EFFECT OF CARIPRAZINE TREATMENT ON SLEEP AND MOOD SYMPTOMS IN PATIENTS WITH BIPOLAR I DEPRESSION AND BASELINE SLEEP DISTURBANCE: POST HOC ANALYSIS OF THREE RANDOMIZED PLACEBO-CONTROLLED CLINICAL TRIALS

Introduction

Sleep disturbance is a core feature of bipolar I disorder (BP-I) and is associated with greater disease severity and worse outcomes, including increased rates of relapse and suicide attempts. Cariprazine (CAR), a dopamine D3-preferring D3/D2 and serotonin 5HT1A receptor partial agonist, is approved for the treatment of manic, mixed, and depressive episodes associated with BP-I. In previous post hoc analyses of three phase 2 or 3, randomized, double-blind, placebo-controlled studies of adults with BP-I depression, CAR demonstrated broad efficacy in treating core depressive symptoms, including a potential benefit for sleep. The present post hoc analysis aimed to further assess the effects of CAR on sleep and mood symptoms in a subset of patients with BP-I depression and reduced sleep at baseline.

Methods

Data were pooled from three randomized, double-blind, placebo-controlled, phase 2 or 3 clinical trials (NCT01396447, NCT02670538, NCT02670551) evaluating CAR 1.5 and 3 mg/d in adults with BP-I depression. To enrich for moderate/severe sleep disturbance, analyses were conducted in patients with a Montgomery–Åsberg Depression Rating Scale (MADRS) item 4 (“Reduced sleep”) score ≥4 at baseline. Effects on sleep, depression, and anxiety symptoms were assessed by change from baseline to week 6 in MADRS item 4, MADRS total score, and Hamilton Depression Rating Scale (HAM-D) anxiety/somatization factor scores, respectively. Additional outcomes included the proportion of patients achieving sleep response (≥2-point decrease in MADRS item 4), MADRS response (≥50% reduction in MADRS total score), and MADRS remission (MADRS total score ≤10) at week 6. For change from baseline outcomes, a mixed-effect model for repeated measures was used to estimate least-square mean (LSM) changes, along with P values corresponding to the between-treatment difference vs placebo. For categorical outcomes, a logistic regression model was used to assess P values vs placebo. Analyses were conducted for individual dose groups (1.5 mg/d, 3 mg/d) and the pooled CAR dose group (combined 1.5-3 mg/d).

Results

At baseline, 984 patients (71.1% of the pooled intent-to-treat population) had moderate or severe sleep disturbance (1.5 mg/d: n=327; 3 mg/d: n=326; placebo: n=331) and were included in the analysis population. At week 6, LSM change from baseline in MADRS item 4 score was significantly > placebo (-1.6) for CAR 3 mg/d (-2.0; P < .01) and the combined 1.5-3 mg/d dose group (-1.9; P < .05), but not for 1.5 mg/d (-1.9; P > .05). A significantly greater proportion of patients treated with CAR 3 mg/d vs placebo achieved sleep response (57.1% vs 48.3%; P < .05). LSM changes in MADRS total score were significantly greater across CAR dose groups (-15.5 to -15.4) vs placebo (-12.3; all P < .001). Significantly more CAR-treated than placebo-treated patients achieved MADRS response (45.6% to 47.9% vs 36.6%; all P < .05) and MADRS remission (30.4% to 31.5% vs 20.8%; all P < .01). CAR was also associated with significantly greater LSM reductions vs placebo in HAM-D anxiety/somatization factor score across dose groups (-3.6 to -3.3 vs -2.7; all P < .01).

Conclusions

In this post hoc analysis, CAR treatment was associated with improvements in sleep and mood symptoms at week 6 among patients with BP-I depression and baseline sleep disturbance. Overall, these results suggest that CAR may be an effective treatment option for this patient population.