Tumour-associated and non-tumour-associated microbiota in colorectal cancerOriginal paper
What was studied?
This prospective study asked whether a consistent colorectal cancer microbiota signature exists and how it links to host immune gene expression. Researchers sampled faeces and mucosa, both on and off the tumour and at proximal and distal sites. Microbiota was profiled by 16S rRNA amplicon sequencing. Expression of cancer and immune genes was measured by real-time quantitative PCR. Bacteria were grouped into co-abundance groups (CAGs) to capture community structure rather than single taxa.
Who was studied?
Participants were adults in Cork, Ireland. The study included 59 patients undergoing surgery for colorectal cancer, 21 individuals with polyps, and 56 healthy controls. Mucosal samples came from 59 cancer patients and 43 stool samples were collected, with both tissue and stool from 32 patients. Faecal controls were 37 healthy age-matched people. People with a prior history of colorectal cancer, inflammatory bowel disease, or irritable bowel syndrome were excluded.
What were the most important findings?
Cancer microbiota differed from controls across the whole colon, not just cancerous tissue, and paired on-tumour and off-tumour samples were similar. Polyp mucosa also differed from controls (p range 0.0002 to 0.03). Six co-abundance groups emerged. Bacteroidetes Cluster 1 and Firmicutes Cluster 1 fell in cancer, while a Pathogen Cluster, Prevotella Cluster, and two others rose. The Pathogen Cluster, rich in likely oral bacteria, correlated positively with IL-17a (0.46), IL-23, and CCL20 expression, suggesting a Th17 response.
What are the greatest implications of this study?
The findings suggest colorectal cancer microbiota is best seen as community-level structures spread across the whole colon, present before overt tumour formation. That challenges a one-organism model and supports microbiota as an early risk factor. Patients stratified into four groups by co-abundance clusters, some tied to inflammatory gene profiles linked to poor prognosis. The authors note bowel preparation and contemporaneous diet as possible confounders. Longitudinal work is needed before these signatures serve as screening biomarkers.