NEW YORK (GenomeWeb News) – Four new melanoma risk loci have been detected through two independent genome-wide association studies reported online yesterday in Nature Genetics.
Along with family history, traits such as fair skin, light hair and eyes, and sun sensitivity are among the best known risk factors for melanoma. But the new studies mark the first time melanoma-associated genetic variants have been identified and confirmed outside of genes with ties to known melanoma-related processes or traits such as skin pigmentation, eye color, or hair color.
Data from the GWASs also verified some of the risk loci found for melanoma in the past.
Members of the GenoMEL consortium found three of the new melanoma-associated regions through a GWAS that relied on genotyping data for thousands of European or Israeli individuals. Two of the three newly detected sites in the genome seem to affect melanoma risk through processes outside of those implicated in the disease in the past.
"We know that overexposure to UV increases the risk of developing melanoma — but this evidence shows that there are new additional genetic faults which can push up the risk further," corresponding author Tim Bishop, with Cancer Research UK, said in a statement.
After sifting through data for thousands of individuals enrolled by researchers from 14 groups participating in GenoMEL, the team ended up doing its main analyses using genotyping data for 2,804 individuals of European or Israeli ancestry with melanoma, 1,835 unaffected controls enrolled through studies by the GenoMEL team, and 5,783 more controls from France and the UK.
Of the seven new candidate loci that they followed up on, in 1,579 cases and 2,036 controls from the UK and Netherlands, the researchers verified statistically significant melanoma associations for SNPs in three regions.
Variants in these regions were found in and around three genes: MX2, a gene previously implicated in narcolepsy, ATM, a gene involved in DNA repair, and CASP8, a gene from a family of protease protein-coding genes that help to control apoptosis and cell proliferation.
Another locus near the chromosome 1 proto-oncogene CCND1 seemed to somewhat influence melanoma risk, but was not conclusively associated with the disease in the validation stage of the study.
In another study, Australian-led research team identified a chromosome 1 locus linked to melanoma through a multi-phase GWAS and meta-analysis that involved thousands of individuals from Australia, the US, and Europe.
For the discovery phase of that study, researchers genotyped DNA from blood or saliva samples from 2,168 cases and 4,387 controls of European descent from Australia.
They then tested nine of the most promising loci from this discovery group in three more case-control cohorts: one that included 2,804 cases and 7,618 controls from Europe, an American cohort with 1,804 cases and 1,026 controls, and a second cohort from the US that was comprised of 585 cases and 6,500 controls.
Two variants, both on chromosome 1, were replicated in this follow-up phase. Meanwhile, a meta-analysis that relied on data from all of the case-control studies pointed to a melanoma risk locus on chromosome 1 that encompasses sequences for up to 10 genes, including the genes ARNT and SETDB1.
None of the variants identified at this chromosome 1 locus has been associated with known melanoma-related mechanisms such as skin pigmentation or melanocyte nevus skin growth patterns.
"Variants at the newly identified locus do not seem to be associated with human pigmentation or measures of nevus density, suggesting that they may influence melanoma risk through distinct mechanisms," Queensland Institute of Medical Research investigator Stuart MacGregor, who was corresponding author on that study, and its co-authors concluded. "Identification of the causal variants at this locus will help refine estimates of risk for this increasingly common cancer."