NCDEU Poster Session 2012

Predictors of Response and Remission With Desvenlafaxine 50 mg/d: A Pooled Analysis of Randomized, Placebo-Controlled Studies in Patients With Major Depressive Disorder

Claudio N. Soares, MD, PhD, FRCPC; Rana S. Fayyad, PhD; Cedric O’Gorman, MD; and Christine J. Guico-Pabia, MD, MBA, MPH
From McMaster University and St. Joseph’s Healthcare, Hamilton, Ontario, Canada (Dr Soares); Pfizer Inc, New York, New York (Drs Fayyad and O’Gorman); and Pfizer Inc, Collegeville, Pennsylvania (Dr. Guico-Pabia).

This poster presentation was supported by Pfizer.

Background: Early identification of likely responders to an antidepressant strategy may guide clinical treatment decisions, minimize the use of ineffective treatments and ultimately improve adherence and treatment outcomes for patients with major depressive disorder. In this study, data on age, gender, baseline depressive and functional measures, duration of current episode as well as improvements of depressive scores at earlier time points were examined to identify possible relevant predictors of treatment outcomes at 8 weeks among patients treated with desvenlafaxine (DVS) 50 mg/d or placebo.

Methods: Data were pooled from double-blind, fixed-dose studies in adult patients with a diagnosis of Major Depressive Disorder (MDD) based on DSM-IV or DSM-IV-TR. Patients were randomly assigned to DVS or placebo. Primary end point was change in HAM-D17 scores from baseline to week 8 (or last observation carried forward [LOCF]). A logistic regression model was used to examine the predictive value on response (defined as ≥50% improvement in HAM-D17 scores) or remission of depression (defined as an achievement of HAM-D17 scores ≤7) of the following variables: age, gender, baseline HAM-D17 total scores, baseline Sheehan Disability Scale total scores, duration of current episode and early HAM-D17 improvements (weeks 2, 3 and 4).

Results: DVS led to significant improvement of depression (HAM-D17 scores from baseline to study endpoint, response and remission) compared with placebo (p<0.0001), and was statistically significant for both the youngest (<40 years of age) and the oldest (>55 years of age) sub-groups. For the 41-54 years age group, DVS also produced significant improvement for both response and improvement in HAM-D17 scores, but not for remission. An equal or greater than 20% improvement on HAM-D17 scores at week 3 strongly predicted response (OR=9.37[7.23,12.15]) and remission (OR=14.7[9.39,23.01] of depression at week 8 (both p<0.0001). A positive predictive value (number of patients remitted at week 8 and with 20% improvement at week 3/those improved at week 3) indicated that 36% of patients that improved at week 3 also remitted at week 8. In addition, a negative predictive value (those who did not remit at week 8 nor improve at week 3/those who did not remit at week 3) showed that 96% of the patients who did not improve at week 3 also did not remit at week 8.

Conclusions: Clinical observations of patients’ early response to the starting/recommended dose of DVS (50 mg/d) may have a clinical value in predicting further outcomes with this antidepressant agent and guide patient management.

Learning Objectives:
To understand that identifying likely responders in patients with major depressive disorder treated with desvenlafaxine (50 mg/d) may guide clinical treatment decisions and minimize the use of ineffective treatments

To understand the positive and negative predictor values for remission and response in patients at week 3 and week 8 of treatment with desvenlafaxine (50 mg/d)

References
Szegedi A, Jansen WT, van Willigenburg APP, et al. Early improvement in the first 2 weeks as a predictor of treatment outcome in patients with major depressive disorder: a meta-analysis including 6562 patients. J Clin Psychiatry. 2009;70(3):344–353. PubMed

Kuk AY, Li J, Rush AJ. Recursive subsetting to identify patients in the STAR*D: a method to enhance the accuracy of early
prediction of treatment outcome and to inform personalized care. J Clin Psychiatry. 2010;71(11):1502–1508. PubMed

Back       Poster