Predicting treatment outcomes of major depressive disorder by early improvement in painful physical symptoms: a pooled analysis of double-blind, placebo-controlled trials of duloxetine

Abstract

Hirofumi Tokuoka,1 Makoto Nishihara,2 Shinji Fujikoshi,3 Aki Yoshikawa,4 Atsushi Kuga1 1Bio-Medicine, Medicines Development Unit Japan, Eli Lilly Japan K.K., Kobe, 2Multidisciplinary Pain Center, Aichi Medical University, Nagakute, Aichi, 3Statistical Science, 4Scientific Communications, Medicines Development Unit Japan, Eli Lilly Japan K.K., Kobe, Japan Objective: We determined if early improvement in painful physical symptoms (PPS) can be a predictor of remission in the treatment of major depressive disorder (MDD).Methods: We included randomized, double-blind, parallel-group clinical trials of duloxetine (40–60 mg/day) versus placebo for the acute treatment of MDD with associated PPS. Only those studies using the Montgomery–Åsberg Depression Rating Scale (MADRS) and the Brief Pain Inventory – Short Form (BPI-SF) were included. Three studies met all criteria and included male or female outpatients aged ≥18 years who met the diagnostic criteria for MDD, had a MADRS total score ≥20, and had at least moderate pain (BPI-SF average pain score ≥3). Positive predictive values (PPVs) and negative predictive values (NPVs) of early improvement in PPS for remission were analyzed. PPVs were the proportion of patients with remission (MADRS total score ≤10) at week 8 out of patients who experienced early improvement in BPI-SF average pain score (≥30% decrease from baseline at week 1, 2, or 4). NPVs were the proportion of patients without remission (MADRS total score >10) at week 8 out of patients who did not experience early improvement in PPS.Results: Data from 1,320 patients were analyzed (duloxetine N=641 and placebo N=679). The overall remission (MADRS total score ≤10 at week 8) rate for the duloxetine group was significantly higher than the placebo group (38.5% vs 21.8%; P<0.0001). For both treatment groups, PPVs of early improvement in BPI-SF (30% improvement from baseline) were higher than the overall remission rate for all weeks examined (weeks 1, 2, and 4); in general, NPVs of early improvement in BPI-SF for nonremission were higher than the overall nonremission rate.Conclusion: Early improvement in PPS can be a useful clinical indicator of subsequent treatment outcome for MDD patients with associated PPS. Keywords: Brief Pain Inventory, Montgomery–Åsberg Depression Rating Scale, pain, predictor, remission

Keywords

• Brief Pain Inventory
• Montgomery–Asberg Depression Rating Scale
• pain
• predictor
• remission