Handling missing covariates in model-Based meta-analysis (Use-case for competitive benchmarking): Rana Jreich and Phyllis Chan
Presented on Thursday, 7th December 11 AM - 12 PM EST (5:00 PM - 6:00 PM CET).
Abstract:
Model-based meta-analysis (MBMA) is a quantitative approach that leverages summary- and/or individual-level of historical clinical trials to inform key drug development decisions. Missingness is a frequent limitation that we may face dealing with MBMA and which may lead to small cohort and biased conclusions. In literature, there exist several statistical approaches to handle missing information in MBMA. In this webinar, we will focus particularly on missing values at the level of baseline characteristics rather than missing outcomes or other quantities. A use case of MBMA with fenebrutinib for competitive benchmarking in rheumatoid arthritis will be presented (https://pubmed.ncbi.nlm.nih.gov/31907670/), and recommendations of best practices will be discussed.
About Our Speakers:
- Rana Jreich:
Rana is an associate director clinical modeling who joined Sanofi since 2020. She has a Ph.D. in applied mathematics Statistics from Paris-Saclay university. Her current work involves providing modeling and simulation support to guide internal decision making, optimize drug development leveraging different source of data (real-world data, historical clinical trials, etc.).
- Phyllis Chan:
Phyllis is a pharmacometrician in the Clinical Pharmacology Modeling & Simulation group at Genentech in South San Francisco. Before joining Genentech in 2017, Phyllis worked at BMS, obtained her PhD in Pharmaceutical Sciences from Auburn University and was a postdoctoral fellow at USC Biomedical Simulations Resource (BMSR). She specializes in population PK, exposure-response, and tumor growth inhibition-overall survival analyses.