EFFECTS OF CARIPRAZINE ON SYMPTOMS OF ANXIETY: A POOLED POST HOC ANALYSIS OF DOUBLE-BLIND CLINICAL TRIALS IN PATIENTS WITH DEPRESSION

Introduction

Anxiety is a common and burdensome co-occurring symptom in both bipolar and unipolar depression. Cariprazine (CAR) is a dopamine D3-preferring D3/D2 and serotonin 5-HT1A receptor partial agonist approved for the treatment of manic/mixed and depressive episodes of bipolar I disorder (BP-I) and the adjunctive treatment of major depressive disorder (aMDD). Separate analyses of BP-I depression and aMDD clinical trials have suggested that CAR has effects on anxiety; however, the assessment of anxiety may be confounded by treatment-emergent adverse events (AEs) that can mimic or mask anxiety symptoms (e.g., akathisia or sedation). This pooled post hoc analysis of BP-I depression and aMDD clinical trials evaluated the effect of CAR on anxiety across indications, including adjustments for the presence of AEs to determine their impact on anxiety scores.

Methods

Data from 2 BP-I depression trials (NCT02670538 and NCT02670551) and 1 aMDD trial (NCT03738215) were pooled and analyzed by treatment groups (placebo [PBO], CAR 1.5 mg/d, and CAR 3 mg/d). The change from baseline to Week 6 in Hamilton Anxiety Rating Scale (HAM-A) total score was evaluated using a mixed model for repeated measures, with treatment group, visit, baseline HAM-A score, treatment-group-by-visit, baseline-by-visit, study, indication (aMDD vs BP-I), and treatment-group-by-indication as covariates. Additional models adjusted for the presence of AEs of akathisia (pooled terms: akathisia, restlessness) or the presence of AEs of sedation (terms: sedation, somnolence, hypersomnolence). Least squares mean differences (LSMDs), 95% CIs, and P values vs PBO were determined in the pooled intent-totreat (ITT) population (patients who took ≥1 dose of study drug and had ≥1 post-baseline assessment). Adjustments were also repeated in individual indications as a sensitivity analysis.

Results

The pooled ITT population consisted of 952 patients from the 2 BP-I clinical trials and 751 from the aMDD clinical trial, for a total of 1,703 patients (PBO, n=568; CAR 1.5 mg/d, n=566; CAR 3 mg/d, n=569). Baseline HAM-A total scores indicated moderate anxiety on average: PBO=19.5, CAR 1.5 mg/d=20.0, CAR 3 mg/d=20.0. Without AE-related adjustments, anxiety improvements based on HAM-A total score were significantly greater at Week 6 with both CAR doses vs PBO (LSMD [95% CI]: 1.5 mg/d, -1.3 [-1.9, -0.7], P < .001; 3 mg/d, -0.7 [1.3, -0.1], P < .05). In AE-adjusted models, change from baseline in HAM-A total score for both CAR doses vs PBO indicated similar treatment effect sizes and significance levels after adjusting for akathisia (1.5 mg/d, -1.3 [-2.0, -0.7], P < .001; 3 mg/d, -0.8 [-1.4, -0.1], P < .05) and sedation (1.5 mg/d, -1.3 [-1.9, -0.7], P < .001; 3 mg/d, -0.7 [-1.3, -0.1], P < .05). In sensitivity analyses of separate indications, results with and without adjustment were consistent for CAR 1.5 mg/d; however, CAR 3 mg/d did not reach statistical significance vs PBO.

Conclusions

CAR was associated with improvements in anxiety in pooled analyses of BP-I depression and aMDD trials. These improvements were observed even after adjusting for potentially confounding effects of treatment-emergent akathisia or sedation; however, analyses were limited by small sample size of patients with AEs due to relatively infrequent incidence rates. Overall, these results suggest that CAR has effects on anxiety across mood disorders, which may be independent of AEs.