An international research collaboration has identified 13 new gene sites associated with the risk of coronary artery disease and validated 10 sites found in previous studies. Several of the novel sites discovered in the study, which is being published online in Nature Genetics, do not appear to relate to known risk factors, suggesting previously unsuspected mechanisms for cardiovascular disease.

"We now have identifed 23 specific genetic 'letters' that appear to confer risk for myocardial infarction and other aspects of coronary artery disease," says Sekar Kathiresan, MD, director of Preventive Cardiology at Massachusetts General Hospital and one of several co-lead authors of the report. "Knowing these sites lays the groundwork for isolating the genes responsible and developing new treatments based on those genes."

Although inherited factors may account for as much as 60 percent of the variation in risk for coronary artery disease, variants identified in genome-wide association studies (GWAS) account for only a small fraction of that risk. Since some of those previous studies may not have been large enough to identify genes with modest effect, 167 investigators at research centers around the world formed the Coronary ARtery DIsease Genomewide Replication and Meta-analysis (CARDIoGRAM) Consortium.

The researchers first assembled data from 14 previous GWAS for meta-analysis – a technique that combines results from several studies into a larger sample – which reviewed data from more than 22,000 individual with heart disease and almost 65,000 controls. The most promising sites identifed in the meta-analysis were then genotyped in another group of more than 56,600 individuals, about half with cardiac disease. The investigators also analyzed potential mechanisms and metabolic pathways by which newly identified variants might affect risk.

Results of the analyses confirmed 10 of 12 previously reported gene variants associated with coronary artery disease and identified 13 sites not previously reported. While most of these variants were most strongly associated with early-onset heart disease, the associations did not vary with the actual clinical condition for which patients were treated – heart attack or coronary artery disease requiring bypass surgery or angioplasty/stenting. Of the 23 variants validated in this study, seven are associated with LDL cholesterol levels and one with hypertension, but the others have no relation to known cardiovascular risk factors.

"The lack of apparent association with the risk factors we know so well is the source of a lot of excitement concerning these results," Kathiresan explains. "If these variants do not act through known mechanisms, how do they confer risk for heart disease? It suggests there are new mechanisms we don't yet understand. Another good thing about these findings is that they are in human patients, not in cells or mice, which gives us a good starting point for figuring out new disease pathways."