M. A. Mederos1, A. McElhany1, J. Petrosino2, N. J. Ajami2, N. Villafane1, S. Mohammed1, E. Oliva1, W. E. Fisher1, G. Van Buren1 1Baylor College Of Medicine,Department Of Surgery, Division Of Surgical Oncology,Houston, TX, USA 2Baylor College Of Medicine,Department Of Molecular Virology And Microbiology,Houston, TX, USA
Introduction: Pancreatic ductal adenocarcinoma (PDAC) is a fatal disease that lacks a method of early detection. Established risk factors for PDAC suggest an inflammatory mechanism of carcinogenesis. Recent epidemiological data portends that certain bacterial populations may increase PDAC susceptibility & progression by diverse mechanisms including modulating inflammation. We aim to characterize the microbiome of pancreatic tissue in those with PDAC & chronic pancreatitis (CP).
Methods: Patients who underwent pancreas resection at our institution from 2004-2016 were identified from our database. Baseline demographics, co-morbid conditions, clinical characteristics, & outcome data were obtained from review of the database. In this cross-sectional study, we used next-generation sequencing protocols & high throughput 16S rRNA gene sequencing to characterize the microbiota in CP/PDAC & normal-adjacent tissue samples (n=6 matched pairs). Shannon diversity indices were used to compare taxonomic richness. Beta-diversity distance comparison was used to compare compositional differences between the two patient groups. P values were calculated using a Wilcoxon matched-pairs rank test & Kruskal-Wallis statistical tests with false discovery rate correction.
Results: Baseline demographics were similar between patient groups. Analysis shows that the microbiome in PDAC & CP tissue samples is dominated by species of lipopolysaccharide (LPS)-rich Proteobacteria followed by Firmicutes, Bacteroidetes, & Actinobacteria with relative abundance means of 53.65%, 31.55%, 5.56%, & 5.04%, respectively. We also observed lower bacterial richness in tumor-associated samples compared to normal-adjacent (p=0.65). This finding was accompanied by an overall increase of LPS-rich Enterobacter species in PDAC tissue compared to normal tissue (p=0.094).
Conclusion: Our analysis showed a dominance of LPS-rich Proteobacteria & a trend to lower microbial richness. Low microbial richness suggests the dominance of a single or few bacteria — a hallmark of microbe-induced inflammatory processes. The trend toward decreased diversity is possibly a result of a multi-factorial dysbiotic state & probably contributes to the pathogenesis of PDAC. Recent studies suggest a role for Proteobacteria in carcinogenesis through inflammatory processes mediated by TLR-activating molecules such as LPS. There is an inherent basal diversity in the microbial structure across multiple body sites due to environmental & genetic factors, thus more specimens are needed to correlate microbial patterns with clinical outcomes. Further studies with more specimens are needed to detect a statistical difference of the microbiome between matched tissue pairs.
