ALKS 3831, currently under development for the treatment of schizophrenia, is composed of olanzapine (OLZ) and a fixed dose of 10mg of samidorphan. In a Phase 2 study in stable patients with schizophrenia, ALKS 3831mitigated OLZ-associated weight gain while maintaining an antipsychotic efficacy profile similar to OLZ.

To assess the efficacy and safety of ALKS 3831 in patients with acute exacerbation of schizophrenia.

This was a Phase 3, 4-week, randomized, double-blind, active and placebo (PBO)-controlled study of ALKS 3831 in patients with acute exacerbation of schizophrenia (ClinicalTrials.gov: NCT02634346). Eligible patients (N=403) were randomized 1:1:1 to receive ALKS 3831, OLZ, or PBO. Patients were treated in an inpatient setting for the first 2weeks of the study and could be treated as inpatients or outpatients for the remaining 2weeks. Patients were excluded if they received OLZ within 6months prior to screening. Antipsychotic efficacy was assessed using the Positive and Negative Syndrome Scale (PANSS), and Clinical Global Impression–Severity (CGI-S) and CGI–Improvement (CGI-I) scales. Safety and tolerability were assessed as adverse events (AEs).

Of 401 randomized patients who received ALKS 3831, OLZ, or PBO, 91%, 89%, and 83% of patients, respectively, completed treatment. The most common reason for discontinuation was withdrawal by patient (6% in both the ALKS 3831and PBO groups, and 7% in the OLZ group). Baseline characteristics were generally similar between groups; however, baseline mean body mass index was higher in the OLZ group than in the ALKS 3831 group. Baseline mean±standard deviation scores were 101.7±11.9 for PANSS total score and 5.1±0.7 for CGI-S scale score. The mean OLZ dose was 18.4mg/day in both active treatment arms. Least squares (LS) mean difference±standard error (SE) vs PBO from baseline to Week 4 in PANSS total score was –6.4±1.8 (P<.001) for the ALKS 3831 group and –5.3±1.8 (P=.004) for the OLZ group. LS mean difference±SE vs PBO from baseline to Week 4 in CGI-S scale score was −0.4±0.1 (P=.002) for the ALKS3831 group and −0.4±0.1 (P<.001) for the OLZ group. The percentage of patients with improvement in PANSS response (≥30% from baseline) at Week 4 was 60%, 54%, and 38% in the ALKS 3831, OLZ, and PBO groups, respectively. The percentage of patients with an improvement in CGI-I scale response (score of ≤2) at Week 4 was 58%, 51%, and 33% in the ALKS 3831, OLZ, and PBO groups, respectively. Discontinuation due to AEs was low in all groups. Common AEs (≥5% in any group) included weight gain, somnolence, dry mouth, anxiety, headache, and schizophrenia.

Treatment with ALKS 3831 was more effective than PBO, as measured by the PANSS and CGI-S scale, and its antipsychotic efficacy was similar to the active control OLZ. The safety profile of ALKS 3831 was similar toOLZ.

Funding Acknowledgements: This study was funded by Alkermes, Inc.