# On Estimation of Optimal Treatment Regimes For Maximizing t-Year Survival Probability

arXiv: https://arxiv.org/abs/1407.7820 · HTML: https://mlsanity.com/paper/1407.7820/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Jul 2014.

## Built on by (most cited first, 9 of 9)

- [A Review of Off-Policy Evaluation in Reinforcement Learning](https://mlsanity.com/paper/2212.06355/) (2022)
- [Doubly Robust Alignment for Large Language Models](https://mlsanity.com/paper/2506.01183/) (2025)
- [Optimizing Pessimism in Dynamic Treatment Regimes: A Bayesian Learning Approach](https://mlsanity.com/paper/2210.14420/) (2022)
- [Efficient and robust transfer learning of optimal individualized treatment regimes with right-censored survival data](https://mlsanity.com/paper/2301.05491/) (2023)
- [Learning Robust Treatment Rules for Censored Data](https://mlsanity.com/paper/2408.09155/) (2024)
- [Evaluating and Learning Optimal Dynamic Treatment Regimes under Truncation by Death](https://mlsanity.com/paper/2510.07501/) (2025)
- [Individualized Multi-Treatment Response Curves Estimation using RBF-net with Shared Neurons](https://mlsanity.com/paper/2401.16571/) (2024)
- [Off-Policy Evaluation and Learning for Survival Outcomes under Censoring](https://mlsanity.com/paper/2603.22900/) (2026)
- [Deep Optimal Individualized Treatment Rules for Bivariate Survival Outcomes via Adaptive Prediction-Powered Learning](https://mlsanity.com/paper/2605.29464/) (2026)

## Abstract (arXiv)

A treatment regime is a deterministic function that dictates personalized treatment based on patients' individual prognostic information. There is a fast-growing interest in finding optimal treatment regimes to maximize expected long-term clinical outcomes of patients for complex diseases, such as cancer and AIDS. For many clinical studies with survival time as a primary endpoint, a main goal is to maximize patients' survival probabilities given treatments. In this article, we first propose two nonparametric estimators for survival function of patients following a given treatment regime. Then, we derive the estimation of the optimal treatment regime based on a value-based searching algorithm within a set of treatment regimes indexed by parameters. The asymptotic properties of the proposed estimators for survival probabilities under derived optimal treatment regimes are established under suitable regularity conditions. Simulations are conducted to evaluate the numerical performance of the proposed estimators under various scenarios. An application to an AIDS clinical trial data is also given to illustrate the methods.