Importance: The prevalence of germline pathogenic variants in familial melanoma genes has primarily been studied in individuals with a personal or family history of cancer, an approach that may introduce ascertainment bias. Objective: To estimate the prevalence of pathogenic variants in clinically actionable familial melanoma genes and their associated cancer risks. Design, Settings, and Participants: This was a genome-first analysis of 2 population-scale genomically ascertained cohorts of individuals registered in the UK Biobank (UKBB) and the US Geisinger MyCode (GMC) databases, each linked with institutional (GMC) or national (UKBB) cancer registry data from 1970 through 2024. Data analyses were conducted from January 2025 through January 2026. Exposures: Germline pathogenic variant status in familial melanoma genes. Main Outcomes and Measures: Prevalence of pathogenic variants in 8 major familial melanoma genes: ACD, BAP1, CDKN2A, CDK4, MITF E318K, POT1, TERF2IP, and TERT promoter; and associated cancer risks among individuals in the UK and US who had been genomically ascertained. Cancer odds ratios and time to cancer analyses were adjusted for sex, birth year, body mass index, and smoking status. Results: The analysis included a total of 696 665 individuals who were genomically ascertained, 227 286 (mean SD age, 57.7 19.6 years) from GMC and 469 379 (mean SD age, 70.0 8.0 years) from UKBB. The cohorts were predominantly female (GMC, 60.6%; UKBB, 54.2%), and of European genetic ancestry (GMC, 93.6%; UKBB, 94.2%). The combined prevalence of the pathogenic variants across all genes evaluated-ACD, BAP1, CDKN2A, CDK4, MITF E318K, POT1, TERF2IP, and TERT promoter-ranged from 0.5% (GMC) to 0.9% (UKBB) in both cohorts. Among individuals with multiple cutaneous melanomas or a first cutaneous melanoma diagnosed before age 40 years, pathogenic variant prevalence exceeded the 2.5% threshold commonly used to recommend germline testing for high- and moderate-penetrance cancer susceptibility genes. Case-control analyses replicated established associations for CDKN2A (brain, cutaneous melanoma, head and neck, pancreas), MITF E318K (cutaneous melanoma, kidney), and POT1 (cutaneous melanoma, hematologic, thyroid). Additional associations, either novel or inconsistently reported, were observed for BAP1 (prostate), CDKN2A (biliary tract, breast, nonmelanoma skin, small intestine), MITF E318K (cervix, nasal cavity and middle ear, nonmelanoma skin), and POT1 (myeloma). In both cohorts, individuals with CDKN2A PVs or MITF E318K developed cutaneous melanoma at younger ages than did noncarriers, although age-of-onset patterns for internal cancers were less consistent. Conclusions and Relevance: In this cohort study, the prevalence of pathogenic variants in familial melanoma genes and their associated cancer risks were estimated, findings that may inform and revise germline testing recommendations and cancer risk counseling. These data also suggest that the spectrum of cancer risk for several genes may be broader than previously recognized.
Goldstein et al. (Wed,) studied this question.