aiqbench analysis
This study applies deep reinforcement learning to optimize chemotherapy dosing strategies for heterogeneous tumors, demonstrating a trade-off between tumor reduction efficacy and inter-patient consistency. The research highlights the potential of closed-loop DRL policies to complement traditional open-loop optimal control benchmarks in high-dimensional biological models. However, the current reliance on full observation assumptions underscores the critical need for future work addressing sparse clinical data and partial observability before clinical translation.
Adaptive Chemotherapy Control under Tumor Heterogeneity via Reinforcement Learning
arXiv:2609.12264v1 Announce Type: new Abstract: Designing effective chemotherapy regimens is hindered by tumor heterogeneity and drug resistance, which complicate the deployment of patient-specific model-based optimal control across diverse populations. We develop and compare closed-loop deep reinforcement learning (DRL) dosing policies with continuous (TD3) and discrete (DQN) action spaces trained on a high-dimensional heterogeneous tumor model. The DRL policies are benchmarked against a Pontryagin's Maximum Principle (PMP)-derived open-loop benchmark. We assess generalization under parametric heterogeneity using a 100-patient virtual cohort with plus or minus 10 percent uniform perturbations in growth and drug-sensitivity parameters. Across this cohort, TD3 achieves higher average tumor reduction, while DQN yields tighter inter-patient dosing consistency, revealing a clear efficacy-consistency trade-off in this study. Our simulations assume full observation of all tumor subpopulations; translation to sparse and noisy clinical measurements will require partial-observability formulations and/or state estimation. Overall, the results show that simulation-trained DRL can learn state-dependent feedback dosing policies that complement open-loop optimal control benchmarks.
Abstract:Designing effective chemotherapy regimens is hindered by tumor heterogeneity and drug resistance, which complicate the deployment of patient-specific model-based optimal control across diverse populations. We develop and compare closed-loop deep reinforcement learning (DRL) dosing policies with continuous (TD3) and discrete (DQN) action spaces trained on a high-dimensional heterogeneous tumor model. The DRL policies are benchmarked against a Pontryagin's Maximum Principle (PMP)-derived open-loop benchmark. We assess generalization under parametric heterogeneity using a 100-patient virtual cohort with plus or minus 10 percent uniform perturbations in growth and drug-sensitivity parameters. Across this cohort, TD3 achieves higher average tumor reduction, while DQN yields tighter inter-patient dosing consistency, revealing a clear efficacy-consistency trade-off in this study. Our simulations assume full observation of all tumor subpopulations; translation to sparse and noisy clinical measurements will require partial-observability formulations and/or state estimation. Overall, the results show that simulation-trained DRL can learn state-dependent feedback dosing policies that complement open-loop optimal control benchmarks.
6 pages. Published in the 2026 American Control Conference (ACC)
Machine Learning (cs.LG); Systems and Control (eess.SY)
arXiv:2609.12264 [cs.LG]
(or arXiv:2609.12264v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2609.12264 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
2026 American Control Conference (ACC), pp. 1203-1208, 2026
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 269 words
Related coverage
AI-Augmented Adaptive Digital Twin Modeling for Brain Tumor Evolution Prediction and Treatment Scheduling
arXiv · Sep 9, 2026
ARGOS: Reinforcement Learning-Driven Multidimensional Elasticity for Service Orchestration in the Computing Continuum
arXiv · Sep 30, 2026
BVR Sim: An Open and High-Throughput Environment for Heterogeneous Air-Combat Reinforcement Learning
arXiv · Aug 27, 2026
SRL-MPC: Shape-Aware Reinforcement Learned Model Predictive Control
arXiv · Aug 24, 2026