2026 ECNP Congress

PS01-1228

Exposure–response relationship of once-monthly subcutaneous olanzapine (TV-44749) for clinical and functional outcomes in schizophrenia

Avia Merenlender Wagner — N. Patel, C.T. Toensing, I. Perlstein, H. Knebel, A. Elgart, R. Singh · Teva Pharmaceutical Industries Ltd., Clinical Pharmacology & Pharmacometrics- Specialty Clinical Development, Netanya, Israel

2026 ECNP Congress

Background

Olanzapine (OLZ) is a second-generation antipsychotic available in oral, intramuscular (IM) immediate release, and IM long-acting injectable (LAI) formulations. Long-term adherence to oral antipsychotics in schizophrenia is challenging and can be improved with LAIs. However, IM LAI OLZ pamoate clinical use is limited by post-injection syndrome risk and post-injection monitoring requirements. TV-44749 is an investigational, once-monthly (q1m), subcutaneous (SC) LAI OLZ designed to provide sustained efficacy and avoid post-injection syndrome risk. SOLARIS (NCT05693935) was a Phase 3 trial in adults with acute exacerbation of schizophrenia. SOLARIS achieved its primary and key secondary efficacy endpoints, demonstrating significant improvements in Positive and Negative Syndrome Scale (PANSS) total score, Clinical Global Impression of Severity (CGI-S) scale score, and Personal and Social Performance (PSP) scale score from baseline to Week 8. This analysis characterised the exposure–response (ER) relationship of TV-44749 for CGI-S and PSP scores using data from SOLARIS.

Methods

Exposure parameters were derived from a previously developed population pharmacokinetic model from completed trials: Phase 1 (SAD-10154 and NPC-10205) and SOLARIS. This analysis included participants from SOLARIS with available exposure data. SOLARIS included an 8-week placebo-controlled period (P1) followed by a ≤48-week open-label period (P2). Participants receiving placebo in P1 were re-randomised to active treatment in P2. ER correlation for CGI-S and PSP were evaluated graphically and using non-linear mixed-effect modelling with exposure metrics (Cavg, Ctrough). CGI-S was scored on a 7-point scale and PSP on a 0–100 scale. Simulations of CGI-S and PSP scores over time were performed to further characterise efficacy.

Results

Overall, 665 participants were included (7808 CGI-S and 2691 PSP observations). Most participants were male (75.2%) and Black or African American (68.1%) and mean (SD) age was 44.3 (11.4) years. Rapid improvements in CGI-S and PSP were observed following initiation of TV-44749, with greater effect than placebo during P1. A proportional odds model that incorporated both placebo and treatment effects (Emax relationship) was used for CGI-S and a multiplicative model for PSP. Model diagnostics and visual predictive checks indicated an adequate model fit. ER relationships for both CGI-S and PSP were flat across the exposure range for the 318, 425, and 531 mg q1m doses, indicating concentrations close to the Emax/plateau of the ER curve; concentration at half of the maximal effect (EC50) could not be reliably estimated. Cavg provided a slightly better fit than Ctrough; however, results were consistent between the two. Baseline CGI-S positively correlated with maximal placebo and treatment responses but negatively with the probability of achieving CGI-S ≤2. Lower baseline PSP score was associated with a larger treatment response. Simulations showed clear differentiation between TV-44749 and placebo, with minimal differences across dose groups.

Conclusions

ER analysis of SOLARIS CGI-S and PSP data revealed significant improvement for TV-44749 versus placebo, with comparable responses across doses indicating exposures near the Emax/plateau of the ER curve. Cavg was a pertinent exposure metric related to response. These data support the efficacy of the SC LAI TV-44749 for treatment of schizophrenia.

References

N/A