Risk-adapted use of thoracic radiotherapy in stage IV non-small cell lung cancer: a multicenter prognostic model study

Abstract

Background Survival benefit of thoracic radiotherapy (TRT) in patients with stage IV non-small cell lung cancer (NSCLC) receiving first-line chemo-immunotherapy remains uncertain. This study developed and validated an overall survival (OS) prediction model and assessed whether TRT benefit varied across risk strata. Methods Patients with stage IV NSCLC treated with first-line chemo-immunotherapy were enrolled. A nomogram was developed using baseline clinical variables to calculate individual risk scores. Patients were classified into high- and low-risk groups according to the median nomogram-derived score. OS was compared between ICI and ICI+TRT groups in the overall cohort and risk subgroups. Treatment-by-risk interaction, 1:1 propensity score matching (PSM), treatment response, TRT pattern, BED10, and metastatic burden subgroup analyses, and safety assessments were performed. Results Among 514 patients, 284 without TRT were assigned to training and internal validation cohorts at a 7:3 ratio, and 84 non-TRT patients formed an external validation cohort. ECOG performance status, alkaline phosphatase, total protein, platelet-to-lymphocyte ratio, and systemic immune-inflammation index were selected as optimal prognostic factors. The nomogram showed good 1- and 2-year OS discrimination, calibration, and clinical utility. Using a median risk score of 103.1, TRT improved OS only in low-risk patients (HR 0.456, 95% CI 0.285–0.730; P = 0.0011), not high-risk patients (HR 0.914, 95% CI 0.556–1.503; P = 0.7239), with consistent findings after PSM. Treatment-by-risk interaction was significant (P = 0.0309). Grade 1–2 pneumonitis was higher with ICI+TRT than ICI alone (28.8% vs. 10.3%; P < 0.001), without significant differences in severe pneumonitis or immune-related adverse events. Conclusion TRT benefit was risk-dependent, improving OS in low-risk but not high-risk patients.