Marshaling the Translational Potential of MC1R for Precision Risk Assessment of Melanoma

The incidence of melanoma has tripled in the last four decades, and the rising prominence of this potentially fatal skin cancer in younger adults is particularly alarming (1, 2). Most melanomas are caused by ultraviolet radiation (UVR), predominately due to sun exposure, and with recent contributions from exposure to artificial UVR sources (3). Primary prevention efforts for melanoma seek to mitigate the development of disease through the reduction of daily UVR exposure through use of hats and clothing to block sun exposure, and use of sunscreen (4). Limiting UVR exposure has the potential to reduce melanoma rates by at least 65% (5). Secondary prevention for melanoma focuses on the early detection of melanocytic lesions by means of thorough skin screening examinations conducted either by a trained professional or completed carefully and routinely at home (6). Although public health messaging promoting sun protection and skin awareness has been available for years, only 35% of the US population uses sunscreen consistently, and only 20% of Americans report having had a total-body skin examination (7–10). Poor prevention behaviors also extend to Hispanics, an important and growing segment of the American population (11). Over the past decade, researchers have considered the communication of genetic test results as a potential means to motivate behavior change in the general population. For melanoma, information on inherited variation in the melanocortin-1 receptor (MC1R) gene may be one risk screening tool that could shape physician–patient communication to motivate melanoma primary and secondary prevention behaviors. The characteristics of MC1R, with nontrivial increased risk of melanoma in persons inheriting specific variants, the high prevalence of variant carriers in the population, and the relevance of carrier status to individuals at low phenotypic risk of disease, make it an ideal genetic marker for feedback to general population subgroups to relay information about heightened disease risk. The time is right for high-quality research focusing on the translation of MC1R genotype into clinical and public health practice.

MC1R encodes a protein central to pathways that signal the production of melanins. It exhibits impressive genetic variability, with the majority (61%) of individuals of European ancestry carrying as least one nonsynonymous variant that has potential to affect native MC1R function (12). In contrast, most MC1R variants related to African ancestry are synonymous changes with little if any potential to impact native MC1R function (13, 14). The nine most prevalent variants at MC1R range in frequency from about 0.5% to 11% in the general population, and the inheritance of one of these variants is a robust marker for increased risk of melanoma (12). This finding is consistently observed across varied populations of European ancestry, including persons identifying as Hispanic and of Mediterranean descent (15, 16). These MC1R variants impart a 1.5- to 2.7-fold increased odds of melanoma (12), which are comparable in strength to several other melanoma risk factors including family history of disease and phenotypic characteristics, although disease odds associated with common or atypical moles are stronger and range upward to 10-fold (17, 18). Importantly, even among individuals with “sun-resistant” phenotypes, including those with darker hair color, fewer freckles, darker skin tones, and/or a proclivity to tan, MC1R risk variants show robust associations with risk of melanoma (12). Thus, MC1R genotype conveys information about inherited risk of melanoma for a broad spectrum of individuals, including those with darker pigmentary phenotypes who may be unaware of their personal susceptibility to melanoma.

Attributable risk refers to the amount of disease reduction possible through the “removal” of a particular exposure. The amount of melanoma risk attributable to the carriage of one of the nine most prevalent MC1R variants ranges from 1.2% to 8.9%, and the attributable risk summed across the nine variants reaches 40% (Table 1; ref. 12). Thus, at least theoretically, if knowledge of MC1R genotype is salient to patients, the potential to affect prevention behaviors across a large portion of the general population is quite possible.

Genetic testing for MC1R could lead to heightened, more precise, and more personalized feedback regarding melanoma risk that is persuasive and motivational, in line with well-established health behavior theories (e.g., The Health Belief Model, The Theory of Planned Behavior; ref. 19). Concerns regarding the potential that MC1R feedback will increase distress and worry about melanoma, result in overuse or inappropriate use of health services, or lead to a false sense of security about melanoma in those found not to be at increased risk will need to be examined empirically, despite the fact that these concerns have not been borne out in other contexts where SNP feedback has been provided (20). It may also be that the low level of risk (∼2- to 3-fold) associated with MC1R variant carriage may not provide a potential motivational impetus for behavior change.

There are several ongoing research studies striving to obtain empirical evidence to inform whether MC1R genotyping can be incorporated into clinical and/or public health practice to impact melanoma risk prevention behaviors. Three intervention studies use randomized designs to test the efficacy of feedback of genetic information. One study set in Australia provides risk information based on a panel of 42 SNP markers in 21 genomic regions known to be associated with melanoma risk that includes eight risk variants in MC1R. A pilot phase assessed behavior changes three months postintervention and found promising results in the mean number of standard erythemal doses (a measure of overall UVR dose) comparing the control arm receiving only nongenetic information about melanoma prevention and early detection to the invention arm receiving information on personalized genomic risk along with nongenetic information (21). Two additional studies, one set in Florida and one in New Mexico (22), are assessing feedback of risk information based on carriage of MC1R genotype alone. Findings from these three studies, all designed to assess behavior changes out to one year postintervention, are expected over the next several years. Of note, the two ongoing studies in the United States have incorporated a core content of shared risk feedback materials, including Spanish translations (23), and there is a healthy cross-fertilization of ideas and practice experience across the investigators of all three ongoing studies.

Testing and feedback of high penetrance genetic risk for melanoma [e.g., genetic testing for rare mutations in the cyclin-dependent kinase inhibitor 2A (CDKN2A) gene among members of melanoma families] requires intensive pre- and posttest education and counseling conducted by a certified genetic counselor. In contrast, MC1R genetic risk screening and feedback to members of the general population will need alternative methods of testing and provision of feedback. Given that risk levels associated with carriage of high risk MC1R variants are comparable to other melanoma risk factors such as family history that are commonly discussed in behavioral interventions and standard clinical practice, it can be argued that feedback of MC1R genotype should not require formal genetic counseling per se. Furthermore, the resource demands of in-person genetic counseling coupled with the limited number of certified genetic counselors reduce the feasibility of this implementation method in community-based clinics and other public health contexts relevant to the general population. The determination and communication of MC1R risk status has potential to improve melanoma prevention efforts given that genetic risk information is novel, may lead to fruitful discussions with physicians and family, and may help individuals understand how genetics and environment together help to inform their skin cancer risk, all potentially leading to risk appreciation and potential behavior change. In the future, MC1R may be able to be combined with other known melanoma risk factors in order to provide individuals with more precise risk levels than are currently available.

Dissemination strategies for MC1R genotype results could include receipt of information via web-based interface or postal mail or via telephone call with a trained (nongenetic counselor) public health professional. In a study offering MC1R testing as part of a multiplex panel of markers for other diseases, participants were receptive to receiving genetic results in the mail followed by a phone call with a trained research educator; notably, although the services of a trained genetic counselor were made available, none of the study participants selected this option (24). Studies are under way that will confirm the comprehensibility and acceptability of these approaches to MC1R test information and feedback (22). Another option is to have MC1R results delivered directly to primary or dermatological health care providers, which will allow for ongoing patient–physician communication focused on skin cancer prevention informed by MC1R status. Regardless of communication channel and whether individuals receive feedback that they have higher risk, average risk, or no MC1R variants, the fundamental goal is to ensure clarity of recommendations regarding primary and secondary prevention behaviors. Ideally, feedback of MC1R genotype will maximize motivation for protection in those at highest risk, and avoid undermining prevention behaviors in those at lower risk, who may indeed have other risk factors for melanoma or who should be reminded of population recommendations for sun protection (4). Although confirmation of efficacy for interventions incorporating precision risk assessment needs to precede dissemination and implementation efforts, the rapidly evolving field of public health genomics and its intersection with skin cancer warrants ongoing discussions regarding these important next steps. Further, the current development of scalable precision risk assessment interventions that reach broad segments of the population will facilitate and speed dissemination and implementation in the coming years.

In the coming years, precision risk assessment of melanoma may include genetic risk screening such as MC1R testing. Genetic information may well provide highly salient information that may raise the immediacy of primary and secondary prevention behaviors in those at heightened genetic risk. However, in other behavioral domains such as diet, physical activity, smoking cessation, and alcohol use, genetic risk feedback does not consistently increase health behavior change (25). For many individuals, genetic risk feedback for melanoma likely requires targeting of other important health behavior change mechanisms such as increasing confidence in performing consistent sun protection, or increasing perceived control over health (26). Many of the existing genetic risk feedback studies oversample research volunteers who were already highly motivated to change their behavior at baseline, even before receiving genetic risk feedback (27); it is a high priority to examine the impact of MC1R testing in individuals with a range of attitudes toward genetic testing and melanoma risk behavior change. Further, specific population subgroups – such as adolescents and young adults who engage in intentional tanning or Hispanics who have suboptimal awareness about skin cancer (28) – may be particularly likely to benefit. If studies show that MC1R genetic risk screening proves to be effective for motivating behavior change, a significant proportion of melanomas may be detected at its earliest stage for which surgical excision is highly curative or a large number of melanomas may be prevented altogether. While other genetic markers for melanoma susceptibility relevant in the general population may certainly emerge in the coming years, the burgeoning research agenda on the public health translational potential of MC1R genetic risk screening will inform and usefully advance current and future precision risk assessment of melanoma.

No potential conflicts of interest were disclosed.

Conception and design: P.A. Kanetsky, J.L. Hay

Development of methodology: J.L. Hay

Writing, review, and/or revision of the manuscript: P.A. Kanetsky, J.L. Hay

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): J.L. Hay

This work was supported, in part, by grant awards from the American Cancer Society to PAK (RSG-14-162-01-CPHPS) and the National Institutes of Health to JLH (R01 CA181241).