Learning Robust Treatment Rules for Censored Data

TL;DR

This paper introduces two robust methods for learning optimal treatment rules with censored survival data, focusing on maximizing truncated mean survival time and buffered survival probabilities, with theoretical and empirical validation.

Contribution

It proposes novel criteria and algorithms for treatment rule optimization in censored survival data, addressing a gap in existing methods.

Findings

Improved performance over existing methods in simulations

Effective application to AIDS clinical trial data

Theoretical guarantees for the proposed algorithms

Abstract

There is a fast-growing literature on estimating optimal treatment rules directly by maximizing the expected outcome. In biomedical studies and operations applications, censored survival outcome is frequently observed, in which case the truncated mean survival time and survival probability are of great interest. In this paper, we propose two robust criteria for learning optimal treatment rules with censored survival outcomes; the former one targets an optimal treatment rule maximizing the truncated mean survival time, where the cutoff is specified by a given quantile such as median; the latter one targets an optimal treatment rule maximizing buffered survival probabilities, where the predetermined threshold is adjusted to account for the truncated mean survival time. We develop a sampling-based difference-of-convex algorithm for learning the proposed optimal treatment rules, and provide theoretical justifications for them. In simulation studies, our estimators show improved performance compared to existing methods. We also demonstrate the proposed method using AIDS clinical trial data.