A genome-wide meta-analysis of actinic keratosis identifies 12 novel loci and overlap with keratinocyte carcinoma and pigmentation

Actinic keratosis (AK) is a premalignant skin condition caused by chronic ultraviolet (UV) radiation exposure. Previous studies have identified several AK-associated loci, but large-scale genome-wide association (GWA) studies are limited. Here, we show a large GWAS meta-analysis across five cohorts (FinnGen, UK Biobank, Kaiser Permanente Genetic Epidemiology Research on Adult Health and Aging [GERA] cohort, Rotterdam Study, and Million Veteran Program [MVP]), totaling 166,553 AK cases and 1,179,021 controls of European ancestry. Replication is performed in the All of Us (AoU) cohort. We identify 89 genome-wide significant loci, uncovering 12 novel loci involved in epigenetic regulation (TET2), immune signaling (LINC00271, APOLD1, ORMDL3), or other pathways (MIR205HG, ANKRD50). We confirm the role of known pigmentation and keratinocyte carcinoma (KC) genes (IRF4, MC1R, TYR) and find additional independent signals at those loci through conditional analyses. Multivariate conditional analysis adjusting for skin color and KC (both BCC and cSCC) genome-wide summary statistics using the QSkin Study shows that AK genetic susceptibility is largely explained by the genetic architecture of these two traits. This shared architecture is supported by in silico functional analyses. Our findings broaden our understanding of AK pathogenesis and susceptibility to improve risk prediction for AK susceptibility and progression to KC. A genome-wide meta-analysis of over 166,000 actinic keratosis cases identifies 89 loci, including 12 novel, and shows that genetic susceptibility to this premalignant skin lesion is largely shared with pigmentation traits and keratinocyte carcinoma.