Refining Cardiovascular Risk

April 2011

Art or genetics?
Art or genetics? In a major advance, the international CARDIoGRAM consortium has discovered 13 new genetic variants associated with increased risk for heart disease. This image shows output from the GeneTitan hardware that the Heart Institute used in its CARDIoGRAM participation. Each square shown in the close-up represents one of the 1.1 million genetic variants each sample is compared against. Black indicates the sample has no copies of a given variant, yellow indicates two copies and other colours indicate one copy.


The 9p21 risk variant is the strongest known common genetic risk factor for heart disease in Caucasians and Asians. Beginning with the discovery of 9p21 in 2007, studies have consistently shown that having one copy of a genetic variation in 9p21 increases a person’s risk of heart disease by 15 to 20 per cent. Having two copies boosts that increase in risk to 30 to 40 per cent.

Somewhat counterintuitively, researchers have not had much success using variations in 9p21 to predict the risk of heart disease among groups of people. Several studies have found that adding information about 9p21 to a panel of traditional risk factors for heart disease, such as high blood pressure, blood cholesterol levels, obesity and smoking, does not add any predictive power.

Much of this has to do with how good the standard risk models are, explained Robbie Davies, a statistical genetic analyst in the Heart Institute’s Cardiovascular Research Methods Centre, led by George Wells, a CARDIoGRAM co-author. “We’re trying to add predictive capabilities on top of a pretty good model. There’s a smaller piece of the risk pie available to predict using genetic information,” he said.

In a paper published recently in Circulation: Cardiovascular Genetics, Davies and his colleagues at the Heart Institute and the Cleveland Clinic attempted to improve the predictive power of genetic information on cardiovascular disease risk by using more of it. They tested traditional risk factors plus 9p21 compared to traditional risk factors plus a panel of 12 genetic variations linked to heart disease or heart attack in previous studies, including 9p21.

They compared the predictive power of the two genetic tests across two large sets of patient samples. Even though the other 11 risk variants in the panel are more weakly associated with heart disease than 9p21, the panel as a whole plus traditional risk factors were significantly better at predicting which patients had heart disease than either traditional risk factors alone or traditional risk factors plus 9p21.

Future research will need to test the ability of additional validated genetic risk factors to improve this risk prediction, as well as the ability to predict risk in other racial groups. “Studies like CARDIoGRAM discover these genes, then it’s up to different studies with different purposes, like ours, to try to make use of them,” concluded Davies.