Early Improvement in Depressive Symptoms With Desvenlafaxine 50 mg/d as a Predictor of Treatment Success in Patients With Major Depressive Disorder

Abstract

This post hoc analysis assessed the predictive value of improvement in depressive scores at early time points for treatment outcomes at week 8 in patients with major depressive disorder treated with desvenlafaxine 50 mg/d or placebo. Pooled data from 6 double-blind, fixed-dose studies in adult patients with major depressive disorder. Patients were randomly assigned to desvenlafaxine or placebo. Primary end point was change in 17-item Hamilton Rating Scale for Depression (HAM-D17) scores from baseline to week 8 (or last observation carried forward). Optimal thresholds of improvement (percent change from baseline HAM-D17) at weeks 2 and 3 for predicting 4 levels of treatment success (≥ 45%, ≥ 50%, and ≥ 65% decrease from baseline HAM-D17, HAM-D17 ≤ 7) at week 8 (last observation carried forward) were determined using receiver operating characteristic analysis. Odds ratios of the predictability of improvement thresholds were computed from a logistic regression model adjusting for significant baseline predictors. Desvenlafaxine 50 mg/d (n = 1207) had significantly greater rates of treatment success for each level of treatment success at 8 weeks compared with placebo (n = 1067). Optimal early improvement thresholds for weeks 2 (20%-30%) and 3 (28%-41%) were highly predictive of all 4 levels of treatment success after adjusting for significant baseline predictors (odds ratios, 0.951-0.960; all P < 0.0001). Negative predictive value of early improvement increased, and positive predictive value decreased, for increasingly stringent definitions of treatment success at week 8. Clinical observations of patients' early response to desvenlafaxine 50 mg/d may have clinical value in predicting treatment success and guiding patient management.