Effect of baseline medications on response to immunotherapy in hepatocellular carcinoma.

Abstract

484 Background: Immunotherapy (IO) response rates in advanced hepatocellular carcinoma (HCC) are less than 20%. The microbiome has been shown to mediate IO response in experimental models, and clinical studies have observed that antibiotics, especially prior to IO initiation, are associated with reduced IO response. We reasoned that commonly prescribed antacid medications, such as proton pump inhibitors (PPIs) and histamine receptor antagonists (H2RAs), which are known to influence the microbiome, may also influence IO response. Methods: This is a retrospective chart review-based study of 95 patients with advanced HCC treated with IO at a single academic medical center. The primary outcome was overall survival (OS). The secondary outcome was overall response rate (ORR). The primary predictors were antibiotic or antacid exposure in the 60 days prior to IO. A secondary predictor was antibiotic or antacid exposure in the 30 days prior to IO. Results: The cohort was predominantly male (84%), was racially diverse (31% White, 23% Black, 23% Asian, 13% Hispanic), and had a median age of 65 years. There were 49 deaths with a median follow up of 0.96 years. The most common underlying liver diseases were HCV (49%), HBV (31%), and NASH (11%). The majority of patients had cirrhosis (80%), with a median Child Pugh score of 6. Within 60 days before IO, 25 patients received antibiotics, 40 received PPIs and 5 received H2RAs. Most patients receiving antibiotics also received a PPI (92%). The median duration of antibiotics was 5 days. Neither antibiotic nor antacid exposure within 60 or 30 days prior to IO was significantly correlated with OS in univariate or multivariate analyses, nor were they correlated with ORR. Conclusions: No significant associations between baseline exposure to antibiotics and antibiotics and OS or ORR were identified in this single-institution study. Larger observational studies or mechanistic studies are needed to clarify interactions between medications, the microbiome, and IO response. [Table: see text]