Abstract
Skin cancer is one of the most common cancers worldwide. Some risk factors including sun exposure and MC1R variants are recognized; however, the identification of additional genetic factors is essential for the development of novel therapeutic strategies. Here, we conducted a proteome-wide Mendelian randomization (MR) using plasma protein quantitative trait loci (pQTLs) from a published study and the UK Biobank genome-wide association study (GWAS) of skin cancers. We replicated the published result of ASIP, which was significantly associated with increased risks of basal cell carcinoma (BCC) and malignant melanoma. Moreover, we newly identified CTSS, which was significantly associated with a decreased risk of BCC. A series of replication analyses using the DeCODE pQTLs and the FinnGen GWAS, and sensitivity analyses including Steiger filtering, reverse MR, and Bayesian colocalization, supported our primary results. Our findings highlighted the possibility of prioritizing proteins for novel therapeutic or preventive targets and biomarkers for skin cancers.