Abstract
Many studies have reported an association between oral microbiota and oral cancer (OC). Yet, the causal relationship between oral microbiota and OC remains undetermined. This Mendelian randomization (MR) study utilized the most recent genome-wide association study of oral microbiota from the ADDITION-PRO study (n = 610, European ancestry), along with the summary statistics for OC from the UK Biobank consortium (643 cases and 406,821 controls, European ancestry). Instrumental variables were selected at P < 1 × 10⁻⁵, clumped for linkage disequilibrium (R² < 0.001, window size = 10,000 kb), filtered for minor allele frequency > 0.01, and F-statistic > 10 to ensure instrument strength. Inverse variance weighted, weighted model, MR-Egger, maximum likelihood, MR-PRESSO, and weighted median were used to assess the causal association between oral microbiota and OC. Reverse MR analysis was conducted on microbes causally linked to OC in the forward MR analysis. In inverse variance weighted estimates, Prevotella had a protective effect on OC (OR = 0.89, 95% CI: 0.81-0.99, P = 0.03), while Veillonella exhibited a deleterious effect on OC (OR = 1.14, 95% CI: 1.01-1.31, P = 0.04). Reverse MR analysis indicated no significant causal effect of OC on oral microbiota. In addition, instrumental variables did not show significant heterogeneity or horizontal pleiotropy. Our findings suggest that Prevotella and Veillonella were causally linked to OC. Further randomized controlled trials are necessary to elucidate the impact of oral microbiota on OC and their underlying mechanism.