Effect measures for comparing paired event times

2025-12-21

AI Summary

The progression-free survival ratio (PFSr) test—widely used in personalized oncology trials—suffers from severe type I error inflation, even under random right-censoring. Method: We propose a novel estimation framework based on paired event-time relative treatment effects. Our approach enables robust nonparametric estimation of PFSr-related probabilities and systematically adapts the paired relative effect methodology to the PFSr setting. Furthermore, we develop a dual-path inferential framework for the restricted mean survival time (RMST), incorporating both difference- and ratio-based estimands. Contribution/Results: Through theoretical derivation, Monte Carlo simulations, and analysis of real molecularly guided oncology trials, we demonstrate that our method maintains strict type I error control (≈0.05) under small sample sizes and high censoring rates, achieves substantially higher statistical power than existing methods, and yields clinically interpretable, statistically robust conclusions.

Technology Category

• Intelligent Robots: State Estimation
• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Security and Privacy: Large-scale security measurements
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

The progression-free survival ratio (PFSr) is a widely used measure in personalized oncology trials. It evaluates the effectiveness of treatment by comparing two consecutive event times - one under standard therapy and one under an experimental treatment. However, most proposed tests based on the PFSr cannot control the nominal type I error rate, even under mild assumptions such as random right-censoring. Consequently the results of these tests are often unreliable. As a remedy, we propose to estimate the relevant probabilities related to the PFSr by adapting recently developed methodology for the relative treatment effect between paired event times. As an additional alternative, we develop inference procedures based on differences and ratios of restricted mean survival times. An extensive simulation study confirms that the proposed novel methodology provides reliable inference, whereas previously proposed techniques break down in many realistic settings. The utility of our methods is further illustrated through an analysis of real data from a molecularly aided tumor trial.

Problem

• Addresses unreliable PFS ratio tests in oncology trials
• Proposes new methods for comparing paired event times
• Ensures valid inference under realistic censoring conditions

Innovation

• Adapting relative treatment effect methodology for PFSr
• Developing inference procedures using restricted mean survival times
• Validating reliability through simulation and real data analysis