2502 - Clinical and Dosimetric Predictors of Radiation Pneumonitis in Pathologic Stage III-N2 Non-Small Cell Lung Cancer Patients Receiving Postoperative Radiotherapy Following Lobectomy: An Analysis of a Prospective Cohort
Presenter(s)
N. Wang1, R. Hou2, C. Wang1, H. Li1, J. Guo1, Q. Zhang1, J. Liu1, W. Yu1, X. Cai3, W. Feng1, and X. L. Fu1; 1Department of Radiation Oncology, Shanghai Chest Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, China, 2Department of Radiation Oncology, Shanghai Chest Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China, 3Department of Radiation Oncology, Shanghai Chest Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China, Shanghai, China
Purpose/Objective(s): Minimizing the risk of severe radiation pneumonitis (RP) is essential for optimizing survival benefits of postoperative radiotherapy (PORT). After surgery, lung tolerance to radiation is reduced compared to individuals with intact lungs. We hypothesize that stricter dose-volume constraints, along with additional clinical predictors, could improve RP risk stratification in patients undergoing PORT after lobectomy. This study aims to identify key risk factors and develop an integrated model for predicting Grade = 2 (G2+) RP in a prospective cohort of patients receiving PORT following lobectomy.
Materials/Methods: The primary cohort was selected from two prospective, randomized clinical trials (NCT02977169 and NCT02974426) initiated in July 2016 at our institution to investigate the value and optimal timing for PORT. Clinical and dosimetric data were collected from patients with completely resected pathologic stage T1-3N2M0 (7th Edition) non-small cell lung cancer undergoing PORT with intensity-modulated radiotherapy. The endpoint of this study was G2+ RP occurrence within 1 year after the initiation of PORT. RP was graded according to CTCAE version 5.0. Dose volume histogram metrics, along with patient, tumor and treatment characteristics were evaluated. LASSO regression, univariate, and multivariate Cox proportional hazards models were used to identify contributing parameters. A nomogram was constructed, and a predictive risk score (PRS) for G2+ RP was developed based on multivariate regression coefficients. Model performance was validated through the Concordance index (C-index), time-dependent C-index curves, calibration curves, and decision curve analysis (DCA).
Results: Data from 178 eligible patients were analyzed. 21 developed G2+ RP, with cumulative incidences of 3.9%, 10.7%, and 11.8% by 3, 6, and 12 months, respectively. Age = 67 years, the percentage of lung volume exceeding 13 Gy (V13) = 25%, and the absolute lung volume spared from a 5 Gy dose (VS5) < 1230 cc were confirmed as independent predictors for G2+ RP. The predictive model integrating age, V13, and VS5 achieved the highest bootstrap-corrected C-index of 0.757. Calibration curves demonstrated strong agreement between 12-month G2+ RP predictions and actual observations. DCA showed our model provided the highest net positive benefits. In addition, PRS could further stratify patients into low (PRS = 5) and high risk (PRS > 5) groups, with cumulative G2+ RP incidences of 6.6% and 29.3%, respectively (p < 0.0001).
Conclusion: PORT-induced lung toxicity observed in our prospective cohort was consistent with other phase III trials. Age = 67 years, V13 = 25%, and VS5 < 1230 cc were independent predictors of G2+ RP in patients receiving PORT after lobectomy. A nomogram incorporating these factors effectively predicted risk of G2+ RP, with high-risk (PRS > 5) demonstrating a significantly higher incidence. These findings offer practical guidance for RP management in the context of PORT following lobectomy.