A Composite Microbe-Metabolite Diagnostic Panel: Further Discussion Is Still Needed.

We read with great interest the article by Gao et al1Gao R. et al.Gastroenterology. 2022; 163: 1024-1037Abstract Full Text Full Text PDF PubMed Scopus (5) Google Scholar that reported on the gut microbial and serum metabolic signatures of colorectal cancer (CRC) with broad population relevance. The authors established a composite microbe–metabolite diagnostic panel for detecting CRC and performed cross-validation in published metagenomic datasets of 7 independent CRC cohorts. This is an impressive study, and we commend the authors for their contributions. However, we have some concerns to share with the authors. First, there were some concerns about the gut microbes identified in the study. Among the 12 gut microbes identified from the entire in-house cohort (Shanghai CRC metagenomic-metabolomic cohort), Parvimonas unclassified, Fusobacterium nucleatum, Roseburia intestinalis, Peptostreptococcus anaerobius, Peptostreptococcus stomatis, Gemella morbillorum, Porphyromonas asaccharolytica, Faecalibacterium prausnitzii, and Parvimonas micra show CRC-related alterations in the feces, which are consistent with previous published results.2Coker O.O. et al.Microbiome. 2022; 10: 35Crossref PubMed Scopus (26) Google Scholar,3Yu J. et al.Gut. 2017; 66: 70-78Crossref PubMed Scopus (571) Google Scholar Gardnerella vaginalis, Lactobacillus iners, and Atopobium vaginae are newly discovered gut microbes associated with CRC in the study. However, these 3 gut microbes are usually isolated from the lower reproductive tract in women, and although these microbes have been reported in male urethral and pulmonary infections, they are mostly case reports.4Wu S. et al.Infect Drug Resist. 2021; 14: 5253-5259Crossref PubMed Scopus (2) Google Scholar At present, no large-scale study has reported the colonization of these microbes in men. The reason the 3 microbes were selected out may be due to the discrepancy of gender in the training cohort, which was the higher proportion of women in the healthy population and the lower proportion of women in CRC and colorectal adenoma (CRA) patients (male/female: CRC, 20/15; CRA, 18/13; healthy control, 6/28). Whether the occasional infection of these microbes in men is related to CRC still has to be investigated. Thus, the inclusion of these microbes in the diagnostic model needs further consideration. Second, the use of drugs in patients as an important determinant of human gut microbiome structure was not exhibited clearly in the study. Several studies have pointed out that commonly used drugs such as nonsteroidal anti-inflammatory drugs, antibiotics, and proton pump inhibitors are associated with specific alterations in gut microbiology and have a cumulative effect.5Nagata N. et al.Gastroenterology. 2022; 163: 1038-1052Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar For F nucleatum in the composite microbe–metabolite diagnostic panel for detecting CRC, it has been proven that aspirin and its primary metabolite salicylic acid could kill both actively growing and stationary F nucleatum and might contribute to the prevention of CRC by modulating gut microbiota.6Brennan C.A. et al.mBio. 2021; 12 (e00547-21)Crossref PubMed Scopus (21) Google Scholar More than 30% patients in the Shanghai cohort had hypertension, diabetes, or cardiac disease, whereas the medications used by these patients were unclear, which may have a potential impact on gut microbiome. Therefore, insufficient analysis of medication use might cause the application of the diagnostic panel to be restricted. Finally, although the authors performed validation of an external cohort to demonstrate good diagnostic efficacy of the diagnostic model, the study was analyzed as a cross-sectional analysis, causing some changes in microbe composition and metabolites to perhaps be the result of cancer or other variable. Therefore, a reliable diagnostic model should be validated in prospective cohorts or large sample sizes in the real world.7Au E.H. et al.Kidney Int. 2020; 97: 877-884Abstract Full Text Full Text PDF PubMed Scopus (8) Google Scholar We note that the authors expect to enroll a population of 2000 in the clinical trial registry, and we expect that future results and interpretation of the model will be performed based on this larger sample size and through specific microbial- and metabolite-based basic research. In addition, comparing the efficiency of detecting CRC or CRA with the existing clinical biomarkers of CRC, such as carcinoembryonic antigen, may provide better evidence to support the clinical application of this diagnostic model. We sincerely look forward to the results of the large sample clinical trial. In conclusion, this study will be of great help for early diagnosis and treatment of CRC. However, further clarification of the aforementioned concerns is needed. Integrated Analysis of Colorectal Cancer Reveals Cross-Cohort Gut Microbial Signatures and Associated Serum MetabolitesGastroenterologyVol. 163Issue 4PreviewCross-cohort gut microbiome signatures and their substantial serum metabolic links for CRC are identified and a composite microbial–metabolic predictive model is established, which may assist its diagnosis and therapeutics. Full-Text PDF ReplyGastroenterologyVol. 164Issue 4PreviewIn our full report,1 we generated a panel of 12 gut bacteria for discriminating patients with colorectal cancer (CRC) and healthy control subjects from our in-house cohort and validated the biomarkers in 7 published metagenomic datasets of CRC. In their comments, Wang et al2 argued that 3 bacteria in the panel, namely Gardnerella vaginalis, Lactobacillus iners, and Atopobium vaginae, are newly discovered biomarkers of CRC in our report and are usually isolated from the lower reproductive tract in women. Full-Text PDF