Translating Treatment Effects between Correlated Endpoints

Fri, Oct 23, 2026, 08:00 – 09:00AM (PDT), 11:00 – 12:00 noon (EST)

Title: Translating Treatment Effects between Correlated Endpoints

Abstract:

Clinical researchers frequently need to answer: 'If a treatment changes endpoint 1 by X%, what change in endpoint 2 should we expect?' Current meta-analytic approaches either produce biased estimates or provide correlation measures without explicit translational coefficients. We demonstrate how multivariate random-effects meta-analysis can derive the Expected Translational Association (η), a coefficient that translates proportional changes between endpoints, simultaneously fitting all correlated endpoints through their shared variance–covariance structure, yielding unbiased estimates and nominal confidence coverage. For convenience, we refer to this stacked implementation as SLIM (Stacked LInear Mixed Effects Model). Extensive simulations showed the multivariate framework significantly outperformed traditional two-stage approaches like Daniels–Hughes, which failed to recover true effects even with large numbers of trials. The framework generalizes naturally to multivariate systems, offering a rigorous and transparent foundation for evidence synthesis in translational science. We also introduce an AI-assisted meta-analysis (AIM) framework for an automated and integrated workflow using SLIM.

Speaker: Dr. Nusrat Rabbee

Dr. Nusrat Rabbee has over 23 years of industry experience in clinical trials and translational research. She currently serves as Head of Statistical Innovation and Multimodal Evidence Synthesis at Regeneron Pharmaceuticals, Inc - where she is responsible for innovative trial design, AI-enabled frameworks for integrating clinical and translational insights from multiple sources of data including RWD across the company’s portfolio. Prior to joining Regeneron Pharmaceuticals, she was the Chief of Statistical and Epidemiology Service at the NIH Clinical Center where her group at NIH focused on various initiatives supporting NIH intramural research including novel AI/machine learning algorithms for detecting MRI image segmentation for tumor detection. Earlier in her career, she held roles of increasing responsibility in R&D at Genentech, Veracyte and Eisai Inc., where she developed molecular diagnostics and high-dimensional signatures for multiple therapeutic areas. She holds several patents on high-dimensional algorithms for combining biomarkers and is the author of the book, “Biomarker Analysis in Clinical Trials Using R” by Talor & Francis. Dr. Rabbee received her Ph.D. in Biostatistics from the Harvard University and was an NSF post-doctoral fellow at the University of California at Berkeley.