Real-world utilization and survival trends of adjuvant immunotherapy in postoperative stage II-III non-small cell lung cancer: A Louisiana population-based study.

Abstract

e20072 Background: Currently adjuvant immunotherapy (AI) is standard of care for patients with resectable non-small cell lung cancer (NSCLC). While clinical trials have established the efficacy of AI, real-world evidence remains limited. This study aimed to evaluate the real-world adoption of AI and its survival benefits in patients with postoperative stage II-III NSCLC using a population-based data. Methods: We analyzed data from the Louisiana Tumor Registry (LTR) for patients diagnosed with stage II-III NSCLC between 2015 and 2023 who underwent definitive surgery. Survival curves were estimated using the Kaplan–Meier method. Multivariable Cox proportional hazards models were employed to assess overall survival (OS), adjusting for sociodemographic factors (age, race, sex, insurance, marital status), clinical variables (stage, grade, tumor number, comorbidities), and treatment (chemotherapy, radiation). Results: This study enrolled 1,473 patients with a mean age of 66.17±9.14 years and median follow-up of 36 months. Of these, 186 (12.63%) received AI, with 51 (27.42%) deaths during follow-up. Kaplan-Meier survival estimates demonstrated that the AI group achieved superior overall survival compared to the non-AI group during the initial three years of follow-up. period. After adjusting for important confounders, the AI group had lower mortality risk compared to the non-AI group, although this did not reach statistical significance (aHR = 0.778; 95% CI = 0.580–1.044, p = 0.094). Additionally, race and marital status were not significantly associated with survival outcomes in the Cox model. However, female sex (aHR: 0.749, p = 0.0004) and having Medicaid insurance (HR: 1.276, p = 0.0463) were identified as significant independent predictors of survival. Conclusions: AI may provide survival benefits for postoperative stage II-III NSCLC patients in a real-world setting. More real-world data with larger sample size and longer follow-up periods would be warranted.