BACKGROUND: Oral microbes detected in feces have been associated with colorectal cancer (CRC) in cross-sectional studies. This study investigated the prospective associations between the oral microbiome and incident CRC in the Agricultural Health Study (AHS), National Institutes of Health-AARP (NIH-AARP) Diet and Health Study, and Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial. METHODS: Individuals with oral samples collected before incident CRC diagnoses were identified in the AHS (N = 331), NIH-AARP (N = 249), and PLCO (N = 446) and compared with referent subcohorts (N = 3431). The V4 region of the 16S ribosomal RNA gene was sequenced from oral wash DNA, and the data were processed with QIIME2. Hazard ratios (HRs) and 95% confidence intervals (CIs) for overall CRC and by anatomic subsite (i.e., proximal colon, distal colon, and rectum) were estimated with Cox proportional hazards models with adjustment for potential confounders by cohort and then meta-analyzed. RESULTS: Overall, no associations were found between microbial characteristics and CRC risk. However, associations were observed with alpha and beta diversity indices and individual genera in analyses stratified by anatomic subsite. For instance, the presence of Olsenella was strongly positively associated with distal colon cancer risk (HR, 2.16
95% CI, 1.59-2.95), whereas the presence of Prevotella 2 was positively associated with rectal cancer risk (HR, 1.68
95% CI, 1.14-2.46). CONCLUSIONS: This large study of the prospective association between the oral microbiome and CRC risk showed numerous site-specific associations, including multiple associations with distal colon and rectal cancer risk.