Abstract
OBJECTIVE: Most patients with sporadic colorectal cancer (SCRC) develop microsatellite instability because of defects in mismatch repair (MMR). Moreover, the gut microbiome plays a vital role in the pathogenesis of SCRC. In this study, we assessed the microbial composition and diversity of SCRC tumors with varying MutL protein homolog 1 (MLH1) status, and the effects of functional genes related to bacterial markers and clinical diagnostic prediction. METHODS: The tumor microbial diversity and composition were profiled using high-throughput sequencing of the 16S ribosomal RNA (rRNA) gene V4 region. Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt2) software and BugBase tool were used to predict the functional roles of the microbiome. We aimed to construct a high-accuracy model to detect and evaluate the area under the receiver operating characteristic curve with candidate biomarkers. RESULTS: The study included 23 patients with negative/defective MLH1 (DM group) and 22 patients with positive/intact MLH1 (IM group). Estimation of alpha diversity indices showed that the Shannon index (p = 0.049) was significantly higher in the DM group than in the controls, while the Simpson index (p = 0.025) was significantly lower. At the genus level, we observed a significant difference in beta diversity in the DM group versus the IM group. Moreover, the abundance of Lachnoclostridium spp. and Coprococcus spp. was significantly more enriched in the DM group than in the IM group (q < 0.01 vs. q < 0.001). When predicting metagenomes, there were 18 Kyoto Encyclopedia of Genes and Genomes pathways and one BugBase function difference in both groups (all q < 0.05). On the basis of the model of diagnostic prediction, we built a simplified optimal model through stepwise selection, consisting of the top two bacterial candidate markers (area under the curve = 0.93). CONCLUSION: In conclusion, the genera Lachnoclostridium and Coprococcus as key species may be crucial biomarkers for non-invasive diagnostic prediction of DM in patients with SCRC in the future.