Abstract
BACKGROUND: There is evidence from observational studies that skin microbiota is linked to skin cancers. Nevertheless, the causal association between skin microbiota and skin cancers is yet to be fully clarified. METHODS: A bidirectional two-sample Mendelian randomization (MR) was performed to determine the causal relationship between skin microbiota and skin cancers. A total of 294 skin microbial taxa were identified from the first genome-wide association study across three skin microenvironments of two German population cohorts. Summary data of three skin cancers (malignant melanoma, squamous cell carcinoma, and basal cell carcinoma) were obtained from the FinnGen consortium. Moreover, sensitivity analysis examined horizontal pleiotropy and heterogeneity, and microenvironment-based meta-analysis confirmed the reliability of the results. RESULTS: We identified 65 nominal causalities and 5 strong causal associations between skin microbiota and skin cancers. Among them, the class Bacilli revealed a bidirectional positive relationship with malignant melanoma. The class Betaproteobacteria and class Gammaproteobacteria demonstrated a causal association with an elevated risk of malignant melanoma and basal cell carcinoma, respectively. In the reverse MR analysis, malignant melanoma was associated with a lower abundance of phylum Bacteroidetes. There were no indications of significant heterogeneity in instrumental variables or evidence of horizontal pleiotropy. CONCLUSION: Our MR analysis indicated bidirectional causal associations between skin microbiota and skin cancers, and had the potential to offer novel perspectives on the mechanistic of microbiota-facilitated carcinogenesis.