonin levels, although exact cut-offs are debatable.
• Prophylactic contralateral lateral neck dissection may even be considered in the presence of ipsilateral N1b disease and calcitonin levels over 200 pg/ml. It may also be considered in cases where a biochemical cure is not achieved after total thyroidectomy and ipsilateral neck surgery.
Mediastinal disease:
• Intra-thoracic disease should be resected, and, when below the level of the brachiocephalic vein, be resected via sternotomy or using robotic surgery where feasible.
Post-operatively, patients with medullary thyroid cancer should aim to have TSH levels in the normal range (i.e. no need for suppression).
Post-operative radiotherapy:
• Patients with macroscopic residual disease, gross extra-thyroidal extension or extensive lymph node metastases should be considered for external beam radiation.
Calcitonin and carcinoembryonic antigen should be measured at two to six weeks and at three to six months after the operation to determine the exact nadir. Those patients with persistent elevation of calcitonin should have further radiological assessment to determine disease persistence. Computed tomography, MRI, ultrasound, selective arteriography, I131-metaiodobenzylguanidine, FDG-PET, indium-111 (In111)-octreotide and DOTA-TATE PET/CT may all be useful in identifying the source of raised calcitonin, but their use in patients with calcitonin levels of less than 400–500 pg/ml is unlikely to identify metastases.
Genetic assessment
Genetic assessment for all patients, including germline RET proto-oncogene mutational analysis, should be performed after surgery, once diagnosis is established, even in the absence of a familial history.
Prophylactic thyroidectomy should be offered to RET-positive family members following specialist counselling by the cancer genetics team. The timing and extent of surgery are dependent on genotype (codon mutation), the calcitonin level and age at detection of RET positivity.
Lifelong follow up is recommended for all patients with medullary thyroid cancer. Screening should include calcitonin and carcinoembryonic antigen. Rising calcitonin levels should trigger investigations to identify potentially treatable metastatic disease.
Patients with recurrent medullary thyroid cancer present a complex problem; they should be managed by teams with experience of such cases, and with access to the full spectrum of surgical and non-surgical approaches that this cohort requires. The primary aim should always be to control locoregional disease.
Calcitonin and carcinoembryonic antigen doubling times are helpful markers of disease aggressiveness and prognosis. A calcitonin doubling time exceeding 6 months is associated with 5-year and 10-year survival rates in excess of 90 per cent and 35 per cent, respectively; shorter doubling times predict markedly worse survival.
In patients with a significant tumour burden and those who are symptomatic of progressive metastatic disease, multikinase inhibitor therapy should be considered. There are two drugs licensed for use in progressive locally advanced and/or metastatic medullary thyroid cancer: vandetanibReference Wells, Robinson, Gagel, Dralle, Fagin and Santoro1064 and cabozantinib.Reference Elisei, Schlumberger, Muller, Schoffski, Brose and Shah1065 Both these drugs were shown to improve progression-free survival over placebo. As with multikinase inhibitors in iodine-refractory thyroid cancer, these drugs have potentially significant side effects, and patients should be managed within a specialist unit. Recent promising data for the specific RET inhibitors selpercatinibReference Wirth, Sherman, Robinson, Solomon, Kang and Lorch1066 and pralsetinibReference Wirth, Sherman, Robinson, Solomon, Kang and Lorch1066 may lead to the licensing and availability of these drugs as a treatment option for patients with tumours harbouring a somatic RET mutation, or those patients with a germline RET mutation.
Supportive care may provide significant symptomatic relief for patients with advanced medullary thyroid cancer. For example, analgesia, anti-diarrheal medication, bisphosphonate or denosumab infusions for extensive bone involvement can all be useful. Patients with isolated metastases or those with differential progression should be considered for locally ablative treatment modalities such as surgical resection, external beam radiotherapy including stereotactic body radiation therapy, radiofrequency ablation, or embolisation.
Recommendations
• All patients with suspected or proven medullary thyroid cancer should have assessments of: serum calcitonin and carcinoembryonic antigen levels, catecholamine and normetanephrine urine estimation (or plasma-free normetanephrine estimation) at 24 hours, serum calcium levels, and PTH levels (evidence-based recommendation (R))
• Routine ultrasound should be performed in all cases, and CT is advised in the presence of advanced local or any nodal disease, to guide the extent of surgery (R)
• All patients with proven medullary thyroid cancer measuring over 5 mm should undergo total thyroidectomy and central compartment neck dissection (R)
• Patients with lateral nodal involvement should undergo selective neck dissection (IIa–Vb) (R)
• All patients with proven medullary thyroid cancer should have genetic screening for germline RET mutation (R)
• All patients with sporadic medullary thyroid cancer should have their histopathology examined for a somatic RET mutation (R)
• Prophylactic thyroidectomy should be offered to RET-positive family members (R)
• Neither FDG-PET/CT nor F-DOPA (fluorodopa) PET/CT is recommended to detect the presence of distant metastases at presentation (good practice point (G))
• Patients with locally advanced or metastatic disease that is progressing and/or symptomatic should be considered for entry into clinical trials, for locally ablative palliative treatments or for systemic targeted therapy (G)
Anaplastic thyroid cancer
The prognosis of patients with anaplastic thyroid cancer is poor, and all patients are considered stage IV at presentation (Table 13).Reference Bible, Kebebew, Brierley, Brito, Cabanillas and Clark1067
Table 13. Disease staging for anaplastic thyroid cancer
* All are considered stage IV
Many patients present with a history of a rapidly enlarging thyroid mass in a long-standing goitre. As soon as a diagnosis of anaplastic thyroid cancer is considered, investigations should be expedited to enable treatment initiation within days not weeks. Core biopsy is preferable to FNAC in achieving a diagnosis if possible, as it will help differentiate anaplastic thyroid cancer from thyroid lymphoma, which can present in a similar manner. Tissue should be sent for urgent immunohistochemistry for BRAF V600E, and next-generation sequencing for BRAF V600E mutation and NTRK, ALK and RET fusions.
If anaplastic thyroid cancer is considered resectable, patients should undergo total thyroidectomy, and be considered for post-operative radiation with or without chemotherapy on an individual case basis. Unfortunately, many patients will present with extremely advanced disease, a poor performance status and extremely limited life expectancy. Such patients require a swift diagnosis and best supportive care. In this situation, tracheostomy should be avoided as it is unlikely to result in a meaningful improvement in overall quality of life.
Select patients who present with advanced disease and who retain a good performance status may be eligible for targeted therapies.
Up to 45 per cent of anaplastic thyroid cancer patients will harbour a BRAF V600E mutation; the combination of a BRAF inhibitor (dabrafenib) and a MEK inhibitor (trametinib) has shown encouraging activity, with a response rate of 61 per cent, partial response rate of 67 per cent and complete response rate of 44 per cent in 23 patients, with 64 per cent maintaining a response for six months or more.Reference Subbiah, Kreitman, Wainberg, Cho, Schellens and Soria1068 These data were presented to the Food and Drug Administration, which led to the approval of this drug combination for this specific group of patients in the USA. It is currently being evaluated in the UK.
In suitable patients, NTRK or RET fusion anaplastic thyroid cancer patients with stage IVC disease, a TRK inhibitor (either larotrectinib or entrectinib) or RET inhibitor (selpercatinib or pralsetinib), preferably inside a clinical trial, if available, may be considered.
In stage IVC anaplastic thyroid cancer, patients with high programmed death-ligand 1 (PD-L1) expression, checkpoint (PD-L1, programmed death-1 (PD1)) inhibitors can be considered in the context of a clinical trial.
In patients without a druggable target or appropriate clinical trial, best supportive care should be pursued, which may include palliative external beam radiotherapy. Cytotoxic chemotherapy has poor response rates, but may be considered in the palliative setting where there are no targeted therapy options and the patient retains a good performance status.
Recommendations
• Initial assessment should focus on identifying the small proportion of patients with localised disease and a good performance status, who may benefit from surgical resection and other adjuvant therapies (good practice point (G))
• Investigations should be expedited to enable the initiation of a treatment plan within days (G)
• The surgical intent, when appropriate, should be gross tumour resection and not merely an attempt at debulking (G)
• Pathology should be sent for molecular profiling, specifically for BRAF V600E mutation, NTRK, ALK and RET fusions (G)
• Targeted therapies should be considered if druggable genomic alteration is present (G)
• If no targeted therapy is appropriate, the patient should be given best supportive care, with consideration of palliative radiotherapy for symptomatic relief (G)
Studies due to report
‘IoN’ trialReference Mallick, Harmer, Hackshaw, Moss and Io1069 – this national multicentre randomised, controlled trial has been designed to address the question of whether the omission of radioactive iodine is non-inferior to its use in low to intermediate risk differentiated thyroid cancer.
‘HoT’ trial1070 – a UK national multicentre randomised, controlled trial designed to address the question of whether lobectomy is non-inferior to total thyroidectomy in selected cases of low-risk differentiated thyroid cancer.
The ‘NIFTy’ trial1071 – this trial seeks to evaluate the role of near-infrared fluorescent imaging in reducing the rate of parathyroid gland injury in thyroid surgery, which may help minimise hypoparathyroidism in patients treated for thyroid cancer.
‘Thy3000’ – a national observational study of the epidemiology and initial management pathway of thyroid nodules. This is a retrospective study of 3000 thyroid nodules in the UK that hopes to increase the understanding of epidemiology of thyroid nodules and the nature of practice across the UK.
Important research questions to be answered
The excellent outcomes associated with differentiated thyroid cancer have made it a challenging subject to study. Large patient numbers and long follow-up periods are required to perform definitive randomised, controlled trials. Therefore, many subjects remain unresolved, including: the extent of surgery and adjuvant therapy required for lower-risk differentiated thyroid cancer, and the duration and frequency of follow up required and how this is influenced by initial treatments (lobectomy vs total thyroidectomy and radioactive iodine ablation, for example).
Although great strides have been made in recent times in terms of treatment for de-differentiated thyroid cancers such as anaplastic thyroid cancers and radio-iodine refractory recurrent cancers, outcomes remain poor. Further work to optimise outcomes in these high-risk groups is required.
The application of risk-stratification strategies for indeterminate cytology (Thy 3) also requires consideration. Such strategies include molecular testing, the presence of nuclear atypia and ultrasound characteristics.
Research in medullary thyroid cancer is limited by the rarity of the disease. The extent of initial treatment including the need for lateral neck dissection, the role of calcitonin in determining initial management, and the targeted treatment of progressive recurrent or distant disease are all important topics for the future.
Introduction
Most head and neck cancers have a propensity to metastasise to regional lymph nodes. The presence, site and size of metastatic neck disease are important prognostic factors in head and neck squamous cell cancer (SCC) (generally, survival in non-human papillomavirus (HPV) associated head and neck SCC decreases by up to 50 per cent with lymph node metastasis, and by the same again with extra-nodal extension). Cancers can present with lymph node metastases. Investigation and treatment of macroscopic and microscopic lymph node disease is applicable to all head and neck cancer sites, and can be treated surgically or non-surgically.
Consequently, the management of lymph node metastases features throughout these guidelines. The assessment of lymph node disease through radiology and cytopathology is covered in Chapter 2 (on clinical assessment and diagnosis). Radiotherapy (RT) and chemoradiotherapy treatment for neck disease is covered in Chapter 4 (on non-surgical oncology). Investigation and management of patients presenting with SCC with unknown primary is discussed in Chapter 27. The management of primary site-specific lymph node disease, both for clinically negative and positive disease, is discussed in each site-specific chapter.
This chapter will pull these areas together, and expand on areas of neck metastatic disease not covered elsewhere.
Assessment of metastatic lymph node disease
For primary diagnosis of cancer
See Chapter 2. Investigations for diagnosis and staging are summarised in Figure 1. For malignant disease other than SCC, further investigations will depend on cytopathology or histopathology. Positron emission tomography computed tomography (PET-CT) should be organised as soon as carcinoma with unknown primary is suspected.
Figure 1. Diagnosis and staging for suspected lymph node metastatic disease. USS = ultrasound scan; FNAC = fine needle aspiration cytology; SCC = squamous cell carcinoma; CUP = carcinoma with unknown primary; PET-CT = positron emission tomography computed tomography; MRI = magnetic resonance imaging; CT = computed tomography
For staging of head and neck cancer
Recommendations
• Magnetic resonance imaging (MRI) or computed tomography (CT) should be used for the clinical staging of neck metastases (evidence-based recommendation (R))
• Positron emission tomography CT should be used in nodal stage N3 disease, and to evaluate the response to chemoradiotherapy (in addition to other indications specific to the primary disease or histology type) (R)
• Positron emission tomography CT should be considered in recurrent disease with lymph node metastases (good practice point (G))
Appropriate staging of neck disease may have been performed as part of the assessment (see above). The choice of cross-sectional imaging (CT and/or MRI) depends upon the imaging preference for primary tumour staging. Computed tomography and MRI show a similar performance in the nodal staging of head and neck SCC.Reference Sun, Li, Li, Li, Su and Gao1072 The presence of regional nodes has been shown to be the strongest independent predictor for distant metastases,Reference van der Kamp, Muntinghe, Iepsma, Plaat, van der Laan and Algassab1073 and, as mentioned above, PET-CT is recommended in N3 stage disease.1074 For this and other standard indications (e.g. tumour stage T4 disease of the hypopharynx or nasopharynx, carcinoma with unknown primary, stage 3 melanoma, and other circumstances with high chance of distant metastasis), PET-CT may increase the accuracy of nodal staging, in addition to MRI or CT, in the setting of a clinically node-negative neck.Reference Lowe, Duan, Subramaniam, Sicks, Romanoff and Bartel1075 Positron emission tomography CT is used to detect persistent disease after chemoradiotherapy and often in suspected recurrent disease (see relevant sections in this chapter).
Lymph node levels
The levels and sublevels of the neck were described by Robbins et al. in 2002, updated in 2008, together with an anatomical description for each one (see Figure 2).Reference Robbins, Shaha, Medina, Califano, Wolf and Ferlito1076
Figure 2. The levels and sublevels of the neck. (a) The anterior or medial border of levels III and IV is formed by the lateral border of the sternohyoid muscle. Key anatomical landmarks dividing levels II, III and IV are the inferior border of the hyoid bone and the inferior border of the cricoid cartilage. (b) The central compartment is further divided in thyroid cancer into (1) pre-laryngeal, (2) pre-tracheal and (3) para-tracheal, with the inferior border of the central compartment defined as the innominate artery on the right and the corresponding axial plane on the left, as inferior lymphatic drainage is contiguous with the anterior superior mediastinum.
Additional regional lymph node sites
The incidence of retropharyngeal lymph node metastases in head and neck SCC is 4–44 per cent, with the highest incidence in nasopharyngeal cancer where the retropharyngeal lymph nodes are the first echelon nodes. In oropharyngeal SCC, the incidence of retropharyngeal lymph nodes is greater in tumours involving the posterior wall, soft palate and contralateral neck. Retropharyngeal lymph node metastases are associated with a poorer prognosis and impact upon therapeutic management.Reference Iyizoba-Ebozue, Murray, Arunsingh, Dyker, Vaidyanathan and Scarsbrook1077 Positron emission tomography CT in combination with CT and MRI improves overall accuracy in detecting metastatic retropharyngeal lymph nodes, and is recommended in equivocal cases.Reference Kim, Choi, Lee, Lee, Park and Yoon1078
Intra-parotid lymph nodes may be first echelon nodes in cutaneous head and neck cancers and temporal bone SCC, and can occasionally be involved when there is cervical lymph node metastasis from mucosal head and neck SCC.Reference Batsakis1079
Buccal, retroauricular and occipital lymph nodes may be involved in cutaneous cancers.
Staging
There were three main areas of change to regional lymph node staging in the eighth edition of the AJCC Cancer Staging Manual.Reference Amin, Edge, Greene, Byrd, Brookland and Washington87
Extra-nodal extension
The presence of extra-nodal extension is a poor prognostic factor. All HPV-negative tumours from any subsite (excluding nasopharyngeal carcinoma) incorporate extra-nodal extension in the nodal staging. For clinical staging, extra-nodal extension should be unequivocal, because the sensitivity of early radiological extra-nodal extension (e.g. indistinct nodal margin) is modest.Reference Park, Guenette, Suh, Hanna, Chung and Baek1080
If there is evidence of pathological extra-nodal extension in a single node sized 3 cm or less, this is upstaged to N2a, and in all other instances to N3b. Pathological extra-nodal extension may be further subdivided into minor or major categories if the metastasis has extended less than 2 mm or more than 2 mm beyond the lymph node capsule respectively, although this subdivision is for documentation purposes only.Reference Amin, Edge, Greene, Byrd, Brookland and Washington87
P16-positive squamous cell carcinoma
In addition, a different staging is adopted for p16-positive oropharyngeal and SCC with unknown primary, based on the better prognosis of HPV-associated oropharyngeal SCC.
It should also be noted that there are differences in the clinico-radiological staging of HPV-related neck disease, and pathological (but post neck dissection only) staging. This is based on the prognostic implications of a neck dissection having been carried out.
Different staging remains in place for the thyroid, nasopharynx and melanoma. The staging for these can be found in the relevant chapters. The main staging (for p16-negative and p16-positive disease) is shown in Tables 1 and 2.
Table 1. Main head and neck cancer nodal (N) staging (including a p16-negative unknown primary and oropharynx)*
* According to the AJCC Cancer Staging Manual (eighth edition).Reference Amin, Edge, Greene, Byrd, Brookland and Washington87
Table 2. Nodal (N) staging for a p16-positive unknown primary (SCC with unknown primary) and oropharynx*
* According to the AJCC Cancer Staging Manual (eighth edition).Reference Amin, Edge, Greene, Byrd, Brookland and Washington87 SCC = squamous cell carcinoma; n/a = not applicable
Nasopharynx
In order to ensure clearer delineation and concordance with imaging, the caudal border of cricoid cartilage is used to define lymph node metastases in low neck levels IV and Vb, requiring upstaging to N3 disease. Nodal masses larger than 6 cm are also classed as N3, with removal of the prior separate N3a and N3b group classification. Nodal stage N3 is seated in the prognostic stage grouping of IVA, rather than IVB.
Classification of neck dissection
Recommendations
• Neck dissection should be classified according to levels or sublevels dissected and the non-lymphatic structures removed, with or without additional nodal groups being dissected (evidence-based recommendation (R))
• There should be an agreed local protocol for neck dissection specimen orientation, to facilitate accurate pathological lymph node staging (good practice point (G))
Historic neck dissection classification can be misleading. The American Academy of Otolaryngology and Head and Neck classification, described in 1991 and updated in 2002, still contains the terms ‘radical’ and ‘modified radical’ neck dissection.Reference Robbins, Shaha, Medina, Califano, Wolf and Ferlito1076 These terms imply the dissection of all levels of the neck. In fact, a full dissection of all levels and sublevels is unusual.
The classification for the commonly surgically removed cervical lymph node groups in the neck is well established and commonly understood.Reference Robbins, Shaha, Medina, Califano, Wolf and Ferlito1076 It is recommended that any type of neck dissection is denoted by ‘ND’, followed by node level or sublevel and non-lymphatic structures removed.Reference Ferlito, Robbins, Shah, Medina, Silver and Al-Tamimi1081 The classification of neck dissection is as shown in Table 3.
Table 3. Classification of neck dissection
* As defined in the AJCC Cancer Staging Manual (eighth edition).Reference Amin, Edge, Greene, Byrd, Brookland and Washington87 IJV = internal jugular vein
There is a lack of clear guidance on distinguishing neck dissection levels for residual or persistent disease after initial RT or chemoradiotherapy from truly recurrent disease. The definition of the latter varies within reports, ranging from six months to two years after primary treatment. This is in contrast to the pathological distinction, in which the ‘y’ prefix indicates those cases that are performed following initial multimodality therapy (neoadjuvant chemotherapy and/or RT). The commonest clinical scenario for neck dissection for residual or persistent disease is when it follows routine surveillance PET-CT, or there is MRI evidence of persistent disease, typically three months after primary chemoradiotherapy. In this setting, neck dissection should be considered as part of the primary multimodality treatment.
There is no consensus regarding a minimum nodal yield from an untreated neck. A minimum nodal yield of 18 lymph nodes is supported by the prognostic therapeutic effect of neck dissection.Reference Divi, Harris, Harari, Cooper, McHugh and Bell1082 The tumour–node–metastasis (TNM) stage refers to 10 and 20 nodes for ‘selective’ and ‘radical’ neck dissections, respectively. The yield from a previously irradiated neck would be expected to be smaller.
In order to facilitate accurate pathological lymph node staging, and to plan post-operative RT volumes and dose, there should be an agreed local protocol for neck dissection specimen orientation. The options are to divide according to neck levels in the operating theatre and send in separate containers, or the orientation on a suitable base and labelling of neck levels with an indelible marking pen.
Treatment of clinically staged node-negative neck
Recommendations
• Elective neck treatment should be offered if the risk of occult metastasis is greater than 20 per cent (evidence-based recommendation (R))
• Surgery and RT are equally effective (R)
• Elective treatment should be given by the same modality used for the primary tumour (R)
• Sentinel node biopsy should be considered for oral cavity SCC (good practice point (G))
A proportion of patients staged as clinically staged node-negative (N0) will have occult metastases. These are not detectable on clinical or radiological assessment, and they include micro-metastases (defined strictly as less than 2 mm). The two strategies in this group are to treat electively, or to observe and then treat, any neck recurrence in the proportion that develop subsequent metastasis. It is now generally accepted that elective neck treatment, rather than observation, should be considered when the risk of occult (undetectable) metastasis is greater than 20 per cent.Reference Weiss, Harrison and Isaacs1083
This is essentially an historic threshold figure. Given the lower morbidity of modern RT techniques and selective neck dissection surgical techniques, there is an argument that the cut-off value could be lower (e.g. 15 per cent). In their landmark clinical trial, D'Cruz et al. showed a survival benefit in elective neck dissection (vs initial surveillance) for N0 oral SCC.Reference D'Cruz, Vaish, Kapre, Dandekar, Gupta and Hawaldar717 However, the pathological metastasis rate was just under 30 per cent in this patient group.
The following are the basic principles that underpin treatment for the clinically node-negative neck:
• If the risk of occult metastasis is greater than 20 per cent, elective neck treatment should be offered. This effectively applies to all mucosal head and neck SCC, with the exception of a T1/2 glottic larynx and T1/2 sinonasal and lip SCC.
• For non-SCC head and neck cancers, the requirement for elective neck treatment is more nuanced and dependent on histological type. In general, low-grade or low-stage cancers may not need elective neck treatment.
• Surgery and RT are thought to be equally effective.Reference Koyfman, Ismaila, Crook, D'Cruz, Rodriguez and Sher1084 Radiotherapy treats retropharyngeal lymph nodes; surgery generally does not.
• Elective treatment should be given by the same modality used for the primary tumour.
• The same principle applies to the contralateral neck – i.e. elective treatment if the chance of occult metastasis is significant.
• For midline tumours, both sides of the neck need to be considered.
The levels or sublevels of the neck that are at risk are determined by the site of the primary tumour. The choice of levels to be treated is explained in the relevant chapters of these guidelines. Surgical dissection of level IIb is associated with higher rates of shoulder dysfunction.Reference Dziegielewski, McNeely, Ashworth, O'Connell, Barber and Courneya1085 There is a high level of evidence to support the omission of level IIb dissection in laryngeal, hypopharyngeal and T1/2 oropharynx primary cancer sites, in cases with no intra-operative findings suggestive of metastatic lymph nodes elsewhere in the neck.Reference Gross, Olsen, Lewis, Kasperbauer, Moore and Olsen1086,Reference Gross, Olsen, Lewis, Kasperbauer, Moore and Olsen1087 The evidence for the omission of level IIb is less convincing for clinically staged N0 oropharyngeal primary cancers with more advanced primary tumours (T3/4).Reference Gross, Olsen, Lewis, Kasperbauer, Moore and Olsen1087 Whilst oral cancers have higher rates of occult metastases than other mucosal primary sites, the rate of occult IIb metastases in clinically staged N0 necks is very low (1 per cent in a meta-analysis of 937 patients), which suggests that level IIb dissection can be omitted in this group also.Reference Ferreli, Festa, Costantino, Malvezzi, Colombo and Spriano1088
For oral cavity cancers in particular, sentinel node biopsy is an alternative to elective neck treatment (see Chapter 17).
Table 4 summarises the neck levels that should be detected for each primary cancer site.
Table 4. Neck level treatment for a clinically node-negative (N0) neck according to primary tumour site
* High-grade histology or tumour (T) stage T3/4. ‘X’ = treatment recommended, evidence-based and/or generally accepted practice; ‘±’ = consider treatment / variable treatment practice; ‘0’ = treatment not necessary
Treatment of clinically staged node-positive neck
Recommendations
• The initial modality used should be the same as for the primary cancer (evidence-based recommendation (R))
• Radiotherapy surgery can be used as a single modality treatment for a single lymph node metastasis sized less than 3 cm (R)
• After neck dissection for SCC, RT or chemoradiotherapy should be considered for patients with adverse pathological features and/or N2+ disease (R)
• After neck dissection for SCC, chemoradiotherapy should be considered for eligible patients with extra-nodal extension and a high chance of residual disease (R)
• When primary chemoradiotherapy or RT is used for cancers with node-positive (N+) neck disease, evaluation of response by PET-CT scan is recommended at three to four months post-treatment (R)
• After post-chemoradiotherapy PET-CT, patients with an incomplete nodal response should be considered for neck dissection (R). (Note that the extent of neck dissection is variable)
• After post-chemoradiotherapy PET-CT, patients with N1–2 HPV-positive cancers and an initial equivocal nodal response should be offered another PET-CT scan in a further three months (R)
A clinically staged N+ neck requires treatment, alongside treatment of the primary cancer.
The modality used is generally the same as for the primary cancer which, in turn, is determined by both the primary and neck staging. When RT is the primary treatment modality, the international consensus guidelines for delineating nodal levels in the neck and selecting lymph node target volume for intensity-modulated RT and volumetric modulated arc therapy, are recommended.Reference Biau, Lapeyre, Troussier, Budach, Giralt and Grau146
The historic principle was that ‘comprehensive’ treatment (equating to all lymph node levels) was required for the N+ neck. However, contemporary practice has moved towards a more selective approach when appropriate.
Generally, the greater the burden of neck disease, the more comprehensive the neck treatment should be. Treatment is also dependent on the site of the primary tumour and clearly the levels shown to be involved on imaging. The nodal groups treated for a small single N1 lymph node metastasis might not differ greatly from those for a N0 neck of the same primary cancer. On the other hand, advanced neck disease (N3), or multiple level involvement in an N2b/c neck, if treated surgically, will generally require dissection of all levels and selected non-lymphatic structures.
Therefore, the choice of lymph node levels is nuanced, and it is almost impossible to arrive at an evidence-based guideline specifying which levels need to be treated and which do not. The described lymph node levels for clinically staged N0 disease (Table 4) can be taken as a minimal or starting point when considering which levels require treatment. Furthermore, image-guided RT utilising fluoro-deoxy-glucose (FDG)-PET/CT based nodal target volume can be used to tailor the dose prescription to different areas of the neck.Reference van den Bosch, Doornaert, Dijkema, Zwijnenburg, Verhoef and Hoeben1089 This has been shown to improve the regional control of disease and survival compared to CT, and supports future potential target volume transformation in therapeutic and elective neck treatment by RT.
For non-surgical primary treatment, synchronous chemoradiotherapy should be given for all suitable patients.Reference Lacas, Carmel, Landais, Wong, Licitra and Tobias109
The following principles apply, however:
• The initial modality used is generally the same as for the primary cancer
• Radiotherapy or surgery can be used as a single modality treatment for single nodal disease sized less than 3 cm without adverse pathological features*
• The minimum neck levels or sublevels to be treated are as for clinically staged N0 disease relevant to the primary site, as well as any level with radiological suspicion of lymph node metastasis (careful pre-operative imaging is essential)
• Radiotherapy or chemoradiotherapy can follow primary surgery, and can be tailored to pathological staging and other histopathological features (see below)
• Surgery can follow primary RT or chemoradiotherapy when there is evidence of possible residual disease (see below)
*Note that this is not the same as clinically staged N1, which, for HPV-positive oropharyngeal SCC, can include multiple nodes measuring up to 6 cm.
Primary neck dissection when the primary tumour is to be treated with RT or chemoradiotherapy, even in cases of resectable N3 disease, is not routinely recommended, but should be considered on a case-by-case basis.Reference Carsuzaa, Gorphe, Vergez, Malard, Fakhry and Righini1090,Reference Boros, Blanchard, Dade, Gorphe, Breuskin and Even1091 One indication may be cases where there is skin ulceration and possible haemorrhage risk.
The definition of resectable N3 disease may vary. Relative contraindications include skull base erosion by metastatic neck disease, prevertebral fascia invasion, and gross carotid involvement. These cases can technically be resected, but will have a low likelihood of cure. In patients with very advanced, bordering unresectable nodal disease, RT or chemoradiotherapy (possibly induction chemotherapy) may be better initial options.Reference Koyfman, Ismaila, Crook, D'Cruz, Rodriguez and Sher1084,Reference Du, Blanchard, Even, Boros, Gorphe and Breuskin1092
Adjuvant treatment after neck dissection
When neck dissection is performed, and there is proven lymph node metastasis, post-operative RT or chemoradiotherapy can be considered. Much of the evidence around the indication for post-operative RT is extrapolated from the oral cavity.Reference Koyfman, Ismaila, Crook, D'Cruz, Rodriguez and Sher1084 The addition of chemotherapy in the presence of extra-nodal extension and/or involved surgical margins is associated with improved locoregional control and overall survival, based on randomised, controlled trials from the European Organisation for Research and Treatment of Cancer trial ‘EORTC 22931’ and Radiation Therapy Oncology Group trial ‘RTOG 9501’ (but not controlled for HPV-positive oropharyngeal SCC).Reference Bernier, Cooper, Pajak, van Glabbeke, Bourhis and Forastiere733
The indications are summarised below, which include primary tumour considerations.
• Pathologically staged N1 without adverse features* – no adjuvant treatment required
• Pathologically staged N1 with adverse features – post-operative RT or chemoradiotherapy
• Pathologically staged N2/3 – post-operative RT or chemoradiotherapy
*Adverse features include positive margin(s) at the primary site and extra-nodal extension, and may include advanced T stage (pathologically staged T3–4), involved nodes in levels IV and Vb, lymphovascular invasion, perineural invasion and a high grade (non-SCC histological tumour types).
For patients aged less than 70 years and fit for chemotherapy, chemoradiotherapy should be considered. The strongest indications are when there is likely to be residual disease after surgery, extra-nodal extension spread,Reference Bernier, Cooper, Pajak, van Glabbeke, Bourhis and Forastiere733 and high nodal burden (e.g. multiple lymph nodes or N3 disease).
Neck dissection for suspected persistent lymph node (only) disease after primary chemoradiotherapy or radiotherapy
When primary chemoradiotherapy or RT is used for cancers with N+ neck disease (most commonly for oropharyngeal SCC), post-treatment surgery may be required for suspected persistent neck disease. Such cases should be discussed in a multidisciplinary team (MDT) meeting.
There is no role for a planned neck dissection in all patients following primary chemoradiotherapy. The PET-Neck trial compared PET–CT-guided active surveillance with planned neck dissection for neck disease staged N2 or N3 treated by chemoradiotherapy, and showed that neck dissection as indicated by post-treatment PET-CT led to fewer neck dissections with no detrimental effect on survival.Reference Mehanna, Wong, McConkey, Rahman, Robinson and Hartley51 The PET-Neck trial had only a small proportion of N3 tumours (6 per cent). However, retrospective N3 stage cohort studies support a PET-CT-guided neck dissection strategy.Reference Adams, Porceddu, Pryor, Panizza, Foote and Rowan1093 In N3 disease, the commonest pattern of failure is distant metastases (17–44 per cent), and there is conflicting evidence regarding the benefit of induction chemotherapy on the impact of distant failure and resultant overall survival.Reference Du, Blanchard, Even, Boros, Gorphe and Breuskin1092
The PET-Neck trial established the standard of care of a PET-CT scan at 12 weeks following chemoradiotherapy. The categorisation of responses as complete, equivocal and incomplete is recommended. There is no consensus regarding the optimum qualitative interpretive criteria (e.g. Neck Imaging Reporting and Data Systems (‘NI-RADS’), Hopkins, and Deauville criteria) in post-treatment PET-CT in head and neck SCC, but the use of scoring systems that minimise indeterminate scores, whilst delivering a high negative predictive value, is recommended.Reference Zhong, Sundersingh, Dyker, Currie, Vaidyanathan and Prestwich1094 Comparison to the pre-treatment PET-CT avidity also may help. When there is evidence of an incomplete response, a neck dissection is required, the extent of which may vary. There is an increasing trend to perform a more selective neck dissection in patients who have completed standard chemoradiotherapy schedules and who have isolated, single-level disease on post-treatment PET-CT imaging.Reference Robbins, Shannon and Vieira1095 However, the extent of pre-treatment neck disease may also be factored into the decision-making, as well as the fact that the post-chemoradiotherapy neck dissection may be the last opportunity for curative neck surgery.
When there is an equivocal response (e.g. mild or no FDG avidity in enlarged nodes, or mild avidity in normal or borderline-sized nodes), there is variance in consequent management. In HPV-positive tumours (the majority in this group), involution continues beyond 12 weeks (especially for cystic nodules). Over 85 per cent of this group with an initial equivocal response do not go on to develop lymph node recurrence. A second-look interval PET-CT at a further three months showed that over 70 per cent of cases convert to a radiological complete response, with no subsequent regional failure.Reference Prestwich, Arunsingh, Zhong, Dyker, Vaidyanathan and Scarsbrook163,Reference Zhou, Chan, Rulach, Dyab, Hendry and Maxfield1096
Ultrasound with fine needle aspiration cytology in the assessment of disease response cannot be relied upon in isolation.Reference Costantino, Mercante, D'Ascoli, Ferreli, Di Tommaso and Franzese1097 Other potential strategies in this situation that require more evaluation include diffusion-weighted imaging MRI,Reference Schouten, de Graaf, Alberts, Hoekstra, Comans and Bloemena1098 or the use of circulating biomarkers such as tumour DNA or HPV antigens to detect disease persistence.Reference Lee, Garcia-Murillas, Cutts, De Castro, Grove and Hurley1099
Currently, a therapeutic neck dissection is generally recommended in advanced N3 HPV-positive or HPV-negative SCC with an equivocal response, because there is little evidence to modify the standard of care with a second, delayed PET-CT scan.
Treatment of contralateral lymph nodes in lateralised primary cancers
Recommendations
• There is no need for contralateral elective neck treatment for lateralised T1/2 cancers with limited or no ipsilateral neck metastases (N0 or single lymph node sized 3 cm or less) (evidence-based recommendation (R))
The contralateral neck should in general be treated according to the same principles as described above, which deal with the ipsilateral neck (or both sides for midline tumours).
Not irradiating the contralateral neck spares treatment effects, and is supported by low contralateral isolated nodal recurrence rates.Reference de Veij Mestdagh, Walraven, Vogel, Schreuder, van Werkhoven and Carbaat1100,Reference McDowell, Casswell, Bressel, Gough, Drosdowsky and Coleman1101 Much of the evidence has focused on oropharyngeal SCC. The evidence supports the fact that contralateral neck irradiation is not indicated for lateralised T1 or T2 oropharyngeal cancers staged as N0 or with a single ipsilateral lymph node sized less than 3 cm (i.e. N1, seventh edition of the TNM classification and staging system). The contralateral regional failure rate with ipsilateral neck (only) treatment is low (1 per cent or less).Reference Al-Mamgani, van Rooij, Fransen and Levendag1102 There is growing evidence that this also applies to cases with multiple ipsilateral lymph nodes (i.e. N2b, seventh edition of the TNM classification and staging system) with low isolated contralateral neck recurrence (around 3 per cent).Reference Taku, Chronowski, Brandon Gunn, Morrison, Gross and Moreno1103 However, practice varies, and MDT discussion is recommended in such cases, especially when there is a high burden of ipsilateral neck disease and more advanced primary cancers.
The use of lymphatic mapping with single-photon emission computed tomography (‘SPECT’)-CT and/or sentinel lymph node biopsy in order to guide neck treatment, including the contralateral neck, may assist with this issue (see the Studies due to report section below).
Recurrent disease
Isolated lymph node recurrent disease
Recommendations
• Salvage neck dissection with curative intent should be offered for recurrent neck disease deemed to be operable (good practice point (G))
When isolated nodal recurrence is found, a PET-CT scan is recommended, to help delineate the recurrence and to assess for local and distant disease. Patients with isolated nodal recurrence that is thought to be resectable with clear margins should be considered for salvage neck dissection, as it provides the best chance of regional control. The reported overall survival outcome for isolated nodal recurrence treated with salvage neck dissection varies from 25 per cent to 56 per cent.Reference Giger, Fink, Demattè, Visini, Elicin and Anschuetz1104,Reference Lindegaard, von Buchwald, Rasmussen, Specht, Vogelius and Zamani1105 Favourable prognostic factors for survival include p16-positive SCC, complete margins, no extra-nodal extension, low-volume disease, no previous neck treatment and a disease-free interval of greater than six months.Reference Chung, Lee, Baek, Bae, Chang and Rho1106 Current evidence suggests regional recurrence occurs generally in the previously treated involved neck levels, regardless of whether the initial modality was surgery or (chemo)radiotherapy. Limited evidence exists regarding the optimal extent of salvage neck dissection in isolated nodal recurrence, with low rates of occult metastases reported in levels I and V.
For unresectable disease or resectable disease resulting in significant morbidity, re-irradiation may be considered in carefully selected patients using high precision conformal RT (intensity-modulated RT, volumetric modulated arc therapy) and novel (stereotactic ablative RT) techniques to reduce severe toxicity.Reference Chopra, Gupta, Agarwal, Budrukkar, Ghosh-Laskar and Dinshaw1107
The combination of chemoradiotherapy and then secondary neck dissection adds to the risk of severe late toxicity (chronic laryngeal or pharyngeal toxicity grade of more than 3, organ dysfunction, or death).Reference Machtay, Moughan, Trotti, Garden, Weber and Cooper1108
Treatment of clinically staged node-negative neck with primary tumour recurrence
Most evidence in this area is from salvage laryngectomy, the commonest salvage surgery scenario. Here, the incidence of occult metastases is low, around 10–15 per cent. Neck dissection in the whole patient group does not improve overall survival, although it does improve locoregional control.Reference Lin, Lam, Warner and Paleri1109,Reference Davies-Husband, Drinnan and King1110
However, the risk of occult metastases is higher in patients with initial N+ disease, a previously untreated neck, advanced T3/4 primary disease or higher risk subsites for occult metastases (e.g. hypopharynx, supraglottis).Reference Lin, Lam, Warner and Paleri1109,Reference Mazerolle, Gorphe, Vairel, Dupret-Bories, Temam and Chaltiel1111 Hence, ipsilateral neck dissection should at least be considered in these circumstances, and may be associated with improved survival.
A concern in the setting of salvage laryngectomy is whether neck dissection increases the risk of pharyngocutaneous fistula, with some supporting evidence for this.
In the setting of salvage surgery for non-laryngeal or non-hypopharyngeal cancers, there are less data, but the principle appears to be broadly similar with generally low rates of occult metastases (around 10 per cent) and no survival advantage as a result of neck dissection.Reference Mazerolle, Gorphe, Vairel, Dupret-Bories, Temam and Chaltiel1111 However, in many cases, the neck will be accessed for microvascular reconstruction, and neck dissection of levels 2–3 adds little morbidity when microvascular access for free-flap reconstruction is required.
Studies due to report
‘Post-operative Adjuvant Treatment for HPV-positive Tumours (PATHOS)’ (ClinicalTrials.gov Identifier: NCT02215265). This study further investigates the significance of extra-nodal extension and omission of chemotherapy for HPV-positive SCC, and the outcomes of ipsilateral neck irradiation for well-lateralised oropharyngeal cancer tumours.
‘Comparing Sentinel Lymph Node (SLN) Biopsy With Standard Neck Dissection for Patients With Early-Stage Oral Cavity Cancer’ (NRG Oncology study identification number: HN006). A phase II and subsequent planned phase III non-inferiority trial of 5 per cent absolute difference in two-year disease-free survival.
‘Lymph Drainage Mapping for Tailoring Elective Nodal Irradiation in Head and Neck Cancer (SUSPECT-2)’ (ClinicalTrials.gov Identifier: NCT03968679). A single-centre prospective trial of single-photon emission computed tomography CT-guided elective nodal irradiation.
‘SPECT-CT Guided ELEctive Contralateral Neck Treatment in Lateralized Oropharyngeal Cancer (SELECT)’ (Canadian Cancer Trials Group study identification number: CCTG-HN11). A phase III randomised, controlled trial.
‘Lymphatic Mapping Of Oropharyngeal Cancer (LOOC)’ (ClinicalTrials.gov Identifier: NCT04498221). A phase II multicentre study to validate the use of sentinel node biopsy.
Research questions
(1) What is the prognostic significance of extra-nodal extension in HPV-related oropharyngeal cancer?
(2) What is/are the ideal radiological modality/modalities for detecting pre-treatment nodal extra-nodal extension?
(3) What are the indications for dose reduction in the surgically staged neck for HPV-related and unrelated head and neck SCC?
(4) What are the indications for dose reduction in the clinically staged neck for HPV-related and unrelated head and neck SCC?
(5) Does sentinel node guided neck treatment offer functional advantages over sentinel node dissection?
Introduction
Producing and interpreting guidelines for the management of head and neck squamous cell carcinoma (SCC) of unknown primary is inherently challenging. Firstly, there is a paucity of robust contemporary evidence on the topic. Many historic studies predate our understanding of the role of human papillomavirus (HPV) in head and neck cancer,Reference Ang, Harris, Wheeler, Weber, Rosenthal and Nguyen-Tan226 and, with the incidence of HPV-related disease rising,Reference Gillison, Chaturvedi, Anderson and Fakhry205 management recommendations must necessarily be updated to be most effective at improving patient care. Secondly, the understanding and definition of what is considered an ‘unknown primary’ evolves during the diagnostic pathway. During this process, clinical examination, imaging investigations and surgical biopsies all may identify a primary disease. As a result, direct inter-study comparisons or meta-analysis are complicated by incongruent cohort definitions and eligibility criteria. Thirdly, true unknown primary disease is not common, and so establishing both a substantial evidence base and reasonable clinical experience regarding its management can be challenging, particularly in single-centre settings.
Despite these limitations, many organisations have produced guidelines covering the management of head and neck SCC of unknown primary, using a variety of methodologies.Reference Mackenzie, Watson, Jankowska, Bhide and Simo1112–Reference Machiels, Rene Leemans, Golusinski, Grau, Licitra and Gregoire871 The present guidelines were produced following a multi-stage meta-consensus initiative that was developed specifically for this work. This incorporated a National Audit of Practice, a National Consensus Day and a National Delphi Exercise. Through this process, novel data were generated, the most up to date published and unpublished studies were considered, and draft statements were generated before being scrutinised by representatives from all UK head and neck multidisciplinary teams (MDTs) to produce these final recommendations. The full outline of this methodology has been published separately.1114
These guidelines follow the patient journey from presentation with unknown primary disease to post-treatment surveillance. Recommendations are included as statements at the beginning of each section, followed by further guidance and commentary to add context. National Institute for Health and Care Excellence (NICE) phraseology has been used when generating the statements, to reflect the strength of evidence and level of certainty in the benefit of the intervention for each recommendation presented.1115 The terms ‘offer’, ‘perform’, ‘refer’ and ‘include’ reflect confidence in a strong patient benefit. Where the evidence offers less certainty in a clear benefit, the term ‘consider’ is used.
These guidelines do not describe the management of non-SCC disease of unknown primary origin.
An illustration of the patient pathway related to proposed minimum required interventions is shown in Figure 1.
Figure 1. Illustration of the patient pathway, related to proposed minimum required intervention patient labels for these groups (each stage assumes the primary site remains unknown). FNE = flexible nasendoscopy; HNSCCUP = head and neck squamous cell carcinoma of unknown primary; HPV = human papillomavirus; PET-CT = positron emission tomography computed tomography; MRI = magnetic resonance imaging; EBV = Epstein–Barr virus
Investigations before diagnostic surgery
Recommendations
• Offer all patients with clinically suspected head and neck SCC of unknown primary ultrasound-guided sampling as a first-line investigation to diagnose cervical metastasis of SCC, which must include p16 and/or HPV subtyping and ancillary tests (evidence-based recommendation (R))
• Do not offer open biopsy to patients with a neck lump as a first-line investigation to diagnose cervical metastasis (R)
• Offer all patients with clinical suspicion of head and neck SCC of unknown primary concurrent magnetic resonance imaging (MRI) and positron emission tomography computed tomography (PET-CT) scanning as first-line cross-sectional imaging investigations (R)
• Consider image-enhancement technology (including narrow-band imaging) as an adjunct to white-light endoscopy in the examination of all patients with clinically suspected head and neck SCC of unknown primary (good practice point (G))
• Refer all patients with clinically suspected pathologically confirmed head and neck SCC of unknown primary to a core member of the head and neck MDT for further investigations (R)
All patients presenting with a neck mass will need a comprehensive history taking, and clinical examination including flexible nasendoscopy (FNE). Alongside FNE, there is good evidence that virtual chromoendoscopy (e.g. narrow-band imaging) can aid the recognition of otherwise occult mucosal lesions, though it is acknowledged that not all UK centres have access to this technology.Reference Di Maio, Iocca, De Virgilio, Giudice, Pellini and D'Ascanio1116
All patients will require cytological or cytopathological confirmation of cancer. Subtyping of HPV is important for the effective management of all clinically suspected head and neck SCC of unknown primary patients; if this is not available on fine needle aspiration cytology then core biopsy should be performed. This is particularly important for patients in whom no primary site is identified by the end of the diagnostic pathway and who do not undergo neck dissection, as they will have no other tissue on which to perform HPV and Epstein–Barr virus analysis. Ultrasound guidance increases the diagnostic accuracy of the biopsy.Reference Robitschek, Straub, Wirtz, Klem and Sniezek1117 Open biopsy is not felt to be an appropriate alternative to ultrasound-guided core biopsy.
Patients presenting with clinically suspected head and neck SCC of unknown primary often experience long diagnostic pathways before starting definitive treatment, with a significant amount of time on the pathway spent awaiting imaging investigations, PET-CT in particular. Current NICE guidance is to consider a PET-CT scan for patients with confirmed metastatic disease in whom no primary is evident on clinical examination.1074 Immediately following cytological or cytopathological confirmation of metastatic disease, concurrent MRI and PET-CT, requested as first-line cross-sectional imaging, would allow synchronous interpretation, cover staging of the chest, and expedite progression to diagnostic surgery in search of a primary site.
Timely referral to a head and neck MDT is deemed essential to ensure appropriate oversight of the diagnostic pathway, as well as subsequent treatment.Reference Licitra, Keilholz, Tahara, Lin, Chomette and Ceruse1118
Diagnostic surgery
Recommendations
• Perform all radiological investigations that aim to identify the primary site prior to discussion at the head and neck MDT and before diagnostic surgery (evidence-based recommendation (R))
• Offer nasopharyngeal biopsies when the cervical node sampling reveals Epstein–Barr virus positive metastasis (R)
• Do not offer biopsies of clinically and radiologically normal upper aerodigestive tract mucosa. This excludes tonsillectomy or tongue base mucosectomy (R)
• Offer ipsilateral tonsillectomy (rather than incisional biopsy) in all patients (R)
• Consider ipsilateral tongue base mucosectomy in all patients (good practice point (G))
• Consider contralateral tonsillectomy (rather than incisional biopsy) in all patients (G)
• Consider contralateral tongue base mucosectomy in all patients (G)
• Perform tongue base mucosectomy using one of the following transoral techniques, when indicated: endoscopic, microscopic or robot-assisted (G)
Strategies for obtaining oropharyngeal biopsies remain contentious. In 2016, NICE guidance included offering surgery to identify the unknown primary.1074 However, as with all surgery, these procedures are associated with their own morbidities and complications; if offered to the patient, each element should be clinically justifiable, and the patient fully informed of the risks and benefits.Reference de Almeida1119,Reference Ozbay, Yumusakhuylu, Sethia, Wei, Old and Agrawal1120
Ipsilateral tonsillectomy is widely accepted as being diagnostically beneficial. However, the detection rate of primary disease from a contralateral tonsillectomy is lower. Consequently, the marginal benefit from removing the contralateral tonsil for diagnostic purposes, as well as the advantage of a symmetrical oropharynx being easier to monitor for future disease, must be weighed against the additional morbidity from the procedure.Reference Pothier, Nankivell, Matthee and Hunasaghatta1121
Removal of the lingual tonsillar tissue (also known as tongue base mucosectomy) as a diagnostic procedure in search of a primary tumour has become more prevalent. This is, in part, due to the advent of robotic technology, although other transoral techniques are available and have proved efficacious. Tongue base mucosectomy has been reported to increase the identification of the primary tumour.Reference Farooq, Khandavilli, Dretzke, Moore, Nankivell and Sharma1122,Reference Winter, Ofo, Meikle, Silva, Fraser and O'Hara1123 There remains debate about the extent of tongue base mucosectomy (whether it should be unilateral or bilateral) as well as the timing of the procedure (whether it should be performed at the same time as palatine tonsillectomy or only following negative histology from palatine tonsillectomy); both factors will affect the apparent detection rate. Practice regarding the extent of oropharyngeal clearance is influenced by concerns of pharyngeal stenosis, though a rate of symptomatic narrowing has not been established in any large-scale cohorts.
Surgical management
Unless otherwise specified, patients with head and neck SCC of unknown primary referred to in this section are assumed to have undergone an adequate diagnostic investigation, as per their MDT, and are due to commence treatment for head and neck SCC of unknown primary.
Recommendations
• Consider ipsilateral tonsillectomy and tongue base mucosectomy, and ipsilateral neck dissection in both HPV-negative and HPV-positive head and neck SCC of unknown primary in cases with a single involved node sized 3 cm or less with no radiological extra-nodal extension (good practice point (G))
• Consider neck dissection prior to treatment in patients with HPV-negative head and neck SCC of unknown primary undergoing radical radiotherapy with advanced disease who are unsuitable for concomitant chemotherapy (G)
• Consider neck dissection prior to radiotherapy, with or without chemotherapy, in HPV-negative head and neck SCC of unknown primary patients with nodal stage N3 neck disease (G)
Single modality surgery may be considered as appropriate treatment for patients undergoing ipsilateral surgery to the oropharynx and the neck with a single involved node measuring 3 cm or less with no radiological extra-nodal extension. Clearance of the contralateral tonsil and/or tongue base may help reassure the MDT that the putative primary sites have been adequately addressed to manage the risk of primary emergence.
Concomitant chemotherapy has been shown to have a significant benefit in patients with HPV-negative disease, which is more commonly associated with an aggressive course.Reference Cheraghlou, Torabi, Husain, Otremba, Osborn and Mehra1124 If concomitant chemotherapy is not felt to be suitable then primary surgery should be considered in these patients, to ensure dual modality therapy is delivered. An alternative is to perform surgery after radiotherapy, depending on PET-CT imaging findings.
In HPV-negative head and neck SCC of unknown primary patients with N3 disease, primary neck dissection should be considered before radiotherapy, regardless of the patients’ suitability for chemotherapy, owing to their poorer survival outcomes.Reference Cheraghlou, Torabi, Husain, Otremba, Osborn and Mehra1124
Some MDTs currently advocate a limited ‘staging’ neck dissection of the clinically negative contralateral neck, with the intention being to show the contralateral neck is histologically disease-free and so to spare this volume from subsequent radiotherapy. However, there is currently a lack of evidence in the literature to support this strategy and so it is not recommended (or opposed) by these guidelines.
Non-surgical management
Unless otherwise specified, patients with head and neck SCC of unknown primary in this section are assumed to have undergone an adequate diagnostic investigation, as per their MDT, and are due to commence treatment for head and neck SCC of unknown primary.
Recommendations
• Consider omitting adjuvant radiotherapy after an ipsilateral neck dissection where there is a single involved node sized 3 cm or less with no extra-nodal extension (good practice point (G))
• Offer adjuvant radiotherapy, with or without chemotherapy, to the ipsilateral neck after an ipsilateral neck dissection where there is a single involved node sized greater than 3 cm, or there are multiple involved nodes, or there is extra-nodal extension (evidence-based recommendation (R))
• Consider adjuvant radiotherapy with or without chemotherapy to the bilateral neck after an ipsilateral neck dissection where there are multiple involved nodes or there is pathological extra-nodal extension (G)
• Consider radiotherapy, with or without chemotherapy, to the bilateral neck if there are multiple involved ipsilateral nodes or there is radiological extra-nodal extension (G)
• Consider including the ipsilateral oropharynx in the treated volume when giving radiotherapy to the neck for unilateral HPV-positive head and neck SCC of unknown primary (G)
• Consider including possible mucosal primary sites when giving radiotherapy to the neck for unilateral HPV-negative head and neck SCC of unknown primary. Decide possible sites based on the pattern of nodal involvement, clinicopathological features and risk factors (e.g. smoking) (G)
• Offer 50 Gy in 2 Gy fractions or equivalent (e.g. 54 Gy in 30 fractions or 56 Gy in 35 fractions) as the radiotherapy dose for possible mucosal primary sites when they are intentionally included in the target volume (R)
• Offer concomitant cisplatin chemotherapy with adjuvant radiotherapy if there is pathological extra-nodal extension and the patient is suitable to receive cisplatin (R)
• Offer concomitant cisplatin chemotherapy with primary radiotherapy if there are multiple involved nodes or radiological extra-nodal extension and the patient is deemed fit to receive cisplatin (R)
• Include the ipsilateral retropharyngeal and retrostyloid nodes in the elective target volume when giving radiotherapy to the ipsilateral neck where level II is involved (R)
Adjuvant radiotherapy to the ipsilateral neck in more advanced head and neck SCC of unknown primary disease is essential.Reference Frank, Rosenthal, Petsuksiri, Ang, Morrison and Weber1125 Contralateral radiation should be considered in the case of extra-nodal extension or where multiple nodes are involved. For the majority of head and neck SCC of unknown primary disease patients who present with level II involvement, retropharyngeal and retrostyloid nodes should be included in the elective target volume.
There is insufficient evidence to support or oppose ipsilateral radiation to the oropharynx in all unilateral HPV-positive head and neck SCC of unknown primaries, or indeed to any putative mucosal sites in HPV-negative disease. Where radiotherapy is given for HPV-negative head and neck SCC of unknown primary disease, the mucosal sites should be chosen based on the pattern of nodal involvement and any other relevant clinicopathological features. In all cases where radiotherapy is given to mucosal target volumes, these guidelines advocate the use of a prophylactic dose, not as high as used in adjuvant or radical dosing regimens, though it is accepted this is based on consensus opinion rather than any high-level evidence. The omission of radiation to putative mucosal sites may be considered under MDT supervision, but assumes an adequate diagnostic investigation and appropriate clinico-radiological surveillance.
Post-treatment surveillance
Recommendations
• Consider adding regular cross-sectional imaging to the regular clinical examination for post-treatment surveillance of patients treated with surgery as a single modality, following bilateral tonsillectomy and tongue base mucosectomy and pathologically staged N1 disease with no extra-nodal extension (good practice point (G))
• Follow up is discussed in a separate chapter, but, because of the nature of unknown primary disease and the effects of treatment on imaging, baseline cross-sectional imaging is recommended, and surveillance imaging should be considered for those who have received single modality surgical treatment, in particular (G)
National audit data have suggested that locoregional control rates may be lower in patients treated by surgery alone. As such, this group has been highlighted for regular imaging surveillance, for primary emergence, in addition to regular clinical review. It is possible that the addition of radiation directed at the neck may give enough dose to putative mucosal sites to treat any occult primary disease that is not clinically, radiologically or histologically evident by the time some patients commence definitive treatment for their head and neck SCC of unknown primary.
Limitations
A complete outline of the methodology used to develop these guidelines is published elsewhere, which outlines the initiative used to generate these consensus recommendations.Reference Hardman, Harrington, Roques, Sood, Jose and Lester1126 The following limitations are highlighted here as particularly relevant to the process. Firstly, attendance at the Consensus Day was self-selecting, giving the potential for disproportionate representation of individual stakeholder groups during the generation of the draft consensus statements. Secondly, during the Delphi exercise, the MDT contact was asked to record the consensus view of their team. However, the true level of consultation with each MDT was not recorded and may have varied. Responses may, therefore, have been biased towards those who engaged in the process locally, and specifically towards ENT team members who were the contact specialty for this exercise. Finally, our Delphi exercise used a binary response to register support either for or against each statement under consideration. As such, there may have been an under-representation of clinical oncology input. This methodology is unable to present the strength of opinion from individual units, and could be seen to misrepresent the views of a minority of respondents who may have had strong opposition to the statement as presented, compared to a majority who felt only weakly in favour.
Acknowledgements
This work was only possible due to contributions of the individuals listed in Appendix 2.
Introduction
Non-melanoma skin cancer is the most common type of malignancy in the Caucasian population and is four times more common than any other cancer in the UK.1127 One in four men and one in six women will develop a non-melanoma skin cancer in their lifetime.Reference Kwiatkowska, Ahmed, Ardern-Jones, Bhatti, Bleiker and Gavin1128 Mortality from non-melanoma skin cancer is rare, but tends to be from local or regional disease. Distant metastasis is rare with the exception of Merkel cell carcinoma. The majority of non-melanoma skin cancers can be cured by adequate initial surgical management. High-risk and complex non-melanoma skin cancer may be challenging to manage, and must be carried out through a skin cancer multidisciplinary team (MDT).
The majority of non-melanoma skin cancers occur in the skin of the head and neck. However, their clinical behaviours do not mirror that of head and neck mucosal malignancies. In order to optimise management and outcomes, all non-melanoma skin cancers (with the exception of specifically defined low-risk basal cell carcinoma (BCC)) should be managed by core members of a skin cancer MDT.
This chapter will describe the recommendations for the investigation and management of non-melanoma skin cancer, drawing on guidelines produced by the British Association of Dermatologists. These are recommended reading, as they are accepted as the national guidelines for the management of non-melanoma skin cancer and are mandated by the National Institute for Health and Care Excellence (NICE).Reference Keohane, Botting, Budny, Dolan, Fife and Harwood1129,Reference Nasr, McGrath, Harwood, Botting, Buckley and Budny1130
Pathology
The majority of non-melanoma skin cancers are keratinocyte cancers, of which approximately 80 per cent are BCCs. Of the remaining non-melanoma skin cancers, the majority are cutaneous squamous cell carcinomas (SCCs), with a very small number of rare tumours such as Merkel cell carcinoma and adnexal tumours.
Premalignant lesions such as actinic keratosis or Bowen’s disease (in situ cutaneous SCC) can undergo malignant transformation to non-melanoma skin cancer. The risk of a single actinic keratosis transforming to SCC is low (around 1 per cent on 10 years). However, patients with actinic keratosis are five times more likely to develop a skin cancer compared to matched controls, with the greatest risk for cutaneous SCC.Reference Guorgis, Anderson, Lyth and Falk1131 Around 3–5 per cent of proven Bowen's disease transforms to cutaneous SCC, higher in high-risk populations such as the immunosuppressed.Reference Cox, Eedy and Morton1132
Epidemiology and aetiology
The UK annual incidence of non-melanoma skin cancer is around 210 000 cases per year, with at least a 5 per cent annual increase observed from 2013 to 2015.Reference Guorgis, Anderson, Lyth and Falk1131 Non-melanoma skin cancer is more common in men, with a mean age of onset in the eighth decade of life. The majority of patients with non-melanoma skin cancer are Caucasian. The major aetiological factors for non-melanoma skin cancer are fair skin type and chronic exposure to sunlight, and indoor tanning or other types of ionising radiation. Genetic syndromes are associated with the risk of non-melanoma skin cancer, such as Gorlin syndrome (BCC), Muir–Torre syndrome which is a phenotypic variant of Lynch syndrome (characterised by keratoacanthomas and sebaceous tumours, but also with an increased risk of cutaneous SCCs), and xeroderma pigmentosa (melanoma and non-melanoma skin cancer).Reference Sowter, Santibanez-Koref, Jackson, Borthwick, Burn and Rajan1133 Immunosuppression is also a significant risk factor. Human papilloma virus infection (beta types), particularly in the immunosuppressed, is also an important aetiological factor in cutaneous SCC.Reference Tampa, Mitran, Mitran, Nicolae, Dumitru and Matei1134
Previous non-melanoma skin cancer is a noteworthy risk factor, with over 40 per cent of patients with BCC and 37 per cent of patients with cutaneous SCC developing further non-melanoma skin cancer. For Merkel cell carcinoma, there is association with ultraviolet (UV) exposure, increasing age and immunosuppression. There is now growing evidence that for some Merkel cell carcinomas, a major aetiological factor is infection with Merkel cell polyomavirus,Reference Spurgeon and Lambert1135 though in Australasia the main factor may be UV exposure.Reference Garneski, Warcola, Feng, Kiviat, Leonard and Nghiem1136,Reference Goh, Walradt, Markarov, Blom, Riaz and Doumani1137
Presentation, diagnosis and multidisciplinary team management
Recommendations
• Measure and record suspected skin cancers (evidence-based recommendation (R))
• Photograph suspected skin cancers before biopsy or excision (R)
• Royal College of Pathology minimum dataset should be used to report all non-melanoma skin cancer pathology (R)
• Cross-sectional imaging should be arranged to assess fixed and/or locally advanced cancers (good practice point (G))
• All head and neck non-melanoma skin cancer (except specifically defined, completely excised low-risk BCC) should be discussed at a skin MDT meeting (R)
Initial patient assessment should include history-taking, a full skin examination, and measurement and photographs of the lesion. For a majority of non-melanoma skin cancer patients, an excisional biopsy will be both diagnostic and therapeutic, planned with a margin likely to completely excise the lesion. It is important where possible to plan a surgical margin that will result in a 1 mm or more histological margin.
Where this is not possible, if there is diagnostic doubt or if the extent of treatment is significant (such as for very large lesions or those invading adjacent structures), and when radiotherapy is being considered as the primary treatment, a diagnostic biopsy may be more appropriate.
Imaging requirements are discussed under the relevant tumour management sections. However, cross-sectional imaging should be undertaken for all primary tumours where deeper tissue involvement is suspected.
In order to optimise management and outcomes, all non-melanoma skin cancers except low-risk BCC cancers should be managed by core members of a skin cancer MDT. This may occur after diagnosis for advanced, rare or metastatic disease, and/or after initial surgery, as well as for recurrent disease.
The following cases do not routinely require MDT discussion in the absence of any other high-risk factors:
• Diagnostics biopsies of BCC
• Diagnostic biopsies of cutaneous SCC
• Completely excised BCC with 1 mm or more histological clearance at all surgical margins following excision with curative intent
• Completely excised low-risk pathologically staged T1 cutaneous SCC in a low-risk patient (these patients can be tabled for inclusion in cancer registries where that is the local mechanism)
All other cases should be discussed and managed under the auspices of a skin cancer MDT. There may be regional variation in the organisation and set up of different levels of skin MDTs. In some parts of the UK, two levels of skin cancer MDT are recognised: local hospital skin cancer MDTs and regional specialist skin cancer MDTs. The British Association of Dermatology have set levels of care to match the complexity of case needs with varying level of clinical expertise (Figure 1), with some variance according to local or regional expertise.
Figure 1. Levels of care for skin cancer multidisciplinary teams. The Improving Outcomes Guidance, either explicitly or by implication, effectively specifies six levels of care, differing in the degree of specialisation, case mix, and the procedures and service consolidation needed, as demonstrated in the table. GP = general practitioner; SCC = squamous cell carcinoma; MDT = multidisciplinary team; LSMDT = local hospital skin cancer multidisciplinary team; SSMDT = specialist skin cancer multidisciplinary team; BCC = basal cell carcinoma; SCG = Specialised Commissioning Group
The importance of specialist pathologists, radiologists, dermatologists, surgeons, clinical oncologists and specialist nurses working together to achieve the best, highest quality outcomes for patients is paramount.
Other specialist input, such as discussion in the head and neck, skull base and/or neuro-oncology MDTs, and collaboration in care, may be required in some cases.
All patients should receive education in self-surveillance and self-examination, UV exposure protection, and vitamin D supplementation.
Staging
The eighth edition of the AJCC Cancer Staging Manual has also clarified the classification for cutaneous SCC of the lip, by defining SCC of the dry vermillion as a cutaneous cancer that as such needs to be managed in line with cutaneous SCC guidance.Reference Califano, Lydiatt, Nehal, O'Sullivan, Schmults, Seethala, Amin, Edge, Green, Byrd, Brookland and Washington1138
There is separate staging for cutaneous SCC of the eyelid and for Merkel cell carcinoma.
Tables 1–3 show the tumour–node–metastasis (TNM) staging for non-melanoma skin cancer (according to the AJCC Cancer Staging Manual, eighth editionReference Califano, Lydiatt, Nehal, O'Sullivan, Schmults, Seethala, Amin, Edge, Green, Byrd, Brookland and Washington1138).
Table 1. Tumour staging for non-melanoma skin cancer*
*According to the AJCC Cancer Staging Manual (eighth edition).Reference Califano, Lydiatt, Nehal, O'Sullivan, Schmults, Seethala, Amin, Edge, Green, Byrd, Brookland and Washington1138 †Beyond subcutaneous fat or more than 6 mm from the granular layer of adjacent normal epidermis
Table 2. Nodal staging for non-melanoma skin cancer*
* According to the AJCC Cancer Staging Manual (eighth edition).Reference Califano, Lydiatt, Nehal, O'Sullivan, Schmults, Seethala, Amin, Edge, Green, Byrd, Brookland and Washington1138
Table 3. Group staging for non-melanoma skin cancer*
* According to the AJCC Cancer Staging Manual (eighth edition).Reference Califano, Lydiatt, Nehal, O'Sullivan, Schmults, Seethala, Amin, Edge, Green, Byrd, Brookland and Washington1138
Management of basal cell carcinoma
Recommendations
• Consider standard surgical excision with a 4–5 mm peripheral margin as the first-line treatment for adults with low-risk BCC (good practice point (G))
• Offer standard surgical excision with immediate reconstruction for adults with BCC with a high-risk factor if the BCC has well defined margins (evidence-based recommendation (R))
• Offer standard surgical excision and delayed reconstruction or Mohs surgery for high-risk or high-risk site BCC with poorly defined margins (R)
• Consider radiotherapy for the treatment of adults with BCC where a patient is unsuitable for, or declines, surgical treatment (R)
• Discuss incompletely excised or complex BCCs within a skin MDT meeting (R)
Basal cell carcinoma is slow-growing and locally invasive, but very rarely metastasises. Although there are many subtypes of BCC, the majority of BCCs can be diagnosed clinically by experienced clinicians with good lighting, skin stretch and dermoscopy.
The British Association of Dermatologists guidelines on the management of BCC have adopted the Royal College of Pathologists dataset for reporting BCCs.Reference Nasr, McGrath, Harwood, Botting, Buckley and Budny1130 They have combined this with guidelines from NICE, the National Comprehensive Cancer Network1139 and the Union for International Cancer Control / American Joint Committee on Cancer AJCC Cancer Staging Manual (eighth edition)Reference Califano, Lydiatt, Nehal, O'Sullivan, Schmults, Seethala, Amin, Edge, Green, Byrd, Brookland and Washington1138 to produce the criteria for high- and low-risk BCCs (Figure 2). A summary of BCC management is shown in Figure 3.
Figure 2. Criteria for low- and high-risk basal cell carcinoma (BCC).
Figure 3. Management of basal cell carcinoma (BCC). GP = general practitioner; MDT = multidisciplinary team; GPwSI = general practitioner with a special interest; GPwER = general practitioner with an extended role; LSMDT = local hospital skin cancer multidisciplinary team; SSMDT = specialist skin cancer multidisciplinary team; NICE = National Institute for Health and Care Excellence; SMC = Scottish Medicines Consortium
Low-risk basal cell carcinomas
Low-risk BCCs of the head and neck should be referred to the hospital setting for management by the local hospital skin cancer or specialist skin cancer MDTs.1140
Whilst destructive techniques such as cryotherapy, curettage and cautery, and photodynamic therapy may safely be used by appropriately trained clinicians in specific circumstances, standard surgical excision remains the mainstay of treatment for low-risk BCCs, using at least a 4 mm peripheral margin.
High-risk basal cell carcinomas
The treatment options are primary surgery, Mohs surgery, radiotherapy or no treatment. Surgery should aim to achieve at least a 5 mm peripheral clearance margin. Mohs micrographic surgery may also be considered, particularly for poorly defined cancers. Radiotherapy may be used in the treatment of primary BCC in patients who are unwilling or unsuitable for surgery, or who have a preference for radiotherapy. Radiotherapy may be of particular benefit in cosmetically sensitive areas. Radiotherapy is contraindicated in previously irradiated areas, in patients with genetic conditions predisposing to BCC (e.g. Gorlin syndrome), and in cases where there is bone or cartilage invasion. Surgery is generally preferred in younger patients, but there may be occasions where, on the balance of risks and benefits, radiotherapy is used.
For recurrent or incompletely excised BCCs, Mohs surgery, standard excision with delayed reconstruction or radiotherapy should be considered following discussion at a specialist skin cancer MDT meeting.
Advanced BCCs – those invading multiple structures, in complex and functionally and cosmetically important areas, with large numbers of co-existing tumours, which are multiply recurrent, a high-risk subtype or require extensive treatment – should be discussed in a specialist skin cancer MDT meeting and treatment carefully planned. The patient's performance status and frailty, and their preferences, should be considered when planning treatment, and it must be accepted that, for some patients, not treating the BCC may be the best option.
Other treatment options may be suitable for patients with complex or advanced disease who are not suitable for standard surgical management or radiotherapy, such as electrochemotherapy, which is NICE-approved in appropriate cases.1141 The use of these modalities must be discussed at a specialist skin cancer MDT meeting. Vismodegib, a hedgehog pathway inhibitor was introduced for use in patients with Gorlin syndrome and unresectable BCCs, but in 2017 approval for vismodegib was withdrawn by NICE. However, it is now available via the National Health Service (NHS) England Cancer Drugs Fund for use in Gorlin syndrome patients where there are six or more BCCs or in a non-Gorlin syndrome patient with more than six non-locally advanced BCCs, and in non-metastatic BCC cases where surgical treatment would be very disfiguring.1142
Patients with a single adequately treated BCC do not require routine follow up and can be discharged with appropriate advice. Patients require follow up if the tumour is at a high risk of recurrence (e.g. after inadequate initial treatment) or there is a high risk of further BCC development (multiple previous BCCs, Gorlin syndrome or immunosuppression).
Cutaneous squamous cell carcinoma
Recommendations
• Request ultrasound with or without fine needle aspiration cytology (FNAC) for clinically suspected lymph node metastases for non-melanoma skin cancers (evidence-based recommendation (R))
• Request axial imaging for T3–4 cancers and include regional nodal basins (good practice point (G))
• Offer standard surgical excision as the first-line option to patients with cutaneous SCC, with peripheral surgical margins based on risk stratification (R)
• Lesions should be excised to the next clear surgical plane; in the scalp, resection should include galea (R)
• Discuss all cutaneous SCC cases, except pathologically staged T1 tumours excised with 1 mm or greater margin, at skin cancer MDT meetings (R)
• Consider a less than 1 mm histological margin in cutaneous SCC as a close margin, and consider wider excision or adjuvant radiotherapy (R)
• Offer adjuvant radiotherapy to patients with pathologically staged T3 cutaneous SCC with significant perineural invasion involving one or more nerves of greater than 0.1 mm in diameter, or involving a nerve beyond the dermis or a named nerve (R)
• Offer parotidectomy and neck dissection to patients with nodal metastasis and no evidence of systemic disease (R)
• Offer adjuvant radiotherapy following nodal dissection for patients with higher than N1 disease or high-risk pathological features (R)
• Consider immunotherapy for patients with locally and/or regionally advanced disease that is not suitable for surgery or radiotherapy, or those patients with systemic metastases (R)
Suspected cutaneous SCC should be managed by clinicians who are core members of a skin cancer MDT. The British Association of Dermatologists published guidelines on the management of patients with cutaneous SCC in October 2020.Reference Keohane, Botting, Budny, Dolan, Fife and Harwood1129 These divide cutaneous SCC into low-, high- and very high-risk status, by integrating clinical, pathological, TNM and margin criteria (Figure 4).
Figure 4. Criteria for risk stratification in cutaneous squamous cell carcinoma (cSCC). pT = pathological tumour stage; HIV = human immunodeficiency virus; HAART = highly active anti-retroviral therapy; MDT = multidisciplinary team; LSMDT = local hospital skin cancer multidisciplinary team; SSMDT = specialist skin cancer multidisciplinary team; GP = general practitioner
In cases where there is diagnostic uncertainty or complex treatment is required, diagnostic biopsy should be undertaken before definitive treatment.
Imaging for cutaneous squamous cell carcinoma
Cancers staged T3–4 need assessment of local extension with appropriate axial imaging (magnetic resonance imaging (MRI) and/or computed tomography (CT)). The regional nodes should be included in this imaging. Magnetic resonance imaging can highlight pathologically involved cranial (or less commonly peripheral) nerves, and should be carried out if there is clinical evidence of nerve involvement (dysaesthesia, formication, motor nerve palsy or severe neuralgic pain).
Lymph node assessment (by ultrasound scan and FNAC) should be conducted if there is evidence of possible lymph node metastasis, but can be considered for high-risk lesions without clinical evidence, such as greater than stage T2 lip SCC. Patients with regional metastatic disease should undergo CT of the thorax or positron emission tomography (PET)-CT when indicated.
Surgery for cutaneous squamous cell carcinoma
For the majority of patients, standard surgical excision will provide adequate treatment. Excision should be undertaken by appropriately trained clinicians, under bright light and magnification or dermoscopy. Peripheral margins should be:
• At least 4 mm for low-risk lesions
• At least 6 mm for high-risk lesions
• At least 10 mm for very high-risk lesions
Where tumours are mobile, the deep margin of excision should be at the next anatomically clear plane. On the scalp, at least the galea should be excised with the specimen. The aim should be to achieve clear surgical margins; this may require extensive resection with fascia, muscle, bone and other structures, which may have significant aesthetic and functional ramifications. Where perineural involvement is clinically or radiologically apparent, surgical resection of an involved nerve should be included in the primary treatment if feasible. Where possible, histologically clear margins should be confirmed prior to complex reconstruction. Clearly, in the head and neck this may not always be possible, as some defects will require immediate reconstruction (e.g. in skull base resection or where one or more major vessel(s) are exposed).
Non-surgical options for cutaneous squamous cell carcinoma
Where surgical excision is not feasible, or is likely to result in a poor functional or aesthetic outcome, or where the patient's preference is to avoid surgery, radiotherapy may be considered.
Locally advanced cutaneous SCC in particular requires careful consideration, as surgical excision (with or without adjuvant radiotherapy) may not be the best treatment modality. Locally advanced cutaneous SCCs can be defined as: tumours with symptomatic or radiological evidence of perineurial spread, pathologically staged T4 tumours, tumours requiring complex or extensive resection deep to subcutaneous fat, and tumours with two or more local recurrences. These cases, and those in which surgery may not be preferred or possible, should be discussed in the specialist skin cancer MDT meeting and the patients seen in a multidisciplinary clinic to ensure the best treatment modalities are selected. Although not explicitly included in the definition, patient factors such as frailty or medical co-morbidities may require a similar approach.
Post-operative management and adjuvant therapy
Non-recurrent isolated pathologically staged T1 cutaneous SCCs with at least 1 mm histological clearance at all margins, in an immunocompetent patient, do not require skin cancer MDT discussion, and the patient can be discharged. Other cutaneous SCC cases should be discussed at the MDT meeting. Patients with one or more clear but close margin may be suitable for observation if the tumour and patient are low risk. For patients with higher risk factors or tumours, and/or one or more close (less than 1 mm) or involved margin, further treatment should be considered and discussed with the patient. This may involve further surgery, including the use of Mohs surgery where appropriate, or adjuvant radiotherapy. An individually tailored approach is required for the immunosuppressed patient, and these patients will require regular surveillance for the long term. Patients with pathologically staged T3–4 tumours with histologically significant perineural invasion or other high-risk features will require specialist skin cancer MDT discussion regarding the possibility of adjuvant radiotherapy.Reference Keohane, Botting, Budny, Dolan, Fife and Harwood1129,1139
Regional lymph node metastasis in cutaneous squamous cell carcinoma
Lymph node metastases may rarely be present at the time of diagnosis or occur later. The UK frequency of metastatic cutaneous SCC has been shown to be 1.1 per cent for women and 2.4 per cent for men.Reference Venables, Autier, Nijsten, Wong, Langan and Rous1143 Other studies have shown higher rates of up to 6.7 per cent, but these may be in preselected higher-risk patients being treated in secondary care.Reference Khan, Mykula, Kerstein, Rabey, Bragg and Crick1144
The mean age of metastatic cutaneous SCC diagnosis is 80 years, and metastases mostly present within 2 years of the primary cutaneous SCC. Risk factors for developing metastatic disease are age over 80 years, being male, immunosuppression and deprivation. Tumours on the ear and lip are also at higher risk of metastasis. This patient group are often older adults, frail and have multiple co-morbidities, and may be challenging to manage.
Survival rates with metastatic cutaneous SCC are poor, with a three-year survival rate of 46 per cent in men and 29 per cent in women, though these data pre-date the emerging role of immune checkpoint inhibition in metastatic cutaneous SCC.Reference Venables, Autier, Nijsten, Wong, Langan and Rous1143 These poor outcomes may reflect the fact that cutaneous SCC is a disease of the older adult. The management of complex metastatic disease is often complicated by multiple co-morbidities and frailty.
Neck dissection and lymphadenectomy
Nodal spread is unpredictable, and varies from the well demonstrated patterns of spread from mucosal aero digestive tract malignancies. The parotid is the most common site of metastasis, followed by level 2.
The evidence for patterns of nodal spread in cutaneous SCC is based in part on sentinel node mapping both for melanomaReference Brennan, Kelemen and Matteucci1145 and from cutaneous SCC patients who have undergone neck dissection.Reference Vauterin, Veness, Morgan, Poulsen and O'Brien1146
Each individual case should be managed on its own merits, but, in general:
• Clinically or radiologically suspected lymph node metastases should be included in therapeutic lymphadenectomy
• If the parotid gland alone is involved but no other lymph nodes on clinical or radiological assessment, parotidectomy and dissection of levels 1–3 should be carried out
• Parotidectomy should be included in a therapeutic lymphadenectomy
• For the clinically staged N+ neck in posterior tumours, levels 2–5 should be dissected; level 1 can be omitted if uninvolved
• For the clinically staged N+ neck in anterior tumours, levels 1–4 should be dissected; level 5 can be omitted if uninvolved
As lymphatic drainage of the skin is initially via the superficial system, removing these draining nodes as well as the deep system seems prudent, and would include the superficial pre-auricular, post-auricular and occipital nodes for primaries draining via those regions.
Post-operative radiotherapy for lymph node metastases
Albeit without good evidence, the general consensus is that patients with pathologically staged N1 disease without extra-capsular spread may be spared radiotherapy. For other cases, there is overwhelming evidence that the best outcomes are achieved with the dual-modality treatment of surgical resection and adjuvant radiotherapy.Reference Veness, Porceddu, Palme and Morgan1147,Reference Porceddu, Bressel, Poulsen, Stoneley, Veness and Kenny1148
Immunotherapy
Immunotherapy is effective in palliation, with growing evidence to indicate that long-term cure is achievable. Though rarely fatal, non-melanoma skin cancer is responsible for a great deal of functional morbidity, disfigurement and healthcare costs.
Probably the most significant recent development in the management of cutaneous SCC is the introduction of immunotherapy in the form of cemiplimab. Platinum-based chemotherapy is rarely used, because it is frequently toxic, especially in a frail older adult population, and both overall response rate and duration of response are very poor. Based on the ‘EMPOWER’ (Eliminating Medications Through Patient Ownership of End Results) trial and real world Systemic Anti-Cancer Therapy (‘SACT’) data, cemiplimab offers a well-tolerated, effective treatment for locally advanced or metastatic cutaneous SCC. Whilst there is not a direct comparison, survival and quality of life are likely to be improved with cemiplimab over palliative chemotherapyReference Migden, Rischin, Schmults, Guminski, Hauschild and Lewis955 or best supportive care in the treatment of patients with locally advanced or metastatic cutaneous SCC where curative surgical treatment or radiotherapy is not deemed appropriate or feasible. Cemiplimab is funded on the NHS through the Cancer Drugs Fund.1149 There is early phase II evidence to support the role of neoadjuvant cemiplimab, and further studies are being undertaken; however, currently this is not supported by NICE or the Cancer Drugs Fund.Reference Gross, Ferrarotto, Nagarajan, Bell, El-Naggar and Johnson1150 Decisions regarding the use of immunotherapy must be made in a specialist skin cancer MDT meeting with one or more oncologists in attendance.
Follow up
All patients with cutaneous SCC should be offered a key worker (e.g. a clinical nurse specialist). They should all have a skin and nodal examination after diagnosis (as well as before), and be educated on self-examination and surveillance, and given advice about UV light protection and vitamin D supplementation.
Patients with low-risk, completely excised cutaneous SCC without a previous history of skin malignancy may be discharged at the first follow-up appointment, but other cutaneous SCC patients will require follow up. Patients with high-risk cutaneous SCC should be offered follow-up appointments every 4 months for 12 months, then every 6 months for at least 12 months. Patients with very high-risk cutaneous SCC should be offered follow-up appointments every 4 months for 24 months, then every 6 months for at least 12 months. Patients with a significant risk factor such as immunosuppression should be offered lifelong follow up.
Merkel cell cancer
Recommendations
• Discuss all cases of Merkel cell carcinoma at the specialist skin MDT meeting (evidence-based recommendation (R))
• Wide local excision of Merkel cell carcinoma should be performed with a 1–2 cm margin down to fascia (R)
• Offer sentinel node biopsy to patients who are staged N0 and M0 (node and metastasis negative) at presentation, at the time of wide local excision (R)
• Offer patients adjuvant radiotherapy to the primary tumour site and involved lymph node basins (R)
• Consider immunotherapy for patients with locally advanced or metastatic disease (R)
Merkel cell cancer is a rare cutaneous neuroendocrine tumour. It occurs predominantly in the Caucasian population, and is commonly found on the head and neck (42.6 per cent). There is growing evidence to support an association with infection with Merkel cell polyomavirus. Despite the name of the tumour, there is still discussion in the literature as to the cell of origin, with Merkel cells (mechanoreceptor cells), dermal and epidermal stem cells, and lymphoid progenitor cells all postulated. The extent of disease at presentation (local, nodal and distant) is predictive of five-year survival, and this has been incorporated into the AJCC Cancer Staging Manual (eighth edition)Reference Califano, Lydiatt, Nehal, O'Sullivan, Schmults, Seethala, Amin, Edge, Green, Byrd, Brookland and Washington1138 Merkel cell carcinoma staging (Tables 4–8).
Table 4. Tumour staging for Merkel cell cancer*
* According to the AJCC Cancer Staging Manual (eighth edition).Reference Califano, Lydiatt, Nehal, O'Sullivan, Schmults, Seethala, Amin, Edge, Green, Byrd, Brookland and Washington1138
Table 5. Nodal staging for Merkel cell cancer*
*According to the AJCC Cancer Staging Manual (eighth edition).Reference Califano, Lydiatt, Nehal, O'Sullivan, Schmults, Seethala, Amin, Edge, Green, Byrd, Brookland and Washington1138 †In-transit metastasis: discontinuous tumour distinct from the primary lesion, and located either between it and regional lymph node or distal to the primary lesion.
Table 6. Metastasis staging for Merkel cell cancer*
* According to the AJCC Cancer Staging Manual (eighth edition).Reference Califano, Lydiatt, Nehal, O'Sullivan, Schmults, Seethala, Amin, Edge, Green, Byrd, Brookland and Washington1138
Table 7. Clinical group staging for Merkel cell cancer*
* According to the AJCC Cancer Staging Manual (eighth edition).Reference Califano, Lydiatt, Nehal, O'Sullivan, Schmults, Seethala, Amin, Edge, Green, Byrd, Brookland and Washington1138
Table 8. Pathological group staging for Merkel cell cancer*
* According to the AJCC Cancer Staging Manual (eighth edition).Reference Califano, Lydiatt, Nehal, O'Sullivan, Schmults, Seethala, Amin, Edge, Green, Byrd, Brookland and Washington1138 †Detected on sentinel node biopsy
Initial assessment of patients with Merkel cell carcinoma should include thorough history-taking and examination, including evaluation of the primary site for satellite lesions, dermal seeding and palpable lymphadenopathy. Macroscopic and microscopic nodal involvement is common, and cross-sectional imaging (CT and/or MRI) should be carried out.
The management of Merkel cell carcinoma can be complex, and must be planned and discussed in a specialist skin cancer MDT meeting, and the treatment carried out by core members.
Treatment of the primary tumour is usually surgical with wide local excision, aiming to achieve clear primary lesion surgical margins, followed in most cases by adjuvant radiotherapy. A 1–2 cm margin down to fascia is recommended. Sentinel node biopsy is indicated for patients with a histologically proven primary Merkel cell carcinoma, who are staged N0M0 at presentation, after discussion in a specialist skin cancer MDT meeting. Merkel cell carcinoma is very radiosensitive, and adjuvant radiotherapy is recommended for most primary tumours and for lymph node metastasis. For patients with local advanced disease or metastatic disease, immunotherapy (avelumab) can be highly effective, with a reported objective response rate of 62.1 per cent.Reference Yusuf, McKenzie, Rattani, Tennant, Bumpous and Miller1151
Guidelines for follow up of Merkel cell carcinoma vary, but the National Comprehensive Cancer Network guidelines recommend complete skin and nodal examination every 3–6 months for 3 years, then every 6–12 months thereafter.Reference Bichakjian, Olencki, Aasi, Alam, Andersen and Blitzblau1152
Non-melanoma skin cancer of the eyelids
Non-melanoma skin cancer of the eyelids warrants special mention as it has its own AJCC Cancer Staging Manual (eighth edition)Reference Califano, Lydiatt, Nehal, O'Sullivan, Schmults, Seethala, Amin, Edge, Green, Byrd, Brookland and Washington1138 staging system, and its own very specific anatomical and reconstructive constraints. There is also variation of the types of non-melanoma skin cancer that present on the eyelids compared to other parts of the body. These tumours should be managed through the specialist skin cancer MDT by clinicians appropriately trained in occuloplastic techniques and skin cancer management.
Studies due to report
‘Rational treatment selection for Merkel Cell Carcinoma’ – this is a trial comparing radiotherapy with surgery for Merkel cell carcinoma (in: https://doi.org/10.1186/ISRCTN16290169).
Cemiplimab, an anti-programmed cell death 1 immune checkpoint inhibitor, has shown promise in the treatment of non-resectable advanced cutaneous SCC. It is currently being investigated in a phase 2 clinical trial in a neo-adjuvant setting prior to surgery.Reference Migden, Rischin, Schmults, Guminski, Hauschild and Lewis955
Pre-operative vismodegib has been investigated for downstaging the extent of surgery.Reference Bertrand, Guerreschi, Basset-Seguin, Saiag, Dupuy and Dalac-Rat956
Important research questions
Randomised, controlled trials comparing standard surgical re-excision of high-risk BCC excised with close (less than 1 mm) or involved histological margins versus Mohs micrographic surgery, radiotherapy or no treatment.
Randomised, controlled trials comparing standard surgical excision versus Mohs micrographic surgery for high-risk BCC.
Investigation of the role of sentinel node biopsy in cutaneous SCC.
There is some evidence that immunotherapy may serve as an effective treatment for advanced BCC. Investigation of the role of immunotherapy in advanced BCC may be important, given the limitations of use on the hedgehog pathway inhibitors.
Introduction
Head and neck mucosal melanomas are rare, accounting for only 1–4 per cent of all melanomas,Reference López, Rodrigo, Cardesa, Triantafyllou, Devaney and Mendenhall1153 and are an aggressive subtype of melanocytic malignancies, with a very poor prognosis. They are neural crest-derived malignancies, thought to arise from errors in melanocyte migration to the mucosa of the upper aerodigestive tract during fetal development.Reference Tyrrell and Payne1154 They most commonly arise in the nasal cavity and paranasal sinuses (80 per cent), followed by the oral cavity, pharynx and larynx (in decreasing order of frequency).Reference Ascierto, Accorona, Botti, Farina, Fossati and Gatta1155 The incidence of head and neck mucosal melanoma has remained stable.Reference Patrick, Fenske and Messina1156 There are, however, no clear modifiable risk factors for head and neck mucosal melanoma. Their aetiology is still poorly understood.Reference Holmstrom and Lund1157,Reference Reuter and Woodruff1158
Because of their rarity, there is little evidence to guide management, and consequently few evidence-based clinical guidelines. Five-year survival rates remain very low at 20–28 per cent, with little improvement in modern times.Reference Ascierto, Accorona, Botti, Farina, Fossati and Gatta1155,Reference Bachar, Loh, O'Sullivan, Goldstein, Wood and Brown1159,Reference Paleri, Kerawala, Winter, Robinson, Jarrom and Prettyjohns1160
Presentation and diagnosis
Clinical presentation is dependent on the subsite of the lesion. The most common presenting symptoms for sinonasal head and neck mucosal melanoma are nasal obstruction and unilateral epistaxis, seen in combination in up to 95 per cent of patients.Reference Ascierto, Accorona, Botti, Farina, Fossati and Gatta1155,Reference Crawford, Tron, Ma and Rivers1161,Reference Letievant, Poupart, Ambrun, Colin and Pignat1162
These, and other sinonasal symptoms, are also common presentations of benign sinonasal disease, which may explain why it has been reported that sinonasal head and neck mucosal melanoma has a longer duration of symptoms prior to diagnosis (6.7 months) when compared to oral cavity head and neck mucosal melanoma (3.1 months).Reference McLean, Tighiouart and Muller1163 Patients with oral cavity head and neck mucosal melanoma are often diagnosed by dental practitioners due to pain, bleeding, ulceration, ill-fitting dentures or the identification of incidental pigmented lesions, and present earlier (3.1 months) when compared to patients with sinonasal head and neck mucosal melanoma (6.7 months).Reference McLean, Tighiouart and Muller1163,Reference Guevara-Canales, Gutiérrez-Morales, Sacsaquispe-Contreras, Sánchez-Lihón and Morales-Vadillo1164 More rarely, head and neck mucosal melanoma can present in the oropharynx or larynx, and these cases present in a similar manner to squamous cell carcinoma (SCC) in those sites.Reference Wenig1165
Within the nasal cavity, the lateral nasal wall and septum are the most common subsites.Reference Moreno, Roberts, Kupferman, DeMonte, El-Naggar and Williams1166 The overwhelming majority of oral cavity head and neck mucosal melanomas arise in the upper maxillary alveolar ridge and hard palate.Reference Meleti, Leemans, Mooi, Vescovi and van der Waal1167 It is worth noting that up to 30 per cent of oral head and neck mucosal melanomas are amelanotic.Reference McLean, Tighiouart and Muller1163 On top of this, they can vary in their macroscopic appearance, being described as red or white, as well as characteristic dark brown.Reference Yde, Sjoegren, Heje and Stolle1168
Patient evaluation is performed according to the primary site, as for other upper aerodigestive tract SCCs. Biopsy can be carried out under general or local anaesthetic, depending on the nature of the disease and the patients’ co-morbidities. Attempts at excisional biopsy may compromise further surgical treatment, hence incisional biopsy or sampling is advised. At the time of biopsy, it is important to make an assessment of resectability, and mapping biopsies of adjacent areas may be useful to determine the extent of disease.
Imaging
Imaging is the same as for other cancers arising from the affected site (see Chapter 17 on oral cavity cancer and Chapter 23 on sinonasal cancer). However, to stage potential regional and distant metastatic sites, contrast-enhanced computed tomography (CT) of the neck, chest, thorax, abdomen and pelvis (or positron emission tomography (PET)-CT), and contrast-enhanced magnetic resonance imaging of the brain, are recommended.
Recommendations
• Patients with head and neck mucosal melanoma present with symptoms dependent on the mucosal site of origin. Some lesions may be amelanocytic, particularly in the oral cavity (evidence-based recommendation (R))
• At the time of biopsies, an assessment of resectability should be made, and mapping biopsies of adjacent areas may help make this assessment (good practice point (G))
• Systemic imaging including brain imaging is recommended (R)
Staging
It is important to note that there is no early stage head and neck mucosal melanoma. All disease is classified as stage III or IV. Clinicians should use the most recent Union for International Cancer Control (‘UICC’) TNM Classification of Malignant Tumours, eighth edition (2017) (Tables 1 and 2).Reference Brierley, Gospodarowicz and Wittekind1169
Table 1. TNM classification of head and neck mucosal melanoma*
* According to the TNM Classification of Malignant Tumours Reference Brierley, Gospodarowicz and Wittekind1169
Table 2. Disease staging of head and neck mucosal melanoma*
* According to the TNM Classification of Malignant Tumours.Reference Brierley, Gospodarowicz and Wittekind1169 TNM = tumour–node–metastasis
Pathology
The diagnostic pathology specimen should be reviewed by pathologists with relevant expertise, usually supplemented by molecular pathology reports. It is worth noting that neither tumour thickness (Breslow depth) nor depth of invasion (Clark's level) inform prognosis, and, therefore, these do not need to be documented. It is recommended that molecular analysis for BRAF V600 and KIT mutations is performed routinely, at the time of first diagnosis, in line with both local and national genomic guidelines and pathways, as it may offer patients treatment options in both adjuvant and metastatic settings. Both genes encode for growth kinase proteins that can be targeted by specific therapies. Over time, it is possible that other mutations may be identified which may represent clinically relevant actionable therapeutic targets.
Recommendations
• There is no early stage head and neck mucosal melanoma, which reflects the aggressive nature of the disease (evidence-based recommendation (R))
• Molecular analysis for all known current targetable mutations (currently BRAF and KIT) should be undertaken (R)
• Unlike in cutaneous melanoma, depth of invasion or tumour thickness do not inform prognosis (R)
Treatment
Localised tumours (stage III disease)
Primary surgery
Surgery with the aim of achieving clear margins is still the standard of care and first-line management option for localised tumours. There are no datasets to use for relative indications or contraindications for surgery, however several series demonstrate improved disease-free survival and overall survival where negative margins can be achieved (R0 resection).Reference Konuthula, Khan, Parasher, Del Signore, Genden and Govindaraj1170,Reference Heppt, Roesch, Weide, Gutzmer, Meier and Loquai1171
In the context of head and neck mucosal melanoma of the paranasal sinuses, historically it was thought that the only method to achieve an R0 resection was by radical transfacial resections with subsequent reconstruction. However, some series report that an endoscopic approach gives comparable disease-free survival and overall survival to open approaches.Reference Sayed, Migliacci, Cracchiolo, Barker, Lee and McBride1172 These findings need to be interpreted in the context of the selection bias that results from smaller lesions being amenable to endoscopic resection. The anatomical constraints of the paranasal sinuses and proximity of vital anatomical structures make achieving 5 mm margins impossible in many patients. Equally, en bloc resection may not be possible, and therefore a mosaic approach with marginal biopsies is a widely accepted endoscopic technique.Reference Amit, Na'ara and Hanna1173 With regard to oral cavity, oropharyngeal, hypopharyngeal and laryngeal head and neck mucosal melanoma, traditional surgical techniques used for the resection of upper aerodigestive tract cancers are accepted as the ‘gold standard’.
Multidisciplinary teams (MDTs) need to weigh the morbidity of surgery with the likelihood of achieving clear margins. Surgery that is thought to carry unacceptable morbidity or have a significant negative impact on a patient's quality of life is clearly contraindicated. What is deemed unacceptable morbidity is patient-specific, and these decisions must be made with the patient at the centre of the process. It may be necessary to involve a local skull base MDT in specific cases.
Post-operative radiotherapy
There is a lack of good evidence to support the routine use of post-operative radiotherapy with respect to improvement in disease-free survival or overall survival. At the time of writing, there are four relevant meta-analyses available in the literature. Three demonstrated no improvement in overall survival, but did demonstrate improvements in locoregional control.Reference Li, Yu, Wang, Yan and Jiang1174–Reference Wushou, Hou, Zhao and Miao1176 The final meta-analysis did demonstrate a significant improvement for three-year overall survival and a mild improvement for five-year overall survival.Reference Hu and Yang1177 The MDTs may still wish to consider post-operative intensity-modulated radiotherapy in tumours with traditional features that denote a high chance of recurrence, such as close or positive margins, stage T4 tumours, sinonasal tumours and/or multifocal tumours. It is essential to inform patients that this treatment would expose them to toxicities associated with radiotherapy, with uncertain benefits in overall survival.
Role of neck dissection and sentinel lymph node biopsy in the node-negative neck
In cutaneous melanoma, regional lymph node disease is a significant factor in predicting prognosis as well as treatment modality. However, there is no evidence that, in head and neck mucosal melanoma cases, treatment of regional lymph nodes in the clinical and radiological node-negative (N0) setting or in known regional lymph node metastases has any impact on disease-free survival or overall survival.Reference Amit, Tam, Abdelmeguid, Roberts, Raza and Su1178,Reference Oldenburg and Price1179 One series from Wu et al. demonstrated a slight improvement in five-year overall survival in patients undergoing selective neck dissection versus observation (18 per cent vs 4 per cent, p = 0.001).Reference Wu, Zhong, Li, Song, Guo and Ren1180 However, this series was limited to 67 patients with nodular oral cavity head and neck mucosal melanoma, and it is not clear how patients were selected for selective neck dissection or observation. Other series available did not report any benefit in disease-free survival or overall survival. Therefore, in the recently published UK guidelines commissioned by Melanoma Focus, it is recommended to only consider surgery to regional lymph nodes in the context where the detection of occult metastatic disease would make a patient eligible for adjuvant systemic treatment (standard of care or clinical trial).Reference Nenclares, Ap Dafydd, Bagwan, Begg, Kerawala and King1181 In the context of oral cavity head and neck mucosal melanoma, it is easier to perform sentinel lymph node biopsy given the easy accessibility of the primary site for injection and first-echelon nodes. This is more difficult in head and neck mucosal melanoma of the paranasal sinuses and therefore a selective neck dissection can be considered as an alternative.
Recommendations
• Surgical resection achieving clear margins gives the best chance of cure. It is vital to weigh the likelihood of achieving clear margins with the morbidity and impact on quality of life of any proposed surgery (evidence-based recommendation (R))
• In sinonasal head and neck mucosal melanoma, endoscopic approaches should be used where possible (good practice point (G))
• There is no evidence to support the use of selective neck dissection or sentinel lymph node biopsy in the N0 neck, and these are only advised in the context of upstaging a patient to make them eligible for adjuvant therapy or a clinical trial (G)
• There is no evidence to support the use of post-operative radiotherapy in terms of improved overall survival or disease-free survival. However, post-operative radiotherapy may help with regional control, and MDTs may still wish to consider its use in select patients with adverse pathological features (G)
More advanced tumours (stage IV disease)
Unfortunately, the prognosis for patients with unresectable tumours or metastatic disease is extremely poor (median overall survival of 9.1 months from diagnosis). Immune checkpoint inhibitors such as anti-programmed death 1 (anti-PD-1) (e.g. nivolumab or pembrolizumab) and T-lymphocyte-associated protein 4 (anti-CTLA-4) inhibitors (e.g. ipilimumab) (either in combination or monotherapy) are now considered the standard of care in metastatic melanoma in the UK.Reference Nenclares, Ap Dafydd, Bagwan, Begg, Kerawala and King1181 As with any immunotherapy, the risk of adverse events, particularly in those with co-morbidities, is significant. As outlined earlier, tumourogenic mutations such as BRAF V600 and KIT are targets for additional systemic therapies, which can be considered in the context of failure on immunotherapy or first-line management where urgent symptomatic benefit is required.Reference Nenclares, Ap Dafydd, Bagwan, Begg, Kerawala and King1181 The BRAF V600 mutation is less common in mucosal melanoma (3–15 per cent) compared to cutaneous melanoma (50 per cent).Reference Griffin, Scotto, Josephs, Mele, Crescioli and Bax1182,Reference Karoulia, Gavathiotis and Poulikakos1183 The KIT alteration is identified in 7–17 per cent of mucosal melanomas.Reference Öztürk Sari, Yilmaz, Taşkin, Narli, Şen and Çomoğlu1184,Reference Lyu, Wu, Li, Wang, Song and Ren1185 Objective response rates in mucosal melanoma in single-agent immunotherapy have been quoted as 23–25 per cent and 37.1 per cent in combination treatment, which is less than in cutaneous melanoma (40.9 per cent and 60.4 per cent respectively for single and combination immunotherapy).Reference D'Angelo, Larkin, Sosman, Lebbé, Brady and Neyns1186 It is not fully understood why mucosal melanoma is less responsive to immunotherapy. In addition to the differences in targetable mutations detailed above, fewer patients with mucosal melanoma have high levels of programmed death-ligand 1 (PD-L1) expression (21 per cent vs 35 per cent).Reference D'Angelo, Larkin, Sosman, Lebbé, Brady and Neyns1186 These factors in combination probably explain the difference in objective response rates noted.
In patients where immunotherapy is felt to be contraindicated because of pre-existing auto-immune conditions, immunosuppression or other co-morbidities, cytotoxic chemotherapy can be considered. Prior to the advent of the immunotherapies outlined above, dacarbazine had been the standard of care for metastatic melanoma, and there is ongoing research investigating the use of this in combination with immunotherapy.Reference Apetoh, Ladoire, Coukos and Ghiringhelli1187 In addition, radiotherapy with palliative intent can be considered in select cases where it is thought it would offer symptomatic benefit.
Recommendations
• Combination immunotherapy should be offered as a first-line management option to patients with a good performance status and who are willing to accept the risk of adverse events (evidence-based recommendation (R))
• Monotherapy should be offered to patients considered insufficiently fit or those who find the risks of combination therapy unacceptable (good practice point (G))
• BRAF or KIT targeted agents can be offered to those with appropriate mutations as a first-line management option if urgent symptomatic relief is desired or in the context of failure on immunotherapy (G)
• Chemotherapy and radiotherapy with palliative intent are other treatment modalities that can be considered in select cases (G)
Follow up
Table 3 outlines the follow-up schedule currently advocated in the UK.
Table 3. Follow-up schedule currently advocated in the UK
*Clinical examination should include examination of the upper aerodigestive tract, supplemented by flexible nasendoscopic examination and palpation of the neck. †Imaging should include cross-sectional imaging of the upper aerodigestive tract, neck, chest, abdomen and pelvis (centres with access to magnetic resonance imaging (MRI) may wish to use MRI to image the neck and paranasal sinuses in combination with computed tomography (CT) for the chest, abdomen and pelvis; otherwise, CT can be used alone). ‡Cross-sectional imaging of the brain refers to MRI or CT (with MRI being preferable in centres with access; otherwise, CT is acceptable).
Recurrence
Head and neck mucosal melanoma unfortunately has the tendency to recur distally as part of the natural history of the disease. In these circumstances, the treatment options are limited to immunotherapy and chemotherapy. Salvage surgery is rarely indicated in the recurrent setting. If salvage surgery were to be undertaken, it should be under the same paradigm as primary surgery, addressing the question of whether surgery is likely to achieve clear margins with acceptable morbidity. Palliative debulking surgery could be considered in, for example, an obstructing sinonasal tumour.
Studies due to report
(1) ‘Neoadjuvant Treatment Associated With Maintenance Therapy by Anti-PD1 Immunotherapy in Patients With Resectable Head and Neck Melanoma (IMMUQ)’ (ClinicalTrials.gov Identifier: NCT03313206).
(2) ‘Mucosal Melanoma of Head and Neck in Intensity-modulated Radiotherapy Era’. A prospective phase II study in patients with mucosal melanoma of the head and neck in the intensity-modulated radiotherapy era (ClinicalTrials.gov Identifier: NCT03138642).
Research questions
• Is there a role for adjuvant or neoadjuvant immunotherapy in patients undergoing surgical resection for head and neck mucosal melanoma?
• What other targetable genetic mutations are there that are specific to head and neck mucosal melanoma?
Introduction
Paragangliomas are rare neuroendocrine neoplasms that arise from neural crest derivatives close to sympathetic ganglia. Paraganglioma supersedes the old term, glomus tumour. Paragangliomas of the head and neck arise from parasympathetic glomus cells in close association with the glossopharyngeal and vagus nerves. Paragangliomas also arise along the aorta, and phaeochromocytoma in the adrenal glands.
Paraganglioma of the head and neck was classified by the World Health Organization according to the site of origin as:
• Carotid body
• Jugulotympanic
• Vagal
• Laryngeal
• Miscellaneous
The larynx, nose, paranasal sinuses, parotid gland, pharynx, sympathetic trunk, thyroid and trachea are very rare sites for head and neck paraganglioma.
The population incidence of head and neck paraganglioma is about 1:30 000, accounting for 3 per cent of all paragangliomas. Women are affected more than men. A familial tendency is seen in about 40 per cent of paragangliomas, more than any other tumour in adults.Reference Williams and Tischler1188
It is thought that only about 5 per cent of head and neck paragangliomas are malignant. Malignant paragangliomas are not defined histologically, but through evidence of regional lymph node spread or distant metastases in non-endocrine tissue. Therefore, paragangliomas are classified as primary and metastatic hereafter.Reference Asa, La Rosa and Mete1189 Metastatic paragangliomas must be distinguished from multifocal primary disease.Reference Asa, La Rosa and Mete1189
Under 5 per cent of head and neck paragangliomas secrete catecholamines, usually noradrenaline.
Surveillance, surgery, and radiotherapy including stereotactic radiotherapy should be considered in each case. Published evidence is mainly retrospective. Paragangliomas are best managed at specialist centres.
Presentation and diagnosis
Recommendations
• Examination should include: flexible nasopharyngolaryngoscopy, cranial nerve function assessment and otoscopy (evidence-based recommendation (R))
• Magnetic resonance imaging (MRI) is mandated in suspected head and neck paraganglioma cases (R)
• All patients should have a plasma metanephrines test at the time of diagnosis, and all patients with raised metanephrine levels should be referred to endocrinology (R)
• All patients should be offered genetic testing (R)
Paragangliomas of the neck can present to several clinical specialties with a neck mass (70 per cent), hoarseness (17 per cent) or dysphagia (11 per cent), or based on incidental imaging findings (13 per cent).Reference Valero, Ganly and Shah1190 Rarely, head and neck paragangliomas are vasoactive, and present with agitation, flushing, hypertension and palpitations. Paragangliomas of the temporal bone usually present with hearing loss or pulsatile tinnitus, but may also present with cranial nerve palsy. Head and neck paraganglioma should be considered in the differential diagnosis of a neck mass.
Clinical examination should include flexible nasopharyngolaryngoscopy, cranial nerve function assessment, otoscopy and auscultation for bruit. Audiometry should be performed for temporal bone paragangliomas.
Head and neck paraganglioma may be suspected at ultrasound imaging. The diagnosis is made using contrast-MRI (see Imaging section below). Fine needle aspiration cytology is not usually necessary.
The secretion of metanephrines may cause labile hypertension, and correlates with a poorer prognosis. All patients with head and neck paraganglioma should be tested for metanephrines, either in plasma or urine, following local protocols. When testing in urine, it is important to consider possible inaccuracies in 24-hour urine collection. All patients with raised metanephrine levels should be referred to endocrinology services.
All patients should be offered genetic testing with clinical genetics or endocrinology services.
Imaging
Recommendations
• All patients should undergo contrast-MRI scanning of the head and neck (evidence-based recommendation (R))
• In jugular and tympanic paragangliomas, patients should also undergo computed tomography (CT) of the temporal bone with fine cuts (R)
• All patients should undergo whole-body imaging to examine for multifocal paragangliomas, phaeochromocytoma and distant metastases (R)
• Patients should undergo MRI (of the thorax, abdomen and pelvis)
• Ga-68-Dotatate positron emission tomography (PET)-CT can be considered instead of MRI for whole-body imaging (good practice point (G))
• Metastatic and functional head and neck paragangliomas are staged using the TNM Classification of Malignant Tumours (eighth edition) for phaeochromocytomaReference Brierley, Gospodarowicz and Wittekind1191 (G)
Contrast-MRI of the head and neck is the preferred mode of imaging to diagnose and assess the extent of a head and neck paraganglioma. High-resolution CT imaging of the base of the skull is also necessary in jugular and tympanic paragangliomas.
Of patients, 10–15 per cent have synchronous multifocal, functional or metastatic paragangliomas. Therefore, all patients should have radiological screening with whole-body MRI of the thorax, abdomen and pelvis, as well as the head and neck. Whole-body MRI is preferable to whole-body CT, because whole-body CT adds a significant dose of ionising radiation. Ga-68-Dotatate PET-CT is more accurate in detecting paragangliomas than other forms of functional imaging,Reference Han, Suh, Woo, Kim and Lee1192 and therefore might be the preferred means of detecting metastatic or multifocal paragangliomas.
Staging
Metastatic and functional head and neck paragangliomas are staged using the Union for International Cancer Control / American Joint Committee on Cancer TNM Classification of Malignant Tumours (eighth edition) for phaeochromocytoma, as below.Reference Brierley, Gospodarowicz and Wittekind1191
Primary tumour (T):
• T1: phaeochromocytoma sized less than 5 cm
• T2: functional (sympathetic) paragangliomas of any size, or phaeochromocytoma sized 5 cm or greater
• T3: local extension to surrounding organs
Regional lymph node (N):
• N0: no regional lymph node metastasis
• N1: regional lymph node metastasis
Distant metastasis (M):
• M0: no distant metastasis
• M1a: metastasis to bone only
• M1b: metastasis to distant lymph nodes, liver or lung
• M1c: metastasis to bone and multiple other sites
The tumour–node–metastasis (TNM) prognostic staging groups are given in Table 1. No histology grading system is established.
Table 1. TNM prognostic staging
Carotid body paragangliomas were classified by Shamblin et al. (1971) into three groups assigned intra-operatively,Reference Shamblin, ReMine, Sheps and Harrison1193 so the utility of the classification for planning surgery is very limited:
• Group 1: tumours resectable without significant trauma to the vessel wall or tumour capsule
• Group 2: tumours partially surround the vessel wall and are more adherent to adventitia
• Group 3: tumours have an intimate and adherent relationship to the entire circumference of the carotid bifurcation
Modifications to the Shamblin classification have been proposed in order to reflect the risks of neurological and vascular injury from carotid body paraganglioma surgery, such as subgroups 3a (complete carotid encasement) and 3b (infiltration of the carotid wall irrespective of tumour size).Reference Luna-Ortiz, Rascon-Ortiz, Villavicencio-Valencia and Herrera-Gomez1194 More recently, Mehanna et al. reported and validated a classification system based on the highest anatomical landmark reached by the upper surface of carotid body paraganglioma.Reference Mehanna, Mistry, Golusinski, Di Maio, Nankivell and Snider1195
The British Society of Otology guideline group supported the use of the Fisch classification of temporal bone paragangliomas with some caveats:Reference Lloyd, Obholzer and Tysome1196
• A: Mesotympanic
• B: Tympanomastoid
• C: Carotid canal involvement: (1) limited; (2) vertical portion; (3) horizontal portion; and (4) extension to foramen lacerum with or without cavernous sinus thrombosis
• D: Intracranial extension: (1) extradural; and (2) intradural
Type A and most type B are tympanic paragangliomas because they arise from the tympanic plexus. Large tympanic paragangliomas may involve the jugular bulb and are thus jugulotympanic. Small jugular paragangliomas may extend superiorly without carotid artery involvement and are thus type B. There may not be stepwise progression through classes.Reference Lloyd, Obholzer and Tysome1196
Genetics
About 40 per cent of paragangliomas are caused by pathogenic variants of cancer predisposition genes, especially succinate dehydrogenase (SDH) genes. The probability of a genetic aetiology is increased by a family history, younger age and tumour multifocality.
All patients with a diagnosis of head and neck paraganglioma should be offered genetic testing; there is no maximum age limit for testing.1197 The referral for genetic testing should be made to clinical genetics or subspecialist endocrinology services, and be concurrent with other investigations. The timing of testing depends on the management of the presenting lesion, patient choice and family circumstances. If treatment does not need to be urgent, management can sometimes wait until the results of germline testing become available, which in the UK may be three months. Patients should undergo germline testing prior to discharge from head and neck or skull base services.
In the UK, 11 genes are screened for pathogenic variants (R223 analysis) (Table 2). Variants that are pathogenic, or likely to be pathogenic, are considered to be actionable, whereas variants of unknown significance are not actionable and should not be used for testing the patient's family. Reassessment is undertaken three to five years later in case a variant of unknown significance has become actionable.
Table 2. Genes screened in the UK*
* According to the National Test Directory, October 2021.1197 (MEN1 is also included in the R223 analysis.) H&N = head and neck; AD = autosomal dominant; VHL = Von Hippel–Lindau syndrome
After a pathogenic, or likely pathogenic, variant has been identified in an individual, predictive genetic testing may be offered to unaffected individuals in the family.
In relatives identified as having one of the more pathogenic variants, surveillance is recommended, typically with annual plasma metanephrines tests from the age of 8 years, whole-body MRI every 3 years from age 15 years, and abdominal ultrasound in the years between MRI scans.Reference Lloyd, Obholzer and Tysome1196 There are separate surveillance protocols for individuals with pathogenic variants of NF1 and VHL. In asymptomatic individuals with a maternally-inherited SDHC variant, surveillance is not currently recommended.
The following description is limited to genes in which pathogenic variants are commonly identified in individuals with head and neck paraganglioma.
SDHD
Pathogenic variants in SDHD are the most common germline variant in head and neck paraganglioma. The tumours are rarely metastatic but frequently bilateral. The majority of patients have head and neck paraganglioma, though phaeochromocytoma can occur. The penetrance is estimated at 80–90 per cent by 50 years of age.Reference Ricketts, Forman, Rattenberry, Bradshaw, Lalloo and Izatt1198
SDHD mutations are inherited in autosomal dominant fashion, but the phenotype demonstrates parent-of-origin effect, with tumour development only if the germline mutation is inherited by the paternal line. With rare exception, the overwhelming majority of carriers of maternally-transmitted SDHD mutations remain tumour-free.Reference Bayley, Oldenburg, Nuk, Hoekstra, van der Meer and Korpershoek1199
SDHB
Pathogenic variants in SDHB are more likely to cause phaeochromocytoma and extra-adrenal functioning tumours than head and neck paraganglioma. SDHB pathogenic variants are the most common, accounting for up to 40 per cent of families with familial paraganglioma syndrome. The mean age of diagnosis is about 30 years.
Earlier estimates of the penetrance of SDHB pathogenic variants were about 50 per cent, but genetic testing in families indicated the penetrance is 20–30 per cent.Reference Andrews, Ascher, Pires, Barnes, Vialard and Casey1200 There is a higher rate of metastatic disease associated with tumours in individuals who have an SDHB pathogenic variant.Reference Burnichon, Rohmer, Amar, Herman, Leboulleux and Darrouzet1201 SDHB pathogenic variants have also been associated with: renal cell carcinoma (often with sarcomatoid features), thyroid malignancies and wild-type gastrointestinal stromal tumours.
SDHC
SDHC pathogenic variants are identified less commonly than variants in the other SDHx genes. A UK-wide series of SDHC cases demonstrated that 65 per cent presented with head and neck paraganglioma.Reference Williams, Chatzikyriakou, Carroll, McGowan, Velusamy and White1202 Around 20 per cent of head and neck paragangliomas and 30 per cent of phaeochromocytomas were metastatic. As with the other genes, penetrance is difficult to estimate. In the UK series, the cumulative risk in probands was 94 per cent by the age of 60 years, and 16 per cent in non-probands.
Surveillance and treatment
The rarity, complexity and potential sequelae of head and neck paragangliomas necessitates early referral to specialist centres, to ensure higher case volumes and robust, personalised care.
A diverse range of specialists is involved. The choices of management are: (1) surveillance, (2) surgery, and (3) radiotherapy alone or in combination. The natural history of head and neck paraganglioma is long, and there is a paucity of prospective data to support the optimal choice of management.
Recommendations
• Patients with head and neck paraganglioma should be referred early to a specialist centre with a regional multidisciplinary team (MDT) (desirable (D))
• Surveillance with serial imaging is often appropriate (good practice point (G))
• Active treatment may be by surgery or radiation, alone or in combination (evidence-based recommendation (R))
• Early treatment is appropriate for primary and metastatic tumours causing symptoms, tympanic paragangliomas, functional tumours, and rapid growth (R)
• Stereotactic radiotherapy is optimal first-line radiotherapy treatment for intracranial tumours sized less than 3 cm (R)
• Carotid body paragangliomas sized less than 4 cm, especially if inferior to the level of the hyoid bone, and tympanic paragangliomas classified as Fisch A and B have low risks of post-operative cranial nerve impairment and are most suited to surgical resection if appropriate (G)
• Non-functional and non-metastatic vagal paragangliomas are best managed through surveillance or radiotherapy (G)
• Patients with cranial nerve palsies should receive MDT input from speech and language therapy and dietetics services; liaison with specialist facial nerve palsy and laryngology teams may be needed (essential (E))
• If surgery is offered for jugulotympanic paragangliomas, subtotal resection with or without adjuvant radiotherapy should be considered, to minimise cranial nerve deficit (G)
• Vascular surgery expertise should be involved if there is a chance of carotid artery reconstruction being required or atherosclerosis, and to facilitate sharing of expertise (E)
• Functional tumours should be treated with surgery (R)
• Pre-operative endocrinology input and alpha-adrenoreceptor antagonist therapy are required before surgery for functional head and neck paragangliomas (E)
• Potentially functional non-head and neck paragangliomas should be removed prior to surgery for head and neck paraganglioma, to avoid the risk of hypertensive crisis (E)
General principles of management
Surveillance
Surveillance is an appropriate option for head and neck paragangliomas that are benign, solitary, stable, non-functional, asymptomatic and non-tympanic. In head and neck paragangliomas under surveillance, more than half remain static in size over a five-year period according to a retrospective study of 109 patients and 258 tumours.Reference Jansen, Timmers, Marres and Kunst1203 Among the tumours that grew (44 per cent), the maximum dimension increased by an average of 2.7 mm per year in jugulotympanic paragangliomas, 6.4 mm per year in carotid body paragangliomas, and 11 mm per year in vagal paragangliomas. These data provide a point of comparison when assessing the rate of growth and considering the offer of treatment for head and neck paragangliomas under observation.
By using surveillance in appropriate head and neck paraganglioma cases, the potential complications of treatment may be postponed or avoided. Hence, management has become more conservative. In a series of 103 patients at Memorial Sloan Kettering Cancer Center, the use of surveillance increased from 5 to 36 per cent in new head and neck paraganglioma cases, while surgery reduced from 95 to 55 per cent, and radiation therapy alone increased from 0 to 9 per cent in the initial management of 103 patients between 1986 and 2017.Reference Valero, Ganly and Shah1190 Six of the 14 patients (43 per cent) under observation were subsequently treated by surgery and/or radiation.
Radiotherapy
Radiotherapy may be used alone or in combination with surgery in head and neck paraganglioma. Radiotherapy alters the kinetics of the tumour but does not usually result in disappearance. Local control rates for head and neck paraganglioma are over 90 per cent with radiotherapy.Reference Lieberson, Adler, Soltys, Choi, Gibbs and Chang1204–Reference Suárez, Rodrigo, Mendenhall, Hamoir, Silver and Grégoire1206
Traditionally, external beam radiotherapy has been used mostly in unresectable head and neck paraganglioma cases to stop or limit tumour growth. Symptomatic improvement was achieved in more than 70 per cent of cases in a review of 34 published series.Reference Lieberson, Adler, Soltys, Choi, Gibbs and Chang1204 The typical dose is 45 Gy over five weeks.Reference Persky, Hu, Harrison, Hong and Sessions1207
In jugular and vagal paragangliomas, major complications of external beam radiotherapy occurred in 12 per cent of patients (57 out of 461) at systematic review, including osteoradionecrosis, cranial neuropathy, brain necrosis, deafness and mortality.Reference Suárez, Rodrigo, Bödeker, Llorente, Silver and Jansen1205 The risk of a second malignancy is an ongoing concern from external beam radiotherapy (0.28 per cent).Reference Lalwani, Jackler and Gutin1208,Reference Springate and Weichselbaum1209 Smaller, more precise treatment volumes achieved by integrating three-dimensional imaging (CT, MRI or PET-CT) may decrease complications in surrounding tissue.
Stereotactic radiotherapy is the most precise form of therapeutic radiation. It incurs a low rate of side effects without compromising local tumour control. The risk of a second malignancy is estimated at less than 0.001 per cent.Reference Muracciole and Régis1210
Stereotactic doses of 12–15 Gy, or hypofractionation with 21 Gy in three fractions, or 25 Gy in five fractions, are typical.Reference Persky, Hu, Harrison, Hong and Sessions1207 Intracranial tumours measuring less than 3 cm are the best candidates for stereotactic approaches, whereas larger tumours or those with extracranial spread are better suited to external beam radiotherapy.Reference Persky, Hu, Harrison, Hong and Sessions1207
In functional head and neck paragangliomas, the published evidence does not support radiotherapy over surgical resection because of the delay in reducing catecholamine secretion.
Surgery
The goals of surgery are complete surgical resection and the preservation of important anatomical structures. However, subtotal resection with adjuvant radiotherapy may be preferred in some temporal bone paragangliomas. Any co-existing functional paragangliomas outside the head and neck should be removed prior to surgery for head and neck paraganglioma, to avoid the risk of hypertensive crisis.
Carotid body and vagal paragangliomas are usually approached transcervically. Jugulotympanic paragangliomas are usually approached transtemporally. Local tumour control rates are about 85–95 per cent.
Post-operative cranial neuropathy is common in larger tumours and vagal paragangliomas. The lower cranial nerves (VIIth–XIIth) are at risk from injury with surgery for associated head and neck paraganglioma, causing permanent disability. The site and size of head and neck paragangliomas are the main risk factors. Vagus nerve palsy can be expected as a consequence of any surgical excision of a vagal paraganglioma (including attempted nerve-sparing surgery); this causes aspiration, hoarseness and the need for long-term gastrostomy tube feeding. Horner's syndrome – enophthalmos, ptosis, miosis and anhidrosis – may occur and cause cosmetic embarrassment.
The involvement of allied health professionals, and speech and language therapy and dietetics services is essential, and liaison with specialist facial nerve palsy and laryngology teams may also be needed.
Surgery is preferred for smaller head and neck paragangliomas as single-modality treatment when the risk of complications is low (e.g. tympanic and smaller carotid body tumours) and for secreting tumours. In other cases, the decision-making is more complex and nuanced.
Pre-operative embolisation of head and neck paragangliomas is generally considered safe.Reference White, Link and Cloft1211,Reference Helal, Vakharia, Brinjikji, Carlson, Driscoll and Van Gompel1212 Aeta-analysis of 1326 patients with carotid body tumours identified that pre-operative embolisation was associated with significantly lower intra-operative blood loss and reduced length of surgery. However, there were no differences in the rates of cranial nerve injury, stroke, transient ischaemic attack or length of stay.Reference Texakalidis, Charisis, Giannopoulos, Xenos, Rangel-Castilla and Tassiopoulos1213 In another meta-analysis, pre-operative embolisation did not reduce the rate of re-operation for haematoma.Reference Robertson, Poli, Hobson, Saratzis and Ross Naylor1214 Hence, pre-operative embolisation should be considered, but the evidence does not support this as a didactic recommendation.
Regarding vascular investigation and management of vascular injury, if a tumour encases the carotid artery and the aim of surgery is total excision, a balloon occlusion study of cerebral blood flow may indicate whether resection might cause stroke, but the accuracy is limited. When carotid artery resection is inevitable, pre-operative coiling is an option in patients with a normal balloon occlusion test finding. Pre-operative stenting has also been shown in small series to benefit carotid artery integrity. In a small series of Shamblin grade II–III tumours, stenting preserved carotid artery integrity.Reference Prasad, Laus, Al-Ghamdi, Vashishth, Piazza and Sanna1215 Carotid artery integrity should be preserved where possible.
In all cases where there is a risk of carotid injury, a vascular surgeon should be present or immediately available. Shamblin grade II–III carotid paragangliomas are more likely to require vascular reconstruction, where primary repair, patch graft or reverse saphenous vein graft may be reasonable options, though there are increased rates of haemorrhage, cranial neuropathy, stroke and mortality.Reference Smith, Passman, Dattilo, Guzman, Naslund and Netterville1216–Reference Vogel, Mousa, Dombrovskiy, Haser and Graham1218 In order to facilitate the sharing of expertise, joint working on carotid body tumours may be beneficial.
Functional (secretory) tumours are rare and only account for 3–5 per cent of head and neck paragangliomas.Reference Cass, Schopper, Lubin, Fishbein and Gubbels1219 Surgical excision should be undertaken when possible. These patients require MDT management with endocrinology and anaesthesia inputs, for pre-operative optimisation and planning.
The objectives are to achieve pre-operative normalisation of heart rate, blood pressure and fluid balance. Alpha-adrenoceptor antagonists are the mainstay of treatment. Phenoxybenzamine is considered first-line treatment and should be commenced at least one to two weeks before surgery. Intra-operative continuous blood pressure monitoring, alpha and beta blockade, and careful fluid balance can all minimise the risk of significant intra- and post-operative complications including hypotension.Reference Challis, Casey, Simpson and Gurnell1220
Management of more common head and neck paraganglioma
Carotid body paraganglioma
Surveillance, surgery or radiotherapy are options for most tumours.
Surgery offers a high chance of long-term local cure, with cure rates of around 95 per cent.Reference Lloyd, Obholzer and Tysome1196,Reference Kotelis, Rizos, Geisbüsch, Attigah, Ringleb and Hacke1221,Reference Chen, Li, Liu and Yang1222 In carotid body paragangliomas, the risk of hypoglossal and vagus nerve injury depends on size: 15 per cent risk for those sized less than 4 cm and 40 per cent risk for those over 4 cm.Reference Lloyd, Obholzer and Tysome1196 A systematic review reported a post-operative weighted cranial neuropathy rate of 20 per cent temporarily and 11 per cent permanently.Reference Robertson, Poli, Hobson, Saratzis and Ross Naylor1214 The hypoglossal and vagus nerves are at greatest risk and equally affected. The 30-day post-operative risk of stroke is 4 per cent and the 30-day mortality rate is less than 1 per cent. Using the Mehanna classification, the risk of complications from surgery for small carotid body paragangliomas that extend to the hyoid bone is about 3 per cent, whereas in tumours that extend to the mandibular angle the risk is 21 per cent, and for even larger tumours, 44–70 per cent.Reference Mehanna, Mistry, Golusinski, Di Maio, Nankivell and Snider1195
Adjuvant radiotherapy may be considered after subtotal resection or for metastatic cases.
Hence, surgery for carotid body paragangliomas may be attractive for small tumours and non-genetic cases. Compared to the main alternatives of surveillance or radiotherapy, surgery offers closure for the patient. Radiotherapy gives similarly excellent local control but no closure, with a residual lesion to monitor. For larger tumours, the decision-making is more complex, and the risks and benefits of surveillance, surgery and radiotherapy should be considered.
Jugulotympanic paraganglioma
Tympanic paragangliomas should be surgically resected with hearing preservation. For tumours extending to the jugular bulb, management is as for jugular paragangliomas.Reference Lloyd, Obholzer and Tysome1196
For jugular paragangliomas, surveillance, surgery or radiotherapy are options for most tumours. With local control from radiotherapy equal to that from surgery, radiotherapy, with less morbidity, is the mainstay of treatment.Reference Lloyd, Obholzer and Tysome1196,Reference Hinerman, Amdur, Morris, Kirwan and Mendenhall1223
Surgery should be considered for secretory tumours, and for tumours causing troublesome local symptoms, as an alternative to radiotherapy for growing tumours and for growth after radiotherapy. The lower cranial nerves are at risk. Total resection may be appropriate in some cases, such as when the tumour does not involve the jugular foramen. Otherwise, to minimise the risk of cranial nerve damage, subtotal resection with or without adjuvant radiotherapy is the preferred management method.Reference Wanna, Sweeney, Carlson, Latuska, Rivas and Bennett1224
As with carotid body paragangliomas, the risk of cranial neuropathy increases with tumour size (about 10 per cent for Fisch class C and 40–80 per cent for Fisch class D tumours).Reference Cass, Schopper, Lubin, Fishbein and Gubbels1219,Reference Wanna, Sweeney, Haynes and Carlson1225 Hearing loss is another complication. In Fisch class C/D tumours, cochlear preservation facilitates hearing rehabilitation. In a meta-analysis of 1048 patients, risks included meningitis in 33 per cent, cerebrospinal fluid (CSF) leak in 10 per cent, severe aspiration pneumonia in 7 per cent and stroke in 2 per cent. The risk of major complications was 28 per cent and the risk of post-operative mortality was 2 per cent.Reference Suárez, Rodrigo, Bödeker, Llorente, Silver and Jansen1205 Free-flap reconstruction may reduce CSF leak and improve cosmesis; two-stage resection may also reduce CSF leak.Reference Lloyd, Obholzer and Tysome1196
Meta-analysis showed a five-year control rate after surgery of about 85 per cent.Reference Suárez, Rodrigo, Bödeker, Llorente, Silver and Jansen1205 Recurrence may be salvaged by radiotherapy or re-operation, or observation may be appropriate. The risk of tumour progression is increased with subtotal resection, but there is some evidence that if more than 80 per cent of the tumour is excised regrowth may be less likely to occur.Reference Wanna, Sweeney, Carlson, Latuska, Rivas and Bennett1224
Vagal paraganglioma
Most patients with vagal paragangliomas are minimally symptomatic. Surveillance is often initially appropriate. Surgery should be avoided in vagal paragangliomas if other management options are available. Surgery for vagal paragangliomas will usually cause at least vagal nerve palsy, and consequent severe impairments in articulation, voice and swallowing, with possible long-term gastrostomy tube dependence. In patients with vagal nerve dysfunction pre-operatively, surgery causes further worsening of function. In a systematic review of 226 vagal paragangliomas, tumour control was obtained in 98 per cent of complete resections and in 93 per cent of all resections, while the vagus nerve was functionally preserved in only 11 patients (5 per cent).Reference Suárez, Rodrigo, Bödeker, Llorente, Silver and Jansen1205 The published evidence from radiotherapy treatment is limited, but in the University of Florida experience of head and neck paragangliomas treated with radiotherapy over 35 years, local control in 17 patients was 100 per cent.Reference Hinerman, Amdur, Morris, Kirwan and Mendenhall1223
Metastatic head and neck paraganglioma
Recommendations
• All patients should be discussed and managed at a specialist centre by a head and neck cancer and/or skull base MDT with oncological input, and/or a paraganglioma MDT with oncological input (essential (E))
There is a small subset of extra-adrenal paragangliomas with a propensity for regional lymph node spread and distant metastasis. All paragangliomas have malignant potential; therefore, they are not classified as benign or malignant, but as primary and metastatic.Reference Asa, La Rosa and Mete1189 Metastasis must be distinguished from multifocal primary disease.Reference Asa, La Rosa and Mete1189 Distant metastases occur mainly to bone, lung and liver.Reference Persky, Hu, Harrison, Hong and Sessions1207,Reference Hu and Persky1226,Reference Hu and Persky1227
There is a high rate of metastasis (30–70 per cent) in the familial syndrome associated with SDHB mutation.Reference Suárez, Rodrigo, Mendenhall, Hamoir, Silver and Grégoire1206 The greater risk of aggressive disease and metastasis with SDHB mutation may lower the threshold for offering surgery. Mutations including SDHA, TMEM127 and VHL are also higher risk. The occurrence of metastasis depends on the site of origin: orbital and laryngeal paragangliomas are very rare, but have a 25 per cent risk of metastasis.Reference Persky, Hu, Harrison, Hong and Sessions1207,Reference Hu and Persky1226,Reference Hu and Persky1227 The occurrence of metastasis is less than 19 per cent for vagal paragangliomas, 5 per cent for jugulotympanic paragangliomas and 3 per cent for carotid body paragangliomas.Reference Suárez, Rodrigo, Mendenhall, Hamoir, Silver and Grégoire1206
Metastasis is confirmed by the presence of tumours in lymph nodes or distant metastasis to non-endocrine tissues.Reference Persky, Hu, Harrison, Hong and Sessions1207,Reference Hu and Persky1226,Reference Hu and Persky1227 Hence, malignancy is established by whole-body imaging, with whole-body MRI and/or Ga-68-Dotatate PET-CT.Reference Han, Suh, Woo, Kim and Lee1192
Local and regional tumour control are best achieved by surgical resection and adjuvant radiotherapy. The five-year survival rate is 50–80 per cent for patients with regional lymph node spread, but only 11 per cent for those with distant metastases.Reference Pacheco-Ojeda1228 The disease can relapse up to 20 years after first treatment.Reference Persky, Hu, Harrison, Hong and Sessions1207,Reference Hu and Persky1226,Reference Hu and Persky1227
Chemotherapy, with cyclophosphamide, vincristine and dacarbazine, is considered the standard of care in advanced metastatic paragangliomas. Response rates to chemotherapy of 33–41 per cent have been reported, with a biochemical response of 54 per cent.Reference Ayala-Ramirez, Feng, Habra, Rich, Dickson and Perrier1229,Reference Niemeijer, Alblas, van Hulsteijn, Dekkers and Corssmit1230 Some patients with distant metastases may be candidates for peptide receptor radionuclide therapy with iodine-123-metaiodobenzylguanidine (123I-MIBG) for example, on which there are limited data on efficacy.Reference Kong, Grozinsky-Glasberg, Hofman, Callahan, Meirovitz and Maimon1231 Bone marrow and renal toxicities are potential side effects. Temozolomide may show a response in SDHB-associated metastatic paragangliomas.Reference Hadoux, Favier, Scoazec, Leboulleux, Al Ghuzlan and Caramella1232
Follow up
Recommendations
• Most patients should have specialist long-term follow up including serial MRI scans (good practice point (G))
• At five years, patients with definitively treated, isolated, sporadic, and benign head and neck paraganglioma may be discharged from follow up (G)
• Patients, and any relatives, with relevant gene variants should have life-long specialist follow up, including annual plasma metanephrines testing and whole-body imaging (evidence-based recommendation (R))
The nature and duration of surveillance for head and neck paraganglioma depends on genetic status and treatment; this is summarised in Tables 3 and 4, adapted from the British Skull Base Society consensus guidelines (2020).Reference Lloyd, Obholzer and Tysome1196 Patients should have a review in the clinic to assess for clinical progression and cranial neuropathy, in addition to radiological and biochemical follow up. At five years, sporadic cases treated with total surgical resection can be discharged from follow up, depending on individual circumstances.
Table 3. Follow-up guidelines for sporadic head and neck paraganglioma
MRI = magnetic resonance imaging
Table 4. Follow-up guidelines for genetic head and neck paraganglioma (life-long follow up)
MRI = magnetic resonance imaging
Patients with complications of disease and therapy, such as cranial nerve palsy and hearing loss, should have follow up tailored to their needs in a head and neck cancer or skull base MDT clinic.
Studies due to report
No phase III trials were identified.
Important research questions to be answered
The data on the long-term outcome of surveillance for head and neck paragangliomas remain limited, together with identification of predictive factors.
More information is required on the true frequency of metastasis in head and neck paraganglioma and on survival outcomes.
The published evidence on radiotherapy for vagal paraganglioma treatment is limited.
It is not clear whether genetic mutation status should alter management in unifocal head and neck paraganglioma cases (e.g. whether there should be a lower threshold for offering surgery to SDHB mutation cases with a greater chance of aggressive disease and metastasis).
Appendices
Appendix 1. List of reviewers
David Allin, Andrew Barber, Andrea Beech, Zac Cole-Healy, Richard Gan, Kate Garcez, Jarrod Homer, Hiro Ishii, Chris Jennings, Jemy Jose, Shane Lester, Navin Mani, Emma Molena, Elizabeth Neslon, Eugene Omakobia, Vin Paleri, Susannah Penney, Paul Pracy, Costa Repanos, Sajid Sainuddin, Mantegh Sethi, Ricard Simo, Maria Smith, Sanjai Sood, Frank Stafford, Selvam Thavaraj, Hugh Wheatley, Mark Wilke, Stuart Winter, Billy Wong.
Appendix 2. Management of head and neck squamous cell carcinoma of unknown primary (Chapter 27) – initiative collaborators
Syed Ahsan, Basil Al Omari, Arvind Arya, Panagiotis Asimakopoulos, Zaid Awad, Nigel Beasley, Kirsty Beaton, Sandeep Berry, Conor Bowe, Alex Bowen, Grainne Brady, Malcolm Buchanan, Dawn Carnell, Preetha Chengot, Peter Clarke, Andrew Coatesworth, James Constable, Oliver Dale, Charles Daultrey, Katharine Davies, Cameron Davies-Husband, John de Almeida, Sian Dobbs, Mouli Doddi, Olly Donnelly, Brian Fish, Liam Flood, Montey Garg, Ajith George, Andrew Harris, Chris Hogan, Christopher Holsinger, Jarrod Homer, Robert Hone, David Howe, Jonathan P Hughes, Kate Hulse, Andrew Husband, Shahid Iqbal, Hiro Ishii, Zsuzsanna Iyizoba-Ebozue, Macolm Jackson, Sharan Jayaram, Jean-Pierre Jeannon, Chris Jennings, Jemy Jose, Noemi Kelemen, Shivun Khosla, Emma King, B Nirmal Kumar, Rohit Kumar, Shane Lester, Peter Lion, Zi Wei Liu, Gitta Madani, Jaiganesh Manickavasagam, Marcos Martinez Del Pero, Oliver McLaren, Kris Milinis, George Mochloulis, Jennifer Montgomery, James Moor, Iain Nixon, Enyi Ofo, James O'Hara, Oladejo Olaleye, Nashreen Oozeer, Nimesh Patel, Claire Paterson, Jo Patterson, Imran Petkar, Vinod Prabhu, Paul Pracy, Mark Puvanendran, Costa Repanos, Max Robinson, Bindy Sahota, Ann Sandison, Mehmet Sen, Ketan Shah, Neil Sharma, Somiah Siddiq, Ricard Simo, Salil Sood, Dev Srinivasan, Paul Stimpson, Mrinal Supriya, Arun Takhar, Taran Tatla, Racheal Thomas, Anna Thompson, Paul Tierney, Athena Togo, Tommi Tornari, Ben Tudor-Green, Kishan Ubayasiri, Ramzan Ullah, Navdeep Upile, Jagdeep Virk, David Walker, Giles Warner, Hugh Wheatley, Mark Wilkie, Richard Williams, Andrew Williamson, Christina Wilson, Stuart Winter.
Open access
