Pretreatment antibiotic exposure and mortality following immune checkpoint inhibitor therapy in non-small cell lung cancer: an infection-stratified propensity-matched analysis

PURPOSE: Pretreatment antibiotic exposure has been associated with inferior outcomes during immune checkpoint inhibitor (ICI) therapy, but whether this reflects an ICI specific effect or underlying clinical vulnerability remains uncertain. METHODS: Adults with non-small cell lung cancer (NSCLC) treated with ICIs from 2015 through 2024 were identified in the TriNetX Research Network. Recent antibiotic exposure was defined as systemic antibiotic use during days 30 to 1 before ICI initiation. Propensity score matching balanced clinical and treatment-related covariates. Primary outcomes were all cause mortality at 90 and 365 days. Sensitivity analyses included landmarking, remote exposure, infection stratification, and a non-ICI comparator. RESULTS: Among 68,723 patients, 19,080 had recent antibiotic exposure; 17,955 per group remained after matching. Mortality was higher with antibiotic exposure at 90 days (17.7% vs 13.7%; HR 1.34, 95% CI 1.27 to 1.41; P<0.001) and 365 days (43.0% vs 37.2%; HR 1.23, 95% CI 1.19 to 1.27; P<0.001). Associations attenuated after exclusion of early deaths. The association was similar between ICI and non-ICI cohorts at 90 days but modestly stronger with ICI at 365 days. Remote exposure did not reproduce the adverse association. CONCLUSIONS: Recent antibiotic exposure was associated with increased mortality during ICI therapy, but the non-ICI comparison and landmark analyses suggest that much of the short-term excess mortality reflects clinical vulnerability rather than an ICI specific effect. The modestly stronger 1-year association with ICI therapy is hypothesis generating.