Diet–Microbiome Interactions in Colorectal Cancer: Associations Between Meat Consumption and the Intratumoral Microbiome

Abstract

Background : The gut microbiome has emerged as a critical mediator of colorectal cancer (CRC) risk, and dietary meat consumption is among the most consistently implicated environmental risk factors for CRC development. However, the relationship between pre-diagnostic meat consumption and the intratumoral microbiome remains largely unexplored. This study examined associations between pre-diagnostic meat consumption and the intratumoral microbiome in CRC patients, integrating predicted functional metabolic pathway analysis and genus-level bacterial presence data.Methods: Participants were drawn from the Puget Sound Colorectal Cancer Cohort (N= 655). Pre-diagnostic red meat, poultry, and total meat consumption were assessed via self-report and dichotomized as high or low based on established thresholds. Intratumoral microbiome data were characterized using 16S rRNA gene sequencing, with functional pathway abundances predicted via PICRUSt2 and CLR-transformed prior to analysis. Nine MetaCyc pathways were examined in relation to meat consumption using linear regression models, selected a priori based on their connections to dietary metabolic pathways involving meat. Associations between meat consumption and the presence of eleven bacterial genera involved in these pathways were examined using Firth’s penalized logistic regression. All models were adjusted for age, sex, BMI, tumor site, and smoking history. Effect modification by tumor site and age at diagnosis were also evaluated. Given the exploratory, hypothesis-generating nature of this analysis, corrections for multiple comparisons were not applied. Results: Among the nine pathways examined, higher total meat consumption was significantly associated with lower abundance of lysine fermentation to acetate and butanoate (β = -0.5997 p = 0.010). At the genus level, high poultry meat consumption was associated with significantly lower odds of Peptoclostridium presence in tumor tissue (OR = 0.57, 95% CI: 0.35–0.93, p = 0.024), with this association markedly stronger among those with colon tumors (OR = 0.46, p = 0.007) than with rectal tumors (OR = 1.08, p = 0.873). Conclusions: Pre-diagnostic meat consumption was associated with alterations in the intratumoral microbiome and predicted metabolic pathway abundances in CRC patients, with several associations demonstrating specificity to colon tumors. The inverse association between total meat consumption and lysine fermentation to acetate and butanoate suggests that dietary patterns may deplete key anaerobic fermentative functions in the tumor microenvironment. The inverse association between high poultry consumption and intratumoral Peptoclostridium presence, while unexpected, may reflect broader dietary patterns that shape microbial abundance of the tumor microenvironment in ways not fully captured in this study. These findings are hypothesis-generating and highlight the potential importance of diet-microbiome interactions in CRC biology, warranting validation with further studies on direct metagenomic data.