2026 ECNP Congress

PS02-2118

Predictors of response with adjunctive lumateperone in patients with major depressive disorder: a pooled analysis of 2 phase 3 trials

Andrea Wilson — M.K. Singh, S. Durgam, W.R. Earley, J.J. Sheehan, I. Turkoz · Intra-Cellular Therapies- a Johnson & Johnson company, Senior Medical Director- Medical Affairs, Tinton Falls, United States

2026 ECNP Congress

Background

Evidence shows that a lack of treatment response occurs in approximately 50% of patients with major depressive disorder (MDD) and can be associated with functional impairment and poor quality of life. Furthermore, research supports that certain patient characteristics and outcomes may help predict a treatment response in patients with depression. Thus, identifying predictors of response is an important measure to further evaluate the efficacy of a treatment and help guide treatment decisions.

Lumateperone is an atypical antipsychotic indicated for schizophrenia, depressive episodes associated with bipolar I or II disorder (bipolar depression), and as adjunct to antidepressant therapy (ADT) for MDD. This post hoc analysis investigated potential predictors of response to lumateperone + ADT at Day 43 in pooled Studies 501 and 502 in patients with MDD with an inadequate response to ADT.

Methods

Pooled data were analysed from 2 positive, Phase 3, randomised, double-blind, placebo-controlled trials (Study 501 [NCT04985942] and Study 502 [NCT05061706]) in adults aged 18-65 years with DSM-5–defined MDD and an inadequate response to 1-2 ADT in the current depressive episode. Patients were randomised (1:1) to lumateperone 42 mg + ADT or placebo + ADT for 6 weeks. Baseline demographics and clinical characteristics of the overall population were evaluated as potential predictors of response (≥50% decrease in Montgomery-Asberg Depression Rating Scale [MADRS] Total score) at Day 43 using penalised logistic regression (elastic net and lasso) and multivariable logistic regression with stepwise selection models. To identify predictors that were consistent across modelling approaches, random forest and Bayesian model averaging were also performed.

Results

Of ~30 demographic and clinical characteristics (eg, age, ethnicity, lifetime depressive episodes) analysed in the pooled population (n=950), the 2 characteristics confirmed as predictors of Day 43 response across all statistical models (elastic net regression, lasso regression, and stepwise logistic regression) were treatment with lumateperone + ADT and early Clinical Global Impression–Severity (CGI-S) score improvement at Day 8. According to the stepwise logistic regression, treatment with lumateperone + ADT led to a 2.2× greater likelihood of response at Day 43 compared with placebo + ADT (odds ratio [OR], 2.2; 95% CI, 1.64-2.93; P<.001). In addition, reduction in CGI-S score by 1 point at Day 8 led to a 2.7× greater likelihood of response at Day 43 compared with no reduction (OR, 2.7; 95% CI, 2.04-3.57; P<.001). Other baseline demographics and clinical characteristics (eg, history of treatment failure, type of concurrent ADT, anxious distress) did not consistently predict a response across models.

Conclusions

Lumateperone 42 mg + ADT treatment and early CGI-S score improvement were the only consistent, significant predictors of response at Day 43 in this pooled analysis of 2 Phase 3 MDD trials. These results further support the use of adjunctive lumateperone 42 mg treatment in patients with MDD with an inadequate response to ADT.

References

No references