INTRODUCTION
Hepatitis C virus (HCV) infects millions of people worldwide and is a leading cause of cirrhosis and liver cancer [Reference Messina1]. Chronic infection is associated with hepatic inflammation and fibrosis [Reference Antonov2]. The combination of pegylated interferon alpha (PEG-IFN-α) and ribavirin is the approved standard, effective and well accepted treatment for chronic hepatitis C [Reference Hoofnagle and Seeff3, Reference Parruti4]. However, the use of ribavirin is controversial in haemoglobinopathic patients because of the haemolytic complications; therefore, interferon alpha (INF-α) monotherapy is used as the best treatment for haemoglobinopathic patients [Reference Mirmomen and Alavian5]. New antiviral drugs have not been approved yet for the treatment of patients with haemoglobinopathy.
INF-α induces the JAK/STAT pathway, which up-regulates genes with antiviral effects against HCV [Reference Samuel6, Reference Lin7]. The essential objective of HCV therapy is to cure the infection leading to the elimination of detectable circulating HCV after stopping the treatment. Rapid virological response (RVR) is defined as undetected HCV RNA level at week 4 of treatment [Reference Poordad, Reddy and Martin8], while sustained virological response (SVR) is defined as an undetectable HCV RNA level 24 weeks after the end of treatment. HCV genotype has been reported as an important predictor in determining SVR in HCV patients. Treatment has shown greater efficacy in genotypes 2 and 3 than in genotypes 1 and 4 [Reference Martin-Carbonero9, Reference Legrand-Abravanel10]. Genotype 4 is common in the Middle East with an SVR achievement of up to 70% [Reference Legrand-Abravanel10, Reference Kamal11]. The SVR rate is about 50–79% for HCV genotype 1 and 75–94% for HCV genotypes 2 and 3 with the advised regimens, in Western and Eastern countries, respectively [Reference Huang12]. This indicates that HCV embodies genetic diversity which demands association study of RVR and SVR with HCV genotypes at the country level.
Moreover, genome-wide association studies (GWAS) have revealed a single nucleotide polymorphism (SNP) located upstream of IL-28B (also called interferon-lambda 3) to be associated with SVR [Reference Tanaka13, Reference Ge14]. The SNP plausibly represents a functional variant influencing response of HCV to IFN-α and therefore it also needs to be studied at the country level.
As inference of positive predictive factors of SVR in HCV patients is of clinical importance [Reference Antonov2], several predictors of SVR were analysed in this study. The study was designed specifically to study HCV treatment effectiveness using PEG-INF-α in patients with mixed HCV genotype and IL-28B polymorphism, suffering from haemoglobinopathy (thalassaemia or sickle cell anaemia) in the Iraqi population.
MATERIAL AND METHODS
Ethical approval
The study protocol was approved by the Ethics and Research Committee of Azadi Teaching Hospital and the School of Medicine, University of Duhok. Written informed consent was obtained from the participants (or their guardians when aged <18 years) of this study.
Patients and antiviral treatment
We recruited all patients with thalassaemia or sickle cell anaemia and HCV who were referred to the infectious disease unit in Azadi Teaching Hospital, Duhok, Kurdistan Region, Iraq during the period from June 2011 to June 2012. In this period, 47 patients visited the unit and all were treated with recombinant interferon monotherapy (recombinant PEG-IFN-α-2a) at a dose of 180 µg/1·73 m2. No patients were treated with PEG-IFN plus ribavirin. During treatment, RVR, early virological response (EVR) (as defined by >2 log fall in the viral load by week 12) and as well as post-treatment SVR rates were determined. Three patients did not attend for treatment and one patient refused to continue treatment due to severe myalgia, those patients were therefore excluded from further study. All patients infected with HCV genotypes 1 and 4 received treatment for 48 weeks, while patients infected with genotypes 2 and 3 received treatment for 24 weeks. All patients were followed up by measuring the viral load, alanine transaminase (ALT) and aspartate transaminase (AST) levels at weeks 4, 12, 24 (for genotypes 2 and 3) and week 48 (for genotypes 1 and 4). These investigations were performed at week 24 after stopping treatment. All the patients who achieved SVR were tested for HCV RNA by reverse transcriptase–polymerase chain reaction (RT–PCR) 6 months later (48 weeks after ceasing treatment).
HCV quantification and genotyping
Automated nucleic acid purification was performed using the Qiagen QIAsymphony (Qiagen, USA). The Qiagen RT–PCR assay was performed using the artus HCV QS-RGQ kit (Qiagen) and was run on a Rotor-Gene Q thermocycler. The value for the 95% lower limit of detection given by the manufacturer was 36·2 IU/ml for HCV. The linear ranges of quantification were 67·6 to 1·8 × 107 IU/ml. HCV genotyping was performed by GEN-C 2.0 reverse hybridization strip assay (Nuclear Laser Medicine, Italy). The assay discriminates between HCV genotypes on the basis of variations in the 5′-UTR and core regions.
DNA extraction and IL-28B genotyping
Frozen whole blood was used for IL-28B genotyping after genomic DNA extraction using QIAamp DNA Blood Mini kit (Qiagen). IL-28B genotyping was performed by PCR with confronting two-pair primers (CTPP) PCR as described previously [Reference Ferreira15]. The primer sequences used for IL-28B were (F1: GACGAGAGGGCGTTAGAGCG, R1: GGAGTGCAATTCAACCCTGGTTCG; F2: GAGCTCCCCGAAGGCGT, R2: AACGCAGGCTCAGGGTCAAT) [Reference Ferreira15]. The reactions consisted of a total volume of 50 µl containing 1·5 U Hotstart DNA polymerase (Takara, China), 0·5 µm of primers F1, F2, and R2, 0·75 µm of primer R1, 1·0 µm dNTP and 50 ng genomic DNA. Thermocycler conditions were as follows: preheating at 95 °C for 5 min followed by 40 cycles of 95 °C for 30 s, 69·2 °C for 1 min, and 72 °C for 3 min. PCR products were separated by standard electrophoresis on 2% agarose gels containing ethidium bromide. The homozygous C/C gives bands of 400 bp and 312 bp, the homozygous T/T gives bands of 400 bp and 128 bp, and the heterozygote genotype yields bands of 400 bp, 312 bp, and 128 bp.
Statistical analysis
Obtained data were completely anonymized before statistical analysis. Univariate and multivariate analysis was conducted by logistic regression. All variables that achieved a P value <0·2 were included in the multivariate study. Step-down regression was used where the variable with the lowest partial correlation was eliminated first. Variable confounder was determined if its deletion changed the coefficient by 20% of a statistically significant predictor. The statistical analysis was conducted with SPSS version 21 (SPSS Inc., USA).
RESULTS
Analysis was performed for patients [males (n = 27, 57·45%); females (n = 20, 42·55%)] suffering from HCV (genotypes 1–4) and treated with recombinant PEG-INF-α-2a monotherapy (Table 1). The ages of the subjects ranged from 6 to 36 years (mean 14·4 years).
Table 1. The predictive variables in accordance with the response achievement (SVR)
SVR, Sustained virological response; RVR, rapid virological response.
SVR achievement
Of our patients, 53·49% (23/43) achieved SVR as defined as undetectable levels of viral load 24 weeks after the end of the treatment. All the patients who achieved SVR tested negative for HCV RNA via RT–PCR results 24 weeks later (at 48 weeks post-treatment).
The association between different factors and SVR achievement
Univariate analysis
The data of SVR achievement were stratified according to initial viral load, ALT, AST, RVR, gender, IL-28 polymorphism, HCV genotype, blood group and ultrasound findings. Univariate logistic regression analysis was conducted to study the association between individual factor and SVR (Table 2).
Table 2. The relationship between response achievement (SVR) and the study independent variables by univariate regression analysis
OR, Odds ratio; CI, confidence interval; SVR, sustained virological response; ALT, alanine transaminase; AST, aspartate aminotransferase; RVR, rapid virological response;
RVR and EVR
All the patients who achieved EVR achieved SVR. RVR for all subjects was 55·81% while SVR was 53·49%. SVR in patients that achieved RVR was 82·61% (P = 0·0004) (Table 3).
Table 3. Significance of sustained virological response (SVR) achievement in HCV genotypes
CI, Confidence interval.
A significant association was found between genotype 3 and SVR achievement. One patient had genotype 2 infection. RVR was achieved but not SVR.
* SVR achieved = (no. of patients/total no. of patients with specified genotype).
IL-28B polymorphisms
Thirty-two (74%) out of 43 of our sample were genotyped as CC and 20/32 (62%) of these CC samples achieved RVR and along with SVR. Eleven (25%) out of 43 of our samples were typed as TT/CT and 3/11 (27%) of these subjects achieved RVR and SVR. A significant difference was found between CC group and TT/CT group in SVR achievement (P = 0·04) (Table 2).
HCV genotype
Twenty-three (53%) out of 43 of our samples were typed as HCV genotype 4, 10/43 (23%) as genotype 1, 9/43 (20·9%) as genotype 3 and 1/43 (2·3%) as genotype 2. SVR was achieved in 88·89% of the patients infected with HCV genotype 3 (P = 0·006) (Table 3) (Fig. 1).
Fig. 1. The RNA log10 IU/ml changes in subjects chronically infected with HCV genotype 1 (a), genotype 2 (b), genotype 3 (c) and genotype 4 (d) following PEG-IFN monotherapy.
Baseline viral load, ALT and AST
Regression analysis was conducted for elucidating independent associations between each of ALT, AST or viral load prior to the initiation of treatment and SVR. There was a significant association between ALT levels at baseline and the attainment of SVR (mean ± s.d.: 67 ± 68 vs. 144 ± 83·7, P = 0·025) (Fig. 2) while no significant association was found between AST and SVR achievement (83 ± 53·4 vs. 114 ± 55, P > 0·05). In addition, a significant association was found between the levels of viral load and SVR achievement (log mean ± s.d.: 4·5 ± 0·8 vs. 5·2 ± 0·83, P = 0·004) (Fig. 3).
Fig. 2. Boxplot depicting serum levels of alanine transaminase (ALT) according to sustained virological response (SVR) achievement. ALT levels were measured before starting the treatment and stratified according to SVR achievement. Significant differences were found in the levels of ALT between the group of patients who achieved SVR and those who did not achieve SVR.
Fig. 3. Boxplot depicting viral load log according to sustained virological response (SVR) achievement. Viral load log levels were measured before starting the treatment and stratified according to SVR achievement. Significant differences were found in the levels of viral load log between the group of patients who achieved SVR and those who did not achieve SVR.
Multivariate analysis
Logistic regression was used to perform the multivariate analysis. In this test a dichotomous outcome is predicted by one or more variables. We used logistic regression to model the effects of RVR, AST, ALT, viral load, IL-28 polymorphism and genotype on SVR achievement. It was found that RVR is the only independent factor associated with SVR achievement (P = 0·00) (Table 4). None of the variables could achieve the 20% reduction in the coeffiecient of relationship and therefore could not be regarded as a confounder.
Table 4. The relationship between response achievement (SVR) and the study independent variables by multivariate regression analysis
OR, Odds ratio; CI, confidence interval; ALT, alanine transaminase; AST, aspartate aminotransferase; RVR, rapid virological response; SVR, sustained virological response.
DISCUSSION
The relationship between SVR achievement and the baseline characteristics blood group, hepatomegaly, gender, genotype, IL-28B polymorphism and viral load was assessed in the recruited patients.
Several previous clinical studies [Reference Di Marco16] showed that IFN monotherapy administered for 6–15 months induced a sustained biochemical response and SVR in 40–50% of thalassemia patients with HCV-related chronic hepatitis. In a study conducted in Iran comprising thalassaemia patients SVR was achieved, ranging from 24% for IFN monotherapy to 51% in patients receiving combined therapy [Reference Mirmomen and Alavian5, Reference Tabatabaei17]. In our study SVR was achieved in about 50% of our patients. Our results are encouraging; however, our data mean that about half of our patients failed to achieve SVR. This warrants clinical trials of the developing direct-acting antiviral drugs (DAA) in patients with haemoglobinopathy.
Previous studies conducted in the central region in Iraq showed that 35% of the samples typed as genotype 4 while 50% of the recruited samples typed as genotype 1 [Reference Al-Kubaisy and Habib18]. No genotypes 2 and 3 were reported in that study. In our study, 53% of our samples typed as HCV genotype 4 followed by 23% for genotype 1, 20·9% for genotype 3 and 2·3% for genotype 2. A previous study conducted in Iran showed that the majority of chronically HCV-infected Iranian patients were infected with HCV genotype 1 (47%) followed by genotype 3 (36%) [Reference Samimi-Rad19]. Previous studies from Saudi Arabia, Kuwait [Reference Koshy20] and Yemen [Reference Ohno21] found that genotype 4 was the most prevalent HCV genotype in those countries. It is probable that differences in race, routes of transmission, and different socioeconomic factors might explain this variation.
In agreement with other studies [Reference Di Marco16], RVR proved to be the strongest independent predictor of SVR. No RVR achievement is an unfavourable marker for SVR. It was previously shown that HCV genotype, pre-treatment viral load and IL-28B polymorphisms played a significant role in SVR achievement [Reference Tanaka13, Reference Ge14, Reference Suppiah22–Reference Rivero-Juarez24]. In our study, most patients with IL-28B CC genotype and those infected with HCV genotype 3 achieved SVR. In agreement with previous studies [Reference Di Marco16], we found a significant association between ALT or viral load levels and SVR achievement while no association was found between AST levels and SVR achievement. However, when multivariate analysis was performed, RVR was the only factor that showed association with SVR achievement. This might be due to the small sample size used in this study. Until other studies are performed with larger sample sizes, these results can be used to counsel patients on the likelihood of achieving SVR. As the new DDAs are expensive and not available for all patients, these predictive factors can be used as parameters for selecting patients for these new regimens. No association was found between gender, ABO blood group or ultrasound findings and SVR achievement.
To conclude, little is known about the response rate to PEG-INF treatment of HCV in Iraq. This study elucidated whether early changes in HCV RNA levels during treatment with PEG-IFN could be used to accurately predict treatment response and identify the factors that influence response rate to antiviral therapy. Therapy is very expensive in Iraq and not all patients can afford it, therefore early discontinuation of treatment in non-responders could avoid the expense and inconvenience of continuing unnecessary treatment allowing recourse to other alternatives such as DAA.
ACKNOWLEDGEMENTS
The authors thank all the doctors who helped in data extraction.
DECLARATION OF INTEREST
None.
