BackgroundThe association between recurrent aphthous stomatitis (RAS) and neoplasms of the mouth and pharynx (NOMAP) has been reported in some previous observational studies. However, causality is still confused. Our research aims to explore the relationship between RAS and NOMAP through a Mendelian randomization (MR) analysis and to explore whether RAS can serve as a risk factor for NOMAP to provide a reference for the clinical strategy.MethodsAn exposure dataset for RAS were collected from a published study based on the UK Biobank (UKB). Outcome datasets included Genome-wide association studies (GWAS) summary statistics of NOMAP from the FinnGen datasets. The core method was inverse variance weighting (IVW). The Bonferroni correction, MR-Egger, weighted median, weighted mode, Cochcan’s Q test, MR-PRESSO, and leave-one-out methods served as complementary methods.ResultsWe found no significant evidence of causal relationships between RAS and NOMAP. After applying the Bonferroni correction, the corrected P was equal to 0.00625 (0.05/1/8). The IVW method provided the sole evidence for RAS on Benign neoplasm of floor of mouth (BNFM) (OR = 2.509, 95% CI: 1.296–4.857, P = 0.006), but the subsequent MR-Egger regression method showed that this result may be due to horizontal pleiotropy (P = 0.035). The Cochran Q-test, MR-Egger regression, and MR-PRESSO did not reveal any heterogeneity or directional pleiotropy for the other outcomes.ConclusionsIn conclusion, this is the first MR analysis to investigate the relationship between RAS and NOMAP. Our research confirmed at the genetic level that no causal association has been identified between RAS and NOMAP, therefore facilitating a logical therapeutic perspective and the development of clinical therapies for them.