ESTRO meets Asia 2024 - Abstract Book

S212

Interdisciplinary – Lung

ESTRO meets Asia 2024

90

Digital Poster

Predictors for pneumonitis for stage II-III non-small cell lung cancer after definitive radiotherapy

Xiaohong Xu, Jian He

Department of Radiation Oncology, Zhongshan Hospital, Fudan University, Shanghai, China

Purpose/Objective:

Consolidative immunotherapy after concurrent chemoradiotherapy has become the standard of care for patients with unresectable stage II-III non-small cell lung cancer (NSCLC) (1, 2). However, it also raises concerns for pneumonitis(3-5). Risk factors for pneumonitis in the era of immunotherapy require reassessment. This retrospective study aimed to investigate the predictors for pneumonitis of unresectable stage II–III NSCLC after radiotherapy with or without immunotherapy.

Material/Methods:

This single-institutional study included patients (N = 104) who were diagnosed with stage II-III NSCLC (American Joint Committee on Cancer 8 th ) and received definitive radiotherapy between March 2018 and May 2023. The primary endpoint was grade ≥2 pneumonitis (G2P), and the secondary endpoint was grade ≥3 pneumonitis (G3P), as defined by Common Terminology Criteria for Adverse Events version 5.0. Time to pneumonitis was defined as time from radiotherapy completion to pneumonitis development. Univariable and multivariable competing risk regression models were used to determine clinical and dosimetric predictors. Only one of the lung dosimetry parameters was input into each multivariable model to avoid multicollinearity between them. Variance inflation factor (VIF) was used to evaluate multicollinearity among independent factors in multivariate logistic regression models, and a VIF > 5 indicates multicollinearity. Mann-Whitney U test was used to compare continuous variables. A multivariable prediction model for G2P was developed. The k-fold cross validation method was utilized for the model's internal validation. A total of 104 patients were included for analysis. Twenty-one patients received consolidative immunotherapy; six of them received additional induction chemoimmunotherapy and one of them received concurrent immunotherapy combined with paclitaxel chemotherapy. In 28% (29/104) and 6% (7/104) of patients, respectively, G2P and G3P developed. The median time to G2P was 2 (range, 0-6) months. In univariable analyses, carbon monoxide diffusing capacity (DLCO), lung volume receiving ≥30 Gy (V30), mean lung dose (MLD), and immune checkpoint inhibitors (ICI) were significantly associated with G2P, while fractionation schedule (conventional fractionated radiotherapy or hypofractionated radiotherapy with dose per fraction < 2.3 Gy and ≥2.3-3Gy, respectively) showed a trend towards significance. Multivariable analyses revealed that lung V30 (hazard ratio [HR]: 1.08; 95% confidence interval [CI]: 1.02– 1.14; P = 0.006), MLD (HR: 1.17; 95% CI: 1.07–1.27; P = 0.001), and ICI (HR: 4.12; 95% CI: 2.01-8.44; P < 0.001) were risk factors for G2P, while hypofrationated radiotherapy was protective factor (HR: 0.34; 95% CI: 0.17-0.67; P = 0.002). To further investigate the possibility of selection bias, we compared the lung dosimetry parameters (volume receiving ≥5 Gy [V5], volume receiving ≥10 Gy [V10], volume receiving ≥20 Gy [V20], V30 and MLD) between the hypofractionated Results: