Melanoma sometimes occurs in families, indicative of genetic susceptibility. Around 1% of melanoma patients report a family history in the UK[1], 7% of Canadian patients[2] and 11% in an Australian study[3]. The reason for the observation of different rates of family clustering in countries at different latitudes is uncertain. One explanation however, may be that it indicates an effect both of shared genes and environment (sun exposure).

Genetic susceptibility to anything may be variably potent. For example, there are predisposing genes with near 100% penetrance such as Huntington’s Chorea (everyone with the predisposition will manifest the syndrome if they live long enough), while there are many other genes, which have a low penetrance. The majority of people who carry these low penetrance genes will not develop the disease. Further, whether they do develop the disease or not may depend on their lifestyle or other genes. For melanoma it is easiest to group susceptibility genes into rare high penetrance genes and more common lower penetrance genes but the truth is that this distinction is not absolute. There is likely to be a continuous range of genes with variable penetrance all of which probably interact with other genes and with the environment.

Rare high penetrance genes for melanoma

By definition these genes have a high penetrance and therefore there are usually multiple cases of cancer in the family.

In some families there appears to be a pre-disposition to a variety of different cancers. For example melanoma may occur as a second tumour in familial retinoblastoma[4, 5]. There is also a possible association with the Li-Fraumeni syndrome[6] in which the predisposition is predominantly to brain tumours, sarcomas and breast cancer and in which inherited mutations in the p53 gene occur[7, 8]. Melanoma is a major risk factor for patients with xeroderma pigmentosum. Thus melanoma occurs as a part of family cancer syndromes where the susceptibility is to a variety of different cancers. In practice, then, taking a good family history, looking for recognisable patterns of cancers and seeking the opinion of Clinical Geneticists if in doubt, is the key approach.

There are in addition reports in the literature of clustering of melanoma in families and individuals with a variety of other cancers such as testicular tumours, lymphoma and brain tumours, the significance of which remains to be established [9-11] but which probably represents shared genetic and environmental aetiology.

In the majority of families with high penetrance melanoma susceptibility genes however the family history will reveal melanoma predominantly. To date there are probably at least 4 genes which underlie predisposition to melanoma in these families. The commonest is a gene on chromosome 9, called CDKN2A, which codes for a cell cycle control protein called p16. Families carrying mutations in this gene have been reported from Europe, the US and Australia. In UK studies, the likelihood of finding a mutation in this gene in melanoma families is much greater in families with larger numbers of affected family members. 50% of English families with 3 or more melanoma cases have CDKN2A mutations but only 12% of families with 2 or 3 cases[12-15]. In most CDKN2A families in the UK, the predisposition appears to be to melanoma alone. In other series however, particularly from the Netherlands and the US, there also appears to be an increased risk of pancreatic cancer[16, 17]. Valid estimates of risk of non-melanoma cancer in CDKN2A families are not yet available.

The second is the CDK4 gene[18], which codes for the protein to which p16 binds. Families with mutations in this gene are very rare indeed.

The third is p14ARF, which is adjacent to and continuous with CDKN2A on chromosome 9 and is coded by exon 1ß, alternately spliced with CDKN2A exon 2. p14ARF is therefore a second product of the same CDKN2A locus. Deletions of this gene have been shown to underlie susceptibility to melanoma and neural tumours[19]. These deletions appear to be very rare. More recently mutations in exon 1ß have been reported.

The commonest gene yet identified, which predisposes to melanoma is the gene which underlies most if not all red hair[20] and freckles[21] , known as MC1R. Variation in this gene has been shown to be associated with melanoma risk even in patients without red hair[22, 23]. The gene codes for the receptor for Melanocyte Stimulating Hormone, through which signaling for melanin production occurs. Variants in the gene appear to result in differing proportions of the black pigment eumelanin, and the redder pigment phaeomelanin, being produced. Some variants appear then to result in the production of phaeomelanin predominantly, with resultant red hair, freckles and sun sensitivity. The precise molecular determinants and biology of the susceptibility to melanoma remain to be elucidated but there appears to be no doubt that MC1R underlies weak susceptibility to melanoma in white populations.

There are other possible low penetrance genes promoting susceptibility to melanoma which Melanoma Genetics Consortium and others are exploring.

The hypothesis then, is that inheritance of these putative low penetrance genes predisposes white skinned peoples to melanoma but that as this predisposition is weak, there may be only one or occasionally two cases in the family. It is possible that the co-inheritance of more than one such low penetrance gene may result in more marked clustering of melanoma in some families. It is also possible that families carrying these low penetrance genes may have more cases of melanoma if they live in areas of the world where the environmental exposures are more extreme such as Queensland: that is, as a result of gene/environment interaction. This may at least in part explain why a family history of melanoma is more common in melanoma patients in hot climates.

In counselling patients as to what is the perceived risk of melanoma these considerations are taken into account and this is discussed below.

Assessing the significance of a family history of melanoma in the clinic

The families who require specialist counselling about risk then are those
With 2 first degree relatives with melanoma.
With 2 cases (even if more distant relatives) if one or more have had multiple primaries or the cases have the atypical mole syndrome (dysplastic nevi) as self examination is then rather more difficult.
With 3 or more cases of melanoma.

Counselling families about risk

In families with 2 cases of melanoma in first or second degree relatives, the melanoma may have occurred either as a result of a high risk gene or as a result of low penetrance susceptibility genes and a shared environment. For example, the patients may both be very fair skinned and have had a lot of sun exposure over the years. If that is the case, then if other family members keep out of the sun then their risks may be reduced. Reassuring advice about sun protection is therefore reasonable here. If one of the cases has had multiple primaries, which is more common with higher penetrance genes however, then one would be more suspicious that there was indeed a higher penetrance gene segregating in the family.

In families with 3 or more cases then the assumption must be that there is indeed a high penetrance susceptibility gene in the family. Cancer susceptibility genes are usually autosomal dominant and therefore the risk of carrying the gene for offspring of cases is 50%.

When familial melanoma was first explored by Clark and his group[26] it was the early data seemed to suggest that gene carriers could be predicted by the presence of the atypical mole syndrome (AMS also known as the dysplastic nevus syndrome or the FAMMM syndrome): that is, an abnormal naevus phenotype characterised by the presence of multiple moles and moles which are atypical (5mm or larger in diameter with a macular component and an irregular or indistinct margin). See “Coping with a diagnosis of melanoma” and image below. However UK studies have shown that this is not the case[27]. That is, that although CDKN2A mutation carriers were more likely to have the AMS, many carriers did not. Non-carriers also had large numbers of naevi. The presence of the AMS cannot therefore be used to predict gene carrier status. All family members should therefore be considered at increased risk and advised about sun protection and naevus surveillance accordingly.

What is the penetrance of high penetrance genes?

CDKN2A is the only gene we have any real data on. These data come from the Melanoma Genetics Consortium[28] and they show that penetrance is lower in Europe perhaps than might have been supposed. Thus, only 5% of gene carriers have developed a melanoma by the age of 40 years, 25% by the age of 60 years and 60% by the age of 80 years. The estimates are, moreover, likely, if anything, to be too high as they are derived from families selected for greater numbers of melanomas, which may therefore have additional risk factors. This might be reassuring news to family members. The Consortium data also showed that the penetrance is much higher in Australia as one would expect because of the effect of UV exposure. Whilst this is probably worrying news for families living there it might paradoxically offer hope. That is, the fact that gene penetrance appears to vary with latitude may suggest that gene carriers will be able to reduce their risk of developing a melanoma by avoiding the sun.

The positive message given to families being counseled therefore is
That children of melanoma patients have only a 50% chance of carrying the causal gene.
That if they carry such a gene but live in Europe they still have a good chance of not developing melanoma as the penetrance is only 60% by the age of 80 years.
That they may be able to reduce that risk by avoiding the sun.
That melanomas after the first two cases in a given family tend to be lower risk lesions with a higher probability of cure, perhaps because of surveillance and other interventions, but also because of greater awareness of the disease in other family members (1).

What advice should we give specifically about moles?

All family members (moley or otherwise) should self-examine their moles so that they know them and can spot a mole which is changing. They need to know the natural history of moles (how they change normally through life from being junctional to dermal) and what changes would be considered worrying. It is often useful to teach family members together and there is a need to ensure that family members take a responsibility for backs which are difficult to see for the patient themselves. A once a month inspection is sufficient.

Photographs are important. The patients need to have photographs of melanomas to take home and prints of their own more atypical moles.

The necessary frequency of follow up by a dermatologist is difficult to define as it depends upon the number of moles and the degree of atypicality of the moles the patient has (and therefore the difficulty of self monitoring), The stability of the moles also varies through life. Some patients appear to experience rapid changes in naevi over a period of years but they then settle down. Some have suggested that more regular follow up should be arranged during pregnancy although there is little evidence for an effect of pregnancy on melanoma. The personality of the patient and their skill in self-examination also varies. Some patients always struggle to understand the concepts of self-examination. Others become extremely adept at spotting change and will seek advice when and where it is needed. Others struggle with balancing anxiety. That is, that some patients worry so much that they seek advice very frequently. Others are so stoical that they don’t seek advice early enough whatever one does. The necessity for follow up and the frequency of that follow up should then be judged on the basis both of the phenotype of the patient and their personalities. What is crucial to remember however is that it should be remembered that all family members should consider themselves to be at risk and must have very prompt access to a dermatologist when they need it.

What should we say about sun avoidance?

Never sunbathe.
Never burn.
Protect yourself from the sun from 11am till 3pm.
Keep your clothes on: they are likely to be more effective than sunblock.
Wear sunblock which is SPF 15 or more and protects against UVA (four stars) as well as UVB.
Do not use sunblock in order to “safely” stay out in the sun longer.

What about gene testing for CDKN2A mutations?

Testing should we expect become widely available with time for those who want it: when the implications of a positive test are known. We hope then that the net result of testing for the majority will be helpful. That is, that a counselled patient with a positive test will then have access to accurate information about their risks. A person who is tested negative will have some justification then for feeling reassured. Finally it is important to remember that testing is only part of genetic counselling: indeed testing is only offered to families when the advantages and disadvantages of testing are fully understood. Many will then choose not to be tested.

Therefore we would suggest that testing is premature at the present time for the following reasons:
The risk of other cancers in CDKN2A families is not yet known.
The penetrance, and the factors which moderate the penetrance of CDKN2A will be much better understood in a few years.
The lack of correlation between the AMS and gene carrier status may mean that the risk to non-CDKN2A gene carriers in families who still have the AMS is probably above population estimates and some have anxieties about false reassurance of these family members.

References

1. Newton, J.A., V. Bataille, K. Griffiths, et al.,
How common is the atypical mole syndrome phenotype in apparently sporadic melanoma?
J Am Acad Dermatol, 1993. 29: p. 989-996.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=8245266

2. Cutler, C., W. Foulkes, J.-S. Brunet, et al.,
Cutaneous malignant melanoma in women is uncommonly associated with a family history of melanoma in first-degree relatives: a case control study.
Melanoma Research, 1996. 6: p. 435-440.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=9013481

3. Aitken, J.F., D.L. Duffy, A. Green, et al.,
Heterogeneity of melanoma risk in families of melanoma patients.
American Journal of Epidemiology, 1994. 140(11): p. 961-973.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=7985658

4. Bataille, V., R. Hiles, and J. Newton Bishop,
Retinoblastoma, melanoma and the atypical mole syndrome.
British Journal of Dermatology, 1995. 60: p. 622-626.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=7756125

5. Traboulsi, E.I., L.E. Zimmerman, and H.J. Manz,
Cutaneous Malignant Melanoma in Survivors of Heritable Retinoblastoma.
Arch Ophthalmology, 1988. 106: p. 1059-1061.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=3041943

6. McKusick, V.,
Online Mendelian Inheritance in Man.
2002.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=OMIM

7. Li, F.P. and J.F. Fraumeni Jr,
Soft-Tissue Sarcomas, Breast Cancer, and other Neoplasms. A Familial Syndrome? Ann.
Intern. Med., 1969. 71: p. 747-752.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=5360287

8. Santibanez-Koref, M.F., J.M. Birch, A.L. Hartley, et al.,
p53 germline mutations in Li-Fraumeni syndrome.
Lancet, 1991. 338: p. 1490-1491.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=1683921

9. Travis, L.B., R.E. Curtis, H. Storm, et al.,
Risk of second malignant neoplasms among long-term survivors of testicular cancer.
J Natl Cancer Inst, 1997. 89(19): p. 1429-39.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=9326912

10. Paunu, N., E. Pukkala, P. Laippala, et al.,
Cancer incidence in families with multiple glioma patients.
Int J Cancer, 2002. 97(6): p. 819-22.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=11857361

11. Hisada, M., R.J. Biggar, M.H. Greene, et al.,
Solid tumors after chronic lymphocytic leukemia.
Blood, 2001. 98(6): p. 1979-81.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=11535538

12. Harland, M, R. Meloni, N. Gruis, et al.,
Germline mutations of the CDKN2 gene in UK melanoma families, in Human Molecular Genetics.
1997. p. 2061-2067.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=9328469

13. Harland, M., E.A. Holland, P. Ghiorzo, et al.,
Mutation screening of the CDKN2A promoter in melanoma families.
Genes Chromosomes Cancer, 2000. 28(1): p. 45-57.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=10738302

14. Harland, M., S. Mistry, D.T. Bishop, et al.,
A deep intronic mutation in CDKN2A is associated with disease in a subset of melanoma pedigrees.
Hum Mol Genet, 2001. 10(23): p. 2679-86.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=11726555

15. Newton Bishop, J.A., M. Harland, D.C. Bennett, et al.,
Mutation testing in melanoma families: INK4A, CDK4 and INK4D.
Br J Cancer, 1999. 80(1-2): p. 295-300.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=10390011

16. Bergman, W., P. Watson, J. de Jong, et al.,
Systemic cancer and the FAMMM syndrome.
British Journal of Cancer, 1990. 61: p. 932-936.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=2372499

17. Lynch, H.T. and R.M. Fusaro,
Pancreatic cancer and the familial atypical multiple mole melanoma (FAMMM) syndrome.
Pancreas, 1991. 6(2): p. 127-131.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=1886881

18. Zuo, L., J. Weger, Q. Yang, et al.,
Germline mutations in the p16INK4a binding domain of CDK4 in familial melanoma.
Nature Genetics, 1996. 12(1): p. 97-99.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=8528263

19. Randerson-Moor, J.A., M. Harland, S. Williams, et al.,
A germline deletion of p14(ARF) but not CDKN2A in a melanoma-neural system tumour syndrome family.
Hum Mol Genet, 2001. 10(1): p. 55-62.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=111367140

20. Valverde, P., E. Healy, I. Jackson, et al.,
Variants of the melanocyte stimulating hormone receptor gene are associated with red hair and fair skin in humans.
Nature Gentics, 1995. 11: p. 328-330.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=7581459

21. Bastiaens, M., J. ter Huurne, N. Gruis, et al.,
The melanocortin-1-receptor gene is the major freckle gene.
Hum Mol Genet, 2001. 10(16): p. 1701-8.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=11487574

22. Valverde, P., E. Healy, S. Sikkink, et al.,
The Asp84Glu variant of the melanocortin 1 receptor (MC1R) is associated with melanoma.
Human Molec Genet, 1996. 5(10): p. 1663-1666.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=8894704

23. Palmer, J.S., D.L. Duffy, N.F. Box, et al.,
Melanocortin-1 receptor polymorphisms and risk of melanoma: is the association explained solely by pigmentation phenotype?
Am J Hum Genet, 2000. 66(1): p. 176-86.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=10631149

24. Winsey, S.L., N.A. Haldar, H.P. Marsh, et al.,
A variant within the DNA repair gene XRCC3 is associated with the development of melanoma skin cancer.
Cancer Res, 2000. 60(20): p. 5612-6.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=11059748

25. Shahbazi, M., V. Pravica, N. Nasreen, et al.,
Association between functional polymorphism in EGF gene and malignant melanoma.
Lancet, 2002. 359(9304): p. 397-401.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=11844511

26. Clark, W.H., Jr., R.R. Reimer, M. Greene, et al.,
Origin of familial malignant melanomas from heritable melanocytic lesions. 'The B-K mole syndrome'.
Arch Dermatol, 1978. 114(5): p. 732-8.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=646394

27. Newton Bishop, J., M. Harland, R. Wachsmuth, et al.,
Genotype/phenotype and penetrance studies in melanoma families with germline CDKN2A mutations.
J Invest Dermatol, 2000. 114: p. 28-33.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=10620111

28. Bishop, D.T., F. Demenais, A.M. Goldstein, et al.,
Geographical variation in the penetrance of CDKN2A mutations for melanoma.
J Natl Cancer Inst, 2002. 94(12): p. 894-903.
www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=12072543

Figure 1: an atypical naevus

Figure 2: the atypical naevus syndrome which is associated with susceptibility to melanoma but not predictive ofCDKN2A gene carrier status.

Links to websites

Association of Clinical Cytogeneticists
http://www.cytogenetics.org.uk

Cambridge Genetics Knowledge Park
http://www.cgkp.org.uk/index.php

e-Melanobase, database of germline mutations in melanoma susceptability genes
http://www.wmi.usyd.edu.au:8080/melanoma.html