Yu Duo, Kane Michael J, Chen Yiqing, Lin Steven H, Mohan Radhe, Hobbs Brian P
Division of Biostatistics, Data Science Institute, Medical College of Wisconsin, Milwaukee, WI.
Department of Biostatistics, Yale School of Public Health, New Haven, CT.
JCO Clin Cancer Inform. 2025 Aug;9:e2500058. doi: 10.1200/CCI-25-00058. Epub 2025 Sep 8.
Lymphocytes play critical roles in cancer immunity and tumor surveillance. Radiation-induced lymphopenia (RIL) is a common side effect observed in patients with cancer undergoing chemoradiation therapy (CRT), leading to impaired immunity and worse clinical outcomes. Although proton beam therapy (PBT) has been suggested to reduce RIL risk compared with intensity-modulated radiation therapy (IMRT), this study used Bayesian counterfactual machine learning to identify distinct patient profiles and inform personalized radiation modality choice.
A novel Bayesian causal inferential technique is introduced and applied to a matched retrospective cohort of 510 patients with esophageal cancer undergoing CRT to identify patient profiles for which immunosuppression could have been mitigated from radiation modality selection.
BMI, age, baseline absolute lymphocyte count (ALC), and planning target volume determined the extent to which reductions in ALCs varied by radiation modality. Five patient profiles were identified. Significant variation in ALC nadir between PBT and IMRT was observed in three of the patient subtypes. Notably, older patients (age >69 years) with normal weight experienced a two-fold reduction in mean ALC nadir when treated with IMRT versus PBT. Mean ALC nadir was reduced significantly for IMRT patients with lower ALC at baseline (<1.6 k/µL) who were overweight or obese when compared with PBT, whereas overweight patients with higher baseline ALC showed clinical equipoise between modalities.
Individualized radiation therapy selection can be an important tool to minimize immunosuppression for high-risk patients. The Bayesian counterfactual modeling techniques presented in this article are flexible enough to capture complex, nonlinear patterns while estimating interpretable patient profiles for translation into clinical protocols.
淋巴细胞在癌症免疫和肿瘤监测中发挥着关键作用。辐射诱导的淋巴细胞减少症（RIL）是接受放化疗（CRT）的癌症患者中常见的副作用，会导致免疫力受损和更差的临床结果。尽管与调强放射治疗（IMRT）相比，质子束治疗（PBT）被认为可降低RIL风险，但本研究使用贝叶斯反事实机器学习来识别不同的患者特征，并为个性化放射治疗方式的选择提供依据。
引入一种新颖的贝叶斯因果推断技术，并将其应用于510例接受CRT的食管癌患者的匹配回顾性队列中，以确定通过放射治疗方式选择可以减轻免疫抑制的患者特征。
体重指数（BMI）、年龄、基线绝对淋巴细胞计数（ALC）和计划靶体积决定了ALC降低程度因放射治疗方式而异的程度。识别出了五种患者特征。在三种患者亚型中观察到PBT和IMRT之间ALC最低点存在显著差异。值得注意的是，体重正常的老年患者（年龄>69岁）接受IMRT治疗时，平均ALC最低点比接受PBT治疗时降低了两倍。与PBT相比，基线ALC较低（<1.6 k/µL）且超重或肥胖的IMRT患者的平均ALC最低点显著降低，而基线ALC较高的超重患者在两种治疗方式之间显示出临床等效性。
个体化放射治疗选择可以成为将高危患者免疫抑制降至最低的重要工具。本文介绍的贝叶斯反事实建模技术足够灵活，能够捕捉复杂的非线性模式，同时估计可解释的患者特征，以便转化为临床方案。
