Robust Bayesian Sequential Borrowing for Multi-Population Clinical Programmes

2026-04-24

AI Summary

This study addresses evidence borrowing bias and efficiency loss in multi-population clinical trials arising from the neglect of gradient similarity among populations. The authors propose a sequential Bayesian evidence borrowing framework that leverages the clinical proximity ordering of populations, incorporating a path-dependent robust mixture prior and closed-form posterior weights to dynamically adjust the contribution of historical data. This approach enables transparent attenuation of heterogeneity and safe information transfer across populations. The framework further supports prospective evaluation of Bayesian Type I error rate and power at both study and program levels. Simulation studies demonstrate that, compared to full pooling, the method substantially improves control of false positives while retaining considerable efficiency gains over independent analyses. The approach is successfully validated using data from adult, adolescent, and pediatric populations in the START trial.

Technology Category

• Machine Learning: Bayesian Learning
• Search and Optimization: Sampling/Simulation-based Search
• Reasoning under Uncertainty: Sequential Decision Making

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Security and Privacy: Data transparency and provenance

Abstract

We introduce Robust Bayesian Sequential Borrowing (RBSB), a framework for extrapolating evidence across adjacent subgroups in multi-population clinical programmes where studies are conducted in sequence and populations are ordered by clinical proximity. Conventional approaches weight all historical sources uniformly or exclude distant populations entirely, failing to reflect the natural gradient of similarity in such programmes. RBSB encodes the programme order through path-dependent borrowing via robust mixture priors that combine an informative component with a unit-information component to guard against prior-data conflict. Posterior weights, derived in closed form from marginal likelihood ratios, provide transparent dynamic attenuation when heterogeneity arises between sequential populations. The framework supports prospective evaluation of Bayesian Type I error, power, and extends naturally to assurance at both the study and programme level. Simulation studies demonstrate superior false-positive control relative to full pooling, while preserving substantial efficiency gains over standalone analyses. A case study of the START trial illustrates the approach across adult, adolescent, and paediatric populations. RBSB offers a practical, regulator-aligned method for disciplined evidence borrowing that exploits temporal and biological proximity while preventing implausible extrapolation across distant populations.

Problem

• multi-population clinical programmes
• evidence borrowing
• clinical proximity
• heterogeneity
• Bayesian extrapolation

Innovation

• Robust Bayesian Sequential Borrowing
• mixture priors
• posterior weighting
• evidence borrowing
• multi-population clinical trials