Abstract
OBJECTIVE: We aimed to compare overall survival (OS) in immune checkpoint inhibitor (ICI)-treated patients with metastatic non-small cell lung cancer (mNSCLC) with pre-existing rheumatoid arthritis (RA) versus those without RA. METHODS: A retrospective cohort study using Medicare claims data was performed. Participants included patients aged ≥66 years with a diagnosis of a malignant neoplasm of lung and bronchus who initiated nivolumab, pembrolizumab, or atezolizumab between March 4, 2015 and April 11, 2019, which is after the US Food and Drug Administration (FDA) approval of ICIs for mNSCLC but before the first FDA approval for stage III disease. Survival analysis using Kaplan-Meier and adjusted Cox proportional hazard models was performed. RESULTS: A total of 2,732 people with mNSCLC (N = 790 RA and N = 1,942 non-RA) were in the analytic cohort. Patients with RA were more likely to be female and had more comorbidities than patients without RA. Patients with RA were more likely to be taking steroids than those without RA (63% vs 45%) but equally likely to be taking dexamethasone, usually prescribed for cancer palliation, specifically (27% vs 28%) before ICI initiation. There was no difference in OS between the RA and non-RA NSCLC Kaplan-Meier survival curves (log-rank P = 0.08) and in adjusted models (hazard ratio 0.92, 95% confidence interval 0.78-1.09). Male sex, having more comorbidities, and steroid dose before ICI initiation were associated with worse OS. In a sensitivity analysis omitting patients receiving baseline dexamethasone, steroid dose before ICI initiation was no longer associated with worse OS. CONCLUSION: After controlling for demographics and comorbid conditions, ICI-treated patients with RA with mNSCLC had no difference in OS compared with patients without RA. After excluding patients receiving dexamethasone, steroid dose was not associated with worse OS.