Abstract
The incidence of colorectal cancer (CRC) has been increasing in recent decades. Current methods for CRC screening have their own drawbacks, thus there is an urgent need to identify the key microbes that drive the development of CRC for wider application in the early detection and prevention of CRC. To address this issue, we performed fecal microbiome analysis by high-throughput sequencing of 16S rRNA gene combined with blood biochemical indicators in patients with CRC stages I, II, III, and IV, healthy people, and patients with polyps. Fecal microbiota of patients with CRC was disturbed, as evidenced by significantly reduced α-diversity in patients with CRC stage IV and markedly different β-diversity. The random forest model identified the top 25 genera from 174 training data, resulting in a diagnostic accuracy of 87.95%. Further, by combining with differential genera analysis, we screened out 11 biomarkers that significantly changed in different groups. Peptostreptococcus, Parvimonas, Shewanella, Oscillibacter, Eggerthella, and Gemella associated with the development of CRC were significantly enriched, while Fenollaria, Staphylococcus, Ezakiella, Finegoldia, and Neisseria associated with the remission of CRC were significantly suppressed in patients with CRC. Importantly, carcinoembryonic antigen (CEA) was significantly correlated with these 11 microbial biomarkers, and carbohydrate antigen 19-9 (CA 19-9) was markedly correlated with Oscillibacter. Notably, co-occurrence network analysis at the genus level exhibited that the microbial co-occurrence network of CRC IV was the most complex and stable. These results suggested that CEA, CA 19-9 and 11 microbial biomarkers may be co-biomarkers for the disease occurrence and development, and non-invasive diagnosis of CRC. IMPORTANCE: Identifying the key microbes that drive the development of colorectal cancer (CRC) has been important in this field. We delved into the research on the association between CRC and fecal microbiota in this study, providing a detailed analysis of the characteristics of fecal microbiota during the transition from normal intestine to polyps to cancer. Fecal bacterial biomarkers and blood biochemical indicators may be co-biomarkers in the development of CRC.