A retrospective analysis of antibiotics usage and effect on overall survival and progressive free survival in patients with metastatic pancreatic cancer.

e15781 Background: Recent studies have shown that the pancreatic ductal adenocarcinoma (PDAC)-associated gut microbiome can play a major in modulating responses to therapies. Antibiotics (ATB) can potentially alter the tumor and gut microbiota diversity and composition leading to modified responses to chemotherapeutic/immune therapy regimens and hence the survival. Methods: We retrospectively analyzed the clinical data of 148 patients (pts) with documented metastatic PDAC seen at MD Anderson Cancer (MDACC) from 2009 to 2017. Along with demographic and chemotherapy regimen details the duration, type and reason for antibiotic consumption for more than 3 days were recorded. Overall survival (OS) and Progression free survival (PFS) were calculated. Log-rank test and Gehan-Breslow-Wilcoxon test were used to check the statistical significance for OS and PFS. Confounding variables were also accounted. Results: We analyzed the data of 148 metastatic pancreatic cancer pts[mean age 62.73, 50.67% males 49.32% females, 75.6% white] out of which 135 patients received antibiotics. The infectious sources consisted of intraabdominal (n = 68), urinary (n = 36), respiratory (n = 57), skin/soft tissue infections (n = 26), blood related (n = 24), and others (n = 76). Beta lactams (n = 96) and Quinolones (n = 96) were the most commonly prescribed antibiotics. When comparing outcomes, we found out that the median OS for pts taking macrolides (n = 24) was 541 days compared to 341 days for pts not taking macrolides (n = 144) (HR = 0.6384, p value = 0.0191) . Median PFS for pts taking macrolides was 178 days versus 124 days in those not taking macrolides (HR = 0.6331, p value = 0.0188). The potential confounders were having a respiratory infection or the type of chemotherapy regimen. Conclusions: Macrolide consumption for > 3 days leads to a prolonged OS and PFS. Consistent with the preclinical evidence our data suggests a potential role for some antibiotics in modulating PDAC-associated gut microbiome. Hence, having implications on the survival of PDAC pts. [Table: see text]