Services on Demand
Print version ISSN 0365-0596On-line version ISSN 1806-4841
An. Bras. Dermatol. vol.79 no.1 Rio de Janeiro Jan./Feb. 2004
CLINICAL, LABORATORY AND THERAPEUTIC INVESTIGATION
Hereditary melanoma: prevalence of risk factors in a group of patients in Southern Brazil*
Charles André CarvalhoI; Mauricio Estrela da CunhaI; Roberto GiuglianiII; Lucio BakosIII; Patrícia Ashton-ProllaIV
students, Universidade Federal do Rio Grande do Sul
IITitular Professor of Medical Genetics, Universidade Federal do Rio Grande do Sul
IIITitular Professor of Dermatology, Universidade Federal do Rio Grande do Sul
IVProfessor of Post-Graduation Program in Genetic and Molecular Biology, Universidade Federal do Rio Grande do Sul
Approximately 10% of melanoma cases are attributable to cell line mutations
in predisposition genes, and are therefore hereditary.
OBJECTIVES: The present study evaluates the prevalence of hereditary melanoma risk factors in a group of patients diagnosed with the disorder in Southern Brazil.
PATIENTS AND METHODS: One-hundred and ninety-five patients diagnosed with melanoma who were consecutively admitted to the Hospital de Clinicas de Porto Alegre between January 1999 and June 2000 were studied.
RESULTS: The mean age at diagnosis of melanoma was 48 years in men and 45 years in women. From the 133 patients who had their family history registered in the hospital records, 13 (9.8%) had a family history of melanoma. The presence of at least one important risk factor for hereditary melanoma was observed in 31 patients (16%; CI 95%; 10.9-21.1%).
CONCLUSIONS: The possibility of a genetic influence in melanoma cases of this region should stimulate the active search for specific hereditary melanoma risk factors, since the identification of at-risk patients may have an impact on disease morbidity and mortality.
Key-words: genes, tumor suppressor; melanoma; neoplastic syndromes, hereditary.
Melanoma - a malignant neoplasia originating in the melanocytes, the cells responsible for cutaneous pigmentation - is the primary cause of death among skin diseases, and represents about 1% of all malignant tumors. Associated with a high morbidity and mortality rate, its incidence has been increasing dramatically in the last four decades in various industrialized countries.1 In the USA, 44,200 new cases were reported in 1999, resulting in 7,300 deaths.2 In 2002, 53,000 new cases are expected.3 Between 1973 and 1998, the incidence and the mortality due to melanoma in that country increased continually, at rates superior to those of most of the other preventable neoplasias.4 In Brazil, it is estimated that about 4,000 new cases5 are diagnosed every year, which is probably an underestimation of the real numbers.5,6 The average age at the diagnosis of melanoma in the general population is 57 years for men and 50 years for women.7 In individuals at high risk for hereditary melanoma, on the other hand, the average age at diagnosis is earlier reaching 36 years in men and 29 years in women.8 Besides the much earlier age at the time of diagnosis, patients with familial melanoma have a greater incidence of multiple primary melanomas.
In this manifestation several risk factors are known that are related basically with cutaneous and pigmental characteristics, such as presence of numerous nevi, atypical nevi, freckles, red hair, incapacity to tan and a propensity to sunburns.9,10 Over many years, several epidemiological studies have shown that solar exposure, particularly during childhood, is the main environmental cause of melanoma.11 It has been estimated that up to 65% of the cases of melanoma may be related to solar exposure.12 In a case-controlled study accomplished with patients in the south of Brazil, Bakos et al.13 found that the relative risk for melanoma was 2.7 (CI 95% 1.3 - 5.6) in patients with phototypes I and II when compared to patients with phototypes III and IV. The same study showed that the color of the eyes and the color of the hair did not represent independent risk factors for melanoma. Additionally, it was demonstrated that the number of nevi (>30) and the presence of dysplastic nevi constitute a factor of moderate risk. Finally, the most important risk factor reported in that work was a history of numerous (30 or more) episodes of solar burn during a lifetime.
Formal genetic analyses accomplished in the decade of the 1980s identified a pattern of autosomal dominant inheritance in some families with multiple cases of melanoma. Nowadays, it is believed that most of the cases of this neoplasia result from the interaction of genetic and environmental risk factors, constituting a model of a multifactorial disease.1 An example of that interaction is demonstrated by the influence of geographical variation on the penetrance of mutations in the CDKN2A gene, in other words, the penetrance increases in proportion to the degree of solar exposure to which the population is submitted, as inferred by the latitude.14
About 8 to 14% of the patients who are diagnosed with melanoma present a family history (FH) that is positive for this neoplasia.15 In addition, it is estimated that about 10% of the patients diagnosed with more than one primary melanoma present a hereditary predisposition to mutation of this tumor. Although groupings of cases of melanoma and/or dysplastic nevi may occur in a family due to common patterns of solar exposure and not to genetic factors, today, 10% of all cases of melanoma are considered to be caused by germinative mutations in susceptible genes, and are thus hereditary.16 In these cases, a single genetic alteration (mutation) suggests a significantly greater risk of developing this and other neoplasias in the affected individual, in his descendants and in other members of his family. Until now, two genes associated with the family type melanoma have been identified: CDKN2A and CDK4.1 The first of these genes is altered in about 25% of the cases of family melanoma, but alterations in the second were identified in only three families. There is strong suspicion that a third gene - p14ARF - should be included in this category, as well as other as yet unidentified genes. Recently, alterations in the MC1R gene have also been associated with a slight increase in the risk of melanoma. This gene codifies a key-protein in the process of cutaneous pigmentation, evidence of a connection between the skin type and the risk of melanoma.1 Finally, mutations in other tumor suppressor genes, such as, for example, BRCA2 and NF1, can also be associated with a hereditary predisposition to melanoma.17,18,19
Phenotypically, the families and individuals with mutations in CDKN2A may present the condition in various forms: multiple cases of melanoma in the family, multiple primary melanomas in an individual and the presence of melanoma and other neoplasias (especially cancer of the pancreas and tumors of the central nervous system) in the same family. The prevalence of mutations in CDKN2A in the general population is not known.1
Most of the melanomas diagnosed in the early stages are curable surgically. However, the mortality rate is high among patients diagnosed in more advanced stages, even after aggressive therapeutic interventions. In this context, the identification of individuals genetically predisposed to melanoma can be important in targeting them for interventions of surveillance, early diagnosis, and treatment in that high-risk group, thus raising the hope of significantly reducing the morbidity and mortality rate of the disease.
The objective of this study was to analyze a sample of patients, attended at a tertiary care hospital in Rio Grande do Sul (Brazil), with a confirmed diagnosis of melanoma, for the prevalence of factors suggestive of hereditary susceptibility to this neoplasia.
PATIENTS AND METHODS
The present study describes a sample composed of all the patients attended for clinical and/or surgical treatment of melanoma at the Hospital de Clinicas de Porto Alegre (HCPA) between January 1, 1999 and June 4, 2000. These cases were identified by a search of hospital records done by the data-processing service of the institution. All of the hospitalizations and hospital procedures were selected when the main characteristics corresponded to the definitions set out by the International Classification of Diseases (ICD) regarding melanoma. The codes considered were: 17200-17299 (ICD9: malignant melanoma of the skin), D03 (ICD 10: melanoma in situ) and C43 (ICD 10: malignant melanoma of the skin). Subsequently, the records of these patients were obtained from the Archive Service of Medical and Health Information (ASMHI). Each record was thoroughly reviewed, studying the confirmation of the diagnosis in an attempt to identify risk factors associated with family melanoma. Specifically investigated were: a family history of melanoma, a family history of pancreatic cancer, a history of pancreatic cancer and melanoma in the same patient, a history of melanoma combined with other malignant neoplasias, a diagnosis of a genetic syndrome associated with melanoma, a diagnosis of dysplastic nevus syndrome and a history of multiple primary melanomas.
Accompanying the risk factors of each case, other characteristics were listed, such as sex, origin, age at the time of the diagnosis and the location of the melanomas.
This project was approved by the Committee of Ethics and Research of HCPA.
The initial computer search identified 253 patients, in 195 (77%) of which the diagnosis of melanoma was confirmed in the review of the records, and were included in this analysis. The other patients did not present sufficient evidence for this diagnosis in their medical records. The principal characteristics of this sample are shown in table 1.
The age at the time the melanoma was diagnosed varied between 13 and 87 years, with a mean of 46.7 years, a standard deviation (SD) of 15.9 years, and a median of 44 years. Among the men, the mean age at diagnosis was 48 years (SD 15.75), while among women it was 44.9 years (SD 15.94).
Of the 133 patients whose family history was registered in the hospital records, 13 (9.8%) presented a family history of melanoma and one (0.75%), of pancreatic cancer. The family history as mentioned in the records was considered, negative or positive, when it was evident that at some time the patient had been questioned regarding a family history of cancer. The presence of cancer of the pancreas along with melanoma, was verified in one (0.5%) patient and the presence of other malignant neoplasias in patients already diagnosed with melanoma, in 15 (7.7%) patients. These other malignant neoplasias are described in table 2. One patient (0.5%) had been diagnosed as having a genetic syndrome associated with melanoma. Dysplastic nevus syndrome had been diagnosed in six (3.1%) patients, and the occurrence of more than one primary melanoma in the same patient, a characteristic associated with family melanoma, was verified in 14 (7.2%) patients, 12 of them with two primary melanomas, and two with three.
In this work a hereditary predisposition to melanoma was considered suggestive in the presence of at least one of the following conditions, in addition to the diagnosis of melanoma in the individual:
1. positive family history (at least one case);
2. multiple primary melanomas;
3. occurrence of cancer of the pancreas and melanoma in the same patient;
4. patient with melanoma and family history of cancer of the pancreas;
5. dysplastic nevus syndrome.
The prevalence of each one of the risk factors above is described in table 3.
Thirty-one patients (16%) presented at least one of these conditions, being considered, therefore, at increased risk due to the presence of hereditary susceptibility to melanoma (CI 95% 10.9% - 21.1%). the mean age at the diagnosis of melanoma in this subgroup of patients was 46.2 years (SD 16.3).
The Hospital de Clinicas de Porto Alegre (HCPA) attends patients referred from the entire state of Rio Grande do Sul and also from other areas of the southern part of Brazil, as it is a regional center for the treatment of melanomas, which could, in itself, introduce a bias into the selective process of this work, in the sense of increasing the representation of the more serious cases, such as those with risk factors for a hereditary predisposition. Generally speaking, the State of Rio Grande do Sul presents a greater incidence of melanoma in comparison with the remainder of Brazil. This may be explained partly by an underreporting of cases in other states. But it also should be noted that an important proportion of the southern population presents phenotypical characteristics (color of the skin, hair and eyes) similar to those of the European and North American populations, which causes them to be particularly susceptible to melanoma.
Another interesting finding was the diagnosis of neurofibromas (along with melanoma) in three patients of the sampling. To date, a definitive association between these tumors has not been described. However, some studies have suggested such a possibility, considering that the melanocytes are responsible for the most preeminent characteristics of neurofibromatosis, such as the café au lait spots and the giant congenital nevus. Furthermore, the melanocyte has the same embryonic origin as most of the tumors associated with neurofibromatosis, in other words, the neuroectoderma.17 Andersen et al.20 demonstrated a deletion in both alleles of the NF1 gene in most of the cell lineage of melanoma, suggesting that such a gene could work as a tumoral suppressor in the development or in the progression of the melanoma.
The central question underlying this work is whether there is a greater prevalence (than previously thought) of specific risk factors for hereditary melanoma in the cases of melanoma diagnosed in Rio Grande do Sul. In this pilot study, the findings suggest that hereditary factors had an important significance in the sample of patients. This evidence is based on significantly younger ages at the time of diagnosis as observed in this sample, compared to the literature, and on the presence of an important risk factor in at least 16% of the cases, compared to 10% in the international literature. In reference to the age at diagnosis, it was observed that more than 55% of the studied cases were identified before 45 years of age, while such a proportion is 35% in the North American statistics.7 The average age at diagnosis in women was five years earlier than that described for the female population and, in men, it was nine years earlier than that described for their male population as a whole. There was basically no difference in the average age at diagnosis between the patients with risk factors for hereditary melanoma and those without, in this sample (46.2 and 46.7 years, respectively).
The patients at risk of hereditary melanoma identified in this study and also those identified at the clinic of Dermatology of HCPA have been invited to participate in a more detailed study, in which they will be offered the molecular test for research of mutations in the CDKN2A gene to be followed by the appropriate genetic counseling. It should be pointed out that, nowadays, the genetic test to identify mutations associated with melanoma is not yet ready to be indicated for use outside of the research context, according to a recent consensus of the Consortium for Genetics of Melanoma, an entity that brings together most of the groups of research in the world dedicated to the study of familial melanoma.21
The results of this study in a group of patients in the south of Brazil indicate that a significant proportion of individuals diagnosed with melanoma presented a factor of genetic risk for this neoplasia. The active search for factors of genetic risk in these patients is relevant in that the identification of high-risk families may reduce the morbidity and mortality of their members. The impact of hereditary factors on the prevalence of melanoma in the population of southern Brazil needs to be evaluated in full detail and in a larger and more representative sample of the various regions of the state.
Financial support: CAC received a scientific initiation grant from PIBIC-UFRGS, MEC received a scientific initiation grant from FAPERGS, PAP received a post-Ph.D. grant from CNPq and currently a ProDoc grant from CAPES. Additional financial support was received from FAPERGS for the realization of this work.
1. Bressac-de-Paillerets B, Avril MF, Chompret A, Demenais F. Genetic and environmental factors in cutaneous malignant melanoma. Biochimie 2002; 84: 67-74. [ Links ]
3. Melanoma: epidemic or better detection? Harv Health Lett. 2002 Jul;27(9):7. [ Links ]
4. Howe HL, Wingo PA, Thun MJ, et al. Annual report to the Nation on the Status of Cancer (1973 through 1998), featuring cancers with recent increasing trends. J Natl Cancer Inst. 2001;93:824-842. [ Links ]
6. Azevedo G, Mendonça S. Risco crescente de melanoma de pele no Brasil. Rev Saude Publ 26:290-294; 1992. [ Links ]
7. Miller BA et al. Cancer Statistics Review 1973 - 1989. NIH Publ No 92 - 2789. Bethesda, Md: NCI 1992. [ Links ]
8. Greene MH. The genetics of hereditary melanoma and nevi. 1998 Update. 1999 Cancer 86:1644-57. [ Links ]
9. Demenais F, Chompret A, Guilloud-Bataille M, Feingold N, Grange F, Avril MF. Interactions between genetic and epidemiological risk factors in cutaneous malignant melanoma in France. Am J Hum Genet 1996; 59: A28. [ Links ]
10. Berwick M, Halpern A. Melanoma epidemiology. Curr Opin Oncol 1997; 9: 178-182. [ Links ]
11. Whiteman DC, Whiteman CA, Green AC. Childhood sun exposure as a risk factor for melanoma: a systematic review of epidemiologic studies. Cancer Causes Control 2001; 12: 69-82. [ Links ]
12. Armstrong BK, Kricker A. How much melanoma is caused by sun exposure? Melanoma Res 1993; 3: 395-401. [ Links ]
13. Bakos L, Wagner M, Bakos RM et al. Sunburn, sunscreens, and phenotypes: some risk factors for cutaneous melanoma in southern Brazil. Int J Dermatol 2002 Sep;41(9):557-62. [ Links ]
14. Bishop DT, Demenais F, Goldstein AM et al. Geographical variation in the penetrance of CDKN2A mutations from melanoma. J Natl Cancer Inst 2002; 94(12): 872-3. [ Links ]
15. Grange F, Chompret A, Guilloud-Bataille M et al. Comparison between familial and non- familial melanoma in France. Arch Dermatol 1995; 131:1154-1159. [ Links ]
16. Lindor NM, Greene MH et al. The concise handbook of family cancer syndromes. J Natl Cancer Inst 1998; 90:1039-1071. [ Links ]
17. Karakayali G, Güngör E, Lenk N et al. Neurofibromatosis and cutaneous melanoma: coincidence or association?. J Eur Acad Dermatol Venereol 1999; 12(2): 190-2. [ Links ]
18. Neuhausen S, Godwin AK, Gershoni-Baruch R et al. Haplotype and phenotype analysis of nine recurrent BRCA2 mutations in 111 families: results of an international study. Am J Hum Genet 1998; 62: 1381-1388. [ Links ]
19. Wooster R et al. Localization of a breast cancer susceptibility gene, BRCA 2 to chromosome 13q12-13. Science 1994; 265: 2088-2090. [ Links ]
20. Andersen LB, Fountain JW, Gutmann DH et al. Mutations in the neurofibromatosis 1 gene in sporadic malignant melanoma cell lines. Nat Genet 1993; 3(2): 118-121. [ Links ]
21. Kefford R, Bishop JN, Tucker M et al. Genetic testing for melanoma. Lancet Oncol 2002 Nov;3(11):653-4. [ Links ]
Dra. Patricia Ashton-Prolla
Serviço de Genética Médica, Hospital de Clínicas de Porto Alegre
Rua Ramiro Barcelos, 2350
90035-003 Porto Alegre RS
Tel/Fax: (51) 3316-8011 / 3316-8010
in April, 28th of 2003
Approved by the Consultive Council and accepted for publication in September, 10th of 2003
* Work done at the Medical Genetics and Dermatology Services of the Hospital de Clinicas de Porto Alegre