Abstract
BACKGROUND: Melanoma is the deadliest of skin cancers and has an increasing annual incidence worldwide. It is a multi-factorial disease most likely arising from both genetic predisposition and environmental exposure to ultraviolet light. Genetic variability of the components of the biological circadian clock is recognized to be a risk factor for different type of cancers. Moreover, two variants of a clock gene, RORA, have been associated with melanoma patient's prognosis. Our aim is to test the hypothesis that specific single nucleotide polymorphisms (SNPs) of the circadian clock genes may significantly influence the predisposition to develop cutaneous melanoma or the outcome of melanoma patients. METHODS: We genotyped 1239 subjects, 629 cases of melanoma and 610 healthy controls in 14 known SNPs of seven selected clock genes: AANAT, CLOCK, NPAS2, PER1, PER2, RORA, and TIMELESS. Genotyping was conducted by q-PCR. Multivariate logistic regression was employed for susceptibility of melanoma assessment, modeled additively. Subgroup analysis was performed by gender. For the female subgroup, a further discrimination was performed by age. For prognosis of melanoma assessment, multivariate Cox proportional hazard regression was employed. The Benjamini-Hochberg method was utilized as adjustment for multiple comparisons. RESULTS: We identified two RORA SNPs statistically significant with respect to the association with melanoma susceptibility. Considering the putative role of RORA as a nuclear steroid hormone receptor, we conducted a subgroup analysis by gender. Interestingly, the RORA rs339972 C allele was associated with a decreased predisposition to develop melanoma only in the female subgroup (OR 0.67; 95% CI 0.51-0.88; P = 0.003) while RORA rs10519097 T allele was associated with a decreased predisposition to develop melanoma only in the male subgroup (OR 0.62; 95% CI 0.44-0.87; P = 0.005). Moreover, the RORA rs339972 C allele had a decreased susceptibility to develop melanoma only in females aged over 50 years old (OR 0.67; 95% CI 0.54-0.83; P = 0.0002). None of the studied SNPs were significantly associated with the prognosis. CONCLUSIONS: Overall, we cannot ascertain that circadian pathway genetic variation is involved in melanoma susceptibility or prognosis. Nevertheless, we identified an interesting relationship between melanoma susceptibility and RORA polymorphisms acting in sex-specific manner and which is worth further future investigation.