Longitudinal Assessment of the Nevus Phenotype in a Melanoma Kindred

Transcription

1 Longitudinal Assessment of the Nevus Phenotype in a Melanoma Kindred See related Commentary on page iv Scott R. Florell,, w Laurence J. Meyer,, w Kenneth M. Boucher,z Patricia A. Porter-Gill, Marybeth Hart, Jennica Erickson, Lisa A. Cannon-Albright,y Lynn K. Pershing,, w Ronald M. Harris,, w Wolfram E. Samlowski,, z John J. Zone,, w and Sancy A. Leachman, w The Melanoma Program, Huntsman Cancer Institute; wdepartments of Dermatology; zoncological Sciences; ymedical Informatics; zinternal Medicine, Division of Oncology, University of Utah, Salt Lake City, Utah, United States of America Phenotypic characteristics of members of a melanoma prone kindred with a V126D CDKN2A gene mutation were monitored over approximately 15 y. Thirty-eight previously studied subjects were recruited. Participants underwent a complete skin examination by the same dermatologist who examined them initially. The size and location of all were recorded on a body map diagram. Total nevus number (TNN) and total nevus density (TND) were determined. CDKN2A sequencing verified 13 mutation carriers and 16 non-carriers. Nine participants were spouse controls without a history of melanoma and did not carry a CDKN2A mutation. Mutation carriers demonstrated a greater mean TNN and TND at initial and follow-up examinations compared with non-carriers and continued to develop rather than show nevus regression seen in non-carriers and spouse controls. Non-carriers showed an intermediate nevus phenotype between mutation carriers and spouse controls. Four of the 13 mutation carriers and one non-carrier have developed invasive melanoma. Over a 15-y interval, TNN and TND were increased in mutation carriers compared with non-carriers and spouse controls. Continued accumulation of in mutation carriers supports a nevogenic role for this CDKN2A mutation. An intermediate nevus phenotype in non-carrier family members suggests the presence of additional modifier genes. Key words: familial/p16/total nevus density/total nevus number J Invest Dermatol 123: , 2004 Both genetic and environmental factors influence the development of melanoma and cutaneous but the greatest risk factor for development of cutaneous melanoma is being a member of a melanoma-prone family (Goldstein and Tucker, 1995). Recognized risk factors include personal history of melanoma (Tucker et al, 1985), multiple and atypical (moles) (Newton et al, 1993; Grulich et al, 1996; Tucker et al, 1997), red hair, freckling, history of blistering sunburns, blue eyes (Marrett et al, 1992; Bliss et al, 1995), and Fitzpatrick type I skin (Marrett et al, 1992) (reviewed in Kefford et al, 1999). With the exception of a single long-term follow-up study of individual melanocytic lesions (Tucker et al, 2002), the natural history of nevus number and density in melanoma kindreds is unknown. In the late 1980s and early 1990s, we recruited members of melanoma prone families for phenotypic and genotypic studies that culminated in the identification of a melanomasusceptibility locus (Cannon-Albright et al, 1992), later identified as CDKN2A (p16) (Kamb et al, 1994), an integral member of the retinoblastoma cell cycle regulation pathway (Piepkorn, 2000). Confirmation of the importance of this tumor suppressor gene in familial melanoma followed and CDKN2A mutations have been discovered in 20 40% of melanoma kindreds (Piepkorn, 2000). Abbreviations: TND, total nevus density; TNN, total nevus number Individuals inheriting a CDKN2A mutation have an increased risk of melanoma (Cannon-Albright et al, 1994; Goldstein and Tucker, 1995) and an increased number of (Cannon-Albright et al, 1994; Zhu et al, 1999; Bishop et al, 2000). This suggests that CDKN2A, in addition to conferring melanoma susceptibility, may participate in nevus development. But correlation of the number of with the genotypic status is problematic because of considerable phenotypic overlap between CDKN2A mutation carriers and non-carriers (Bishop et al, 2000). One possible explanation for this overlap is the co-inheritance of additional genes involved in nevus development. This possibility is supported by the observation that increased numbers of seen in the absence of germline CDKN2A mutations are an independent risk factor for melanoma (Grulich et al, 1996; Kefford et al, 1999). Defining the natural history of nevus development in melanoma kindreds is an important step in understanding the relationship between CDKN2A mutations and other undiscovered modifier genes. Here we report longitudinal phenotypic observations of 38 previously examined Utah melanoma kindred members and spouse control subjects over an average interval of 15 y. This kindred was instrumental in establishing CDKN2A as a major melanoma susceptibility gene (Cannon-Albright et al, 1992) and carries a well-characterized CDKN2A gene mutation, V126D, that co-segregates with melanoma susceptibility (Kamb et al, 1994). The aim of this longitudinal Copyright r 2004 by The Society for Investigative Dermatology, Inc. 576

2 123 : 3 SEPTEMBER 2004 investigation was to evaluate the effect of a single point mutation in CDKN2A on the nevus phenotype and the development of melanoma. Results Twenty-nine kindred members and nine spouse control subjects were re-examined. The mean follow-up period was 15.8, 15.7, and 14.7 y for the mutation carrier, non-carrier, and spouse control subjects, respectively. Thirteen were heterozygous for the 377 T! A mutation and 25 were noncarriers, including nine spouse control subjects. None of the subjects examined had additional promoter or coding region mutations of CDKN2A. There were no mutations in the ARF coding region. The Utah Population Database, which links multi-generation pedigree records of members of the Church of Jesus Christ of Latter-day Saints to Utah Cancer Registry data, was queried to determine if any of the spouse control subjects were members of other known Utah melanoma kindreds or had a personal or family history of melanoma. No association was identified. Demographic data are summarized in Table S1. NEVUS PHENOTYPE IN A MELANOMA KINDRED 577 The goal of data collection was to evaluate the major risk factors for melanoma in the context of an individual s CDKN2A mutation status. The following risk factors were assessed: total nevus number (TNN) and total nevus density (TND), constitutional phenotypic characteristics (hair color, eye color, freckling, skin type, and natural skin pigmentation), environmental factors (including markers of chronic sun exposure and self-reported sunburn history) and personal history of melanoma (Table I). Mutation carriers have significantly more than noncarriers and continue to develop additional The mean and standard deviation of TNN and TND for all genotype groups are shown in Table S1. Mutation carriers demonstrated significantly larger TNN and TND at the first examination than the non-carrier subjects, confirming Figure 1 Mutation carriers developed more (A) and have a larger TND (B) that increased over the follow-up interval. Box-and-whiskers plot representing median (horizontal line within box), upper quartile, lower quartile, and range of values for TNN and TND by carrier group. I, initial examination, F/U, follow-up examination. Figure 2 Mutation carriers demonstrate a constant increase in TNN with age. Scatterplot of subject age at initial and follow-up examinations versus TNN. As can be seen, the mutation carriers (A) show an increase in TNN at a constant rate with age, whereas the non-carrier (B) and spouse control subjects (C) tend to decrease TNN with age (best fitting line second-order polynomial).

3 578 FLORELL ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY Figure 3 Mutation carriers show an increased number of in all three size categories. As can be seen, mutation carriers have increased numbers of small, intermediate, and larger over time, whereas non-carriers and spouse control subjects show no significant change over the follow-up interval. previous findings (Cannon-Albright et al, 1994). Notably, the differences between these genotypic groups were magnified over the follow-up interval, with mutation carriers continuing to develop at an average of per y, (95% CI ), whereas both parameters tended to decrease in non-carriers and spouse control subjects. The striking phenotypic differences over the follow-up interval among all genotypic groups are shown in Figs 1 and 2. Mutation carriers had significantly more clinically atypical (average 10 8) than non-carriers (average 2 2, p ¼ ) or spouse controls (1 2, p ¼ ). TNN was stratified into o4 mm, 4 5 mm, and 45 mm revealing that mutation carriers increased the number of in all three size categories over the follow-up interval. A significant difference in the size distribution of was not identified in non-carrier or spouse control subjects over time (Fig 3). At the initial examination, mutation carriers had a significantly larger number of 46 mm compared with non-carriers (p ¼ 0.03) or spouse control subjects (p ¼ 0.05). At follow-up, mutation carriers showed significantly more in all size groups as compared with non-carriers (pp0.03) or spouse control subjects (pp0.04). A significant difference was not identified between non-carriers or spouse controls in any nevus size category in the initial or follow-up examination. Figure 4 Subjects in all genotypic groups show variability in TNN and TND over the follow-up interval. Each point on these scatterplot diagrams represents the difference in TNN or TND (follow-up value initial value) for each participant over the follow-up period. As can be seen, there is significant variability in nevus development over time; subjects demonstrate either an increase, a decrease, or similar numbers of counts. These data show that the variability in TNN and TND increases from the spouse control to mutation carrier group, suggesting the possibility of co-inheritance of a nevogenic modifier gene in this kindred. Non-carriers show an intermediate nevus phenotype Spouse control subjects, on average, were older than non-carriers because they were married at the initial examination. After adjusting for the age difference, non-carriers appear to have an intermediate nevus phenotype between mutation carriers and spouse control subjects. Although non-carriers have a larger TNN and TND than spouse controls, the change in these parameters over the follow-up interval is similar (Figs 1 and 2). Nevus phenotype in CDKN2A family members is variable Nevus phenotype varied considerably within genotypic groups, as reflected by the standard deviation for TNN and TND. The standard deviation was greatest in the mutation carrier group, followed by the non-carrier group, and was least in the spouse control group (Table S1 and Fig 4). Furthermore, the mutation carrier and non-carrier groups can be subdivided into those individuals that increased the TNN/TND, those that did not change, and those that decreased these parameters over the follow-up interval. These data support the possible existence of a modifying nevogenic gene in this kindred that could account for the

4 123 : 3 SEPTEMBER 2004 NEVUS PHENOTYPE IN A MELANOMA KINDRED 579 Table I. Phenotypic data among participants by genotype status Mutation Carrier Status V126D Mutation Carriers Average (SD) V126D Mutation Non- Carriers Average (SD) Spouse Controls Average (SD) Subject Personal history of melanoma # o4 mm Initial exam Follow-up exam 4 5 mm 45 mm TND Age TNN o4 mm 4 5 mm 45 mm TND Clinically Atypical Sunburn History c Reflectance Spectroscopy c Hair Childhood 0 10 y y 420 y Eye Skin Age TNN Nevi a Color b Freckling c of Age of Age of Age Color d Type c Red None Green/Hazel Dark Few Green/Hazel Dark Many Blue Dark Few Green/Hazel Blonde Few Blue Dark Many Blue Dark Many Green/Hazel Dark Many Green/Hazel Dark Few Green/Hazel BCC Red ND ND ND ND Blue Red None Green/Hazel SCC Red None Blue SCC Dark Many Green/Hazel (33) (190) (56) (280) 10 (8) Red Few Dark Dark Few Green/Hazel Red Few Dark Red None Green/Hazel Red Few Dark Blonde None Blue 1 NA 2 2 BCC Dark Few Blue Dark Few Green/Hazel Dark Many Blue Red None Green/Hazel Dark None Blue Dark None Dark Dark Many Dark Dark Few Green/Hazel Dark Few Dark Red None Green/Hazel (30) (125) (26) (105) 2 (2) Dark Few Green/Hazel Blonde Few Blue Dark Many Blue Blonde Few Green/Hazel Dark None Blue BCC Blonde Few Blue Dark Few Blue Dark Few Blue Blonde Many Blue (12) (50) (10) (47) 1 (2) 2.4 (0.5) 2.2 (0.6) 0 ¼ pt unsure 2.1 (0.9) 1 ¼ zero 2 ¼ 1to3 3 ¼ 4to10 4 ¼ (3.0) 19.6 (3.8) 20.0 (1.7) Overall Assessment of Sun Damage c 1.3 (0.5) 1.7 (0.5) 1.7 (0.8) Number of Other Signs of Chronic Sun Damage c 1.2 (1.2) 1.1 (1.0) 1.6 (1.1) Non- Melanoma Skin Carcinoma a A clinically atypical nevus was defined as having one or more of the following clinical features: border irregularity, color asymmetry, presence of multiple colors, diffusion of pigment, and irregular contour. Number of atypical available for follow-up examination only. b Gene carrier versus non-carrier p ¼ 0.93; non-carrier versus spouse control p ¼ 0.02; gene carrier versus spouse control p ¼ 0.05 (Pearson s w 2 ). c Statistically significant differences were not identified between genotypic groups. d Gene carrier versus non-carrier p ¼ 0.05 (Pearson s w 2 ); non-carrier versus spouse control p ¼ 0.02 (Pearson s w 2 ); gene carrier versus spouse control p ¼ 0.10 (Fisher s exact test). BCC, basal cell carcinoma; SCC, squamous cell carcinoma.

5 580 FLORELL ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY intermediate nevus phenotype in non-carrier subjects and accentuate further the difference in nevus development in carrier subjects. Constitutional phenotypic characteristics and environmental factors Other investigators have demonstrated both genetic and environmental components influencing the development of and melanoma in CDKN2A melanomaprone families (Meyer et al, 1992; Cannon-Albright et al, 1994; Goldstein et al, 1998; Bishop et al, 2000, 2002). Because nevus development has been associated with constitutional phenotype (light hair, eye color, Fitzpatrick skin type) (Green et al, 1995) as well as environmental factors, including acute (Richard et al, 1993; Bataille et al, 1998) and chronic sun exposure (Richard et al, 1993; Wachsmuth et al, 2001), we sought to determine if differences in these parameters could partially explain the dramatic phenotypic differences observed among the genotypic groups. Presence of red hair or light-colored eyes was not significantly associated with increased number of at the initial or follow-up examination (red hair: p ¼ 0.17 for TNN at first visit; p ¼ 0.52 for TNN at follow-up; light eyes: p ¼ 0.42 at first visit; p ¼ 0.49 at follow-up, Mann Whitney U test). No significant differences in skin type, natural skin pigmentation measured by reflectance spectroscopy, self-reported sunburn history, the degree of chronic sun exposure, or the number of markers of chronic sun exposure were identified between carriers, non-carriers, and spouse control subjects. Development of melanoma in the study population Four of the 13 mutation carriers developed invasive melanoma, two of whom were diagnosed during the follow-up interval. Mutation carriers were stratified into subjects with and without melanoma. Three of four mutation carriers with melanoma had red hair (p ¼ 0.11, Mann Whitney U test). Differences in other constitutional or environmental factors or skin pigmentation were not identified. Subjects with melanoma showed a larger mean TNN and TND (at initial examination and at follow-up) but the differences did not achieve statistical significance. Three of the four subjects with melanoma had more than 80 (range 83 97) at the initial examination. At follow-up, the TNN in these three individuals was more than 100 (range ). In contrast, the fourth subject had a non-carrier-like nevus phenotype with only one nevus at the initial examination and five at follow-up. Differences in the nevus size distribution (p3 mm, 4 5 mm, and X6 mm) were not identified among mutation carriers with and without melanoma, at either the initial (px0.15) or follow-up (px0.33) examinations. The mean age of the four male subjects with melanoma was 31 y at the time of diagnosis, as compared with a mean of 58 y in Caucasian men in Utah (Utah Cancer Registry data, personal communication, Charles Wiggins, PhD, May 2003). Two of the four participants with melanoma developed multiple primary melanomas, and one of those also developed pancreatic adenocarcinoma (Parker et al, 2003). The subject with the largest TNN and TND developed six invasive melanomas, five melanoma in situ lesions, and multiple atypical spindle cell and dysplastic. In sum, the mutation carrier group developed 11 melanomas and four of these lesions were diagnosed during the follow-up interval. One non-carrier subject was discovered to have an invasive melanoma at age 53 y, during the follow-up interval. This individual had a mutation carrier-like phenotype with 75 and the largest TND of the non-carriers at follow-up. Discussion A major risk factor for melanoma, regardless of the CDKN2A mutational status, is the number of present on the skin (Swerdlow et al, 1986; Bataille et al, 1996, 1998; Kefford and Mann, 2003). Whereas much is known about the natural history of during the first two decades of life, only a few reports of nevus history in healthy adult populations are available and suggest that: (1) are rare at birth and begin to develop in early childhood (MacKie et al, 1998; Harrison et al, 2000); (2) peak in number at approximately during the second through fourth decades of life (Nicholls, 1973; Cooke et al, 1985; MacKie et al, 1985); and (3) decline substantially to very few by the eighth decade (Nicholls, 1973; MacKie et al, 1985). We have thoroughly characterized a large family with a pathogenic V126D CDKN2A mutation. The V126D CDKN2A mutation present in this particular melanoma kindred has a profound effect on nevus development that is exaggerated over a mean follow-up period of 15 y. Individuals carrying this mutation, on average, have increased numbers of, whereas non-carrier and spouse control subjects have decreased numbers of over time in a manner similar to previous population-based studies (Nicholls, 1973; Cooke et al, 1985; MacKie et al, 1985). Moreover, mutation carriers had increased numbers of in all size categories over time, suggesting that these subjects continue to develop new and that existing increase in size with time. Unfortunately, the previous body map diagrams were not sufficiently detailed to allow direct comparison of individual at the follow-up examination. Cellular and molecular evidence is accumulating to suggest that CDKN2A is a prominent regulator of melanocyte senescence (Bennett, 2003). It is possible that regression of in non-carriers and spouse controls (and lack of regression in carriers with diminished CDKN2A function) represents a clinical correlate to the cellular senescence function of CDKN2A in melanocytes. The identification of an intermediate phenotype in noncarrier kindred members compared with spouse control subjects suggests the presence of other nevogenic or modifier genes that elevate the number of in noncarriers and may increase the penetrance of and melanomas in mutation carriers. In addition, the degree of variability in the TNN and TND is striking and shows considerable overlap among the mutation carriers and noncarriers, also suggesting the presence of a nevogenic modifier gene(s). Although mutation carrier subjects had significantly more atypical than the other groups, the clinical phenotype did not discriminate mutation carriers from non-carriers, a finding corroborated by others (Bishop et al, 2000) and illustrated by two participants with outlier nevus phenotypes who developed melanoma (one each in the mutation carrier and non-carrier groups), Thus, among kindreds that carry CDKN2A mutations, the most reliable

6 123 : 3 SEPTEMBER 2004 method of identifying individuals at highest risk is DNA sequence analysis of CDKN2A. Kindred subjects and spouse controls in our study are all of northern European descent; thus, most phenotypic characteristics among the mutation carriers, non-carriers and spouse control subjects were similar because they share a common genetic background. We were unable to identify significant differences among the three groups for any of the well-established melanoma risk factors. Therefore, the differences in nevus phenotype among the genotypic groups could not be attributed to known environmental or constitutional factors. It was notable, however, that three of the four mutation carriers that developed melanoma had red hair and all of the mutation carriers had light-colored eyes. Both red hair and light eye color have previously been associated with increased risk for cutaneous melanoma (Marrett et al, 1992; Bliss et al, 1995) and melanocortin 1 receptor (MC1R) variants (red hair phenotype) are associated with increased melanoma risk in CDKN2A kindreds (Box et al, 2001). Our data support the notion that coinheritance of additional established risk factors (i.e., red hair, light eyes, etc.) superimposes further risk upon the CDKN2A mutation status. Our study represents a longitudinal evaluation of a wellcharacterized melanoma kindred carrying a CDKN2A mutation with respect to the natural history of nevus development. The somewhat low rate of recruitment in our study raises the possibility of self-selection bias for those individuals with large numbers of or a history of melanoma. Although we are unable to exclude self-selection bias or assess its impact on our data, the effect is likely to be limited because the kindred participants were unaware of their CDKN2A mutation status. Furthermore, the rate of recruitment for kindred members and spouse controls was similar (45% for kindred members and 56% of spouse control subjects), despite significant differences in number of. We have confirmed previously reported findings that mutation carriers develop more than non-carriers or spouse controls (Cannon-Albright et al, 1994). Furthermore, we have shown that, over time, mutation carriers continue to develop, unlike the non-carrier or spouse control subjects. Further study is required to determine if non-carrier subjects are at increased risk of developing melanoma, a likely possibility given the evidence of an intermediate nevus phenotype in these subjects and data from the Melanoma Genetics Consortium suggesting a higher incidence of melanoma in noncarrier members of high-risk pedigrees compared with the general population (Kefford and Mann, 2003). These data provide additional support that CDKN2A is important not only in the development of melanoma, but also in growth regulation of benign melanocytic lesions. In addition, the existence of other nevogenic or modifying genes is likely. Materials and Methods Recruitment of subjects This study was approved by the Institutional Review Board at the University of Utah. All clinical investigation was conducted according to Declaration of Helsinki principles. Sixty-four members of this melanoma kindred and 16 spouse control subjects were contacted by letter inviting them to participate in a follow-up study. Phenotypic data regarding these individuals were NEVUS PHENOTYPE IN A MELANOMA KINDRED 581 not part of the recruitment strategy and were not available to recruitment personnel. Informed consent was obtained from kindred members agreeing to participate. Twenty-nine kindred members and nine spouse control subjects were recruited for this study. Subjects were not aware of their CDKN2A mutation status. Phenotypic data collection Phenotypic data were obtained from the following sources: (1) family history and epidemiologic questionnaires; (2) comprehensive total-body skin examination; and (3) reflectance spectrophotometry. History and epidemiology questionnaires All but one of the participants completed family history and epidemiology questionnaires. Participants self-reported natural hair color, natural eye color, childhood freckling on a scale of 1 3 (1 ¼ none, 2 ¼ few, and 3 ¼ many), and the approximate number of sunburns experienced during the first, second, and third decades of life (0 ¼ subject unsure, 1 ¼ zero, 2 ¼ 1 3, 3 ¼ 4 10, and 4 ¼ more than 10 sunburns). Reported melanomas were confirmed through the Utah Cancer Registry, by obtaining the pathology report, and by review of histologic slides by dermatopathologists (SRF and RMH). Total body skin examination All participants were examined by the same dermatologist (LJM) in the late 1980s and in this study. The dermatologist remained blinded to subject genotype status and had not examined or followed any subject in clinic in the intervening time period between the initial and follow-up examinations. Both examinations included a comprehensive total body skin examination in which the location and size of each clinically detectable nevus X2 mm in diameter was recorded on a body map diagram by the same protocol utilized in the previous study (Meyer et al, 1992). The number of clinically atypical (defined as having one or more atypical clinical features including border irregularity, color asymmetry, presence of multiple colors, diffusion of pigment, or irregular contour) was recorded (for the follow-up examination only). In addition, a subjective assessment of overall and body site-specific chronic sun exposure was made (scale 0 3; 0 ¼ not detectable, 3 ¼ severe) based on the presence of actinic keratoses, solar elastosis, poikiloderma, and solar lentigines. The Fitzpatrick skin type was also assigned by the examiner. Reflectance spectrophotometry An objective measure of natural skin pigmentation was determined by reflectance spectrophotometry (Green and Martin, 1990). Skin reflectance between wavelengths 450 and 850 nm was measured with a fiberoptic computerbased visible light spectrophotometer (S2000 Miniature Fiber Optic Spectrometer, Ocean Optics, Dundedin, Florida) at the inner upper arm, a sun-protected location. The calculated area under the curve between 450 and 615 nm was recorded for each subject. CDKN2A mutational status To verify genotype, DNA was extracted from peripheral blood leukocytes using a commercial kit (Gentra Systems, Minneapolis, Minnesota) followed by PCR-based DNA sequencing. The promoter region and exons 1a 3 of CDKN2A and exon 1b (ARF) were sequenced on an automated sequencer (University of Utah Sequencing Core Facility and Yale Diagnostic Laboratory, New Haven, Connecticut) to exclude the possibility of other CDKN2A or ARF mutations. Statistical analysis The TNN was defined as the number of clinically detectable X2 mm in diameter. Lesions o2 mm in diameter were excluded to reduce the possibility of erroneously including freckles in the TNN (Nicholls, 1973). TND was estimated by calculating the area of all using the formula Spr 2 divided by the estimated body surface area of the individual, as described previously (Goldgar et al, 1991; Meyer et al, 1992). Statistical analysis was performed using Statistica software (Releases 5.1 and 6.0, StatSoft, Inc). Comparison of medians was done using the exact Mann Whitney U test. Multiple regression analyses were performed to adjust for age and sex, with TNN, change in number of, TND, or change in TND as the single dependent variable, and age at initial visit, sex, and carrier status

7 582 FLORELL ET AL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY as dependent variables. Sex and carrier status were incorporated through the use of indicator variables. To compare the rates of nevus formation during two time periods, two ratios were computed: R 1 ¼ (# at initial visit)/(initial age), and R 2 ¼ (# at follow-up # at initial visit)/(age at follow-up age at initial visit). R 1 and R 2 were compared using the exact Wilcoxon matched-pair test. For change in TNN, the squared deviation from the genotypic group mean was computed, and the variability among the genotypic groups was evaluated by applying the Kruskal Wallis ANOVA. All reported p-values are two-tailed. This work was supported by grants from The Skin Cancer Foundation (SRF), the Dermatology Foundation Leaders Society Dermatologist Investigator Research Fellowship and Clinical Career Development Award (SRF), National Institutes of Health grant K23 RR (SRF), Doris Duke Charitable Foundation (SAL), Fellowship-To-Faculty Transition Award from the University of Utah funded in part by the Howard Hughes Medical Institute (SAL), The Huntsman Cancer Foundation (SAL), the Tom C. Mathews Jr Familial Melanoma Research Clinic at Huntsman Cancer Institute, the Huntsman General Clinical Research Center Public Health Service grant (MO1 RR00064), National Cancer Institute (NCI) Cancer Center Support Grant 5P30CA420-14, and the Utah Cancer Registry, funded by Contract #NCI-CN from the NCI with additional support from the Utah Department of Health and the University of Utah. Support for all datasets within the Utah Population Database was provided by Huntsman Cancer Institute. We thank Harold M. Erickson, PhD, and Jason Brinton, BS, for technical assistance, and Charles Wiggins, PhD, for providing Utah Cancer Registry SEER data. We gratefully acknowledge the willing participation of all of the family members in this study. Supplementary Material The following material is available from com / products / journals / suppmat / JID / JID23312 / JID23312sm.htm Table S1. Age, follow-up period, and nevus phenotype characteristics for Utah melanoma kindred 1 follow-up participants. DOI: /j X x Manuscript received March 6, 2004; revised April 7, 2004; accepted for publication May 5, 2004 Address correspondence to: Sancy A. Leachman, MD, PhD, Huntsman Cancer Institute, 2000 Circle of Hope, University of Utah Health Sciences Center, Salt Lake City, UT References Bataille V, Bishop JA, Sasieni P, Swerdlow AJ, Pinney E, Griffiths K, Cuzick J: Risk of cutaneous melanoma in relation to the numbers, types and sites of naevi: A case control study. Br J Cancer 73: , 1996 Bataille V, Grulich A, Sasieni P, et al: The association between naevi and melanoma in populations with different levels of sun exposure: A joint case-control study of melanoma in the UK and Australia. Br J Cancer 77: , 1998 Bennett DC: Human melanocyte senescence and melanoma susceptibility genes. Oncogene 22: , 2003 Bishop DT, Demenais F, Goldstein AM, et al: Geographical variation in the penetrance of CDKN2A mutations for melanoma. J Natl Cancer Inst 94: , 2002 Bishop JA, Wachsmuth RC, Harland M, et al: Genotype/phenotype and penetrance studies in melanoma families with germline CDKN2A mutations. J Invest Dermatol 114:28 33, 2000 Bliss JM, Ford D, Swerdlow AJ, et al: Risk of cutaneous melanoma associated with pigmentation characteristics and freckling: Systematic overview of 10 case control studies. The International Melanoma Analysis Group (IMAGE). Int J Cancer 62: , 1995 Box NF, Duffy DL, Chen W, et al: MC1R genotype modifies risk of melanoma in families segregating CDKN2A mutations. Am J Hum Genet 69: , 2001 Cannon-Albright LA, Goldgar DE, Meyer LJ, et al: Assignment of a locus for familial melanoma, MLM, to chromosome 9p13 p22. Science 258: , 1992 Cannon-Albright LA, Meyer LJ, Goldgar DE, et al: Penetrance and expressivity of the chromosome 9p melanoma susceptibility locus (MLM). Cancer Res 54: , 1994 Cooke KR, Spears GF, Skegg DC: Frequency of moles in a defined population. J Epidemiol Community Health 39:48 52, 1985 Goldgar DE, Cannon-Albright LA, Meyer LJ, Piepkorn MW, Zone JJ, Skolnick MH: Inheritance of nevus number and size in melanoma and dysplastic nevus syndrome kindreds. J Natl Cancer Inst 83: , 1991 Goldstein AM, Falk RT, Fraser MC, Dracopoli NC, Sikorski RS, Clark WH Jr, Tucker MA: Sun-related risk factors in melanoma-prone families with CDKN2A mutations. J Natl Cancer Inst 90: , 1998 Goldstein AM, Tucker MA: Genetic epidemiology of familial melanoma. Dermatol Clin 13: , 1995 Green A, Martin NG: Measurement and perception of skin colour in a skin cancer survey. Br J Dermatol 123:77 84, 1990 Green A, Siskind V, Green L: The incidence of melanocytic naevi in adolescent children in Queensland, Australia. Melanoma Res 5: , 1995 Grulich AE, Bataille V, Swerdlow AJ, Newton-Bishop JA, Cuzick J, Hersey P, McCarthy WH: Naevi and pigmentary characteristics as risk factors for melanoma in a high-risk population: A case control study in New South Wales, Australia. Int J Cancer 67: , 1996 Harrison SL, MacKie RM, MacLennan R: Development of melanocytic in the first three years of life. J Natl Cancer Inst 92: , 2000 Kamb A, Shattuck-Eidens D, Eeles R, et al: Analysis of the p16 gene (CDKN2) as a candidate for the chromosome 9p melanoma susceptibility locus. Nat Genet 8:23 26, 1994 Kefford RF, Mann GJ: Is there a role for genetic testing in patients with melanoma? Curr Opin Oncol 15: , 2003 Kefford RF, Newton Bishop JA, Bergman W, Tucker MA: Counseling and DNA testing for individuals perceived to be genetically predisposed to melanoma: A consensus statement of the Melanoma Genetics Consortium. J Clin Oncol 17: , 1999 MacKie RM, Andrew N, Lanyon WG, Connor JM: CDKN2A germline mutations in UK patients with familial melanoma and multiple primary melanomas. J Invest Dermatol 111: , 1998 MacKie RM, English J, Aitchison TC, Fitzsimons CP, Wilson P: The number and distribution of benign pigmented moles (melanocytic naevi) in a healthy British population. Br J Dermatol 113: , 1985 Marrett LD, King WD, Walter SD, From L: Use of host factors to identify people at high risk for cutaneous malignant melanoma. Can Med Assoc J 147: , 1992 Meyer LJ, Goldgar DE, Cannon-Albright LA, Piepkorn MW, Zone JJ, Risman MB, Skolnick MH: Number, size, and histopathology of in Utah kindreds. Cytogenet Cell Genet 59: , 1992 Newton JA, Bataille V, Griffiths K, et al: How common is the atypical mole syndrome phenotype in apparently sporadic melanoma? J Am Acad Dermatol 29: , 1993 Nicholls EM: Development and elimination of pigmented moles, and the anatomical distribution of primary malignant melanoma. Cancer 32: , 1973 Parker JF, Florell SR, Alexander A, DeSario JA, Shami PJ, Leachman SA: Pancreatic carcinoma surveillance in familial melanoma patients. Arch Dermatol 139: , 2003 Piepkorn M: Melanoma genetics: An update with focus on the CDKN2A(p16)/ARF tumor suppressors. J Am Acad Dermatol 42: , 2000 Richard MA, Grob JJ, Gouvernet J, Culat J, Normand P, Zarour H, Bonerandi JJ: Role of sun exposure on nevus: First study in age-sex phenotype-controlled populations. Arch Dermatol 129: , 1993 Swerdlow AJ, English J, MacKie RM, O Doherty CJ, Hunter JA, Clark J, Hole DJ: Benign melanocytic naevi as a risk factor for malignant melanoma. Br Med J (Clin Res Ed) 292: , 1986 Tucker MA, Boice JD Jr, Hoffman DA: Second cancer following cutaneous melanoma and cancers of the brain, thyroid, connective tissue, bone, and eye in Connecticut, Natl Cancer Inst Monogr 68: , 1985 Tucker MA, Fraser MC, Goldstein AM, et al: A natural history of melanomas and dysplastic : An atlas of lesions in melanoma-prone families. Cancer 94: , 2002 Tucker MA, Halpern A, Holly EA, et al: Clinically recognized dysplastic. A central risk factor for cutaneous melanoma. J Am Med Assoc 277: , 1997 Wachsmuth RC, Gaut RM, Barrett JH, et al: Heritability and gene-environment interactions for melanocytic nevus density examined in a UK adolescent twin study. J Invest Dermatol 117: , 2001 Zhu G, Duffy DL, Eldridge A, et al: A major quantitative-trait locus for mole density is linked to the familial melanoma gene CDKN2A: A maximum-likelihood combined linkage and association analysis in twins and their sibs. Am J Hum Genet 65: , 1999