Pancreatic cancer intratumoral microbiome and characteristics within paired patient samples.

Abstract

744 Background: While most studies evaluating the microbiome in gastrointestinal cancers analyze stool, little is known about the microbiota of the peri-tumoral and intra-tumoral environment. Here, we evaluated the intra-tumoral and peri-tumoral (duodenum and normal pancreas) microbiome for paired duodenal, normal pancreas and resected tumor specimens from pancreatic cancer patients. The purpose of this study was to describe the similarities and differences within patient microbiota. Methods: Fifteen specimens from 5 patients with pancreatic cancer were collected during surgical resection. Genomic bacterial DNA was extracted from these specimens and underwent 16S rRNA sequencing. Alpha (Inverse Simpson) and beta diversity were calculated, and relative abundances of individual bacterial species were compared. Sorensen distance was used the evaluate the spread in beta diversity between paired sample types. Results: Of the five patients who underwent resection, the following baseline characteristics were obtained: median age = 65 years (range 55 -80 years), 2/5 patients were treated with gemcitabine/abraxane, 3/5 patients were treated with oxaliplatin, irinotecan, fluorouracil, and leucovorin (FOLFIRINOX); 4/5 patients received pre-operative radiation. 16s sequencing analysis of the pancreatic tumor showed the dominant genus to be Escherichia/Shigella (10.6%). Bradyrhizobium (10.1 %) was dominant in the normal pancreas . Escherichia/Shigella (14.3%) was abundant in the duodenum. There was a trend towards higher alpha diversity in tumor vs. normal duodenum/ pancreas (p = 0.12). Sorensen distance was statistically different between sample types (p = 0.004), with duodenal samples most consistent (distance = 67.82), and tumor vs. normal pancreas (81.86) and tumor vs. other tumor samples the most heterogeneous (78.5). Conclusions: This pilot data suggests that the pancreatic tumor microbiome is distinct from the normal pancreas and duodenal microbiome, which indicates tumor specific bacteria should be studied. In future studies, intra-tumoral microbiome may be more relevant to associations with outcomes and treatment response than stool or intestinal microbiome studies.