Bayesian integration G-formula for platform SMART designs allowing for adding new treatments

2026-04-28

AI Summary

This study addresses the limitations of traditional SMART trials, which struggle to dynamically incorporate new treatment options and lack robust methods for comparing non-concurrent interventions. To overcome these challenges, the authors propose an innovative platform SMART framework that integrates the adaptive nature of platform trials with sequential multiple assignment randomized trial (SMART) design, enabling dynamic expansion of treatment arms. They further develop a Bayesian Integrated G-formula (BIG) estimator to enable unbiased causal inference for dynamic treatment strategies under non-concurrent controls. Through Bayesian inference, G-formula theory, and master-protocol-driven simulation studies, the BIG estimator demonstrates superior estimation accuracy and robustness compared to existing approaches. The method’s practical utility is validated through successful application in the SNAP trial, confirming its effectiveness in real-world settings.

Abstract

Dynamic treatment regimes (DTRs) are sequences of decision rules to guide treatment assignments in response to a patient's evolving, time-varying disease status. Sequential multiple assignment randomized trials (SMARTs) are considered the gold standard experimental design for evaluating DTRs. However, SMARTs often require more time to complete compared with a single stage RCT and new candidate treatments may become available or feasible during the trial. Platform trials are an adaptive trial design that allow new treatments to be added to the ongoing study according to a prespecified master protocol. In this paper, we introduce a novel platform SMART that integrates features from both platform trials and SMARTs, allowing new treatments to be added during the trial. Additionally, we propose the Bayesian integration G-formula (BIG) estimators for platform SMARTs to account for non-concurrent treatment comparisons. Extensive simulations are conducted to evaluate the performance of different BIG estimators against benchmark methods. We demonstrate the proposed BIG estimators based on the S. aureus Network Adaptive Platform (SNAP) trial.

Problem

• dynamic treatment regimes
• sequential multiple assignment randomized trials
• platform trials
• non-concurrent treatment comparisons
• adaptive trial design

Innovation

• platform SMART
• Bayesian integration G-formula
• dynamic treatment regimes
• non-concurrent control
• adaptive trial design