An international group of researchers announced on June 24 that they have identified a robust microbial signature associated with colorectal cancer through one of the most comprehensive analyses of the gut microbiome to date. The study, published in Cell Host & Microbe, involved reanalysis of data from 27 studies and included 6,779 publicly available gut microbiome sequencing profiles as well as 906 intestinal tissue samples.
The research team, which includes members from Germany, Switzerland, and the Netherlands as part of the Mi-EOCRC consortium and EMBL Heidelberg's Zeller and Zimmermann groups, developed computational approaches to integrate datasets generated using different sequencing methods. "The key tool is a machine learning algorithm that is trained to distinguish cancer from non-cancer microbiomes," said Zeller. "It outputs a score of how 'cancer-like' a microbiome is. We can apply this to any existing human gut microbiome dataset, including from dietary intervention studies."
Their findings showed that the colorectal cancer-associated microbial signature was consistent across various cohorts regardless of geography, sequencing method, or age at diagnosis. Analysis also revealed that microbes enriched in tumor tissue were similar to those observed in fecal samples from patients with colorectal cancer. Importantly, these microbes could already be detected in early-stage tumors within tissue samples.
However, detection accuracy was lower for early-stage cancers and tumors located further upstream in the colon when using stool samples. Michael Zimmermann said, "These results suggest that colorectal cancer-associated changes in the microbiome may appear early in disease development and raise the question of how the tumor shapes the microbiome and how the microbes impact the tumor microenvironment through signalling, metabolic, and other interactions." He also noted limitations: "This limitation is important for future clinical translation... It suggests that more sensitive approaches, larger datasets, or combinations with other measurements may be needed before microbiome-based tools could contribute to reliable detection of early pre-cancerous lesions."
The study found a link between diet—particularly fiber intake—and strength of the colorectal cancer-associated pattern; higher fiber intake was associated with reduced scores on their machine-learning classifier for this pattern.
Researchers highlighted differences among Fusobacterium subspecies related to geography and gene content within their analysis—a level of detail considered important for understanding potential health impacts.
While noting its promise as a reference point for future risk assessment or prevention strategies based on microbial signatures identified by machine learning algorithms, researchers emphasized that current methods do not yet match existing non-invasive screening tests such as fecal immunochemical tests.
