Learning Personalized Treatment Strategies with Predictive and Prognostic Covariates in Adaptive Clinical Trials
Management Science · 2026 · Andres Alban、Stephen E. Chick、Spyros I. Zoumpoulis
中文摘要
生物医学研究可能揭示疾病治疗与患者协变量之间交互作用的洞见。我们展示了如何通过拓展最优贝叶斯学习的思想，利用此类洞见来提高精准医学适应性临床试验的效率。我们提出一个响应适应性多臂临床试验模型，该模型利用关于预测性—预后性协变量结构的知识，加速学习能为试验后患者带来最佳期望结果的个体化治疗策略。我们的基础模型是用于最优臂识别的上下文线性老虎机模型，且结果可能被延迟观测。我们刻画了向试验内患者序贯分配治疗的最优策略，并因该最优策略难以计算，提出了几种基于贝叶斯一步前瞻技术的可计算启发式方法。我们证明，所提出的若干启发式方法在学习治疗策略方面是渐近最优的。基于两个受脓毒症管理启发的案例研究的数值结果表明，我们的启发式方法能够显著提高学习精准医学治疗策略的临床试验效率。我们提供了若干扩展：允许考虑来自试验内患者结果的奖励（从而化解探索—利用权衡），以及在缺乏关于协变量结构的生物医学洞见时利用 Lasso 推断协变量结构。我们提出的试验设计可引起临床试验资助者、设计者与管理者的兴趣。在可获得关于协变量—治疗交互作用洞见的情形下，该方法也可能适用于其他上下文老虎机问题。本文由 Stefan Scholtes 接收（医疗健康管理领域）。资助：本研究得到欧盟“地平线 2020”研究与创新计划下设玛丽·斯克沃多夫斯卡-居里行动的支持[资助号 676129]。S. E. Chick 感谢 Simba Gill 博士与 Sabi Dau 对 INSEAD 医疗健康管理倡议的支持。补充材料：在线附录与数据文件可通过 https://doi.org/10.1287/mnsc.2022.02048 获取。
Abstract
Biomedical research may uncover insights regarding the interaction of the treatments of a disease with patient covariates. We show how to use such insights to improve the efficiency of adaptive clinical trials for precision medicine by extending ideas from optimal Bayesian learning. We present a model for response-adaptive multi-arm clinical trials that leverages knowledge about the predictive-prognostic covariate structure to accelerate the learning of personalized treatment strategies that obtain the best expected outcomes for post-trial patients. Our base model is a contextual linear bandit for best-arm identification, and outcomes may be observed with delay. We characterize the optimal policy for sequentially allocating treatments to in-trial patients and, because it is hard to compute, propose several computable heuristics based on Bayesian one-step look-ahead techniques. We prove that several of our proposed heuristics are asymptotically optimal in learning treatment strategies. Numerical results based on two case studies motivated by sepsis management show that our heuristics can significantly improve clinical trial efficiency to learn a treatment strategy for precision medicine. We provide extensions that allow for rewards from outcomes of in-trial patients (resolving the exploration-exploitation tradeoff) and for inferring covariate structure using Lasso when biomedical insights on covariate structure are lacking. Our proposed trial design is of interest to funders, designers, and managers of clinical trials. It may also apply to other contextual bandit problems in settings where insights about covariate-treatment interactions are available. This paper was accepted by Stefan Scholtes, healthcare management. Funding: This work was supported by the Marie Skłodowska-Curie Actions under the European Union’s Horizon 2020 research and innovation programme [Grant 676129]. S. E. Chick gratefully acknowledges the support of Dr Simba Gill and Sabi Dau to the INSEAD Healthcare Management Initiative. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2022.02048 .