Gene–Environment Analyses in a UK Biobank Skin Cancer Cohort Identifies Important SNPs in DNA Repair Genes That May Help Prognosticate Disease Risk

Abstract

Abstract Background: Despite well-established relationships between sun exposure and skin cancer pathogenesis/progression, specific gene–environment interactions in at-risk individuals remain poorly-understood. Methods: We leveraged a UK Biobank cohort of basal cell carcinoma (BCC, n = 17,221), cutaneous squamous cell carcinoma (cSCC, n = 2,331), melanoma in situ (M-is, n = 1,158), invasive melanoma (M-inv, n = 3,798), and healthy controls (n = 448,164) to quantify the synergistic involvement of genetic and environmental factors influencing disease risk. We surveyed 8,798 SNPs from 190 DNA repair genes, and 11 demographic/behavioral risk factors. Results: Clinical analysis identified darker skin (RR = 0.01–0.65) and hair (RR = 0.27–0.63) colors as protective factors. Eleven SNPs were significantly associated with BCC, three of which were also associated with M-inv. Gene–environment analysis yielded 201 SNP–environment interactions across 90 genes (FDR-adjusted q < 0.05). SNPs from the FANCA gene showed interactions with at least one clinical factor in all cancer groups, of which three (rs9926296, rs3743860, rs2376883) showed interaction with nearly every factor in BCC and M-inv. Conclusions: We identified novel risk factors for keratinocyte carcinomas and melanoma, highlighted the prognostic value of several FANCA alleles among individuals with a history of sunlamp use and childhood sunburns, and demonstrated the importance of combining genetic and clinical data in disease risk stratification. Impact: This study revealed genome-wide associations with important implications for understanding skin cancer risk in the context of the rapidly-evolving field of precision medicine. Major individual factors (including sex, hair and skin color, and sun protection use) were significant mediators for all skin cancers, interacting with >200 SNPs across four skin cancer types.