Systematic review and meta-analysis evaluating the impact of antibiotic use on the clinical outcomes of patients with cancer treated with immune checkpoint inhibitors.

Abstract

2585 Background: In recent years, the gut microbiome has increasingly emerged as influencing the response to immune checkpoint inhibitors (ICIs) and antibiotic (ABX) exposure has repeatedly been shown to impair clinical outcomes of patients suffering from different cancer types and treated with ICIs. We published in 2020 a meta-analysis confirming that ABX use hampered survival of non-small cell lung cancer (NSCLC) patients treated with ICIs. The present study aims to determine whether ABX use also reduces survival of patients receiving ICIs for other cancers. Methods: PubMed and major oncology conferences’ proceedings were systematically searched to identify studies assessing the impact of ABX on the clinical outcomes of cancer patients treated with ICIs. Studies were included when reporting data on Overall Survival (OS), Progression-Free Survival (PFS), Overall Response Rate (ORR) and Progressive Disease Rate (PD), according to ABX exposure. Pooled Hazard Ratios (HRs) for OS and PFS and Odds Ratios (ORs) for ORR and PD were calculated, as well as HRs for OS and PFS according to different cancer types and different ABX exposure time windows (TWs). Results: Overall, 94 independent cohorts were included, representing 26,174 patients suffering from various types of cancer. The pooled HRs for PFS (61 cohorts, 13,224 patients) and OS (88 cohorts, 25,480 patients) were 1.47 [95% Confidence Interval (CI) 1.31-1.66] and 1.66 [95% CI 1.50-1.83], respectively, confirming a significant harmful impact of ABX on patient’ survival, observed across all cancer types (Table). The analyses of OS and PFS based on ABX exposure TWs suggested a stronger deleterious effect of ABX when taken around ICI treatment initiation. The response to treatment among ABX users was also impaired: the pooled ORs for ORR (30 cohorts, 4,590 patients) and PD (33 cohorts, 4,972 patients) were 0.55 [95% CI 0.39-0.77] and 1.97 [95% CI 1.48-2.64], respectively. Conclusions: ABX were shown to impair the clinical outcomes of cancer patients treated with ICIs, regardless of cancer type. [Table: see text]