Prediction of Symptomatic Radiation Pneumonitis in Lung Cancer Patients: A Radiomics and Dosiomics Machine Learning Approach Using the Prospective Multicenter RTOG 0617 and REQUITE trials.
Authors
Affiliations (6)
Affiliations (6)
- Department of Radiation Oncology, School of Medicine, TUM Klinikum Rechts der Isar, Technical University of Munich (TUM), Munich 81675, Germany; Institute of Machine Learning in Biomedical Imaging (IML), Helmholtz Zentrum München (HMGU) GmbH, German Research Center for Environmental Health, Neuherberg 85764, Germany. Electronic address: [email protected].
- Department of Radiation Oncology, School of Medicine, TUM Klinikum Rechts der Isar, Technical University of Munich (TUM), Munich 81675, Germany; Institute of Radiation Medicine (IRM), Helmholtz Zentrum München (HMGU) GmbH, German Research Center for Environmental Health, Neuherberg 85764, Germany; Partner Site Munich and German Cancer Research Center (DKFZ), Heidelberg, German Cancer Consortium (DKTK), Munich 80336, Germany.
- Department of Radiation Oncology, School of Medicine, TUM Klinikum Rechts der Isar, Technical University of Munich (TUM), Munich 81675, Germany; Institute of Machine Learning in Biomedical Imaging (IML), Helmholtz Zentrum München (HMGU) GmbH, German Research Center for Environmental Health, Neuherberg 85764, Germany; School of Computation, Information and Technology, Technical University of Munich (TUM), Munich 81675, Germany.
- Department of Radiation Oncology, School of Medicine, TUM Klinikum Rechts der Isar, Technical University of Munich (TUM), Munich 81675, Germany; Institute of Machine Learning in Biomedical Imaging (IML), Helmholtz Zentrum München (HMGU) GmbH, German Research Center for Environmental Health, Neuherberg 85764, Germany.
- Department of Radiation Oncology, School of Medicine, TUM Klinikum Rechts der Isar, Technical University of Munich (TUM), Munich 81675, Germany.
- Institute of Machine Learning in Biomedical Imaging (IML), Helmholtz Zentrum München (HMGU) GmbH, German Research Center for Environmental Health, Neuherberg 85764, Germany; School of Computation, Information and Technology, Technical University of Munich (TUM), Munich 81675, Germany; School of Biomedical Engineering and Imaging Sciences, King's College London, WC2R 2LS London, United Kingdom.
Abstract
Radiation-induced pneumonitis (RP) is a side effect after thoracic radiotherapy (RT). The ability to predict RP would facilitate treatment modifications. This study investigates the predictive capacity for symptomatic RP (CTCAE≥2) employing Radiomics and Dosiomics models. Computed tomography (CT) scans, along with physical and 2-Gy equivalent dose volumes (EQD2), dose-volume histograms (DVH), and clinical parameters, were evaluated for 708 multicenter lung cancer patients, among whom 89 developed RP≥2. The training cohort consisted of 441 patients from the prospective RTOG 0617 trial. External validation was carried out on 267 patients from the prospective REQUITE study. A Random Forest classifier was employed, with feature selection executed within the inner loop of a 10x5-fold nested cross-validation (nCV) utilizing the minimum-redundancy-maximum-relevance algorithm. To address class imbalances, synthetic oversampling and undersampling were implemented using SMOTE-Tomek. The QUANTEC Normal Tissue Complication Probability (NTCP) model served as a reference. Additionally, the experiments were stratified by subgroups (standard/high-dose and 3D-conformal RT (3D-CRT)/intensity-modulated RT (IMRT)). The best radiomics model identified in the nCV was trained on the standard-dose subgroup achieved a test ROC-AUC of 0.56. The baseline NTCP model showed a predictive performance with a ROC-AUC of 0.56, which was largely dependent on radiation technique (ROC-AUCS: 3D-CRT: 0.75, IMRT: 0.50). The Dosiomics<sub>EQD2</sub> model, trained on the full training cohort, attained the second-best performance in the nCV, demonstrating the same technique-dependence (ROC-AUC of 0.75 vs. 0.39). Using a Dosiomics<sub>EQD2</sub> ensemble model trained separately on 3D-CRT and IMRT subgroups increased overall performance to a testing ROC-AUC of 0.61, outperforming other modeling strategies for IMRT, while being outperformed by clinical models for 3D-CRT. This prospective trial-based study reveals an overall limited predictive capacity of radiomics and dosiomics models and a large influence of radiation technique. IMRT-specific models should be investigated further.