GENETICS OF MELANOMA

nheriting specific variants of the melanocortin-1 receptor (MC1R) gene that confers red hair, freckles, and pale skin doubles the risk for melanoma, independent of the effect of exposure to ultraviolet (UV) radiation, according to a case–control study published online on April 6 in JAMA Dermatology.

The link between sun exposure and skin cancer is well established, but it is weaker for melanoma than for squamous cell or basal cell carcinomas. Preclinical studies demonstrate elevated melanoma risk in the absence of UV exposure in mice that have certain variants (alleles) of MC1R. In humans, similar variants lie behind Fitzpatrick type 1 skin (people with skin that always burns, never tans, and is pale white, and with blond or red hair, blue eyes, and freckles).

In their study, Judith Wendt, MD, PhD, from the Medical University of Vienna, and her colleagues extend the findings of an independent genetic predisposition to humans, indicating that sun exposure alone is not sufficient to identify high-risk patients.

Clues From Mice

Separating nature from nurture in melanoma risk proved more challenging for humans than for mice, because "our red-haired friends and relatives walk around outside, exposed to UV," said David E. Fisher, MD, PhD, director of the Cutaneous Biology Research Center at the Massachusetts General Hospital in Boston. He was involved in some of the mouse studies (Nature. 2012;491:449-453) and, with Elisabeth M. Roider, MD, also from the Massachusetts General Hospital, wrote an accompanying editorial.

Dr Fisher told Medscape Medical News that the new study was "cleverly designed to look at independent signs of UV radiation damage in the skin, and from that, normalize for sunburn history and actinic skin damage as evidence of UV injury. Then they asked, do redheads no longer have an elevated risk of melanoma compared with nonredheads?"

The melanocortin-1 receptor, produced in response to the melanocyte-stimulating hormone, which is secreted in response to UV exposure, controls the balance of eumelanin and pheomelanin in the skin. Deficiency of MC1R favors production of pheomelanin, which confers the red hair/pale skin phenotype and is associated with less sun protection and production of reactive oxygen species. Since the 1990s, studies have associated MC1R "loss-of-function" gene variants with increased risk for melanoma but attributed the link to UV exposure.

An Independent Risk Factor

For their study, Dr Wendt and her colleagues evaluated 991 melanoma patients and 800 control subjects from the Molecular Markers of Melanoma Study. The protocol genotyped participants for five high-risk variants (R) and five low-risk variants (r); other (wild-type) variants were classified as 0.

The researchers first considered association of sunburn history (12 sunburns over the lifetime or 10 or more before age 20) and signs of severe actinic skin damage with melanoma risk. Then they used a multivariate analysis to evaluate the contribution of the MC1R genotype to melanoma risk, adjusting for sunburns and actinic damage. All calculations were adjusted for age and sex.

More than 12 previous sunburns increased melanoma risk (adjusted odds ratio [OR], 2.20; 95% confidence interval [CI], 1.63 - 2.96; P < .001), as did 10 or more sunburns before age 20 (adjusted OR, 2.19; 95% CI, 1.73 - 2.78; P < .001). Actinic skin damage was also associated with increased melanoma risk. The strongest associations were for severe skin damage on the back and neck (adjusted OR, 3.83; 95% CI, 2.74 - 5.35) and for wrinkling (adjusted OR, 3.62; 95% CI, 2.64 - 4.95)

In addition, the researchers found that the number of R and r variants significantly increased melanoma risk. Individuals with two or more (R or r) faced a melanoma risk twofold higher than individuals with wild-type variants (OR, 2.13; 95% CI, 1.66 - 2.75; P < .001).

R was associated with increased melanoma risk, whether it was present in two copies in homozygotes (R/R) (OR, 1.77; 95% CI, 1.12 - 2.79), a heterozygote with one r variant (R/r) (OR, 2.93; 95% CI, 2.07 - 4.16), or a wild-type variant (R/0) (OR, 1.94; 95% CI, 1.45 - 2.60).

Variant r was associated with increased risk when homozygous (r/r) (OR, 1.64; 95% CI, 1.13 - 2.38) and when paired with a wild-type variant (r/0) (OR, 1.35; 95% CI, 1.05 - 1.73).

The MC1R genotype emerged as a significant risk factor for melanoma after adjustment for sunburns and skin damage. The associations persisted for different risk variant combinations, as well as for different sites of skin damage.

Table. MC1R Genotype as a Risk Factor for Melanoma

Variants	Odds Ratio	95% Confidence Interval
R/R	1.65	1.02–2.67
R/r	2.63	1.82–3.81
R/0	1.83	1.36–2.48
r/r	1.50	1.01–2.21

Clinical Implications

Discovery of a second mechanism "doesn't mean that UV radiation is unimportant. We believe that the red pigment as a stressor amplifies the effect, making UV exposure even worse," Dr Fisher told Medscape Medical News.

Although genetic testing probably is not necessary because people who are fair skinned and burn easily already know it, Dr Fisher said, the UV-independent link "means that even in parts of the skin where the sun doesn't shine, if you are a redhead, the risk of melanoma is higher there too. People need to do self-exams and have their partners examine them, even in places that clothing covers," he explained. He advises avoiding the midday sun rather than relying solely on sunscreens, because they protect against sunburn, not cancer.

Limitations of the study include recall bias concerning previous sunburns and selection bias in using control subjects who were also hospital patients.

Dr Wendt, Dr Fisher, and Dr Roider have disclosed no relevant financial relationships.

JAMA Dermatol. Published online April 6, 2016