IDDF2021-ABS-0210 Altered gut metabolites and bacterial interactions are implicated in colorectal carcinogenesis and can detect precancerous and cancerous lesions

2021 
Background Gut microbiota contributes to colorectal cancer (CRC) pathogenesis through specific pathogens and their metabolites. We aim to determine gut microbiota-associated metabolites and their linkage to CRC. Methods We performed metabolomics by gas chromatography coupled to time-of-flight mass spectrometry, and shotgun metagenomics analysis on fecal samples from 386 subjects [118 CRC patients, 140 colorectal adenomas (CRA) patients and 128 normal controls (NC)]. Stepwise logistic regression models were used to identify metabolites and bacterial markers discriminating CRC stages. Associations among CRC-associated metabolites and bacteria were estimated with zero-inflated negative binomial regressions analysis. Results Principal component analysis and partial least squares-discriminant analysis showed differences in the gut metabolite profiles among CRC, CRA and NC groups. Norvaline and myristic acid showed increasing trends from NC, through CRA, to CRC. CRC-associated metabolites were enriched in branched-chain amino acids and aminoacyl-tRNA biosynthesis pathways. Twenty metabolites classified CRC from NC subjects with an area under the curve (AUC) of 0.80, and CRC from CRA with AUC of 0.79. Combinations of metabolites and bacterial markers improved the diagnostic performances (CRC vs NC, AUC: 0.94; CRC vs CRA, AUC 0.92; CRA vs NC, AUC: 0.86), indicating a potential for early diagnosis of colorectal neoplasia. Moreover, relationships among CRC-associated metabolites and bacteria were altered across CRC stages; certain associations exhibited increasing or decreasing strengths while some were reversed from negative to positive or vice versa. Conclusions Gut metabolites are altered in colorectal carcinogenesis. The combination of metabolites and bacterial species can increase the chance of a non-invasive diagnosis of colorectal neoplasia.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    0
    Citations
    NaN
    KQI
    []