Background: Comparative evidence for second-step treatment strategies in severe depression is scarce. Up-titrating a well tolerated selective serotonin reuptake inhibitor (SSRI) versus switching to a serotonin norepinephrine reuptake inhibitor (SNRI) after initial SSRI non-response are possible treatment options. It is often unclear whether relevant tolerability and efficacy differences exist between SSRI up-titration versus switch to an SNRI.
Objective: The objective of this study was to evaluate tolerability and efficacy of up-titration of escitalopram versus switch to duloxetine in patients who failed to respond to escitalopram 10 mg/day.
Methods: This was an active-controlled, parallel-group, double-blind, randomized study in a general community comparing escitalopram and duloxetine in patients with severe depression; patients who did not respond (<50% Montgomery-Åsberg Depression Rating Scale [MADRS] improvement) to 2 weeks of single-blind escitalopram 10 mg/day during the lead-in period were randomized to 8 weeks of double-blind treatment. 571 male and female outpatients aged 18-65 years with severe depression (MADRS total score ≥30) participated in the study and received at least one dose of escitalopram 10 mg/day in the single-blind lead-in phase. During the double-blind randomized phase, 474 patients who did not respond to lead-in escitalopram were randomized and received treatment with escitalopram 20 mg (n = 229) or duloxetine 60 mg (n = 245). Treatment was single-blind escitalopram 10 mg/day during a 2-week lead-in followed by 8-week double-blind escitalopram 20 mg/day or duloxetine 60 mg/day. The main outcome measure was time to all-cause premature study discontinuation.
Results: There was no difference in time to all-cause discontinuation between groups (hazard ratio escitalopram/duloxetine = 0.95 [95% CI 0.64, 1.41]; p = 0.727). Treatment with escitalopram compared with duloxetine resulted in significant improvement in MADRS total score at the end of week 8 (least squares mean difference [LSMD] = -1.87 [95% CI -3.60, -0.14]; p = 0.034) using last observation carried forward (LOCF) analysis. Significantly more escitalopram (54%) than duloxetine (42%) patients achieved remission (MADRS ≤10) by week 8 (p = 0.013). Adverse events were similar between the two treatment groups.
Conclusion: In initial non-responders to escitalopram 10 mg/day, dose escalation to 20 mg/day provided better efficacy than switching to duloxetine 60 mg/day, while discontinuations for any reasons and adverse events were similar.
Clinical trial registration: Registered at ClinicalTrials.gov as NCT00384436.