514. Association Between Antibiotic Exposure and Clinical Outcomes of Immune Checkpoint Inhibition

Abstract

Abstract Background Recent studies have suggested that antibiotic (ABX) use and resulting changes in the gut microbiome alter the efficacy of immune checkpoint inhibitors (ICIs) in cancer treatment, but detailed data are lacking. We performed a retrospective analysis of the impact of ABX timeframe of administration on overall survival (OS) and progression-free survival (PFS) in non-small cell lung cancer (NSCLC) patients treated with ICIs. Methods We assembled a cohort of stage IV NSCLC patients treated with ICIs at Johns Hopkins Medicine between 1/1/2013 and 12/31/2019 (n=154). Patient demographics, ICI and ABX (intravenous and/or oral, n=100) treatments, tumor characteristics, and progression and survival data was extracted from the electronic medical record (EMR) and verified by chart review. Progression was defined by radiological evidence or provider notes. Univariate analysis was performed with Kaplan-Meier curves with log-rank tests and multivariate analysis was performed with Cox regression. Results Demographic data indicated that most patients were white (n=109), smokers (n=130), and had Eastern Cooperative Oncology Group (ECOG) performance statuses of zero to one (n=128). The median age was 67 years old and slightly under half of patients (n=65) received ICIs as first line therapy. Analysis with Kaplan-Meier curves for OS and PFS revealed significantly worse OS for patients with exposure to any ABX vs. not exposed in the 2.8 years before to 4.8 years after ICI start (Median OS: 14 vs. 21.5 months, p = 0.019; Median PFS: 4.4 vs. 7.7 months, p = 0.24). ABX exposure in the 2 months before to 2 months after ICI start also negatively impacted OS (Median OS: 7.5 vs. 17.4 months, p = 0.05; Median PFS: 3.3 vs. 5.8 months, p = 0.39). Decreased OS with any ABX use in the 2.8 years before to 4.8 years after ICI start was supported by multivariate analysis controlling for age, race, ECOG performance status, tumor histology, prior lines of therapy, and prior treatment types (Adjusted Hazard Ratio: 2.0, 95% CI: 1.3-3.3). Kaplan-Meier survival analysis of antibiotic exposure Overall survival (A) and progression-free survival (B) in stage IV NSCLC patients treated with ICIs, with or without exposure to ABX in the 2.8 years before to 4.8 years after ICI start. P-values calculated with log-rank tests. Conclusion Our study indicates a negative association of ABX on OS of ICI-treated NSCLC patients in both a broad and narrow timeframe around ICI start. Further work defining ABX impact on therapeutic outcomes in a larger cohort of patients across multiple tumor types is ongoing. Disclosures Grant Wilson, BS, Infectious Diseases Society of America G.E.R.M. Grant: Grant/Research Support Cynthia L. Sears, MD, Bristol Myers Squibb: Grant/Research Support Fyza Shaikh, MD, PhD, Bristol Myers Squibb: Grant/Research Support.