Elastic Priors to Dynamically Borrow Information from Historical Data in Clinical Trials

TL;DR

This paper introduces an elastic prior method for clinical trials that adaptively borrows historical data based on data congruence, improving power while controlling type I error.

Contribution

The elastic prior approach uniquely controls information borrowing and error rates using a monotonic elastic function, enhancing existing methods.

Findings

Better control of type I error compared to existing methods.

Higher or comparable statistical power in simulations.

Effective adaptation to data congruence levels.

Abstract

Use of historical data and real-world evidence holds great potential to improve the efficiency of clinical trials. One major challenge is how to effectively borrow information from historical data while maintaining a reasonable type I error. We propose the elastic prior approach to address this challenge and achieve dynamic information borrowing. Unlike existing approaches, this method proactively controls the behavior of dynamic information borrowing and type I errors by incorporating a well-known concept of clinically meaningful difference through an elastic function, defined as a monotonic function of a congruence measure between historical data and trial data. The elastic function is constructed to satisfy a set of information-borrowing constraints prespecified by researchers or regulatory agencies, such that the prior will borrow information when historical and trial data are congruent, but refrain from information borrowing when historical and trial data are incongruent. In doing so, the elastic prior improves power and reduces the risk of data dredging and bias. The elastic prior is information borrowing consistent, i.e. asymptotically controls type I and II errors at the nominal values when historical data and trial data are not congruent, a unique characteristics of the elastic prior approach. Our simulation study that evaluates the finite sample characteristic confirms that, compared to existing methods, the elastic prior has better type I error control and yields competitive or higher power.