Nonparametric Bayesian Multi-Treatment Mixture Cure Survival Model with Application in Pediatric Oncology

2025-08-12

AI Summary

In multi-arm clinical trials, heterogeneous causal effects arise from overlapping treatment components and the presence of long-term survivors (cured patients), complicating causal identification. To address this, we propose a covariate-dependent nonparametric Bayesian mixture cure model that jointly models treatment arms via latent variable–driven shared mechanisms. The model simultaneously estimates individualized survival curves, causal treatment effects, and cure proportions. Its key innovation lies in integrating nonparametric Bayesian priors with multi-treatment joint modeling, enabling cross-arm information sharing through latent functions and posterior inference via MCMC. Simulation studies demonstrate robustness under model misspecification and high-dimensional covariates. Applied to the AALL0434 trial, the method successfully detects survival differences across methotrexate administration regimens and quantifies their interactions with critical clinical covariates—including age and genetic biomarkers—providing an interpretable, causally grounded framework for precision oncology.

Technology Category

• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality
• Multiagent Systems: Mechanism Design

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Heterogeneous treatment effect estimation is critical in oncology, particularly in multi-arm trials with overlapping therapeutic components and long-term survivors. These shared mechanisms pose a central challenge to identifying causal effects in precision medicine. We propose a novel covariate-dependent nonparametric Bayesian multi-treatment cure survival model that jointly accounts for common structures among treatments and cure fractions. Through latent link functions, our model leverages sharing among treatments through a flexible modeling approach, enabling individualized survival inference. We adopt a Bayesian route for inference and implement an efficient MCMC algorithm for approximating the posterior. Simulation studies demonstrate the method's robustness and superiority in various specification scenarios. Finally, application to the AALL0434 trial reveals clinically meaningful differences in survival across methotrexate-based regimens and their associations with different covariates, underscoring its practical utility for learning treatment effects in real-world pediatric oncology data.

Problem

• Estimating heterogeneous treatment effects in multi-arm oncology trials
• Identifying causal effects with overlapping treatments and long-term survivors
• Modeling individualized survival with shared treatment structures and cure fractions

Innovation

• Nonparametric Bayesian multi-treatment cure survival model
• Latent link functions for treatment sharing
• Efficient MCMC for posterior approximation