MBMA Bridging Models as a Tool for Exploration of Clinical Endpoints in Unstudied Indications

Model-based meta-analysis (MBMA) offers a powerful framework for quantitatively integrating clinical trial data to inform drug development and decision-making. In this study, a novel quantitative modeling approach using MBMA was proposed to project key efficacies of treatment in future Phase 2 or Phase 3 studies for an untested indication. As a case example, aggregate level data from a total of 67 published randomized-controlled clinical trials were analyzed to build dose-response and longitudinal models, enabling us to bridge efficacy endpoints across psoriasis (PsO) and psoriatic arthritis (PsA), two related conditions within the psoriatic disease spectrum. The models incorporated both placebo and treatment effects while accounting for trial-level heterogeneity, such as differences in study phase and population. Specifically, modeling has been done in two steps: (1) bridging across PsO and PsA indications and (2) bridging across endpoints within the PsA indication. Multiple model evaluations and assessments including external validation, by leaving one study out, were conducted to ensure robustness and confirm the predictive accuracy of the framework. Outcomes supported the potential of the proposed approach not only to characterize treatment effects across diseases but also to support evidence-based decisions on trial design for the new indication. This approach holds strong promises for accelerating development timelines and improving success rates in drug development.