- Bayesian network meta-analysis of 10 randomized, placebo‑controlled adjunctive MDD trials compared lumateperone (CAPLYTA) versus aripiprazole, brexpiprazole, cariprazine and quetiapine XR.
- Lumateperone showed the largest effects on efficacy: MADRS change MD -4.71 (95% CrI -5.78, -3.63), MADRS response OR 2.33 (95% CrI 1.77, 3.05), MADRS remission OR 2.22 (95% CrI 1.57, 3.07), and CGI‑S change MD -0.60 (95% CrI -0.74, -0.46).
- On weight outcomes lumateperone had no statistically significant mean weight gain versus placebo plus ADT (MD -0.08; 95% CrI -0.30, 0.13) and a lower odds of ≥7% weight increase (OR 0.41; 95% CrI 0.04, 1.42).
- The NMA was funded by Janssen Research & Development, LLC.
Study and methods
A Bayesian network meta-analysis pooled doses from 10 randomized, double‑blind, placebo‑controlled adjunctive MDD trials to compare five FDA‑approved atypical antipsychotics—aripiprazole, brexpiprazole, cariprazine, lumateperone, and quetiapine XR—using a star‑shaped network anchored on antidepressant therapy (ADT) plus placebo; outcomes assessed included MADRS change, response, remission, CGI‑S, and select safety measures (weight, ≥7% weight gain, akathisia, somnolence).
Efficacy results
Lumateperone showed the largest effects across efficacy endpoints: MADRS change MD −4.71 (95% CrI −5.78, −3.63); MADRS response OR 2.33 (95% CrI 1.77, 3.05); MADRS remission OR 2.22 (95% CrI 1.57, 3.07); and CGI‑S change MD −0.60 (95% CrI −0.74, −0.46); pairwise comparisons anchored to lumateperone favored it versus most comparators for MADRS and CGI‑S and versus all but one comparator for response and remission.
Safety and tolerability
Lumateperone showed no statistically significant mean weight gain versus placebo+ADT (MD −0.08; 95% CrI −0.30, 0.13) and had lower odds of ≥7% weight increase (OR 0.41; 95% CrI 0.04, 1.42), ranking most favorably on weight outcomes (100% probability of superiority on mean weight change); it was the only treatment judged comparable to placebo+ADT for akathisia risk (OR 3.78; 95% CrI 0.40, 17.17), while somnolence risk was higher than placebo across all treatments including lumateperone (OR 5.90; 95% CrI 2.86, 11.50).
Context and limitations
The analysis used indirect comparisons across trials that may differ in design and populations, so results should be interpreted alongside individual trial data and clinical judgement; the NMA was funded by Janssen Research & Development, LLC and presented at the Neuroscience Education Institute Spring Congress (May 1–3, 2026).
