Development of a microbiome- and metabolite-based prediction tool for colonic polyposis with and without a known genetic driver

Background and Aims: Although the microbiome and metabolome of sporadic colon polyps and colorectal cancer has been extensively studied, analysis of individuals with colonic polyposis has been limited.Methods: Right/left colon biopsies, stool, polyps, and biopsies of uninvolved tissue surrounding polyps were collected from individuals with polyposis (gene-positive [Gene-pos], gene-negative [Gene-neg], serrated polyposis syndrome [SPS]) and without polyposis to assess for mucosal and stool microbiome and metabolomic signatures.Results: First, we demonstrate that the mucosa-associated microbiota in individuals with colonic polyposis was representative of the polyp associated microbiota. Next, taxonomic analysis showed differential microbiota populations between the Gene-pos and Gene-neg polyposis cohorts as well as SPS compared to other Gene-neg subjects. We used these differential microbiota taxa to perform linear discriminant analysis, allowing us to differentiate Gene-pos from Gene-neg subjects and SPS from Gene-neg subjects with an accuracy of 89% and 93% respectively. Targeted metabolomics using 1H NMR to quantitatively measure stool metabolites found an increase in alanine and valerate in SPS subjects. To identify potential biomarkers for polyposis, we generated receiver operating characteristic (ROC) curves and found that the proportion of leucine to tyrosine and 3-hydroxyisovalerate to leucine in fecal samples had AUC of 0.844 and 0.811 respectively, making these ratios potential biomarkers for differentiating SPS from non-polyposis subjects.Conclusion: We characterized the microbiota and metabolomic of individuals with colonic polyposis, demonstrating relevant compositional differences, and allowing for prediction of the presence of polyposis using microbiome and metabolomic signatures which may pave the way for better diagnostic tools for individuals with colonic polyposis and their families.