Practical Implementation of Bayesian Approaches in Drug Development
Frank Harrell
Discussant
Vanderbilt University School of Medicine
Sunday, Aug 3: 2:00 PM - 3:50 PM
0664
Topic-Contributed Paper Session
Music City Center
Room: CC-209B
Applied
Yes
Main Sponsor
Biopharmaceutical Section
Co Sponsors
International Society for Bayesian Analysis (ISBA)
Section on Bayesian Statistical Science
Presentations
The power prior is a widely used Bayesian approach for incorporating historical data in clinical trials, but its application to time-to-event data presents challenges, particularly in defining prior-data compatibility within the contexts of interpolation and extrapolation and ensuring appropriate borrowing. We propose a variation of the power prior for time-to-event analysis under a piecewise constant hazard model, using outcome similarity-driven bounded weighting approach to dynamically adjust historical data contribution. We evaluate its performance through simulations assessing type I error, power, bias, and effective sample size (ESS) and compare it with Normalized Power Prior (NPP) and Commensurate Power Prior (CPP) methods. A case study further illustrates its practical implementation.
In this session, we will present some challenges and opportunities of implementing Bayesian approaches in practice. We will examine the value of using such approaches in clinical trials including borrowing strength from external data, implementing Bayesian adaptive designs and using posterior probabilities for prediction and decision making. We will finalize with lessons learned and propose future opportunities to use Bayesian approaches.
The past decade has seen a steady increase in the number and variety of Bayesian approaches in drug and biologic development. Bayesian methods are now used to some degree in early and late phase trial design, pediatric drug development, subgroup analyses, pharmacovigilance, and non-clinical studies. In this talk, I will review the current state of Bayesian statistics in regulatory science, present case studies from a variety of applications, and discuss what the future of Bayesian methods might look like at FDA after the expected publication of a draft guidance document on the topic in September, 2025.
You have unsaved changes.