From Guidance to Practice: Evolving Approaches to Covariate Adjustment in Clinical Trials

Rerandomization is an effective treatment allocation procedure to control for baseline covariate imbalance. For estimating the average treatment effect, rerandomization has been previously shown to improve the precision of the unadjusted and the linearly-adjusted estimators over simple randomization without compromising consistency. However, it remains unclear whether such results apply more generally to the class of M-estimators, including the g-computation formula with generalized linear regression and doubly-robust methods, and more broadly, to efficient estimators with data-adaptive machine learners. In this paper, using a super-population framework, we develop the asymptotic theory for a more general class of covariate-adjusted estimators under rerandomization and its stratified extension. We prove that the asymptotic linearity and the influence function remain identical for any M-estimator under simple randomization and rerandomization, but rerandomization may lead to a non-Gaussian asymptotic distribution. We further explain, drawing examples from several common M-estimators, that asymptotic normality can be achieved if rerandomization variables are appropriately adjusted for in the final estimator. These results are extended to stratified rerandomization. Finally, we study the asymptotic theory for efficient estimators based on data-adaptive machine learners, and prove their efficiency optimality under rerandomization and stratified rerandomization. Our results are demonstrated via simulations and re-analyses of a cluster-randomized experiment that used stratified rerandomization. 

The win ratio, first popularized by Pocock et al. (2012), has emerged as a powerful tool for evaluating composite and hierarchical outcomes. As adoption grows, covariate adjustment can help further increase precision and reduce bias. Yet integrating covariate information into a hierarchical comparison framework presents theoretical and practical challenges, from defining suitable risk strata to ensuring valid inference under complex study designs. In this talk, we discuss state-of-the-art strategies for covariate adjustment—including model-based and stratified techniques—and highlight practical guidance for implementation in real-world clinical trials. Real examples will be provided as illustrations. 

The area under the curve (AUC) of the mean cumulative function (MCF) has recently been introduced as a novel estimand for evaluating treatment effects in recurrent event settings, offering an alternative to the commonly used Lin-Wei-Yang-Ying (LWYY) model. The AUC of the MCF provides a clinically interpretable summary measure that captures the overall burden of disease progression, regardless of whether the proportionality assumption holds. To improve the precision of the AUC estimation while preserving its unconditional interpretability, we propose a nonparametric covariate adjustment approach. This approach guarantees efficiency gain compared to unadjusted analysis, as demonstrated by theoretical asymptotic distributions, and is universally applicable to various randomization schemes, including both simple and covariate-adaptive designs. Extensive simulations across different scenarios further support its advantage in increasing statistical power. Our findings highlight the importance of covariate adjustment for the analysis of AUC in recurrent event settings, offering practical guidance for its application in randomized clinical trials. 

In the digital era it is easier than ever to collect and exploit rich covariate information in trials. Recent work explores how to use this information to integrate external controls, including the use of hybrid control trials (HCTs) where a randomized controlled trial is augmented with external controls. HCTs are of particular interest due to their ability to preserve partial randomization while also improving trial efficiency.
However, most HCT estimators rely on an unrealistic assumption: mean exchangeability of the controls. Literature has focused on the development of these estimators, but few, if any, have discussed a formal approach to address the inevitable bias introduced from a violation of this assumption, slowing the acceptance of this study design.
To address this, we introduce a non-parametric sensitivity analysis that recognizes that the assumption can be reframed as a 'no unobserved confounders' assumption. We leverage omitted variable bias methodologies to estimate the maximum bias introduced from unmeasured covariates, allowing for a critical evaluation of the causal gap that invalidates significant findings. We show that with sufficient understanding of the covariate-outcome relationship, this method reliably bounds bias while also allowing for gains in efficiency.
By enhancing the credibility of HCT findings, our method aims to facilitate the broader adoption of HCTs, thereby safely accelerating the development of new therapies, especially in underfunded or underrepresented areas of research.







 

The importance of covariate adjustment has been increasingly recognized in drug development, driven by the recent FDA guideline. Despite the recommendations therein, the best practices and pitfalls of covariate adjustment are still not explicitly clear to statisticians working on clinical trials. This presentation introduces the development of RobinCar2, a validated R package for ROBust estimation and INference for Covariate Adjustment in Randomized clinical trials. This package has been developed jointly with ASA BIOP Covariate Adjustment SWG and the conditional and marginal effect task force within the ASA Oncology Estimand SWG. Examples will be illustrated to handle continuous and binary endpoints in randomized clinical trials.