Inherited susceptibility to melanoma

INTRODUCTION — The etiology of all cancers depends upon the interplay between environmental and genetic factors.

For melanoma, the most significant environmental risk factor is solar ultraviolet (UV) radiation exposure [1]. However, this risk is greatly influenced by genetic factors. As an example, skin type, a heritable trait, modifies the risk presented by a given amount of solar exposure. Dark-skinned populations have a much lower incidence of melanoma than fairer skinned populations exposed to equivalent sunlight. In the United States, the risk for melanoma in the White population is approximately 20 times higher than in the Black population [2].

However, there are other inherited factors, aside from those that define pigmentation, that contribute to hereditary risk. A study of familial risk and hereditability identified 23,980 people who had cancer and were twins [3]. The risk of concordance for melanoma was much greater than in the general population, and the increase was more pronounced in monozygotic compared with dizygotic twins. From this data, the heritability (ie, the proportion of variation of risk that can be attributed to genetic variation) of melanoma was estimated to be 58 percent.

Significant progress has been made toward understanding the genes that contribute to inherited susceptibility for melanoma in some patients [4]. Uncommon but high-risk alleles contribute to the hereditary cancer phenotype that includes multiple cases of the associated cancer or cancers on one side of the family, multiple primary cancers in a given individual, and early age of onset for a given cancer. With advances in genomic technologies and the conceptual framework to isolate more prevalent but lower risk alleles, the spectrum of genetic variants that contribute to melanoma risk has expanded.

The hereditary risk factors for melanoma are discussed here, along with potential implications for genetic screening. Other risk factors associated with the development of melanoma are discussed separately. (See "Melanoma: Epidemiology and risk factors".)

HEREDITARY VERSUS SPORADIC MELANOMA — Melanomas can be divided into those that are sporadic and those with a hereditary component. Sporadic melanomas are due to random acquired pathogenic variants in melanocytes. Acquired pathogenic variants may be due to the environment, aging, chance events, other nonheritable factors, or possibly low-penetrance-risk single-nucleotide polymorphisms that specify risk traits (eg, skin color).

An estimated 8 to 12 percent of patients with melanoma have a family history of the disease. However, not all of these individuals have hereditary melanoma [5,6]. In some cases, the apparent familial inheritance pattern may be due to clustering of sporadic cases in families with common heavy sun exposure and a susceptible skin type. In other situations, co-inheritance of modifying genes may enhance or inhibit the apparent penetrance of a high-risk pathogenic variant and thus dictate the strength of the family history.

In areas of moderate to high incidence of melanoma, families with three relatives with melanoma, individuals with three or more primary melanomas, or individuals with melanoma diagnosed at an earlier age (before age 40) are candidates for hereditary melanoma susceptibility testing [7,8]. Twenty to 60 percent of these families harbor a pathogenic variant in CDKN2A/p16 [7,9-11].

Less common melanoma susceptibly genes include CDKN2A/ARF, CDK4, TERT, MITF, BAP1, and POT1. Other genes that may soon be included in hereditary melanoma testing include ACD, MC1R, RB1, and TERF2IP [11,12].

In addition, there are other cancer predisposition syndromes in which melanoma is part of a broader spectrum of tumors but is not the most common malignancy. An example of this is BRCA-associated hereditary breast and ovarian cancer. In this condition, the primary risks are for breast and ovarian cancers, but the risks of some other malignancies are also increased, including melanoma.

These cancer predisposition syndromes may be classified into those in which melanoma is the most frequent manifestation and those in which melanoma is less common than other tumor types [11]. The following section discusses the genes associated with syndromes where melanoma is the dominant feature.

HIGH-RISK SUSCEPTIBILITY LOCI ASSOCIATED WITH HEREDITARY MELANOMA — Melanoma is the most common malignancy associated with the genes discussed in this section. Other types of tumors may also be seen but are less frequent [11].

CDKN2A gene — CDKN2A is the most frequently identified pathogenic variant in familial forms of melanoma. Pathogenic variants in the CDKN2A and CDK4 (refer below) genes can be found in a subset of families with familial atypical multiple mole melanoma syndrome (FAMMM).

Pathogenesis — Initial clues that chromosome 9p21 contained a melanoma susceptibility locus came from cytogenetic investigations and studies of loss of heterozygosity (LOH) [13]. Strong evidence for linkage to this region came from an analysis of 11 melanoma pedigrees in Texas and Utah [14]. Subsequent studies using positional cloning methods isolated a candidate locus that was ultimately designated CDKN2A [15,16]. This gene has also been called MTS1, p16INK4A, and CDKN2. The gene encodes two proteins, p16 and p14ARF, that are transcribed in alternate reading frames through the use of alternative first exons [17]. Germline CDKN2A pathogenic variants in melanoma families are usually missense or nonsense changes that impair the function of p16, although rare pathogenic variants in p14ARF have also been reported [9,10,18].

The p16 protein is a negative regulator of cell cycle progression at the G1/S checkpoint [19,20]. It interacts with the cyclin-dependent kinases (CDKs; CDK4 or CDK6), enzymes that control early events in the cell cycle, to catalyze phosphorylation of the retinoblastoma family of proteins. A complex of cyclin D and CDK together phosphorylate the retinoblastoma gene protein (RB1), thereby releasing the transcription factor E2F1 from RB1 and allowing E2F1 to induce S phase genes. Consequently, the cell proceeds from G1 arrest through S phase. The p16 protein binds to and inhibits CDK4, and thus serves as a brake on cell cycle progression.

Inactivating pathogenic variants of p16 disrupt its inhibitory function on CDK4, thereby permitting inappropriate progression through the cell cycle [21]. Tumorigenesis may result from an impairment in senescence, cell differentiation, or cell death [22,23]. (See 'CDK4 gene' below.)

Prevalence of pathogenic variants — The frequency of germline CDKN2A pathogenic variants in patients with hereditary melanoma is highly variable.

The most extensive analysis of melanoma families found that the major features associated with an increased frequency of germline CDKN2A pathogenic variants were multiple cases of melanoma in a family, early age at diagnosis, and family members with multiple primary melanomas or pancreatic cancer [10]. Using these criteria, 385 families with three or more patients with melanoma from the Melanoma Genetics Consortium (GenoMEL) were compared across various geographical locales. Overall, 39 percent of these selected families had CDKN2A pathogenic variants, ranging from 20 percent (32 of 162) in Australia to 45 percent (29 of 65) in North America to 57 percent (89 of 157) in Europe. The presence of pancreatic cancer in a given family also predicted CDKN2A pathogenic variants, except for in Australian families.

The prevalence of germline CDKN2A pathogenic variants in the general population with melanoma is much lower than in studies of multiple-case families, which may reflect the presence of other co-inherited risk alleles, increased detection bias, or unknown confounders in these cohorts. This was illustrated by a study of 3550 patients with melanoma in which 65 CDKN2A pathogenic variant carriers were identified (1.8 percent) [24]; the rates of CDKN2A pathogenic variants among individuals with a single primary melanoma and multiple primary melanomas were 1.2 and 2.9 percent, respectively [25].

Pathogenic variant penetrance — Estimates of the penetrance of CDKN2A pathogenic variants among carriers vary depending upon the methods of population ascertainment [24,26].

●In the multicenter Genes Environment and Melanoma (GEM) study, histories were obtained from 429 first-degree relatives of index cases, seeking cases of melanoma [24]. In these individuals, the incidence of melanoma in CDKN2A carriers was estimated to be 14, 24, and 28 percent at 50, 70, and 80 years of age, respectively. This compared to an incidence of melanoma of 1.5, 4.5, and 6.2 percent in the relatives of non-CDKN2A carriers and an estimated 0.6, 1.6, and 2.6 percent in the baseline population.

●In the GenoMEL study, which was based upon 80 families with documented CDKN2A pathogenic variants and multiple affected family members, the incidence of melanoma was 30 percent by age 50 years and 67 percent by age 80 years [26]. The higher penetrance observed in this study at least in part reflects the bias induced by only analyzing families with two or more cases. An interaction between genetic predisposition and environmental factors was suggested in this series by geographic variations. In Europe, the United States, and Australia, penetrance at age 80 years was 58, 76, and 91 percent, respectively.

The same factors that influence the incidence of melanoma in the general population (eg, total nevus count, presence of dysplastic nevi, sunburn) also affect CDKN2A penetrance. This was illustrated in a study of 53 melanoma-prone families and 295 families ascertained through probands diagnosed with melanoma but unselected for family history [27]. The presence of both sunburn and a melanoma predisposing gene increased the risk of melanoma 15 times more than in noncarriers. In families with CDKN2A pathogenic variants, total nevus count, dysplastic nevi, and sunburn all significantly increased the risk of developing melanoma. In addition, the presence of variants in the MC1R gene increases the risk of developing melanoma in CDKN2A carriers. (See 'Melanocortin-1 receptor' below and "Melanoma: Epidemiology and risk factors".)

The generalizability of these observations to other geographic areas is uncertain since most of the mutated CDKN2A genes were derived from a single founder pathogenic variant. These observations require confirmation.

CDK4 gene — p16 inhibits both CDK4 and CDK6, thereby regulating cell cycle progression. Several melanoma families have been described who lack pathogenic variants in CDKN2A but have germline pathogenic variants in the CDK4 gene [28-30]. In these families, the alterations were on arginine 24 of CDK4, resulting in a CDK4 protein that was insensitive to inhibition by p16. There are no apparent differences in the phenotype (eg, age at diagnosis, number of melanomas) of families carrying either CDKN2A or CDK4 pathogenic variants [31].

Shelterin complex genes — The shelterin complex protects chromosomal ends by regulating how the telomerase complex interacts with telomeres [32]. This complex includes multiple genes, three of which have been associated with familial melanoma:

●POT1 – POT1 controls telomerase-mediated telomere elongation and is part of the shelterin complex. Four percent of familial melanoma pedigrees negative for CDKN2A and CDK4 pathogenic variants were found to have a variant in POT1. Two of 34 (5.8 percent) families with five or more melanoma cases carried a POT1 variant [33].

●ACD and TERF2IP – Following the reports of POT1 involvement in familial melanoma, five other proteins involved in the shelterin complex (ACD, TERF2IP, TERF1, TERF2, TINF2) were studied [32]. Six melanoma families had pathogenic variants in the ACD gene. There were five distinct pathogenic variants in ACD, and four of those were in the POT1 binding domain. Four families carried TERF2IP variants. Novel alterations were identified in TERF1, TERF2, and TINF2; however, they were not significantly associated with melanoma.

Certain alterations in POT1, ACD, and TERF21P have also been associated with hereditary chronic lymphocytic leukemia and familial glioma [34,35].

TERT — A promoter mutation in TERT (c.-57T>G) has been identified in two unrelated families with multiple cases of melanoma. The TERT promoter mutation was first identified in a German family with 14 cases of melanoma [36]. A United Kingdom family with the same promoter mutation had seven cases of melanoma [37]. Both families had carriers with bladder cancer. Although rare, the TERT promoter mutation appears to be associated with a high risk for early onset melanoma and possibly other cancers. Variants in the promoter region are thought to increase recognition sites for Ets/Tcf transcription factors and enhance TERT expression [36].

BAP1 gene — Germline variants in the BRCA1-associated protein 1 (BAP1) gene are linked to the BAP1-tumor predisposition syndrome (BAP1-TPDS). BAP1 encodes a deubiquitinating enzyme, which functions as a tumor suppressor.

Germline pathogenic variants in the BAP1 gene have been identified in families with cutaneous and uveal melanoma [38,39]. In an Australian cutaneous melanoma registry, BAP1 variants were identified in 7 of 1109 cases (0.6 percent). The prevalence in other populations has yet to be determined [40]. (See "The molecular biology of melanoma", section on 'Mutations with prognostic relevance'.)

In addition to cutaneous and ocular melanoma, mesothelioma (both pleural and peritoneal) and renal cancer have been observed in BAP1 families and are proposed to be part of the core tumor spectrum linked to germline BAP1 pathogenic variants; furthermore, somatic BAP1 pathogenic variants have been described in renal cell carcinoma [38,39,41,42]. (See "Malignant peritoneal mesothelioma: Epidemiology, risk factors, clinical presentation, diagnosis, and staging" and "Epidemiology, pathology, and pathogenesis of renal cell carcinoma", section on 'BAP1 gene' and "Hereditary kidney cancer syndromes", section on 'Hereditary BAP-1-associated renal cell carcinoma'.)

The presence of unique highly atypical nevoid melanoma-like melanocytic proliferations has also been observed in some individuals with BAP1 pathogenic variants [38]. BAP1-absent dermal lesions called melanocytic BAP1-mutated atypical intradermal tumors (MBAITs) can be found adjacent to these lesions [43]. The lesions may be the most common feature in BAP1 pathogenic variant carriers, identified in 75 percent of BAP1 carriers who underwent total body skin examinations in one study [44]. Prior to the description of MBAITs, these lesions were likely diagnosed as atypical Spitz tumors. (See "Spitz nevus and atypical Spitz tumors".)

Other tumor types may be part of the BAP1-TPDS spectrum, including basal cell carcinoma, meningioma, and cholangiocarcinoma [45].

MODERATE RISK POLYMORPHISMS ASSOCIATED WITH SPORADIC MELANOMA

MITF (E318K) variant — A rare germline missense mutation in the microphthalmia-associated transcription factor (MITF) gene, E318K, confers a 2.2- to 4.8-fold increased risk of melanoma. The E318K variant is thought to enable MITF to act as an oncogene by abolishing a sumoylation site, thereby altering the transcriptional function of the MITF protein [46-48].

The prevalence (carrier frequency) of the MITF E318K variant in patients with personal histories of melanoma has been reported to be 0.017 in an Australian/United Kingdom population [46], 0.014 in a French population [47], and 0.009 in an Italian population [49]. The MITF E318K variant was identified in 1.9 percent of all p16INK4A wild-type melanoma patients in a Spanish cohort. The prevalence of MITF increased to 2.6 percent in those with multiple primary melanomas [50]. In an Italian cohort, carriers had a threefold higher risk of melanoma compared with controls and a 6.4-fold higher risk for multiple primary melanomas [49].

Some studies have identified a relationship of this variant with renal cell carcinoma, pancreatic cancer, and pheochromocytoma/paraganglioma [48,50,51]. However, it is not yet clear whether the association between this variant and other malignancies is indeed related to the MITF E318K variant or a result of shared environmental or polygenic risk factors. (See "Epidemiology, pathology, and pathogenesis of renal cell carcinoma", section on 'Translocation renal cell carcinoma (MiT/TFE-related RCC)' and "Pheochromocytoma in genetic disorders" and "Paragangliomas: Epidemiology, clinical presentation, diagnosis, and histology".)

A meta-analysis of data compiled from nine published studies, which were conducted on a mixture of hereditary and sporadic melanoma populations, demonstrated that MITF E318K was significantly correlated with melanoma. Additionally, this study systematically evaluated the prevalence of the MITF E318K variant in multiple cancer cohorts using germline whole-exome sequence data from the TCGA panel and from several genetically-enriched cohorts. Among the 25 cancers tested (including renal cell cancer, pancreatic cancer, and pheochromocytoma/paraganglioma), uterine carcinosarcoma and melanoma demonstrated the strongest associations with the MITF E318K variant [52].

Melanocortin-1 receptor — Variants of the MC1R gene have been associated with the red hair/fair skin phenotype [53], which is more common in subjects with cutaneous melanoma. In addition, analysis of the MC1R gene in patients with melanoma has shown an increased prevalence of MC1R gene variants compared with healthy controls [54-60].

This was illustrated by a study in which the MC1R gene was sequenced in 267 patients with melanoma and 382 control subjects [54]. The presence of MC1R variants was associated with a two- to fourfold increase in the risk of melanoma compared with individuals carrying the wild-type MC1R gene. The association between MC1R variants and melanoma was stronger in individuals with fewer additional risk factors (eg, those with dark skin or few nevi). In addition, patients with a variant MC1R gene were three- to fourfold more likely to have thick melanomas.

There also is evidence that variants of MC1R can interact with other genes involved in the pathogenesis of melanoma. MC1R variants increase the penetrance of CDKN2A [61]. In a Melanoma Genetics Consortium (GenoMEL) study of 815 CDKN2A pathogenic variant carriers from 186 families, the presence of one of the four most common MC1R variants was associated with a statistically significant increased risk of melanoma; the presence of two or more variants further increased that risk compared with having just one variant (odds ratios compared with those without an MC1R variant were 5.8 and 2.3, respectively).

The BRAF oncogene is the most common site for somatic pathogenic variants in melanoma [62]. Such molecular alterations are more common in melanomas arising in skin that has little chronic sun-induced damage, and they are less frequent in melanomas associated with severe, chronic solar damage [63]. In patients with melanoma originating in skin with limited sun-induced damage, somatic BRAF mutations are strongly associated with the presence of inherited MC1R variants and are less common in patients with wild-type MC1R [64,65].

TUMOR SYNDROMES WITH INCREASED MELANOMA RISK — Although melanoma is not the most common malignancy associated with the genes discussed in this section, the risk for melanoma in patients with these genetic alterations remains elevated compared with the general population [11].

BRCA1 and BRCA2 genes — The BRCA1 and BRCA2 genes encode proteins that are important in DNA repair. Germline pathogenic variants are associated primarily with an increased risk of breast, ovarian, and prostate cancers; however, other cancer types have been associated with pathogenic variants. (See "Cancer risks and management of BRCA1/2 carriers without cancer".)

An association of BRCA1 and BRCA2 pathogenic variants with melanoma has been observed in some but not all studies:

●In a report from the Breast Cancer Linkage Consortium (BCLC), BRCA2 carriers had an increased risk of cutaneous melanoma (relative risk 2.6) [66]. BRCA2 pathogenic variants also appear to occur in families with aggregations of ocular melanoma (7 of 62 such families in one series) [67]. Another large single-institution study examined cancer risk in 993 BRCA1 and BRCA2 carriers; within the BRCA1 cohort, there was a trend toward an increased risk for melanoma [68].

●Other studies have not found an association between melanoma and inherited pathogenic variants in BRCA1 and BRCA2. A study of individuals with cutaneous melanoma and Ashkenazi Jewish ancestry detected no germline pathogenic variants in the 92 individuals analyzed for the three BRCA Ashkenazi Jewish founder mutations [69]. With respect to ocular melanoma, no germline pathogenic variants in BRCA1 or BRCA2 were detected in a cohort of 25 individuals with a personal history of ocular melanoma and a personal/family history of breast or ovarian cancers [70].

TP53 — Germline pathogenic variants in TP53 are associated with Li-Fraumeni syndrome. Li-Fraumeni syndrome is characterized by early onset and multiple primary cancers, particularly sarcomas, breast cancers, adrenocortical carcinomas, brain tumors, and leukemias. Increased rates of melanomas have also been reported [71]. (See "Li-Fraumeni syndrome".)

PTEN gene — Cowden syndrome, due to pathogenic variants in the PTEN gene, is classically associated with an increased risk for cancer of the breast, colon, endometrium, and thyroid, as well as benign hamartomas. Individuals with Cowden syndrome often have macrocephaly, autism spectrum disorders, and specific skin lesions, such as trichilemmomas and papillomatous papules. PTEN carriers may also have an increased risk for melanoma [72,73]. (See "PTEN hamartoma tumor syndromes, including Cowden syndrome".)

RB1 gene — The cyclin D/cyclin-dependent kinase (CDK) complex phosphorylates the RB1 protein, allowing cell cycle progression from G1 to S phase; p16 binds to and inhibits the activities of these three enzymes, functioning as a brake on cell cycle progression. (See 'Pathogenesis' above.)

In one large cohort study of retinoblastoma survivors, the mortality from second cancers, including melanoma, was significantly increased [74]. Although germline RB1 pathogenic variants have not been reported in patients with melanoma, alterations or loss of RB1 expression has been described in a limited number of melanoma cell lines [75].

GENETIC TESTING

Who to test — Four clinical scenarios may warrant genetic counseling and/or testing for hereditary melanoma:

●Melanoma diagnosed at earlier than expected ages

●Multiple relatives with melanoma

●Individuals with multiple primary melanomas

●Other types of cancer in the family

The general principles of genetic testing are discussed separately. (See "Genetic testing".)

Early onset melanoma — Earlier than expected age at diagnosis is associated with an increased risk for a hereditary melanoma syndrome. The average age of onset for melanoma in the United States population is 64 years (Surveillance, Epidemiology, and End Results [SEER]), while the average age of onset in families with mutated CDKN2A (the most common pathogenic variant) is 35 years (range 14 to 68 years) [7,26]. However, early onset alone is not necessarily predictive of germline predisposing pathogenic variants, and only a small percentage of patients with early onset melanoma will actually have a CDKN2A pathogenic variant. (See 'CDKN2A gene' above.)

It is also important to note that unlike other hereditary cancer predisposition conditions, standalone early onset melanoma may not be associated with genetic susceptibility. In one cohort of young patients with a personal history of melanoma (median age 32 years) and no family history of melanoma, there was only one patient with a germline CDKN2A pathogenic variant [76]. In another study, a deleterious CDKN2A pathogenic variant was identified in only 1 of 60 patients diagnosed with melanoma prior to age 20 years, and that patient was part of a kindred with a known CDKN2A pathogenic variant [77].

Significant family history of melanoma — Multiple cases of melanoma in a family may suggest that a germline pathogenic variant in the family is playing a role in the cancer development. In the international Melanoma Genetics Consortium (GenoMEL) study, the likelihood of finding a pathogenic variant in the CDKN2A gene increased progressively with the number of involved family members and was approximately 70 percent when six or more members of the kindred had cutaneous melanoma [10].

The criteria that constitute a "significant" family history of melanoma depend in part on the geographic location of the families. Sun exposure is a proven modifier of melanoma risk. Therefore, in areas where there is a high incidence of melanoma, at least three family members would need to be affected to suggest hereditary susceptibility. However, in areas of low melanoma incidence, two affected family members may be sufficient to warrant a referral for genetic counseling and/or testing [11].

Relatedness of the affected family members is also critical in estimating the probably of genetic alterations. For instance, a mother, daughter, and son trio of affected patients is more likely due to shared genetic content than a mother, niece, and distant cousin.

MelaPRO is software program that utilizes best-estimate penetrance along with Mendelian laws of inheritance to arrive at a CDKN2A carrier probability, with separate modules for different geographic areas [78]. MelaPRO is included in CancerGene, a free downloadable software program.

Multiple primary melanomas — Another feature of inherited cancer predisposition is the development of multiple primary cancers, and this appears to be true in hereditary melanoma.

Data on the incidence of germline CDKN2A pathogenic variants in individual patients with multiple primary melanomas are variable. In the Genes Environment and Melanoma (GEM) cohort, 2.9 percent of 1189 patients with multiple primary melanomas had germline CDKN2A pathogenic variants [24]. In several smaller studies, the incidence of pathogenic variants in patients with multiple melanomas ranged between 8 and 16 percent [79-82].

In families with multiple members with melanoma, the likelihood of identifying a CDKN2A pathogenic variant is much higher with increasing numbers of individuals with multiple primary melanomas. In an international study of 385 melanoma families, 70 percent of families had a CDKN2A pathogenic variant when there were three or more relatives with multiple primary melanomas. In contrast, the detection rate was <20 percent in families without individuals with multiple primary melanomas [10].

Familial melanoma and other malignancies — The occurrence of other associated tumors (eg, pancreatic cancer, breast cancer, brain tumors), either in the patient's personal history or family history, should be taken into account when considering possible hereditary predisposition syndromes.

Melanoma and pancreatic cancer have both been associated with pathogenic variants in the CDKN2A gene [29,83-85]. The relationship between CDKN2A pathogenic variants and other cancer types is illustrated by a study based upon 330 high-risk melanoma-prone families that included 236 patients with sporadic multiple primary melanoma and 466 cases with familial melanoma [86]. Pathogenic variants in CDKN2A were identified in 66 families. There was a statistically significant increased incidence of other solid tumors compared with those families with wild-type CDKN2A (prevalence ratio 3.0). Specific tumor types for which there was an increased incidence included pancreatic cancer, lung cancer, and breast cancer (prevalence ratios 3.0, 3.0, and 2.0, respectively). Similar associations have been observed in families with alterations in CDK4 [7,31].

Specific pathogenic variants in the CDKN2A gene may have other associations:

●Melanoma astrocytoma syndrome refers to an increase in brain tumors in some melanoma-prone kindreds [87] with germline alterations of the p14ARF component of CDKN2A [88,89]. One cohort study found an increase in a wide range of malignant and benign neural tumors in patients with alterations in p14ARF including glioblastoma, gliosarcoma, astrocytoma, schwannomas, and malignant peripheral nerve sheath tumors [90]. In addition, many of these patients had a clinical diagnosis of neurofibromatosis, but had negative genetic testing for NF-1 gene mutations. The contribution of these changes to the full spectrum of hereditary melanoma remains to be clarified.

●A CDKN2A missense variant (rs3731249) confers a threefold increased risk for acute lymphoblastic leukemia in children of European (odds ratio 3.0) and Hispanic ancestry (odds ratio 2.8). The variant was not significantly associated with other cancers, suggesting that this variant is specific to acute lymphoblastic leukemia risk [91].

Pathogenic variants that have been associated with various other malignancies include:

●BAP1 – In addition to cutaneous and ocular melanoma, other cancers, including non-asbestos-related mesothelioma and renal cancer, have been reported in families with germline BAP1 pathogenic variants [38,39,41]. The presence of unique nevoid melanomas and highly atypical nevoid melanoma-like melanocytic proliferations has also been observed in some of these individuals [38]. (See 'BAP1 gene' above.)

●MITF – The MITF E318K pathogenic variant is reported to be associated with melanoma, renal cell cancer, breast cancer, pancreatic cancer, and lymphoma [11,47,49]. (See 'MITF (E318K) variant' above.)

●BRCA1 and BRCA2 – Molecular alterations in the BRCA1 and BRCA2 genes are most commonly associated with breast and ovarian cancers but can also be associated with melanoma and other malignancies. (See 'BRCA1 and BRCA2 genes' above and "Cancer risks and management of BRCA1/2 carriers without cancer".)

●PTEN – The PTEN gene is commonly associated with breast, uterine, and non-papillary thyroid cancer, along with other features such as macrocephaly. If these features are present in a family with melanoma, testing for pathogenic variants in the PTEN gene should be considered. (See 'PTEN gene' above and "PTEN hamartoma tumor syndromes, including Cowden syndrome".)

●TP53 – Families with a history of melanoma and breast cancer along with sarcoma, brain cancer, and leukemia should be tested for pathogenic variants in TP53. (See 'TP53' above and "Li-Fraumeni syndrome".)

Single-gene versus multigene panel — The clinical availability of next-generation sequencing technology has led to the development of cancer gene panel testing, which simultaneously analyzes multiple cancer susceptibility genes. Commercially available multigene panel tests may be targeted to a specific type of cancer (eg, breast, colon) or may broadly test cancer genes (as so called "pan-cancer" panels). Melanoma-specific panels, offered by some clinical testing laboratories, typically include an analysis of BAP1, BRCA2, CDK4, CDKN2A, MITF, POT1, PTEN, RB1, and TP53. When the personal and/or family history indicates testing for more than one gene, multigene panel testing may be a more efficient method of testing. For example, a family history of melanoma, pancreatic cancer, and breast cancer may warrant both CDKN2A and BRCA2 testing. Such approaches may also facilitate the identification of actionable pathogenic variants [92].

Given the broad testing available, multigene panel testing may incidentally detect a melanoma susceptibility gene pathogenic variant in an individual or family not phenotypically suggestive of hereditary melanoma risk. Such testing should only be performed once patients are counseled appropriately and made aware of the possibility of incidental findings, the implications for relatives, the psychological consequences, and the potential changes to medical management. (See "Genetic testing".)

Although single-gene CDKN2A testing is used less commonly in the age of multigene panels, the lessons learned from CDKN2A testing may be relevant for other melanoma susceptibility genes. The interpretation of such test results is difficult, especially in the context of a negative result. Importantly, unaffected individuals from hereditary melanoma families who test negative for the familial CDKN2A pathogenic variant are still at increased risk of developing melanoma despite their negative genetic status [93]. This leads to the concern that a negative test result will result in decreased surveillance and vigilance.

Genetic testing for CDKN2A pathogenic variants may increase the motivation for risk-reducing behaviors and frequent surveillance, as well as lower the biopsy threshold for suspicious lesions, which might lead to earlier diagnosis and improved survival [94,95]. On the other hand, genetic testing could potentially cause psychological distress, lead to unnecessary biopsies in carriers, and reduce the motivation for preventive behaviors in those without CDKN2A pathogenic variants [96]. (See "Genetic testing".)

Genetic counseling, with or without genetic testing, may promote awareness in individuals at increased risk for melanoma and increase compliance with photoprotection recommendations in both unaffected carriers and unaffected noncarriers [97]. A study involving two CDKN2A families concluded that educating patients about genetic testing as well as photoprotection increased photoprotective behaviors, regardless of pathogenic variant status [98]. In addition, knowledge of familial melanoma risks in families who decline genetic testing has been suggested to encourage melanoma risk-reducing and early detection behaviors [99]. In another study, genetic test reporting contributed to the understanding of hereditary risk for melanoma regardless of carrier status [100]. (See "Primary prevention of melanoma".)

SURVEILLANCE AND MANAGEMENT — Many of the melanoma-associated genes on multigene panels do not have a precisely defined risk for melanoma. The personal and family history should be used to make surveillance and management decisions. Referral to a high-risk specialist may be valuable in making decisions, particularly since there may be guidelines for management of other cancer risks (eg, breast, colon, or ovarian).

Surveillance in melanoma-prone kindreds — For patients within a melanoma-prone kindred, including those previously treated for melanoma, skin checks by a health care provider trained to detect melanoma are essential and should be performed at least annually. If there are a large number of clinically atypical moles, skin exams should be performed at least twice a year. Skin self-exams should be performed monthly, and individuals should be trained on the proper technique. Sun protective practices should also be engaged, including sun avoidance during midday and the use of sunscreen.

The approaches to screening and early detection of melanoma as well as preventative measures are discussed in detail separately. (See "Screening for melanoma in adults and adolescents" and "Primary prevention of melanoma".)

In families with a known or suspected CDKN2A pathogenic variant associated with a family history of pancreatic cancer, pathogenic variant carriers should undergo surveillance for melanoma because of the risk of developing melanoma. (See "Screening for melanoma in adults and adolescents" and "Primary prevention of melanoma".)

Surveillance for other cancers — In families with a CDKN2A pathogenic variant or another molecular alterations known to be associated with hereditary melanoma, or suspected or known to be associated with a family history of pancreatic cancer, CDKN2A pathogenic variant carriers should also be referred to a gastroenterology specialist to discuss the option of pancreatic cancer surveillance. Although the efficacy of pancreatic cancer surveillance remains under investigation, the International Cancer of the Pancreas Screening (CAPS) Consortium summit on the management of patients with an increased risk for familial pancreatic cancer recommends consideration of pancreatic cancer surveillance for individuals with CDKN2A pathogenic variants and a first-degree relative with pancreatic cancer [101]. There are no chemoprevention regimens or other prophylactic interventions that are known to benefit CDKN2A pathogenic variant carriers.

Patients with germline BAP1 pathogenic variants should receive routine skin and ophthalmological examinations. Given the increased risk of mesothelioma and kidney cancer, a multidisciplinary team is often necessary to screen for BAP1-related visceral malignancies, although there are no formal evidence-based guidelines on how that should be done.

The prevalence of germline pathogenic variants in CDK4, the TERT promoter, and the shelterin complex is sufficiently low that there are no formal guidelines for management. Surveillance schedules should be similar to those for CDKN2A pathogenic variant carriers.

Familial clustering may be due to an inherited genetic risk or common environmental exposure with sporadic clustering. In either case, all family members are at risk and should obtain dermatologic screening and follow-up.

Patients with melanoma — For patients with diagnosed melanoma as well as a positive family history of melanoma, surveillance and follow-up should be guided by the patient's melanoma staging. (See "Surgical management of primary cutaneous melanoma or melanoma at other unusual sites".)

SUMMARY AND RECOMMENDATIONS

●Risk factors – The chief environmental risk factor associated with cutaneous melanoma is exposure to ultraviolet (UV) radiation. This risk is modified by high-risk genetic factors, the most frequent of which are pathogenic variants in the CDKN2A tumor suppressor gene. Other genetic factors associated with an increased risk of melanoma have also been identified. (See "Melanoma: Epidemiology and risk factors".)

●CDKN2A – Families with known germline CDKN2A pathogenic variants are characterized by having multiple family members with melanoma, early age of onset, individuals with multiple primary melanomas, and coexistence with other primary tumors, especially pancreatic cancer. However, even in the presence of these criteria, germline pathogenic variants in CDKN2A are uncommon. (See 'CDKN2A gene' above.)

●Other pathogenic variants – Pathogenic variants in several other genes have also been associated with a hereditary predisposition to melanoma (CDK4, BAP1, MITF, etc) and should be considered when evaluating patients. (See 'CDK4 gene' above and 'BAP1 gene' above and 'MITF (E318K) variant' above.)

●Genetic counseling – For patients thought to be at increased risk of having an inherited susceptibility to melanoma, we suggest genetic counseling with a qualified health care provider to receive education regarding the risks and benefits of genetic testing and to discuss disease expression and the associated changes to medical management (Grade 2C). (See 'Single-gene versus multigene panel' above.)

●Surveillance and management – For individuals with a pathogenic variant in a melanoma susceptibility gene, we recommend close clinical surveillance and education regarding melanoma risk-reducing behaviors (eg, sunscreen use, sun avoidance) (Grade 1C). (See 'Surveillance and management' above.)

For individuals with CDKN2A or CDK4 pathogenic variants, we also recommend consideration of referral to a gastroenterology specialist to discuss the option of pancreatic cancer surveillance for carriers who have a family history of pancreatic cancer in a relative known or suspected to carry the familial pathogenic variant. For patients with a BAP1 pathogenic variant, referral to an ophthalmologist is reasonable. (See 'Surveillance for other cancers' above.)

●Patient education on melanoma primary prevention and screening – For individuals thought to be at increased risk of having an inherited susceptibility to melanoma in whom no germline pathogenic variant is detected, we recommend education regarding melanoma risk-reducing behaviors and surveillance recommendations, which should be based upon the personal and family history (Grade 2C). (See "Primary prevention of melanoma" and "Screening for melanoma in adults and adolescents", section on 'High-risk patients'.)

ACKNOWLEDGMENTS — The UpToDate editorial staff acknowledges Michele Gabree, MS, CGC; Linda H Rodgers, MGC, CGC; and Devanshi Patel, MS, CGC, who contributed to an earlier version of this topic review.