Efficient Testing Using Surrogate Information

2025-04-21

AI Summary

In clinical trials, surrogate markers are often used to infer treatment effects due to cost or time constraints, yet their validity frequently exhibits patient-level heterogeneity; existing methods either rely on strong parametric assumptions or require global surrogacy, limiting practical utility. This paper proposes the first fully nonparametric testing framework that permits surrogacy to hold only within identifiable subpopulations and adaptively selects between surrogate and primary endpoints accordingly. The method integrates kernel smoothing, conditional average treatment effect (CATE) modeling, and doubly robust inference, enabling efficient, unbiased treatment effect estimation and hypothesis testing under stratified randomization. It further supports prospective power and sample size calculations. Simulation studies and analyses of two HIV clinical trials demonstrate that the proposed approach improves statistical power by 15–30% over state-of-the-art methods, maintains strict Type I error control, and accurately identifies subpopulations where the surrogate is valid.

Technology Category

• Machine Learning: Kernel Methods
• Reasoning under Uncertainty: Stochastic Optimization
• Search and Optimization: Non-convex Optimization

Application Category

• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Security and Privacy: Large-scale security measurements
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

In modern clinical trials, there is immense pressure to use surrogate markers in place of an expensive or long-term primary outcome to make more timely decisions about treatment effectiveness. However, using a surrogate marker to test for a treatment effect can be difficult and controversial. Existing methods tend to either rely on fully parametric methods where strict assumptions are made about the relationship between the surrogate and the outcome, or assume the surrogate marker is valid for the entire study population. In this paper, we develop a fully nonparametric method for efficient testing using surrogate information (ETSI). Our approach is specifically designed for settings where there is heterogeneity in the utility of the surrogate marker, i.e., the surrogate is valid for certain patient subgroups and not others. ETSI enables treatment effect estimation and hypothesis testing via kernel-based estimation for a setting where the surrogate is used in place of the primary outcome for individuals for whom the surrogate is valid, and the primary outcome is purposefully only measured in the remaining patients. In addition, we provide a framework for future study design with power and sample size estimates based on our proposed testing procedure. We demonstrate the performance of our methods via a simulation study and application to two distinct HIV clinical trials.

Problem

• Nonparametric testing method for surrogate marker heterogeneity
• Estimating treatment effects using valid surrogate subgroups
• Designing future studies with power and sample size estimates

Innovation

• Nonparametric surrogate marker testing method
• Kernel-based estimation for heterogeneous subgroups
• Framework for power and sample size design