Integrated analysis of gut microbiome compositional and genomic alterations reveals strain diversity and actionable biomarkers across multiple colorectal cancer cohorts

Gut microbiome dysbiosis is considered to play a critical role in the development of colorectal cancer. While numerous studies have identified specific microbial species enriched in cancer patients, knowledge about their genomic variants remains limited. Here, we systematically investigated the compositional and genomic alterations in the gut metagenomic sequencing data using 1723 samples collected from 10 different CRC datasets. Our combined analysis revealed a core set of 250 microbial genomes, 4365 single nucleotide variants and 21 structural variants differed between CRC and healthy control groups. Utilizing the microbial SNV and SV profiles, we identified strain level phylogenetic diversity among 27 CRC related species, such as Faecalibacterium prausnitzii, Coprococcus_B comes, Fusicatenibacter saccharivorans, and Ruminococcus_D bicirculans. We also constructed metagenomic classification models based on these multi-dimensional alterations and observed superior diagnostic capability of the multi-dimensional model, achieving an average AUC of 0.82 and 0.80 in the discovery stage and validation stage, respectively. Compared with fecal occult blood test (FOBT), we identified a combination of 4 species, could increase the AUC of FOBT testing by 13.5 our integrated analysis uncovers how microbial genomic variations, in conjunction with compositional changes, influence cancer progression and suggests combining microbial features with cost-effective FOBT testing may be a promising approach for the mass screening of CRC.