Adaptive Semiparametric Bayesian Borrowing Model in Basket Trials for Robust Inference and Decision-Making

• Activity Number: 318 - Adaptive (and Other) Clinical Trial Designs
• Type: Contributed
• Date/Time: Wednesday, August 11, 2021 : 3:30 PM to 5:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Rachael Liu* and Veronica Bunn and Jianchang Lin and Junjing Lin
• Companies: Takeda Pharmaceuticals and Takeda Pharmaceuticals and Takeda Pharmaceuticals and Takeda Pharmaceuticals
• Keywords: Basket trials; Bayesian hierarchical models; Nonparametric priors; Subgroup analysis; power; familywise type I error rate

Abstract

In I-O and cell therapy, it is often seen to have basket trials in phase II targeting on multiple types of cancers. It is critical to have robust estimation to identify the promising cancer types to plan registration enabling trials. There are many challenges and methodological complications when large biases can be introduced due to small sample size or random variability. We review several recent methods for subgroup analysis in the Bayesian framework to correct for bias in basket trials setting. We proposed a flexible semi-parametric Bayesian borrowing model to improve the estimation. We present simulation results comparing our proposed model to various Bayesian hierarchical models for subgroup analysis in a phase II basket trial. For different scenarios considered, we compare the average total sample size, and frequentist-like operating characteristics of power and familywise type I error rate. We also compare the precision of the model estimates of the treatment effect by assessing average relative bias and the width of the 95% credible interval for the bias. Our proposed model offers improved performance across most of the scenarios and requires least assumptions.