Comparative real-world survival of first-line atezolizumab, nivolumab, and pembrolizumab in older patients with metastatic non–small cell lung cancer.

e20557 Background: While new therapies known as immune checkpoint inhibitors (ICIs) have improved overall survival (OS) in metastatic non–small cell lung cancer (mNSCLC), there is a paucity of comparative effectiveness evidence to inform optimal first-line treatment selection. This study compared OS among older patients with mNSCLC who received first-line atezolizumab, nivolumab, or pembrolizumab. Methods: This retrospective cohort study used the 2014-2020 SEER-Medicare data and included patients aged 66 years or older, diagnosed with mNSCLC, and treated with first-line atezolizumab, nivolumab, or pembrolizumab. OS was compared using the multivariable-adjusted (MV-adjusted), and propensity score matching (PSM) Cox proportional hazards (CPH) models. The Restricted Mean Survival Time (RMST) model was applied when the proportional hazards (PH) assumption was violated. Sensitivity analyses were performed under two conditions: (1) restricting the cohort to patients who continuously used ICIs for at least three months and (2) censoring patients upon switching to another ICI. Subgroup analyses by histology (squamous or nonsquamous) were also conducted. Results: The study cohort included 4,635 patients, with 112 treated with atezolizumab, 251 with nivolumab, and 4,272 with pembrolizumab. Pembrolizumab was associated with a significant survival advantage compared to atezolizumab (MV-adjusted hazard ratio (HR)=0.67; 95% confidence interval (CI)=0.53-0.84; PSM HR=0.69; 95% CI=0.54-0.87) and nivolumab in MV-adjusted CPH model (HR=0.83; 95% CI=0.69-0.99). The survival benefits of pembrolizumab remained consistent in sensitivity analyses when compared to atezolizumab (MV-adjusted HR =0.58; 95% CI=0.43-0.79; PSM HR=0.65; 95% CI=0.48-0.87) and nivolumab (three-year MV-adjusted RMST ratio=1.24; 95% CI=1.08-1.43; three-year PSM RMST ratio=1.17; 95% CI= 1.01-1.34). The OS difference between nivolumab and atezolizumab was inconclusive (MV-adjusted HR=0.73; 95% CI=0.46-1.16; PSM HR=0.65; 95% CI: 0.41-1.04). In subgroup analyses, the OS difference between pembrolizumab and nivolumab was inconclusive among patients with squamous histology (MV-adjusted HR=0.85; 95% CI=0.65-1.11; PSM HR=1.15; 95% CI=0.89-1.49). Conclusions: In this population-based study, first-line pembrolizumab was associated with a significant OS benefit compared with atezolizumab and nivolumab among older adults with mNSCLC. For patients with squamous histology, OS differences between pembrolizumab and nivolumab were inconclusive. These findings support pembrolizumab as the preferred first-line treatment for older patients with mNSCLC, while highlighting the need for further investigation into the effectiveness of nivolumab specifically within squamous histology subgroups.