Germline splicing mutations of CDKN2A predispose to melanoma.

Coding mutations of the CDKN2A gene on chromosome 9p21 cosegregate with 25-60% of familial melanoma cases, but there remains a number of 9p21-linked kindreds that lack germline coding mutations of CDKN2A. We sequenced CDKN2A exons 1alpha, 2, 3, and the adjacent intronic regions in 167 melanoma-prone families (at least two affected first-degree relatives), and detected four splice site variations, three of which cosegregate with the disease. RT-PCR experiments verified that these three variants, including an AGgt to ATgt mutation that demonstrates a founder effect, do affect splicing. While an exon 1alpha splice donor site mutation incompletely abolishes splicing, the correctly spliced mRNA yields a protein (Q50P) that cannot effectively interact with CDK4. We also performed RT-PCR on mRNA from 16 melanoma-prone kindreds to search for cryptic splice sites deep within introns, but identified no splice variants. Meanwhile, we screened 139 affected families using allele-specific PCR for the recently discovered IVS2-105A>G mutation, but found only one family that possesses this alteration. We conclude that splice site mutations do predispose to disease in a subset of melanoma-prone kindreds. Characterization of additional splice site variants and other noncoding alterations of CDKN2A should allow us to detect a wider range of mutations in at-risk patients.

Melanoma in adolescents: a case-control study of risk factors in Queensland, Australia.

The incidence of melanoma increases markedly in the second decade of life but almost nothing is known of the causes of melanoma in this age group. We report on the first population-based case-control study of risk factors for melanoma in adolescents (15-19 years). Data were collected through personal interviews with cases, controls and parents. A single examiner conducted full-body nevus counts and blood samples were collected from cases for analysis of the CDKN2A melanoma predisposition gene. A total of 201 (80%) of the 250 adolescents with melanoma diagnosed between 1987 and 1994 and registered with the Queensland Cancer Registry and 205 (79%) of 258 age-, gender- and location-matched controls who were contacted agreed to participate. The strongest risk factor associated with melanoma in adolescents in a multivariate model was the presence of more than 100 nevi 2 mm or more in diameter (odds ratio [OR] = 46.5, 95% confidence interval [CI] = 11.4-190.8). Other risk factors were red hair (OR = 5.4, 95%CI = 1.0-28.4); blue eyes (OR = 4.5, 95%CI = 1.5-13.6); inability to tan after prolonged sun exposure (OR = 4.7, 95%CI = 0.9-24.6); heavy facial freckling (OR = 3.2, 95% CI = 0.9-12.3); and family history of melanoma (OR = 4.0, 95%CI = 0.8-18.9). Only 2 of 147 cases tested had germline variants or mutations in CDKN2A. There was no association with sunscreen use overall, however, never/rare use of sunscreen at home under the age of 5 years was associated with increased risk (OR = 2.2, 95%CI = 0.7-7.1). There was no difference between cases and controls in cumulative sun exposure in this high-exposure environment. Factors indicating genetic susceptibility to melanoma, in particular, the propensity to develop nevi and freckles, red hair, blue eyes, inability to tan and a family history of the disease are the primary determinants of melanoma among adolescents in this high solar radiation environment. Lack of association with reported sun exposure is consistent with the high genetic susceptibility in this group.

Phenotypic variation in eight extended CDKN2A germline mutation familial atypical multiple mole melanoma-pancreatic carcinoma-prone families: the familial atypical mole melanoma-pancreatic carcinoma syndrome.

BACKGROUND: Hereditary pancreatic carcinoma shows extant phenotypic and genotypic heterogeneity as evidenced by its integral association with a variety of hereditary cancer syndromes inclusive of the familial atypical multiple mole melanoma (FAMMM) syndrome in concert with CDKN2A (p16) germline mutations. METHODS: Creighton University's familial pancreatic carcinoma resource comprises 159 families of which 19 (12%) show the FAMMM cutaneous phenotypes. The authors describe eight families with the FAMMM-pancreatic carcinoma (FAMMM-PC) association in concert with a CDKN2A germline mutation. Each family was thoroughly educated about all facets of the study, including the molecular genetics, reduced penetrance of CDKN2A mutations, and their variable expressivity. Genetic counseling was provided to each patient. RESULTS: Diversity in cancer presentation within and among the families was noteworthy, wherein melanoma predominated in certain of the families whereas pancreatic carcinoma predominated in others. Early-onset pancreatic carcinoma (at ages 35, 45, 46, and 49 years) appeared in some of the families whereas markedly later-onset pancreatic carcinoma occurred in others. There were four incidences of melanoma and pancreatic carcinoma as double primaries in the same individuals. One patient with melanoma and pancreatic carcinoma had a third primary of breast carcinoma. Another patient had sarcoma, esophageal carcinoma, and two melanoma primaries, whereas his daughter had sarcoma and was a carrier of a CDKN2A mutation. CONCLUSIONS: The authors suggest that these tumors may collectively, in concert with CDKN2A mutations, constitute a "new" putative hereditary carcinoma syndrome referred to as FAMMM-PC. More clinical and molecular genetic research on additional families with pancreatic carcinoma in concert with the FAMMM will be required.