Multiple Primary Melanomas in a CDKN2A Mutation Carrier Exposed to Ionizing Radiation FREE

Risk factors for melanoma include environmental insults (eg, blistering sunburns) and genetic factors underlying hair and eye color, skin type, many or atypical moles, and family history. Being a member of a melanoma-prone family is the most significant risk factor, increasing one's likelihood of developing a melanoma to 35 to 70 times that of the general population.¹ The gene for cyclin-dependent kinase inhibitor 2A (CDKN2A or p16) is the major melanoma predisposition gene identified at present. p16 is a key regulator of G₁/S cell cycle entry by controlling phosphorylation of retinoblastoma via inhibition of cyclin-dependent kinases 4 and 6 (CDK4 and CDK6).² Variants in the gene for the melanocortin 1 receptor (MC1R) alter the production of pigment, shifting the equilibrium from eumelanin to pheomelanin. MC1R is a low-penetrance melanoma susceptibility gene; heterozygotes for the R151C allele have a 2-fold increased risk, and homozygotes such as the patient described herein carry an approximate 4-fold risk to develop melanoma compared with the general population.³ Carriage of MC1R variants by those with the CDKN2A mutation confers an additional risk for melanoma development.^{4 - 7}

The relationship between environmental and genetic factors is illustrated by international investigations of CDKN2A mutation carriers. The GenoMel, an international Melanoma Genetics Consortium (http://www.genomel.org/), reported that the lifetime prevalence of melanoma among individuals with CDKN2A mutations varied considerably by geographic locale.⁸ The lifetime incidence of melanoma was 58% in Europe, 76% in the United States, and 91% in Australia among CDKN2A carriers of similar ethnic backgrounds. Bishop et al⁸ suggested that this difference was best explained by the difference in UV exposure and UV sensitivity among the different populations. The CDKN2A mutation carrier population clearly represents the group most likely to develop melanoma, and through careful history and observation, researchers may be able to deduce other melanoma-promoting events that may not be obvious in the general population.

Ionizing radiation is a known mutagen and carcinogen. Long- and short-term exposures have been associated with leukemia and thyroid, breast, and lung cancers.^{9 - 10} Studies that tracked the rates of cancer in the atomic bomb survivors of Nagasaki and Hiroshima reported an increased rate of basal cell and squamous cell carcinomas, but not melanoma, among this exposed Japanese population. In 2005, Fink and Bates¹¹ published a meta-analysis of melanoma in 7 cohorts of individuals with short- or long-term exposure to ionizing radiation. Most of these groups included large white populations, the race most susceptible to melanoma development, and not represented by the work on Nagasaki and Hiroshima survivors. The authors reported a statistically significantly increased relative risk of melanoma in 5 of their 7 cohorts, including a cohort of 3702 subjects involved in the 1957 above-ground nuclear testing.¹¹

A large, retrospective analysis of individuals treated with ionizing radiation for hereditary retinoblastoma demonstrated a significantly elevated rate of melanomas in the treatment field.¹² This would suggest that in genetically predisposed individuals, exposure to ionizing radiation can promote melanoma through the retinoblastoma pathway. The CDKN2A protein product, p16, regulates cell growth through the retinoblastoma pathway,¹³ but the effects of ionizing radiation on CDKN2A mutation carriers have yet to be studied. In a 2005 study of multiple primary melanomas (MPM) and CDKN2A mutations, Puig et al¹⁴ noted that one of their patients with MPM had a CDKN2A mutation and a history of long-term exposure to ionizing radiation. We herein describe a CDKN2A mutation carrier with 7 primary melanomas on the same side of her body and a unique history of exposure to ionizing radiation.

The patient is a white woman who underwent investigation for a history of MPM. Medical records and blood and tissue samples were collected for investigation.

Her medical history was remarkable for the development of a melanoma on her right ear at 28 years of age, treated by excision and a radical neck lymph node dissection (Figure 1). She subsequently developed 6 additional primary melanomas. Of the 7 total melanomas, 6 were found on the right side of her body and 1 was in the midline (Figure 2). A squamous cell carcinoma of her right upper lip was diagnosed in 1997. Other pertinent medical history includes a 46-year history of smoking, coronary artery disease with placement of 3 coronary artery stents, and polycythemia vera diagnosed at 63 years of age.

Her environmental exposure history was significant for 1 blistering sunburn before 20 years of age, with minimal sunbathing and no subsequent sunburns. She also reports 1 episode of intense exposure to ionizing radiation at 27 years of age that occurred in October 1958 in southern Utah and Nevada. The patient and her husband left the car in which they were traveling to observe an atmospheric nuclear detonation from an undetermined distance for 20 to 30 minutes. After leaving the blast site, the patient sat in the passenger side of the car, the side oriented toward the test site, riding with her arm outside the car. They noted a fine, sticky dust on the top of the car when they arrived home, consistent with nuclear fallout. The patient denies the development of nausea or other acute symptoms the day of the exposure.

Her family history includes a cousin with melanoma and melanoma in the children of another cousin. Multiple other cancer types were also described in her extended family.

Pertinent findings included phototype I skin, green eyes, red hair, and mild freckling. On the right side of her body, the patient had 14 moles, 1 of which was atypical. No nevi were larger than 5 mm. She also had 12 well-healed biopsy and treatment scars. On the left side of her body, the patient had 12 moles, 1 larger than 5 mm and 1 with atypical features, and 5 well-healed biopsy scars.

Results of routine laboratory tests, including complete blood cell count, liver function tests, and a lipid panel, were unremarkable other than an elevated hematocrit consistent with polycythemia vera. Sequencing of the CKDN2A gene and promoter region revealed a G−34T 5′ UTR mutation known to be deleterious.¹⁵ No additional mutations were found within the coding region. The alternate reading frame gene was also sequenced and had no mutations. Sequencing of MC1R demonstrated that the patient was homozygous for the R151C mutation. Chromosomal studies were performed on melanocytes, peripheral blood, and fibroblasts. Tissue specimens were collected from both sides of the body. Metaphase cells analyzed from cultures of all these tissues revealed a normal female karyotype. Fluorescence in situ hybridization of all specimen types was performed, and no chromosomal abnormalities were found. Direct sequencing of MC1R was performed as previously described.^{4 ,16 - 17}

This is a unique case report of ionizing radiation exposure in an individual with a deleterious CDKN2A mutation and MPM. The patient had 7 primary melanomas during a course of about 30 years, starting late in her third decade of life. Her laboratory studies are significant for a deleterious CDKN2A mutation, homozygosity for an MC1R red hair color variant, and fluorescence in situ hybridization results that were negative for chromosomal aberrancy on either side of her body. Her physical examination revealed skin phototype I, red hair, numerous biopsy and excisional scars, and a random distribution of 26 nevi, 2 of which had atypical features. Most of her melanomas occurred in traditionally sun-protected areas. Her family history is consistent with hereditary melanoma.¹

Many groups have investigated the correlation between MPM and CDKN2A mutations (Table).^{14 ,18 - 24} Collectively, these groups have described 467 individuals with MPM, 70 of them (15.0%) with CDKN2A mutations. Not all groups specified the number of primary melanomas for each individual. Of the studies that did specify the number of primary melanomas,^{14 ,18 - 19 ,23 - 24} 5 of 311 cases (1.6%) had at least 7 primary melanomas. Even among individuals with CDKN2A mutations, it is rare to have 7 primary melanomas. Puig et al¹⁴ reported that among their 104 patients with MPM, only 3 had a second primary melanoma appear in proximity (defined as on the same extremity as the first melanoma).¹⁴ Our patient was initially seen with a grouping of 4 independent melanomas on the right side, 1 on the inner left thigh, and 3 melanomas close to one another on the right ear, right shoulder, and upper back area (Figure 2).

Her exposure history is significant for 1 blistering sunburn and her report of observing an above-ground nuclear test while traveling in Utah and Nevada. The time of the exposure she describes is chronologically consistent with Operation Hardtack II (September 12 to October 30, 1958), during which atmospheric nuclear testing took place in southern Nevada.^{25 - 26}

Previous studies have linked exposure to ionizing radiation and nonmelanoma skin carcinomas, including leukemia and thyroid, breast, and lung cancers.^{9 ,27 - 28} There are also several brief reports that have described an excess rate of polycythemia rubra vera in subjects exposed to the fallout of nuclear tests.^{29 - 30} Ionizing radiation has not been historically considered a promoting agent for melanoma, because many of the original studies that reviewed the cancer in atomic blast survivors in Japan did not identify melanoma as an outcome of interest.

The studies reviewed by Fink and Bates¹¹ collectively totaled hundreds of thousands of individuals who were predominantly white, the population most susceptible to melanoma. In addition to the rates of melanoma, the authors reported which of these studies demonstrated an excess rate of leukemia as a means to distinguish which cohorts received a significant radiation exposure. Five of the 7 cohorts had elevated rates of leukemia and melanoma. The remaining 2 cohorts did not have excess melanoma or excess leukemia, suggesting that both did not receive sufficient irradiation to induce any malignancy. They conclude that melanoma has not always been considered a significant end point in epidemiological studies on ionizing radiation, which may have hampered identifying a potentially causal relationship between ionizing radiation and melanoma. They also recommend that future epidemiological studies of ionizing radiation consider inclusion of melanoma as an outcome of interest.¹¹

It is not clearly understood why some patients with CDKN2A mutations never develop melanoma, whereas others with the same mutation and even in the same family seem exquisitely prone to the development of numerous primary lesions. We suggest that in our patient, the MPM may have been a result of the superimposition of environmental exposures (ionizing radiation) on her highly vulnerable genetic predisposition. The careful examination of phenotypic and environmental risk factors in the very vulnerable subpopulation of CDKN2A mutation carriers may be a strategy to identify and clinically verify risk factors less obvious in the general population.

Correspondence: Sancy A. Leachman, MD, PhD, Department of Dermatology and Huntsman Cancer Institute Melanoma Program, University of Utah, 2000 Circle of Hope, Room 5242, Salt Lake City, UT 84112-5550 (sancy.leachman@hci.utah.edu).

Financial Disclosure: None reported.

Accepted for Publication: April 20, 2007.

Author Contributions: Dr Leachman had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Hansen, Bowen, Cannon-Albright, and Leachman. Acquisition of data: Eliason, Hart, Florell, Harris, Swinyer, and Leachman. Analysis and interpretation of data: Eliason, Porter-Gill, Chen, Sturm, and Leachman. Drafting of the manuscript: Eliason, Hansen, Hart, and Leachman. Critical revision of the manuscript for important intellectual content: Eliason, Hansen, Hart, Porter-Gill, Chen, Sturm, Bowen, Florell, Harris, Cannon-Albright, Swinyer, and Leachman. Administrative, technical or material support: Chen and Sturm. Study supervision: Leachman.

Funding Support: This study was supported by a grant from the Tom C. Matthews Jr Familial Melanoma Research Clinic, the Huntsman Cancer Foundation, grants R01CA102422 and M01RR00064 from the National Institutes of Health, and grant C06-RR11234 from the National Center for Research Resources.

Additional Contributions: April Alexander Larson provided intellectual contributions to the study. The Pedigree and Population Resource (with funding from the Huntsman Cancer Foundation) provided data and computing support related to the Utah Population Database.