NEW YORK (GenomeWeb) – A genome-wide association meta-analysis has uncovered new and known risk factors for endometrial cancer, including variants in and around expression quantitative trait loci (eQTL) that implicated potential causal genes.
Researchers from the University of Cambridge, QIMR Berghofer Medical Research Institute, and elsewhere brought together data from 17 prior endometrial cancer studies, encompassing more than 12,900 endometrial cancer cases and almost 109,000 age-matched controls. Their meta-analysis highlighted nine new loci with genome-wide significant ties to endometrial cancer, while subsequent analyses led to risk alleles linked to altered expression of genes suspected of contributing to the gynecological cancer.
The work "doubled the number of known endometrial cancer risk loci and revealed candidate causal genes for future study," senior author Deborah Thompson, a public health and primary care researcher at the University of Cambridge, and her co-authors wrote in their Nature Communications paper yesterday, adding that their related expression quantitative trait locus analysis "identified candidate causal genes and pathways related to tumor development for follow-up studies."
Using data generated through studies done for the Endometrial Cancer Association Consortium, Epidemiology of Endometrial Cancer Consortium, and UK Biobank project, the team compared directly genotyped and imputed SNP profiles in 12,906 endometrial cancer cases and 108,979 unaffected controls matched for ancestry, geography, and age.
The search uncovered genome-wide significant associations at all but one of the eight loci implicated in endometrial cancer previously, the researchers reported, along with 125 new endometrial cancer-associated SNPs falling at nine distinct loci. They noted that one of these loci, a site on chromosome 2, appeared to have closer ties to endometrial cancers from non-endometrioid histological subtypes.
The team went on to tease 'credible causal risk' SNPs (ccrSNPs) at the new risk loci, incorporating eQTL data from studies such as GTEx and the Cancer Genome Atlas project to focus in on candidate causal genes such as CDCA8 or RCN1, as well as tissues most likely to show altered gene expression in the presence of the endometrial cancer-associated alleles.
That work was complemented by gene network analyses and available chromatin immunoprecipitation sequencing profiles, which also helped the researchers narrow in on potential causal genes, pathways, and epigenetic features involved in endometrial cancer.
When the researchers considered some of the non-cancer-related traits linked to the endometrial cancer risk variant set, on the other hand, they found that at least some of the risk alleles corresponded with traits such as obesity, type 2 diabetes, or age at first menstruation, or menarche.
"Using this larger GWAS-meta dataset, we were … able to confirm the previously published Mendelian randomization studies finding that higher [body mass index] is causal for endometrial cancer risk, and the protective effect of later age of menarche on endometrial cancer risk," the authors wrote.