Duloxetine versus escitalopram and placebo in the treatment of patients with major depressive disorder: onset of antidepressant action, a non-inferiority study

Objective:The goal of a non-inferiority study is to test whether a new treatment has at least as much efficacy as an established treatment(1). The purpose of this non-inferiority study was to compare the speed of onset of antidepressant efficacy for duloxetine (a dual serotonin and norepinephrine reuptake inhibitor) and escitalopram (a selective serotonin reuptake inhibitor). Research design and methods: This was a randomized, double-blind, placebo- and active comparator-control led study, in which patients (>= 18 years) meeting DSM-IV criteria for Major Depressive Disorder (MOD) received duloxetine 60 mg once daily (QD; N = 273), escitalopram 10 mg OD (N = 274), or placebo (N = 137) for 8 weeks. The primary objective was to compare the onset of antidepressant efficacy, by testing the hypothesis that the percentage of duloxetine-treated patients achieving onset criteria at Week 2 was not inferior to that in the escitalopram group. Main outcome measures: Onset of efficacy was defined as a 20% decrease from baseline on the 17-item Hamilton Rating Scale for Depression (HAMD(17)) Maier subscale that was maintained or exceeded at all subsequent visits. Results: Probabilities of meeting onset criteria at Week 2 for duloxetine- and escitalopram-treated patients were 42.6% versus 35.2%, respectively (treatment difference = 7.4%; 95% confidence interval, -1.3% to 16.2%; p = 0.097). Both drugs showed significant improvement compared with placebo (p <= 0.05) on the primary efficacy measure (Maier subscale) at Week 1 and endpoint (Week 8). No differences were found between duloxetine, escitalopram, and placebo rates of remission or response at 8 weeks. Adverse events that occurred significantly more frequently among duloxetine-treated patients when compared with those receiving escitalopram were nausea, dry mouth, vomiting, yawning, and irritability. The rate of discontinuation due to adverse events did not differ significantly between treatment groups. Limitations: Given the difficulties in constructing appropriate dose comparisons, the results of this study should be interpreted specific to the doses tested and not extrapolated to the drug as a whole. This study employed a fixed-dose design; flexible-dose designs are more likely to find a difference between antidepressants and placebo. Conclusion: In this study, both duloxetine and escitalopram showed significantly greater improvement on the primary efficacy measure than placebo over the 8-week acute treatment period, while no differences were observed between drugs or between drugs and placebo on response and remission rates at 8 weeks. Escitalopram at a starting dose of 10 mg QD was better tolerated than duloxetine at a starting dose of 60 mg OD. This study met its pre-defined primary objective of assessing if duloxetine was non-inferior to escitalopram in antidepressant onset efficacy, and the results show that duloxetine is at least as fast as (non-inferior to) escitalopram.