Colibactin, a genotoxin produced by pks+ E. coli, imprints highly specific mutational signatures SBS88 and ID18 in colorectal cancer (CRC) and even in normal colonic crypts. Population-scale analyses show these signatures are enriched in early-onset CRC, vary geographically, and are imprinted early during tumor evolution, where probabilistic attribution indicates that colibactin contributes to a measurable fraction of APC driver mutations in colibactin-positive cancers. Beyond colibactin, Fusobacterium nucleatum exerts clade-specific effects on tumor ecology and therapy response, with data supporting both chemoresistance and sensitization to anti-PD-1 in microsatellite stable (MSS) CRC. This article covers mechanistic, genomic, and molecular epidemiology evidence, outlines analytic standards for signature detection (whole-genome sequencing (WGS)/whole-exome sequencing (WES), single-sample fitting, and limits at low mutation counts), and charts translational paths spanning noninvasive screening (stool metagenomics + mutational signatures in tissue/circulating tumor DNA (ctDNA)), risk stratification, and microbial-targeted interventions (antibiotics, phages, ClbP inhibitors). Framing microbiome–genome crosstalk as a tractable axis enables testable clinical hypotheses for precision oncology.
Sungwon Jung (Mon,) studied this question.