Genetic risk scores don't predict cardiovascular disease in women

Genetic risk scores constructed from the medical literature were not significantly associated with total cardiovascular disease (CVD) in women, according to a new study.

Researchers used data from the Human Genome Research Institute to create two genetic risk scores, the first including 101 single-nucleotide polymorphisms (SNPs) associated with CVD and its risk factors and the second including 12 SNPs associated with incident CVD. The scores were evaluated in 19,313 white women participating in the Women's Genome Health Study, a subset of the Women's Health Study. Main outcome measures were incident myocardial infarction, stroke, arterial revascularization and cardiovascular death. The study was published in the Feb. 17 Journal of the American Medical Association.

Seven hundred seventy-seven CVD events occurred during a median of 12.3 years of follow-up. Of these, 199 were myocardial infarctions, 203 were strokes, 63 were CVD deaths and 312 were revascularizations. After adjustment for age, both risk scores were associated with increased CVD risk but were unable to distinguish between women who were at risk for cardiovascular events and those who were not (c index, 0.52 for both scores). After adjustment for traditional CVD risk factors, such as total cholesterol and blood pressure, the association between both scores and CVD risk disappeared while an association between self-reported family history and CVD persisted.

The study was limited in part because the authors were not able to include rare alleles in their genetic scores. However, the results have clinical relevance because they reinforce the utility of traditional risk factors and the importance of family history in determining CVD risk, the authors wrote. They noted that although the scores they tested were not helpful for risk prediction, their findings point to areas for additional research that may improve genetic prediction in the future.