Anxiety does not predict response to duloxetine in major depression: results of a pooled analysis of individual patient data from 11 placebo-controlled trials

Abstract

Background: Uncontrolled antidepressant trials suggest that anxious patients with major depressive disorder (MDD) are less responsive to antidepressant treatment than less anxious patients. The objective of this study is to determine whether specific antidepressant effects, estimated by drug-placebo differences, are reduced in anxious depression during treatment of MDD with duloxetine.

Methods: This is a retrospective secondary pooled analysis of all placebo-controlled trials of duloxetine at therapeutic doses conducted by the sponsor in outpatients with nonpsychotic unipolar MDD, using the Hamilton Depression Rating Scale (HAMD). Anxious depression was defined by > or =7 on the anxiety/somatization factor of the HAMD. Response was defined as > or =50% improvement from baseline to endpoint on the HAMD. Remission was defined as an endpoint HAMD< or =7. Analyses were performed in the intent-to-treat sample with at least one post-treatment rating.

Results: Eleven trials included 2,841 patients of whom 1,326 were classified as anxious and 1,515 as nonanxious. Change on the HAMD was greater with duloxetine than placebo in both anxious (9.91 versus 7.55, P<.001) and nonanxious (6.65 versus 5.23, P<.001) patients. Level of anxiety had no effect on the drug-placebo differences. Response and remission rates were significantly greater in duloxetine than placebo-treated patients and drug-placebo differences were unaffected by anxious status. Use of HAMD items psychic and somatic anxiety to define anxious subgroups had similar outcomes.

Conclusions: Duloxetine was more effective than placebo in achieving response and remission in both anxious and nonanxious patients. Anxious status did not affect the magnitude of the drug effect.