Editors' ChoiceCardiovascular Disease

Genetic Risk Scores for Heart Disease: Useful or Useless?

See allHide authors and affiliations

Science Translational Medicine  24 Feb 2010:
Vol. 2, Issue 20, pp. ec29
DOI: 10.1126/scitranslmed.3000982

Coronary artery disease continues to be a leading cause of death worldwide in spite of recent advances in prevention and treatment. In addition, current predictors of cardiovascular disease used in clinical practice—such as high cholesterol, obesity, and hypertension—fail to identify many individuals that ultimately develop this disease. Progress in predicting and understanding heart disease is clearly needed.

Now, Paynter et al. have assessed a large number of single nucleotide polymorphisms (SNPs) from the literature that have been linked to coronary artery disease through genome-wide association studies (GWASs). They gathered results from over 19,000 women, yielding 101 SNPs that were significantly associated with disease. From these data, they calculated a “genetic risk score” for each of these women and then compared it with their subsequent history of cardiovascular disease. Using two separate approaches, the investigators found that although higher risk scores were associated with a heightened risk for coronary artery disease, the genetic scores failed to incrementally add to the predictive ability of the risk factors currently used in clinical practice.

Several additional aspects of this study require mentioning. First, a family history of coronary artery disease continued to be predictive for the development of coronary artery disease even after adjustment for the traditional risk predictors, indicating genetic underpinnings of this disease that have yet to be fully defined. Second, many of the SNPs assessed in the study were from GWASs that used variable definitions for coronary artery disease, which would negatively affect the ability of any SNP to predict clinically significant phenotypes such as heart attack. Third, many of the common SNPs identified are likely not causative but may actually tag a region of the genome that harbors a rare functional genetic variant with a much greater effect on disease risk.

Going forward, we will need to evaluate both common (as done here) and rare genetic variants in carefully designed prospective studies before issuing a final verdict on the utility of genetic risk scores. Most important, although this study shows that we have not yet learned how to use heart disease–associated SNPs as risk predictors, SNPs nevertheless highlight pathways involved in disease that may lead to targeted therapeutic interventions in the future.

N. P. Paynter et al., Association between a literature-based genetic risk score and cardiovascular events in women. JAMA 303, 631–637 (2010). [Abstract]

Navigate This Article