Does the probability of receiving placebo influence clinical trial outcome? A meta-regression of double-blind, randomized clinical trials in MDD

Abstract

Substantial and highly variable placebo response rates represent a major obstacle to antidepressant development in major depressive disorder (MDD). However, whether the likelihood of receiving active treatment or placebo, a proxy of the degree of expectation of improvement, may itself influence clinical trial outcome is unclear. The goal of this work was to examine whether the probability of receiving placebo influences clinical trial outcome antidepressant MDD trials. Medline/Pubmed publication databases were searched for randomized, double-blind, placebo-controlled trials of antidepressants for adults with MDD. 146 manuscripts involving 182 clinical trials were pooled (n = 36,385). Pooled response rates for drug and placebo were 53.8% and 37.3%. A meta-regression (random-effects) established that the probability of receiving placebo, year of publication, and baseline severity were independent predictors of the risk ratio of responding to antidepressants versus placebo. Specifically, a greater probability of receiving placebo, greater baseline severity and an earlier year of publication predicted greater antidepressant-placebo “efficacy separation”. Fixed versus flexible dose design, trial duration and population age did not influence clinical trial outcome.

Introduction

Major depressive disorder (MDD) is a prevalent illness that is often chronic, and associated with significant morbidity, mortality, functional impairment, and patient suffering. Antidepressant drugs, along with certain forms of psychotherapy, have long been the mainstay of treatment for MDD. Historically, several factors have been proposed as potential obstacles to drug development in MDD, including the relatively modest overall efficacy of all contemporary agents, our limited understanding of the underlying biological mechanisms contributing to MDD, the likely variability in underlying pathophysiology of the illness, and our relative lack of precision when assessing clinical improvement in MDD trials. Among these, the presence of a substantial and variable (i.e. difficult to predict) placebo-response is perhaps the best studied (Walsh et al., 2002). Substantial and variable placebo response rates in clinical trials for MDD can hinder the antidepressant development process by increasing the likelihood of obtaining an equivocal finding in studies designed to demonstrate the efficacy of a given treatment (i.e., resulting in “false-negative” pivotal trials). Unfortunately, however, to date only a handful of studies have systematically examined the effects of various elements of study design including the choice of primary outcome measure (Carmody et al., 2006, Khan et al., 2002a, Khan et al., 2004a, Faries et al., 2000), the presence and type of placebo lead-in period (Trivedi and Rush, 1994, Faries et al., 2001, Lee et al., 2004), severity of depression at baseline (Khan et al., 2002b, Khan et al., 2004b, Khan et al., 2007, Stein et al., 2006), dosing schedule (i.e., fixed versus flexible dosing) (Khan et al., 2003), as well as the dispensing of concomitant medications during the trial (Wernicke et al., 1997), on placebo response rates and/or clinical trial outcome.

It has previously been reported that a greater degree of expectation of improvement at baseline resulted in a greater probability of subsequently responding to treatment (Sotsky et al., 1991). However, whether this relationship can be attributed to either the “true” antidepressant or placebo response is unclear. In addition, it is equally unclear whether the likelihood of receiving active treatment or placebo, a proxy of the degree of expectation of improvement, may itself influence clinical trial outcome. In fact, in a preliminary work, Khan et al. and Stein et al. suggested an inverse relationship between the number of treatment arms, a proxy of the probability of receiving placebo, and antidepressant–placebo separation in clinical trials for depression (Khan et al., 2004b, Stein et al., 2006). Identifying elements of study design that can predict both the magnitude of the placebo response as well as overall clinical trial outcome can lead to the design of more effective clinical trials which, in turn, can decrease the likelihood of effective treatments being erroneously deemed as ineffective. Therefore, the goal of this work was to confirm and extend previous work by Khan et al and Stein et al by examining whether the probability of receiving placebo influences clinical trial outcome in double-blind, randomized clinical trials in major depressive disorder MDD. In order to accomplish this while controlling for confounding from other elements of clinical trial design, traditional meta-analytic and meta-regression techniques were employed.