Model Informed Development in Vaccines Immunizes Against Choosing Poor Vaccine Candidates and/or Non-Informative Dose-Levels

• Activity Number: 120 - Recent Advances in Vaccine Dose and Regimen Finding
• Type: Topic Contributed
• Date/Time: Monday, August 3, 2020 : 1:00 PM to 2:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Radha Railkar* and Jeff Sachs and Jos Lommerse and Nele Mueller-Plock and S.Y. Amy Cheung and Brian Maas and Luzelena Caro and Antonios Aliprantis and Kalpit Vora and Amy Espeseth and Andrew Lee
• Companies: Merck & Co., Inc. and Merck and Certara and Certara and Certara and Merck and Merck and Merck and Merck and Merck and Merck
• Keywords: vaccines; meta-analysis; dose selection; modeling; non-linear mixed effects

Abstract

As the landscape for vaccine development has grown increasingly competitive and costs to conduct clinical trials have soared, leveraging all available information has helped make critical development decisions earlier and with more objectivity. Model-based meta-analysis (MBMA) is a tool for integrating prior knowledge, both public and proprietary, to inform discovery and development decisions. As in traditional meta-analysis, MBMA allows quantification of the mean effect, but inclusion of a model further enables quantification and explanation of the variability in observed results across trials using multiple parameters (e.g., treatments, doses, time, population characteristics, etc.), making it a powerful tool for predicting new, potentially interesting clinical scenarios through clinical trial simulation. In this talk we illustrate the use of MBMA to conduct optimal dose selection for vaccine/prophylaxis programs and highlight the advantages of using this method relative to traditional methods for dose selection, especially when coupled with dose-escalation designs and non-linear mixed effects (NLME) methods for dose-response modeling.