Loo et alReference Loo, Glozier, Barton, Baune, Mills and Fitzgerald1 stated ‘ … adequately dosed racemic ketamine [i.e. cohort 2, not cohort 1] was shown to have superior antidepressant efficacy to an [psycho] active [placebo] control drug, with the proportional increase in remission comparing favourably to that of studies of intranasal esketamine tested against a non-active placebo’. An indirect comparison such as this requires that the data being compared are obtained from studies using fundamentally similar methods and similar populations.Reference Phillippo, Ades, Dias, Palmer, Abrams and Welton2 In this regard, the study design of the Ketamine for Adult Depression Study (KADS)Reference Loo, Glozier, Barton, Baune, Mills and Fitzgerald1 and that of the esketamine TRANSFORM-2 study,Reference Popova, Daly, Trivedi, Cooper, Lane and Lim3 which the authors cited for comparison, differed substantially on variables known to affect treatment response (e.g. baseline illness severity, comparator). For example, patients in KADS cohort 2 had less severe baseline depressive symptoms per the Montgomery–Åsberg Depression Rating Scale (MADRS) (baseline mean total score: 30.3, midazolam/antidepressant; 28.9, ketamine/antidepressant) compared to those in TRANSFORM-2 (37.3, inactive placebo/antidepressant; 37.0, esketamine/antidepressant). Despite these differences in baseline MADRS scores, the authors stated that ‘participants in this study [had] a higher level of treatment resistance … than the [TRANSFORM-2] pivotal study of Popova et al’,Reference Popova, Daly, Trivedi, Cooper, Lane and Lim3 which they based on a 24% failure rate to electroconvulsive therapy, an exclusion criterion of TRANSFORM-2.
In the KADS,Reference Loo, Glozier, Barton, Baune, Mills and Fitzgerald1 subcutaneous ketamine was compared to psychoactive placebo (subcutaneous midazolam), each combined with ongoing oral antidepressant initiated ≥4 weeks before study enrolment), whereas esketamine nasal spray was compared to placebo nasal spray, each combined with newly initiated oral antidepressant, with fixed-dose titration to the maximum labelled dose, in TRANSFORM-2.Reference Popova, Daly, Trivedi, Cooper, Lane and Lim3 Thus, patients enrolled into the KADS knew they would continue an antidepressant to which they had not responded and had a 50% likelihood of receiving an anxiolytic drug that lacked antidepressant efficacy, whereas all esketamine-treated patients knew they would receive a new oral antidepressant that they had not previously failed, with resulting differences in expectation bias likely. Furthermore, titration to maximum labelled dose of oral antidepressant in TRANSFORM-2 increased the likelihood of improvement in MADRS score from baseline in the control arm and decreased the likelihood of achieving larger differences between the placebo and esketamine arms. These differences in dosing of oral antidepressant may account for, at least in part, differences in outcomes (e.g. remission rate [defined by MADRS ≤12]: 4.1% and 21.6% in the psychotropic placebo control and ketamine arms, respectively, of the KADS, and 31.0% and 52.5% in the placebo control and esketamine arms, respectively, of TRANSFORM-2).
In the Discussion, Loo et al state ‘these results appear compatible with a prior meta-analysis,Reference Bahji, Vazquez and Zarate4 indicating that racemic ketamine showed a larger treatment effect than esketamine’. However, several research groupsReference Drevets, Popova, Daly, Borentain, Lane and Cepeda5 have noted limitations of the meta-analysisReference Bahji, Vazquez and Zarate4 cited as supportive, most notably that the meta-analysis included data from trials that used markedly differing study designs and patient populations.
Taken together, disparities in study design, including patient population and control arms, make indirect comparisons of efficacy between esketamine and ketamine unfounded. With regard to such comparisons, we agree with Loo et alReference Loo, Glozier, Barton, Baune, Mills and Fitzgerald1 in their statements that ‘ … clearly uncertainty remains on comparative efficacy and large, direct, randomized comparisons are required’ and that ‘future research questions include head-to-head comparisons of racemic, R- and S-ketamine … ’.
Data availability
Data availability is not applicable to this article as no new data were created or analysed in this study.
Author contribution
All authors (G.J., S.S. and J.K.) contributed to the writing and review of this commentary.
Funding
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Declaration of interest
At the time of submission all authors were employees of Janssen-Cilag Pty Ltd (a part of the Janssen Pharmaceutical Companies of Johnson & Johnson, which developed and markets esketamine nasal spray (Spravato®)) and are stockholders of Johnson & Johnson.
eLetters
No eLetters have been published for this article.