Rare melanoma genes common enough to test in high-risk patients, study suggests

Genes linked to familial melanoma are rare in the general population, turning up in less than 1% of people. But among patients who develop multiple melanomas or are diagnosed young, they show up often enough to clear the bar that typically triggers genetic testing, according to a genome-first analysis published online in JAMA Dermatology. The findings may help refine germline testing recommendations and cancer risk counseling as population-level genomic screening continues to expand.

Researchers led by Alisa M. Goldstein, Ph.D., and Michael R. Sargen, M.D., of the National Cancer Institute analyzed 696,665 genomically ascertained individuals across two population-level cohorts: the U.S. Geisinger MyCode (GMC) database (227,286 individuals) and the UK Biobank (UKBB; 469,379 individuals). Both linked germline exome data to cancer registry records spanning 1970 through 2024.

Unlike most prior melanoma genetics research, which selected participants based on personal or family cancer history, this study identified individuals by genetic findings first. The authors noted that the genome-first design reduces the ascertainment bias that can occur when families share environmental and lifestyle exposures alongside genetic risk.

The investigators evaluated ACD, BAP1, CDKN2A, CDK4, MITF E318K, POT1, TERF2IP, and TERT promoter, all of which are well-established familial melanoma genes. Combined pathogenic variant prevalence ranged from 1 in 212 (0.5%) in the GMC database to 1 in 113 (0.9%) in the UKBB. Most of that prevalence was attributable to the MITF E318K variant. CDKN2A, POT1, and BAP1 followed, while variants in ACD, CDK4, and TERF2IP were rare. No TERT promoter variants were detected in the UKBB.

Among clinically high-risk subgroups, prevalence exceeded the 2.5% pretest probability threshold established by the PREMMplus model. Patients with multiple cutaneous melanomas had a variant prevalence of 5.3% in the UKBB and 7% in GMC, while those diagnosed with a first melanoma before age 40 showed rates of 3.2% and 3.7%, respectively. The authors wrote that these findings support consideration of genetic testing in these groups regardless of family history.

Case-control analyses replicated established gene-cancer associations, including CDKN2A with brain, head and neck, and pancreatic cancers; MITF E318K with kidney cancer; and POT1 with thyroid and hematologic malignant neoplasms. The study also identified associations that were novel or inconsistently reported in prior research, including BAP1 with prostate cancer; CDKN2A with biliary tract, breast, nonmelanoma skin, and small intestine cancers; MITF E318K with cervix and nasal cavity cancers; and POT1 with myeloma.

In both cohorts, carriers of CDKN2A and MITF E318K variants were diagnosed with melanoma considerably earlier than people without the variants, with diagnosis rates climbing once carriers reached their 30s. The authors said this pattern argues for starting routine skin surveillance at roughly age 30 for this group.

The investigators cautioned that several associations rested on small case counts and that both cohorts consisted predominantly of individuals of European genetic ancestry, potentially limiting generalizability. They also noted that cancer diagnoses in the GMC database may have been underascertained when care occurred outside the Geisinger system.

“This cohort study and genome-first analysis characterizes the prevalence and cancer risks of PVs in major familial melanoma genes,” the authors concluded. “These findings may help inform genetic testing recommendations and suggest that several familial melanoma genes may increase risk for a broader spectrum of cancers than previously recognized.”