Impact of commonly prescribed medications on immune-related adverse events.

e14704 Background: Commonly used medications may have immune modulating properties that affect the risk of immune-related adverse events (irAEs) after immune checkpoint inhibitor (ICI) therapy. Methods: We identified 27,998 patients who received ICI from 2010 to 2023 in the national Veterans Affairs system, matched to 66,488 historical control patients treated with non-ICI systemic therapies. We extracted medication usage in the year before therapy start. We used propensity-weighted Cox regression analysis to assess the effect of each medication on severe irAE (treatment with prolonged course of high-dose steroids, adrenal insufficiency, or new insulin-dependent diabetes). To identify medication effects specific to ICI vs. other therapies, we report the adjusted hazard ratio (aHR) for each medication stratified by cohort (ICI vs. historical control) and the p-value for the interaction between each medication class and cohort. Results: Among ICI patients, 52% of patients were treated in the first line, 72% received ICI monotherapy, and 95% received PD-1/PD-L1 inhibitors alone. The most common cancer types were NSCLC (46%), melanoma (10%), and liver cancer (8%). Among 15 medication classes, metformin, loop diuretics, and PPIs were associated with small, marginally statistically significant differences in toxicity among ICI patients, but similar associations were observed in historical control patients (p for interaction >0.18; Table). After adjustment for multiple comparisons, no medication classes were associated with increased risk of severe irAE in ICI patients. Conclusions: Commonly prescribed medication classes have no significant effects on toxicity specific to ICI therapy. [Table: see text]