Gut microbiome meta-analysis reveals dysbiosis is independent of body mass index in predicting risk of obesity-associated CRC

Abstract

Obesity is a risk factor for colorectal cancer (CRC), accounting for more than 14% of CRC incidence. Microbial dysbiosis and chronic inflammation are common characteristics in both obesity and CRC. Human and murine studies, together, demonstrate the significant impact of the microbiome on governing energy metabolism and CRC development; yet, little is understood about the contribution of the microbiome to development of obesity-associated CRC as compared to non-obese individuals. In this study, we conducted a meta-analysis using five publicly available stool and tissue-based 16S rRNA and whole genome sequencing (WGS) data sets of CRC microbiome studies. High-resolution analysis was employed for 16S rRNA data using Resphera Insight, which allowed us to achieve species-level information to compare with WGS. Characterization of the confounders between studies, 16S rRNA variable region, and sequencing method, did not reveal any significant effect on alpha diversity in CRC prediction. Both 16S rRNA and WGS were equally variable in their ability to predict CRC. Results from community structure and composition analysis confirmed lower diversity in obese individuals without CRC; however, no universal differences were found in diversity between obese and non-obese individuals with CRC. When examining taxonomic differences, the probability of being classified as CRC did not change significantly in obese individuals for all taxa tested. However, random forest classification was able to distinguish CRC and non-CRC stool when body mass index was added to the model. Overall, microbial dysbiosis was not a significant factor in explaining the higher risk of colon cancer among individuals with obesity.