This exploratory preprint used whole-exome sequencing (WES) data to characterize tumor-colonizing microbiota in primary colorectal tumors from Hispanic/Latino (H/L) patients. The analysis aimed to compare microbial composition and predicted function between early-onset colorectal cancer (EOCRC) and late-onset colorectal cancer (LOCRC) within a small proof-of-concept cohort.
The authors emphasize that this work is preliminary, based on a limited sample set, and has not been peer reviewed. The study reports institutional ethical approval and patient consent and declares no competing interests.
The cohort comprised four primary colorectal tumors collected at City of Hope: two from EOCRC patients (<50 years) and two from LOCRC patients (≥50 years). Researchers analyzed WES reads after removing host-derived sequences to enrich for microbial signal. Taxonomic profiling was performed at family, genus, and species levels; predicted microbial metabolic pathways were inferred from the recovered microbial sequences.
Microbial taxonomic profiles were generated from non-host reads present in tumor WES data. Analyses were conducted at three nested taxonomic resolutions—family, genus, and species—to assess both broad and fine-scale differences in intratumoral microbiota between age groups. Functional profiling was predicted computationally to infer metabolic pathways represented by the detected microbial community.
Across family, genus, and species levels, LOCRC tumors exhibited markedly higher microbial diversity than EOCRC tumors. The LOCRC group contained more than twice the number of unique bacterial families, nearly three times the number of unique genera, and more than twice the number of unique bacterial species compared with the EOCRC group. These results indicate substantially greater taxonomic richness and complexity in late-onset tumors within this cohort.
Despite pronounced compositional differences, a conserved core microbiota was identified across both EOCRC and LOCRC samples. Families and genera present in the core included Fusobacteriaceae, Prevotellaceae, Fusobacterium, and Prevotella. The presence of these taxa in both age groups suggests some consistent elements of the colorectal tumor microenvironment microbiome in this H/L cohort.
LOCRC tumors were enriched for multiple taxa previously associated with colorectal cancer. Reported examples include Fusobacterium nucleatum, Bacteroides fragilis, Parvimonas micra, Porphyromonas asaccharolytica, and Dialister pneumosintes. The enrichment of these CRC-associated species in LOCRC samples contributed to the higher species-level diversity observed in late-onset tumors.
Species-level analysis revealed substantial divergence between age groups: only a single bacterial species was shared between EOCRC and LOCRC. This finding indicates that as taxonomic resolution increases, microbial communities in tumors from early- and late-onset patients become more distinct. The authors interpret this as evidence of progressive microbial divergence with finer taxonomic resolution.
Functional profiling predicted 11 microbial metabolic pathways across the tumors. Nine pathways were shared between EOCRC and LOCRC, two were unique to EOCRC, and none were uniquely associated with LOCRC. Shared core pathways included those involved in energy metabolism, amino acid biosynthesis, phospholipid metabolism, and central carbon metabolism. The similar abundance of these core functions across age groups supports the concept of functional redundancy within the colorectal tumor microenvironment despite marked taxonomic differences.
The study provides an initial characterization of intratumoral microbiota in H/L patients with EOCRC and LOCRC, suggesting taxonomic differences by age but preserved microbial functions. Authors propose that these observations form a foundation for larger multi-omics investigations to evaluate the potential of microbiome-based biomarkers and therapeutic targets in precision oncology for colorectal cancer.
The report is explicitly exploratory and limited by very small sample size (four tumors). Because the source is a preprint, findings have not been validated through peer review. Specific methodological details (for example, exact bioinformatic pipelines, statistical testing, and absolute abundance measures) were not fully reported in the abstract-level source content provided here. The authors recommend larger, multi-omics studies to confirm taxonomic patterns, validate predicted functional conservation, and evaluate clinical relevance for biomarker or therapeutic development.