Department of Radiation Oncology, University of Pennsylvania, Philadelphia, Pennsylvania.
Department of Radiation Oncology, University of Pennsylvania, Philadelphia, Pennsylvania.
Int J Radiat Oncol Biol Phys. 2023 Dec 1;117(5):1270-1286. doi: 10.1016/j.ijrobp.2023.06.009. Epub 2023 Jun 19.
Our objective was to use interpretable machine learning for choosing dose-volume constraints on cardiopulmonary substructures (CPSs) associated with overall survival (OS) in radiation therapy for locally advanced non-small cell lung cancer.
A total of 428 patients with non-small cell lung cancer were randomly divided into training/validation/test subsets (n = 230/149/49) in Radiation Therapy Oncology Group 0617. Manual or automated contouring was performed to segment CPSs, including heart, atria, ventricles, aorta, left/right ventricle/atrium (LV+RV+LA+RA), inferior/superior vena cava, pulmonary artery, and pericardium. Peri (pericardium-heart), rest (heart-[LV+RV+LA+RA]), clinical target volume (CTV), and lungs-CTV contours were also obtained. Dose-volume histogram features were extracted, including minimum/mean dose to the hottest x% volume (Dx%[Gy]/MOHx%[Gy]), minimum/mean/maximum dose, percent volume receiving at least xGy (VxGy[%]), and overlapping volume of each CPS with planning target volume (PTV_Voverlap[%]). Clinical parameters were collected from the National Clinical Trials Network/Community oncology research program data archive. Feature selection was performed using a series of multiblock sparse partial least squares regression, stability selection supervised principal component analysis, and Boruta. Explainable boosting machine (EBM) was trained using a conditional survival distribution-based approach for imputing censored data, treating survival analysis as a regression problem. Harrell's C-index was used to evaluate OS discrimination performance of EBM, Cox proportional hazards (CPH), random survival forest, extreme gradient boosting survival embeddings, and CPH deep neural network (DeepSurv) models in the test set. Dose-volume constraints were selected using the binary change point detection algorithm in Shapley additive explanations-based partial dependence functions.
Selected features included LA_V60Gy(%), pericardium_D30%(Gy), lungs-CTV_PTV_Voverlap(%), RA_V55Gy(%), and received_cons_chemo. All models ranked LA_V60Gy(%) as the most important feature. EBM achieved the best performance for predicting OS, followed by extreme gradient boosting survival embeddings, random survival forest, DeepSurv, and CPH (C-index = 0.653, 0.646, 0.642, 0.638, and 0.632). EBM global explanations suggested that LA_V60Gy(%) < 25.6, lungs-CTV_PTV_Voverlap(%) < 1.1, pericardium_D30%(Gy) < 18.9, RA_V55Gy(%) < 19.5, and received_cons_chemo = 'Yes' for improved OS.
EBM can be used to discriminate OS while also guiding dose-volume constraint selection for optimal management of cardiac toxicity in lung cancer radiation therapy.
我们的目的是使用可解释的机器学习方法，为局部晚期非小细胞肺癌放射治疗中与总生存（OS）相关的心肺亚结构（CPS）选择剂量-体积限制。
共 428 名非小细胞肺癌患者随机分为训练/验证/测试子集（n=230/149/49），来自放射治疗肿瘤学组 0617。手动或自动勾画用于分割 CPS，包括心脏、心房、心室、主动脉、左/右心室/心房（LV+RV+LA+RA）、下/上腔静脉、肺动脉和心包。还获得了心包-心脏（peri[pericardium-heart]）、休息（heart-[LV+RV+LA+RA]）、临床靶区（CTV）和肺-CTV 轮廓。提取了剂量-体积直方图特征，包括最热 x%体积的最小/平均剂量（Dx%[Gy]/MOHx%[Gy]）、最小/平均/最大剂量、至少接受 xGy 的体积百分比（VxGy[%]）以及每个 CPS 与计划靶区（PTV_Voverlap[%]）的重叠体积。临床参数来自国家临床试验网络/社区肿瘤学研究计划数据档案。使用一系列多块稀疏偏最小二乘回归、稳定性选择监督主成分分析和 Boruta 进行特征选择。使用基于条件生存分布的方法训练可解释的提升机（EBM），将生存分析视为回归问题，以对删失数据进行插补。使用 Harrell's C 指数评估 EBM、Cox 比例风险（CPH）、随机生存森林、极端梯度提升生存嵌入和 CPH 深度神经网络（DeepSurv）模型在测试集中的 OS 判别性能。使用 Shapley 加性解释的部分依赖函数中的二分变化点检测算法选择剂量-体积限制。
选定的特征包括 LA_V60Gy（%）、心包 D30%（Gy）、肺-CTV_PTV_Voverlap（%）、RA_V55Gy（%）和 received_cons_chemo。所有模型均将 LA_V60Gy（%）评为最重要的特征。EBM 在预测 OS 方面表现最佳，其次是极端梯度提升生存嵌入、随机生存森林、DeepSurv 和 CPH（C 指数分别为 0.653、0.646、0.642、0.638 和 0.632）。EBM 的全局解释表明，LA_V60Gy（%）<25.6、肺-CTV_PTV_Voverlap（%）<1.1、心包 D30%（Gy）<18.9、RA_V55Gy（%）<19.5 且 received_cons_chemo='Yes'可以改善 OS。
EBM 可用于区分 OS，同时还可以指导剂量-体积限制选择，以优化肺癌放射治疗中心脏毒性的管理。
