Abstract

BackgroundObservational studies and experimental models suggest that use of antibiotics close to the administration of immune checkpoint inhibitors (ICI) can negatively affect tumor response and patient survival. This observation may be attributed to microbiome dysbiosis and the resultant suppression of host immune response against neoplastic cells.MethodsWe conducted a systematic search of PUBMED and EMBASE databases and references of articles retrieved. We included studies published between 1/1/17 and 2/1/20, which evaluated the association between antibiotic use and clinical outcomes in cancer patients treated with ICI. Primary endpoints were overall survival (OS), progression free survival (PFS), response rate (RR) and progressive disease (PD) rate. We performed a study-level random-effects meta-analysis with pooling of hazards ratios (HR) for OS, PFS, and odds ratios (OR) for RR and PD (PROSPERO ID: CRD42020166473).ResultsWe included 41 studies with a total of 10,857 patients. The most common malignancies were lung cancer (59.7%), melanoma (23.1%), renal cell and urothelial carcinomas (8.1%). OS and PFS were shorter, RR lower, and PD higher in patients receiving antibiotics, both in univariate analyses and after adjustment for other confounders. Heterogeneity was significant for all outcomes, less so for adjusted OS and PFS (Table). To our knowledge, this is the largest meta-analysis on the association between antibiotic use and efficacy of ICI, and the only one to address RR and PD to-date.Association between antibiotics and clinical outcomes. ConclusionWe demonstrated a significant association between antibiotic use and unfavorable clinical outcomes in patients with cancer receiving ICI. Such patients may be an important target group for antibiotic stewardship interventions. The high heterogeneity across all outcomes underscores the need for more detailed, patient-level studies with stratification by host, antibacterial and cancer treatment factors.Disclosures All Authors: No reported disclosures

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call