Impact of antibiotic use on clinical outcomes in patients with urothelial cancer receiving a programmed death protein 1 or programmed death ligand 1 (anti-PD-1/L1) antibody

4557 Background: Previous data has suggested that patients treated with anti-PD-1/L1 antibodies who receive antibiotics during their therapy might have dramatically decreased progression-free and overall survival 1,2. This has clinical implications for management of patients with suspected bacterial infection while on treatment with these agents. We assessed the relationship between antibiotic use and tumor response rate, progression-free survival, and overall survival in a large dataset of patients with urothelial cancer treated with anti-PD-1/L1 antibodies. Methods: We examined seven trials that led to drug approval and which included 1747 patients with metastatic or locally advanced urothelial cancer treated with an anti-PD-1/L1 antibody. Five trials enrolled patients who had received prior platinum-based therapy and two enrolled patients who were cisplatin-ineligible. Six were single arm trials and one was a randomized controlled trial whose control arm is not included in these analyses. Concomitant medication datasets were searched for systemic antibiotic used by each patient while on treatment. Results: Overall, 51% of patients (n=892) were exposed to antibiotics (ABX+) and 49% (n=855) were not exposed (ABX-). In these exploratory analyses, small numeric differences in OS, PFS, and ORR were seen in ABX+ vs. ABX- patients. Median OS was 9.23 vs. 9.86 months, median PFS was 105 vs 101 days, and ORR was 20% vs. 21% in ABX+ vs. ABX- patients, respectively. Conclusions: Patients who were treated with antibiotics while on therapy with an anti-PD-1/L1 antibody for urothelial cancer had similar outcomes to those who were not treated with antibiotics. Numeric differences in outcomes were not significant and did not duplicate previous analysis demonstrating a median OS that was doubled in ABX- patients1. Our exploratory analyses do not appear to demonstrate a clear need for practitioners to avoid antibiotic use in patients treated with PD-1/L1 agents for fear of significantly impacting clinical outcomes. References: 1) Tinsley et. al., ASCO annual meeting 2018, abstract 3010 2) Routy et. al., Science 05 Jan 2018: Vol. 359, Issue 6371. [Table: see text]