Metagenomic Analysis of Intratumoral Microbiome Linking to Response to Neoadjuvant Chemoradiotherapy in Rectal Cancer

Purpose To assess taxonomic and functional characteristics of tumor-bearing microbiota and its association with response to neoadjuvant chemoradiation therapy (nCRT) in patients with locally advanced rectal cancer. Methods and Materials We performed metagenomic sequencing of biopsy tumoral tissues from 73 patients with locally advanced rectal cancer before nCRT. Patients were classified into poor responders (PR) and good responders (GR) according to response to nCRT. Subsequent investigation of network alteration, key community, microbial biomarkers, and function related to nCRT responses were carried out. Results The network-driven analysis systematically revealed 2 co-occurring bacteria modules that exhibited opposite relationship with rectal cancer radiosensitivity. In the 2 modules, prominent alteration of global graph properties and community structure was observed between networks of PR and GR group. By quantifying changes in between-group association patterns and abundances, a total of 115 discriminative biomarker species linked to nCRT response were found, and 35 microbial variables were selected to establish the optimal randomForest classifier for nCRT response prediction. It yielded an area under the curve value of 85.5% (95% CI, 73.3%-97.8%) in the training cohort and 88.4% (95% CI, 77.5%-99.4%) in the validation cohort. In a comprehensive consideration, 5 key bacteria showed high relevance with inducing resistance to nCRT, including Streptococcus equinus, Schaalia odontolytica, Clostridium hylemonae, Blautia producta, and Pseudomonas azotoformans. One key hub including several butyrate-formation bacteria involving with driving network alteration from GR to PR indicate that microbiota-derived butyrate may also be involved in reducing the antitumor effects of nCRT, especially Coprococcus. The functional analysis of metagenome linked the nitrate and sulfate-sulfur assimilation, histidine catabolic process, and resistance to cephamycin to the reduced therapeutic response. It also linked to leucine degradation, isoleucine biosynthesis, taurine, and hypotaurine metabolism to the improved response to nCRT. Conclusions Our data offer novel potential microbial factors and shared metagenome function linked to resistance to nCRT. To assess taxonomic and functional characteristics of tumor-bearing microbiota and its association with response to neoadjuvant chemoradiation therapy (nCRT) in patients with locally advanced rectal cancer. We performed metagenomic sequencing of biopsy tumoral tissues from 73 patients with locally advanced rectal cancer before nCRT. Patients were classified into poor responders (PR) and good responders (GR) according to response to nCRT. Subsequent investigation of network alteration, key community, microbial biomarkers, and function related to nCRT responses were carried out. The network-driven analysis systematically revealed 2 co-occurring bacteria modules that exhibited opposite relationship with rectal cancer radiosensitivity. In the 2 modules, prominent alteration of global graph properties and community structure was observed between networks of PR and GR group. By quantifying changes in between-group association patterns and abundances, a total of 115 discriminative biomarker species linked to nCRT response were found, and 35 microbial variables were selected to establish the optimal randomForest classifier for nCRT response prediction. It yielded an area under the curve value of 85.5% (95% CI, 73.3%-97.8%) in the training cohort and 88.4% (95% CI, 77.5%-99.4%) in the validation cohort. In a comprehensive consideration, 5 key bacteria showed high relevance with inducing resistance to nCRT, including Streptococcus equinus, Schaalia odontolytica, Clostridium hylemonae, Blautia producta, and Pseudomonas azotoformans. One key hub including several butyrate-formation bacteria involving with driving network alteration from GR to PR indicate that microbiota-derived butyrate may also be involved in reducing the antitumor effects of nCRT, especially Coprococcus. The functional analysis of metagenome linked the nitrate and sulfate-sulfur assimilation, histidine catabolic process, and resistance to cephamycin to the reduced therapeutic response. It also linked to leucine degradation, isoleucine biosynthesis, taurine, and hypotaurine metabolism to the improved response to nCRT. Our data offer novel potential microbial factors and shared metagenome function linked to resistance to nCRT.