Comparison of Stereotactic Body Radiotherapy and Surgery for Stage I Lung Cancer: A Multidisciplinary Cohort Study Utilizing Propensity Score Overlap Weighting and AI-Based CT Imaging Analysis

Le résumé fourni par la source

Background: With rising life expectancy and widespread lung cancer screening, early-stage non-small cell lung cancer (NSCLC) incidence has increased. While surgery is the standard treatment for operable stage I NSCLC, many patients are ineligible due to age or comorbidities. Stereotactic body radiotherapy (SBRT) has achieved good primary tumor control rates and overall survival. This study compares the outcomes of SBRT and surgery for stage I NSCLC using propensity score overlap-weighted dataset. Methods: This retrospective study analyzed clinical stage I NSCLC patients treated at a tertiary hospital from 2012 to 2021. Baseline differences between SBRT and surgery groups were adjusted using overlap weighting. AI-based CT analysis (CT AI-CAD) assessed tumor characteristics, verified by radiologists. Primary outcomes were 5-year cumulative incidence of recurrence and overall survival, with subgroup analyses based on tumor features. Results: Of 1474 patients, 1258 underwent surgery, and 216 received SBRT. After overlap weighting, baseline characteristics were well balanced. The 5-year cumulative incidence of recurrence and OS showed no statistically significant differences between SBRT and surgery groups (recurrence: 16.2% vs. 16.1%; OS: 80.5% vs. 82.9%). Further AI-based CT subgroup analysis showed no significant differences in recurrence rates across tumor features. A solid tumor diameter associated with a significant increase in recurrence was identified as 16.6 mm for SBRT and 18.6 mm for surgery. Conclusions: After overlap weighting, SBRT and surgery showed no statistically significant differences in treatment outcomes in stage I NSCLC. These findings may help guide the timing and selection of safe and effective treatment approaches.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.