Objective: The authors sought to determine study design factors that may influence clinical trial outcome in augmentation/combination trials for antidepressant partial responders/nonresponders with major depressive disorder (MDD) and to examine whether the use of a prospective treatment phase (lead-in) to assess antidepressant nonresponse may result in a better chance to detect a drug-placebo separation in such trials.
Data Sources: MEDLINE/PubMed publication databases were searched for randomized, double-blind, placebo-controlled trials of adjunctive pharmacologic strategies for antidepressant partial responders/nonresponders with MDD. The search term depression was successively cross-referenced with the terms augmentation, adjunct, and adjunctive to identify pertinent trials. (The search was limited to articles published between January 1980 and October 2010.)
Study Selection: Thirty-five articles involving 40 adjunctive drug versus placebo comparisons were pooled (n = 4,676). Final inclusion of articles was determined by consensus between the authors.
Data Extraction: Data extracted included whether there was a lead-in phase and, if so, the drugs, the doses, and the total duration of the lead-in phase. Additional data extracted included the number of patients enrolled, patient characteristics, methods used to define treatment resistance, drug dosages, duration of the adjunctive trial, response and remission rates, and rates of discontinuation for any reason and for adverse events.
Results: The risk ratio of responding to the adjunctive drug versus placebo was not influenced by any of the study design factors analyzed (probability of receiving placebo, year of publication, severity of depression at baseline). Meta-regression analysis yielded no significant difference in the risk ratio of responding and remitting to the adjunctive drug versus placebo between studies that did versus did not include an antidepressant lead-in phase. However, pooled response/remission rates for adjunctive drug and placebo were statistically significantly lower in trials that did versus did not include a lead-in phase (response rates: for adjunctive drug, 42.6% vs 47.4%, respectively, P = .014; for adjunctive placebo, 29.7% vs 36.2%, respectively, P = .002; remission rates: for adjunctive drug, 31.0% vs 37.3%, respectively, P = .003; and adjunctive placebo, 18.1% vs 24.7%, respectively, P = .001).
Conclusions: These results suggest that the choice to use historical data only to define treatment resistance prior to patient enrollment and randomization rather than requiring patients to first undergo a prospective lead-in phase can be a reasonable and evidence-supported approach to design effective clinical trials on augmentation/combination strategies for partial responders/nonresponders with MDD.
J Clin Psychiatry 2012; 73(5): 676-683
© Copyright 2012 Physicians Postgraduate Press, Inc.