Bayesian Uncertainty Directed Trial Designs

TL;DR

Bayesian uncertainty directed designs (BUD) optimize information gain during trials, balancing exploration and exploitation across various trial types, and adapt to multiple endpoints and biomarkers.

Contribution

This paper introduces BUD, a flexible Bayesian design framework that focuses on maximizing information, differing from traditional response-adaptive methods.

Findings

BUD balances treatment allocation to optimize information gain.

Finite-sample performance of BUD is demonstrated through multiple trial examples.

Asymptotic analysis of patient allocation proportions is provided.

Abstract

Most Bayesian response-adaptive designs unbalance randomization rates towards the most promising arms with the goal of increasing the number of positive treatment outcomes during the study, even though the primary aim of the trial is different. We discuss Bayesian uncertainty directed designs (BUD), a class of Bayesian designs in which the investigator specifies an information measure tailored to the experiment. All decisions during the trial are selected to optimize the available information at the end of the study. The approach can be applied to several designs, ranging from early stage multi-arm trials to biomarker-driven and multi-endpoint studies. We discuss the asymptotic limit of the patient allocation proportion to treatments, and illustrate the finite-sample operating characteristics of BUD designs through examples, including multi-arm trials, biomarker-stratified trials, and trials with multiple co-primary endpoints.