当前位置: 首页 > 期刊 > 《临床肿瘤学》 > 2005年第5期 > 正文
编号:11329533
Role of the CDKN2A Locus in Patients With Multiple Primary Melanomas
http://www.100md.com 《临床肿瘤学》
     the Dermatology Department and Genetics Service, Hospital Clínic, Institut d’Investigacions Biomèdiques August Pi Sunyer, Universitat de Barcelona

    Centre de Genètica Mèdica i Molecular, Institut de Recerca Oncològica Hospital Duran i Reynals, L'Hospitalet de Llobregat, Barcelona

    Genetic Counseling of Cancer

    Dermatology Department, Hospital Sant Joan, Universitat de Rovira i Virgili, Reus, Spain

    ABSTRACT

    PURPOSE: We have studied a consecutive case series of patients with multiple primary melanoma (MPM) for the involvement of the melanoma susceptibility loci CDKN2A and CDK4.

    PATIENTS AND METHODS: One hundred four MPM patients (81 patients with two primary melanomas, 14 with three, five with four, one with five, two with six, and one with seven) were included.

    RESULTS: Seven different CDKN2A germline mutations were identified in 17 patients (16.3%). In total, we identified 15 CDKN2A exon 2, one exon 1 missense mutation, and one exon 1 frameshift mutation. The age of onset was significantly lower and the number of primary melanomas higher in patients with mutations. CDKN2A mutations were more frequent in patients with familial history of melanoma (35.5%) compared with patients without (8.2%), with a relative risk (RR) of 4.32 (95% CI, 1.76 to 10.64; P = .001), and in patients with more than two melanomas (39.1%) compared with patients with only two melanomas (10%) with an RR of 3.29 (95% CI, 1.7 to 6.3; P = .002). The A148T polymorphism was more frequent in patients with MPMs than in the control population (P = .05). A variant of uncertain significance, A127S, was also detected in one patient. No CDK4 mutations were identified, suggesting that it has a low impact in susceptibility to MPM.

    CONCLUSION: MPM patients are good candidates for CDKN2A mutational screening. These patients and some of their siblings should be included in a program of specific follow-up with total body photography and digital dermoscopy, which will result in the early detection of melanoma in this subset of high-risk patients and improve phenotypic characterization.

    INTRODUCTION

    Cutaneous melanoma (CM) is a potentially fatal type of skin cancer that derives from melanocytes. The incidence of CM is continuing to increase; at present it is the cancer with the highest rate of increase in white populations.1 Nearly 10% of CM occurs in a familial setting. Two main genes involved in melanoma susceptibility have been described previously. The first, CDKN2A, located on chromosome 9p21, encodes two distinct proteins: p16INK4A and p14ARF. These proteins are transcribed from different exons 1; they share exons 2 and 3, but in a different reading frame. Both proteins can function as tumor suppressors. The p16INK4A protein inhibits the activity of the cyclin D1-cyclin–dependent kinase 4 (CDK4) complex, the function of which is to drive cell-cycle progression by phosphorylating the retinoblastoma protein, whereas p14ARF interacts with the MDM2 protein, of which the principal function is to associate with and promote the ubiquitin-mediated degradation of the p53 tumor suppressor.2 The implication of p14ARF in melanoma susceptibility has been described only recently.3-5

    The second melanoma susceptibility gene, CDK4, is an oncogene located in 12q13.6 Only three melanoma families worldwide are carriers of mutations in CDK4 exon 2 (Arg24Cys and Arg24His). The two mutations are located in the p16INK4A-binding domain of CDK4, leaving the p16INK4A protein unable to inhibit the D1-cyclin-CDK4 complex.

    Germline CDKN2A mutations have been described in 10% to 50% of families from several countries7-12 and in 17% of Spanish melanoma families we have studied.13 But CDKN2A germline mutations have also been described in patients with more than one primary CM without familial history.14 These patients have an increased risk of carrying mutations in CDKN2A, which are detected in 8.3% to 15% of multiple primary melanoma (MPM) patients irrespective of family history,14-16 in 9% to 12% of sporadic cases,1,17 and in 47.8% of familial cases.18,19

    Our aim in the present work was to assess the role of the two transcripts of the CDKN2A locus in melanoma susceptibility by studying 104 patients with MPMs to reveal the genotype/phenotype correlation in these patients and their families.

    PATIENTS AND METHODS

    Patients, Study Design, and Evaluation of Phenotypes

    A consecutive case series study design was used to analyze clinical and genetic characteristics of patients with MPM.

    One hundred four patients (52 women and 52 men) with more than one primary cutaneous melanoma were included (81 patients with two primary CMs, 14 patients with three, five patients with four, one patient with five, two patients with six, and one patient with seven). All patients were treated and controlled at the Melanoma Unit in the Hospital Clinic of Barcelona (Barcelona, Spain). Total-body exploration with digital image support and digital epiluminiscence microscopy (Mole Max II, Derma Instruments, Vienna, Austria) was performed to improve the likelihood of early diagnosis of melanoma. Age of onset, previous history of familial melanoma (or other cancer), personal history of cancer, presence of dysplastic nevi (DN) confirmed by biopsy, localization of melanomas, and time elapsed between primaries were recorded. The mean age of onset was 43.8 years (standard deviation [SD], ± 16.26 years; range, 9 to 83 years), and in 30 cases two primary melanomas were diagnosed at the same time (synchronic primary melanomas). The age of onset was significantly lower in women compared with men (Tables 1-4), but the number of primary melanomas was nearly the same in both sexes. In only three cases did the second melanoma appear close to the first melanoma (eg, on the same extremity), and in these three cases histology was more suggestive of a new primary melanoma than epidermotropic metastases. At least one dysplastic nevus has been removed in 75 patients (72.1%). Familial melanoma was previously present in 25 cases, but three new familial cases were identified in the exhaustive revision of the pedigrees. When a germinal mutation was identified, further study of the family was then performed, and mutation-carrying relatives were called for total body exploration and epiluminiscence microscopy for early diagnosis of melanoma. This resulted in the diagnosis of three new cases of melanoma and identification of two new melanoma families. When no germ line mutations were identified, first-degree relatives with dysplastic nevus syndrome were also included in the clinical and epiluminiscence program of follow-up, resulting in the diagnosis of one melanoma case and the identification of one new melanoma family. Initial results from analysis of some patients have been previously reported.4,13,19,20

    The study was approved by the ethical committee of the hospital, and informed consent was obtained from all patients.

    Mutational Analysis

    Genomic DNA was obtained from peripheral lymphocytes of all patients included in the study by standard methods.

    Primers for the CDKN2A locus were amplified by polymerase chain reaction (PCR) as previously described.19,21 Allele-specific PCR for the -34G>T and IVS2-105A>G mutations was performed as previously described.22,23 The -34G>T variant generates an aberrant initiation codon and IVS2-105A>G creates a false GT splice donor site. CDK4 exon 2 was amplified in two overlapping fragments as previously described.6

    The technique used to detect mutations was the single strand conformation analysis. Denatured PCR product 3 μL was combined with loading buffer and loaded into GeneGel Excel 12.5 acrylamide gels (Amersham Biosciences, Uppsala, Sweden) and run at 15°C for 2 hours. The gels were silver-stained as previously described.24 DNA samples with abnormally migrating products were sequenced as follows: PCR products were purified using the GFX PCR DNA and Gel Band purification kit (Amersham Biosciences) and automatically sequenced using the BigDye Terminator v3.1 Cycle Sequencing kit (Applied Biosystems, Foster City, CA) and an ABI3100 automatic sequencer (Applied Biosystems). The mutations described in this study were designated following the recommendations of den Dunnen and Antonarakis.25

    Haplotype Analysis

    Haplotype analyses were performed in individuals carrying the recurrent mutations G101W and R87W. Seven 9p21 microsatellites around CDKN2A were used: D9S736, D9S1749, D9S972, D9S942, D9S1748, D9S171, and D9S126. One of the primers was fluorescence labeled. Amplification conditions were as previously described.26 The PCR products were analyzed on an ABI310 (Applied Biosystems, Foster City, CA).

    Statistical Analysis

    To evaluate the association between categoric variables, Pearson’s 2 test was used. If the expected frequency was < 5, Fisher’s exact test was used (Table 4). The 95% CIs were obtained by the exact binomial method, and the relative risk (RR) was calculated. To compare mean of quantitative variable, the Student t test was used.

    RESULTS

    CDKN2A variants detected in this study are summarized in Tables 1 and 2. Five missense and two frameshift CDKN2A mutations were detected in 17 MPM patients (16.3% of 104 patients analyzed). Most were identified in unrelated patients, but six patients belonged to three different MPM-prone families. In 15 patients, the mutation was located in exon 2 (four missense: V59G, L65P, R87W, and G101W; one frameshift: 358delG), in one patient it was located in exon 1 (missense, G35E), and in another patient in exon 1 (frameshift, 60ins16; Table 1). Thirteen patients (76.5%) were carriers of mutations that impair both p16INK4A and p14ARF proteins. In three patients (17.7%) the mutation affected only the p16INK4A protein, and in one case (5.8%) the mutation affected only p14ARF. A previously reported variant of unknown functional significance (A127S) was detected in one patient (Table 1). Various silent polymorphisms were also detected (Table 2). No mutations were identified in CDK4 exon 2.

    Clinical and Molecular Consequences of CDKN2A Mutations

    We detected seven different mutations, one variant of unknown significance (A127S), and four polymorphisms in the 104 MPM patients studied (Tables 1 and 2). Seventeen patients were CDKN2A mutation carriers, and their clinical characteristics are summarized in Table 3.

    An exon 1 frameshift mutation was identified in a woman with two primary melanomas (diagnosed at the ages of 37 and 43 years, respectively). The mutation is a 16-base insertion (60ins16) that creates a premature stop codon in exon 1 of p14ARF, and has no effect in p16INK4A. Two of the three daughters of the proband were also carriers of the mutation, but have not yet developed melanoma (at ages 32 and 38 years, respectively). Clinically atypical nevi were not present in the index patient or in her daughters.

    The frameshift mutation 358delG was identified in a patient belonging to a melanoma family that has been clinically and molecularly characterized previously.19 This mutation produces a premature stop codon in exon 2 of CDKN2A that results in translation of a truncated p16INK4A protein. Because this change affects the first nucleotide after the p14ARF stop codon, it probably does not affect the p14ARF protein. Mutation carriers in this family tend to develop other cancers in addition to melanoma (Table 3). Two new melanoma cases were identified in specific follow-up of carriers in this kindred.

    The missense mutation G35E was identified in a patient with four MPMs without family history of melanoma or DN but with several other cancers in the family (Table 3). This mutation affects only exon 1, and it has not been described previously. In some families another missense mutation affecting the same codon (G35A) has been reported and was related to melanoma susceptibility.9 We have screened 100 control patients and more than 600 melanomas, and this change has not been detected. Moreover, because glycine35 is located in the first ankyrin repeat and is conserved between human and mouse, it is likely to be of functional significance.

    Mutation V59G in p16INK4A affects also p14ARF (S73R). This mutation was identified in a patient affected by three MPMs whose son developed melanoma 1 year after the mutation was identified. Subsequently, two other cases of melanoma have been identified in this family. This family was originally reported before the third melanoma was diagnosed in the proband.13 Recently it has been demonstrated that this mutation derived from a common ancestor with some French and Jewish families.20

    Mutation L65P was identified in a 30-year-old patient with two synchronically diagnosed primary melanomas, and DN syndrome (> 200 moles). His mother, aunt, and sister are also carriers of the mutation, and in the special follow-up of carriers, an early invasive melanoma (Breslow thickness, 0.5 mm) has been diagnosed recently in the 35-year-old sister. The index patient was a professional chronically exposed to ionizing radiation. This mutation only affects the p16INK4A transcript, producing a nonfunctional protein.

    An R87W mutation was identified in two unrelated families (M196 and M927). This mutation has no effect in p14ARF protein. In family M196, the proband was a 20-year-old patient with two synchronic MPMs and DN syndrome. When we visited the family, an in situ melanoma was identified in the patient’s father, who was also a carrier of the mutation. In the specific follow-up of the father, a second melanoma (early invasive; 0.5-mm thickness) was also identified. In family M927, a patient with two MPMs and familial history of cancer was also carrier of the mutation.

    The most common variant detected was G101W in p16INK4A. This mutation also creates a missense mutation in p14ARF (R115L). It was identified in nine patients belonging to seven unrelated families (Table 3). In three cases we found no evidence of family history of melanoma, although in one family DN were reported in some relatives. The other six cases occurred in four unrelated melanoma families. The G101W mutation has been identified in a large number of melanoma families worldwide (also in Spain), and a common founder effect has been demonstrated by haplotype analysis (unpublished data).27

    Polymorphisms and Variants of Unknown Significance

    The previously well-characterized polymorphism 442 G>A (A148T) within exon 2 and the silent 500C>G, 513C>A, and 540C>T variants within exon 3 were also present in this set of patients (Table 2). The A148T transversion was identified in 13.5% of the MPM patients, and in one patient, an exon 2 CDKN2A (G101W) mutation detected in trans with the polymorphism. A148T was statistically more frequent in MPM patients (13.5%) than in controls (5.45%; P = .05). This polymorphism was in linkage disequilibrium with the 500C>G polymorphism in exon 3.

    A variation of unknown significance, A127S, was identified in one patient with two primary melanomas. This variant has been described as both a polymorphism and a possible mutation.28

    Haplotype Analyses

    G101W has been reported in several melanoma families from different countries: Italy, United States, Queensland in Australia, France, Israel, and Spain.13,27,29 It has been suggested that a common founder, rather than a mutational hotspot, is the likely explanation for this finding.27 This notion is reaffirmed by analysis of our MPM patients carrying the variant, who all share the previously reported common founder haplotype (data not shown).

    In addition, analysis of haplotype in two families with the recurrent R87W mutation also identified a common haplotype, in favor of its derivation from a common ancestor rather than a mutation hot spot.

    Likelihood of MPM Patients Carrying CDKN2A Mutations

    The risk to identify a CDKN2A mutation increased with the number of primary melanomas and the presence of familial history of melanoma (Table 4). CDKN2A mutations were more frequent in patients with familial history of melanoma (35.5%) compared with patients without (8.2%), with an RR of 4.32 (95% CI, 1.76 to 10.64; P = .001). Moreover, CDKN2A mutations were also more frequent in patients with more than two melanomas (39.1%) than in patients with only two melanomas (10%), with an RR of 3.29 (95% CI, 1.7 to 6.3; P = .002). The mean age of onset was significantly lower in carriers of mutations (32.88 years ± 12.61 years) compared with noncarriers (45.84 years ± 16.11 years; P = .003; Table 4). The presence of DN was similar between the two groups, with carriers no more likely to have DN than noncarriers (Table 4).

    Clinical Outcome of Melanoma in MPM Patients According to CDKN2A Status

    During a mean period of follow-up of 5 years, eight patients relapsed (Table 4). Two of them were carriers of CDKN2A mutations (358delG and 60ins16). One of the noncarriers is alive without disease after a locoregional relapse. No significant differences were detected between carriers and noncarriers, but this could be caused by the low number of patients who relapsed in each group. The mean Breslow thickness of the first melanoma in carriers was 1.65 mm (SD, ± 2.04) and in noncarriers, 1.054 mm (SD, ± 1.29), but again differences were statistically nonsignificant.

    DISCUSSION

    CDKN2A is the main locus associated with melanoma susceptibility. This locus encodes two different transcripts: p16INK4A and p14ARF, which share exons 2 and 3. However, their first exon is unique ( and ). While the role of p16INK4A mutant proteins in melanoma susceptibility seems to be clear, the involvement of p14ARF in melanomagenesis has been unclear until recently.

    The likelihood of finding a mutation in CDKN2A (p16INK4A/p14ARF) increases with the number of affected family members,30 the presence of MPM,14 and history of pancreatic cancer in the family.31 On the basis of these data, the aim of the present work was to elucidate the role of both proteins in melanoma susceptibility in a group of melanoma patients with a suspected high genetic predisposition. Thus, we analyze the results from clinical and genetic characterization of 104 MPM patients.

    Nearly 17% of the patients studied were carriers of germline mutations at the CDKN2A locus. In contrast, the prevalence of CDKN2A mutations in Spanish melanoma families is only 18%.13 These results suggest that it is just as important to screen MPM patients for CDKN2A mutations as familial melanoma cases. Our data are consistent with previously reported CDKN2A mutation prevalence in MPM in other studies. Mutation frequencies range from 8% to 12% in sporadic MPM patients, but they rise to 47.8% when MPM and positive-melanoma families are considered (Table 5).

    In 17 of 104 MPM patients, we detected seven different CDKN2A mutations. Five are located in exon 2 (one base pair [bp] deletion and four missense), one missense mutation in exon 1, and one exon 1 insertion (Tables 1 and 2). Only three of these mutations affect both p16INK4A and p14ARF (V59G, R87W, and G101W). In total, 13 (76.5%) of 17 mutant carriers were carriers of mutations that impair both proteins; in three patients (17.7%), the mutation affects only p16INK4A protein; and finally in one case (5.8%), the mutation affects only p14ARF. Thus, p16INK4A is affected in 94.2% of cases, which underlines the importance of this transcript in MPM susceptibility.

    However, one patient with no family history of melanoma harbored a germline mutation in exon 1, a 160-bp insertion, that resulted in a frameshift after codon 21 and a premature termination signal at codon 67. We have demonstrated that this p14ARF mutant protein is not functional.4 A variation of unknown significance, A127S, was also identified in one MPM patient. This variant has been described both as a polymorphism and a possible mutation. However the latter view is favored by Soufir et al28 who noted that it is located in the fourth ankyrin domain, and it has been reported as germline mutation in individuals who developed multiple types of cancer (including melanoma). Furthermore, in vitro experiments have shown that it has an intermediate in its ability to cause growth arrest in cultured cells. Thus, A127S is likely to play a role in susceptibility to MPM.

    The A148T polymorphism was commonly detected in the present study (in 13.4% of MPM patients studied). This proportion is significantly higher than that found in the 100 controls analyzed (5.45%). Because our controls were anonymous DNA samples from healthy individuals attending a workplace health program, the differences could be related to the origin of the samples and not only to melanoma susceptibility. Previous studies from the United Kingdom were not able to demonstrate differences between melanoma patients and controls in the proportion of A148T, but demonstrated a high variability in A148T prevalence among different populations.32 Population-based studies should be performed in order to clarify the role of this polymorphism in melanoma susceptibility.

    In contrast, CDK4 mutations were not identified in this series, suggesting its low impact on susceptibility to MPMs. This is consistent with other studies worldwide, in which activated CDK4 has been observed in less than 1% of familial melanoma patients.

    Our study suggests clinical differences between MPM patients who are carriers of CDKN2A mutations and those who are not. For instance, the age of onset is significantly lower in those with mutations (32.88 years ± 12.61 years) compared with noncarriers (45.84 years ± 16.11 years; P = .003). These results are consistent with the data reported from MPM patients in Sweden.16 Similarly, Stam-Posthuma et al33 reported the mean age at first melanoma diagnosis in multiple primary cases in Dutch population as 36 years in familial cases and 45.7 years in patients without familial melanoma. In our series, the age of onset is not statistically different in patients belonging to melanoma families and patients without familial melanoma history, regardless of mutation status.

    As expected, we found that CDKN2A mutations were more frequent in MPM patients with familial history of melanoma (35.5%) compared with patients without familial history (8.2%) and in patients with more than two melanomas (39.1%) compared with patients with only two melanomas (10%; Table 4). DN were also observed in most (80.6%) of our patients. This is probably more frequently than we would expect to observe in sporadic melanoma (reported range, 34% to 59%),34 suggesting that the DN phenotype may be a cofactor that contributes to the development of multiple melanomas (MMs). The presence of DN in almost all p16INK4A mutants with MPM should suggest that this protein could be partially involved in nevogenicity. DN were not observed in any of the carriers of the 60ins16 p14ARF mutation, which suggests that susceptibility to melanoma in p14ARF mutant patients may be independent of the development of nevi. Nevertheless, this proposal should be tested in more patients. Unfortunately, conclusions will be limited by the small number of MM kindreds worldwide that harbor p14ARF-specific mutations.

    Knowledge of the likelihood of detecting CDKN2A mutations has particular relevance for clinical screening of patients and their families. In addition, analysis of these geneotype/phenotype relationships will result in better knowledge of penetrance and clinical variability of the disease. In the present study, the molecular identification of CDKN2A mutation carriers and their clinical screening with epiluminiscence microscopy and digital follow-up quickly resulted in the detection of eight early melanomas in our clinic that would normally have been detected much later. This underlines the importance of mutational screening and regular and thorough dermatologic screening of relatives of individuals harboring germline CDKN2A mutations for earliest MM diagnosis. However, because we detected melanoma in a relative of a MPM patient not carrying a CDKN2A mutation, dermatologic screening, sun exposure avoidance, and self-examination are encouraged for relatives of MPM patients without mutation in CDKN2A. Mutational studies of families with MM detect 18% to 50% of mutation carriers in CDKN2A.2 For MPM, it represents 8% to 48% of cases (Table 5). These data suggest the existence of at least another susceptibility gene related to MM development. Thus, in families of MPM patients in which no mutation has been detected, the risk persists for all family members.

    Several authors reported the ability of total-body photography in the early detection of melanoma in high-risk patients.35 Epiluminiscence microscopy (also called dermoscopy) is a technique that increases the accuracy of melanoma diagnosis in approximately 10% to 27% respect to examination with the naked eye, improving the clinical decision making of dermatologists.36,37 Dermoscopy should be used in the management of melanoma patients and, in general, in the diagnosis of skin tumors.36,37,38 Also, digital dermoscopy follow-up has been reported to allow the early detection of melanomas that otherwise could be misdiagnosed.38,39,40,41 A substantial decrease in unnecessary excisions in the patients and an increase in early melanomas detected have been already demonstrated.38,39,40,41 Our group reported a protocol for the follow-up of high-risk patients that included total-body photography and digital dermoscopy.41 Further studies should be conducted to analyze the real impact of these strategies.

    In our series, prognosis of melanoma in terms of overall survival in carriers and noncarriers was not detected to be different. In a mean period of follow-up of 5 years, nearly 12% of MPM patients with CDKN2A mutations died with melanoma metastasis, whereas only 6.8% of noncarriers died with metastases, but differences were not statistically significant. Furthermore, studies including larger series of patients from different institutions are required to clarify this question.

    MPM patients are at least as likely to harbor mutations as familial melanoma patients. Melanoma patients with early age of onset, familial history, and the presence of more than 2 primary melanomas are the most likely to harbor CDKN2A mutations. Thus, they and their families must be intensively followed in the dermatology clinic.

    Mutations affecting p16INK4A, p14ARF, or both are detected in MPM patients, confirming that both proteins contribute to melanoma susceptibility. The specific follow-up of at-risk patients is important for the early detection of melanoma and increased knowledge of the penetrance and phenotypical characterization of this disorder.

    Authors' Disclosures of Potential Conflicts of Interest

    The authors indicated no potential conflicts of interest.

    Acknowledgment

    The work was performed within the Multidisciplinary Malignant Melanoma Unit, which comprised C. Badenas, A. Campo, T. Castel, C. Conill, R. Gallego, J. Malvehy, M. Milà, J. Palou, S. Puig, R. Rull, M. Sánchez, S. Vidal, A. Vilalta and R. Vilella. We are grateful to Graeme Walker for the review of the manuscript, to the technician Remedios Cervera for her collaboration, to the data manager Rosa Cuadrado, and to the dermatology residents Alex Llambrich and Mauricio Vera.

    NOTES

    Supported by grants 01/1546 and 03/0019 from Fondo de Investigaciones Sanitarias; V2003-REDC03/03 and /07; grant RO-1 CA 83115 (fund 538226 from the National Cancer Institute, Bethesda, MD), and a personal grant to Francisco Cuellar CONACYT, Personal Scholarship 152256/158706, Mexico City, Mexico.

    Authors' current affiliations are as follows: Molecular Haematology, Institute of Child Health, London UK; Dermatology Department, Clínica Platon, Barcelona; Dermatology Department, Hospital General de Catalunya, Sant Cugat; Dermatology Department, Hospital Arnau de Vilanova, Universitat de Lleida, Lleida, Spain.

    Initial results from analysis of some patients have been previously reported (Puig et al, 1997; Ruiz et al, 1999; Rizos et al, 2001; Yakobson et al, 2003).

    Authors' disclosures of potential conflicts of interest are found at the end of this article.

    REFERENCES

    MacKie RM, Andrew N, Lanyon WG, et al: CDKN2A germline mutations in UK patients with familial melanoma and multiple primary melanomas. J Invest Dermatol 111:269-272, 1998

    Piepkorn M: Melanoma genetics: An update with focus on the CDKN2A(p16)/ARF tumor suppressors. J Am Acad Dermatol 42:705-722, 2000

    Randerson-Moor JA, Harland M, Williams S, et al: Germline deletion of p14arf but not CDKN2A in a melanoma-neural system tumour syndrome family. Hum Mol Genet 10:55-62, 2001

    Rizos H, Puig S, Badenas C, et al: A melanoma-associated germline mutation in exon 1beta inactivates p14ARF. Oncogene 20:5543-5547, 2001

    Hewitt C, Lee W, Evans G, et al: Germline mutation of ARF in a melanoma kindred. Hum Mol Genet 11:1273-1279, 2002

    Soufir N, Avril MF, Chompret A, et al: Prevalence of p16 and CDK4 germline mutations in melanoma-prone families in France. Hum Mol Genet 7:209-216, 1998

    Hussussian CJ, Struewing JP, Goldstein AM, et al: Germline p16 mutations in familial melanoma. Nat Genet 8:15-21, 1994

    Holland EA, Beaton SC, Becker TM, et al: Analysis of the p16 gene, CDKN2, in 17 Australian melanoma kindreds. Oncogene 11:2289-2294, 1995

    Walker GJ, Hussussian CJ, Flores JF, et al: Mutations of the CDKN2/p16INK4 gene in australian melanoma kindreds. Hum Mol Genet 4:1845-1852, 1995

    Borg A, Johansson U, Hakansson S, et al: Novel germline p16 mutation in familial melanoma in Southern Sweden. Cancer Res 56:2497-2500, 1996

    Fitzgerald MG, Harkin DP, Silva-Arrieta S, et al: Prevalence of germline mutations in p16, p19ARF, and CDK4 in familial melanoma: Analysis of a clinic-based population. Proc Natl Acad Sci U S A 93:8541-8545, 1996

    Harland M, Meloni R, Gruis N, et al: Germline mutations of the CDKN2 gene in UK melanoma families. Hum Mol Genet 6:2061-2067, 1997

    Ruiz A, Puig S, Malvehy J, et al: CDKN2A mutations in Spanish cutaneous malignant melanoma families and patients with multiple melanomas and other neoplasia. J Med Genet 36:490-494, 1999

    Monzón J, Liu L, Brill H, et al: From L, McLaughlin J, Hogg D, Lassam NJ: Mutations in multiple primary melanomas. N Engl J Med 338:879-887, 1998

    Blackwood MA, Holmes R, Synnestvedt M, et al: Multiple primary melanoma revisited. Cancer 94:2248-2255, 2002

    Hashemi J, Platz A, Ueno T, et al: CDKN2A germline mutations in individuals with multiple cutaneous melanomas. Cancer Res 60:6864-6867, 2000

    Auroy S, Avril MF, Chompret A, et al: Sporadic multiple primary melanoma cases: CDKN2A germline mutations with a founder effect. Genes Chromosomes Cancer 32:195-202, 2001

    Mantelli M, Barile M, Ciotti P, et al: High prevalence of the G101W germline mutation in the CDKN2A (P16(ink4a)) gene in 62 Italian malignant melanoma families. Am J Med Genet 107:214-221, 2002

    Puig S, Ruiz A, Castel T, et al: Inherited susceptibility to several cancers but absence of linkage between dysplastic nevus syndrome and CDKN2A in a melanoma family with a mutation in the CDKN2A (p16INK4A) gene. Hum Genet 101:359-364, 1997

    Yakobson E, Eisenberg S, Isacson R, et al: A single Mediterranean, possibly Jewish, origin for the Val59Gly CDKN2A mutation in four melanoma-prone families. Eur J Hum Genet; 11:288-296, 2003

    Mao L, Merlo A, Gauri B, et al: A novel p16INK4A transcript. Cancer Res 55:2995-2997, 1995

    Liu L, Dilworth D, Gao L, et al: Mutation of the CDKN2A 5' UTR creates an aberrant initiation codon and predisposes to melanoma. Nat Genet 21:128-132, 1999

    Harland M, Mistry S, Bishop DT, et al: A deep intronic mutation in CDKN2A is associated with disease in a subset of melanoma pedigrees. Hum Mol Genet 10:2679-2686, 2001

    Badenas C, Torra R, Lucero L, et al: Mutational analysis within the 3' region of the PKD1 gene. Kidney Int 55:1225-1233, 1999

    den Dunnen JT, Antonarakis E: Mutation nomenclature extensions and suggestions to describe complex mutations: A discusssion. Hum Mutat 15:7-12, 2000

    Ruiz A, Puig S, Lynch M, et al: Retention of the CDKN2A locus and low frequency of point mutations in primary and metastasic cutaneous malignant melanoma. Int J Cancer 76:312-316, 1998

    Ciotti P, Struewing JP, Mantelli M, et al: A single genetic origin for the G101W CDKN2A mutation in 20 melanoma-prone families. Am J Hum Genet 67:311-319, 2000

    Soufir N, Ribojad M, Magnaldo T, et al: Germline and somatic mutations of the INK4a-ARF gene in a xeroderma pigmentosum group C patient. J Invest Dermatol 119:1355-1360, 2002

    Holland EA, Schmid H, Kefford RF, et al: CDKN2A (P16(INK4a)) and CDK4 mutation analysis in 131 Australian melanoma probands: Effect of family history and multiple primary melanomas. Genes Chromosomes Cancer 25:339-348, 1999

    Aitken J, Welch J, Duffy D, et al: CDKN2A variants in a population-based sample of Queensland families with melanoma. J Natl Cancer Inst 91:446-452, 1999

    Goldstein AM, Fraser MC, Struewing JP, et al: Increased risk of pancreatic cancer in melanoma-prone kindreds with p16INK4 mutations. N Engl J Med 333:970-974, 1995

    Bertram CG, Gaut RM, Barrett JH, et al: An assessment of the CDKN2A variant Ala148Thr as a nevus/melanoma susceptibility allele. J Invest Dermatol 119:961-965, 2002

    Stam-Posthuma JJ, van Duinen C, Scheffer E, et al: Multiple primary melanomas. J Am Acad Dermatol 44:22-27, 2001

    Naeyaert JM and Brochez L: Dysplastic Nevi. N Eng J Med 349:2233-2240, 2003

    Halpern AC: Total body skin imaging as an aid to melanoma detection. Semin Cutan Med Surg. 22:2-8, 2003

    Bafounta ML, Beauchet A, Aegerter P, et al: Is dermoscopy (epilumiscence microscopy) useful for the diagnosis of melanoma Results of a meta-analysis using techniques adapted to the evaluation of diagnostic tests. Arch Dermatol 137:1343-1350, 2001

    Carli P, de Giorgi V, Chiarugi A, et al: Addition of dermoscopy to conventional naked-eye examination in melanoma screening: A randomized study. J Am Acad Dermatol 50:683-689, 2004

    Haenssle HA, Vente C, Bertsch HP, et al: Results of a surveillance programme for patients at high risk of malignant melanoma using digital and conventional dermoscopy. Eur J Cancer Prev 13:133-138, 2004

    Kittler H, Perhamberger H, Wolff K, et al: Follow-up of melanocytic skin lesions with digital epiluminiscence microscopy: Patterns of modifications observed in early melanoma, atypical nevi, and common nevi. J Am Acad Dermatol 43:467-476, 2000

    Menzies SW, Gutenev A, Avramidis M, et al: Short-term digital surface microscopic monitoring of atypical or changing melanocytic lesions. Arch Dermatol. 137:1583-1589, 2001

    Malvehy J, Puig S: Follow-up of melanocytic skin lesions with digital total-body photography and digital dermoscopy: A two-step method. Clin Dermatol 20:297-304, 2002(Susana Puig, Josep Malveh)