Careers  |  Sign In  |  Register  |   Twitter

Genetic risk scores don’t predict CVD

Reuters Health • The Doctor's Channel Daily Newscast

NEW YORK (Reuters Health) – Contrary to expectations, individual genetic markers don’t add up to a genetic risk score that’s any better than traditional risk factors for predicting cardiovascular disease, researchers are reporting.

The study, which involved more than 19,000 women, “looked at combinations of alleles, and our results (were) aimed at evaluating our ability to predict risk using our total knowledge to date rather than focusing on the effects of particular alleles,” lead author Dr. Nina P. Paynter, from Brigham and Women’s Hospital, Boston, explained in a phone call with Reuters Health.

“A genetic risk score could certainly be very useful,” she said. “There are people of intermediate risk that we could be doing a better job of classifying.”

In the Journal of the American Medical Association for February 17, Dr. Paynter and her colleagues point out that genome-wide association studies have identified genes associated with cardiovascular disease and intermediate phenotypes such as blood pressure and cholesterol level. The individual genes have small effects on cardiovascular disease risk prediction, however, and recently identified markers have not yet been tested.

For their studies, the investigators created 2 genetic risk scores using single nucleotide polymorphisms (SNPs) from the National Human Genome Research Institute catalog of study results published between 2005 and June 2009.

The first risk score involved 101 SNPs associated with cardiovascular disease or with risk factors for cardiovascular disease. The second included 12 SNPs linked to cardiovascular disease only.

Both risk scores used simple counts of the total number of risk alleles (the allele associated with an increased level or probability), assuming additive and independent effects for each risk allele.

Their study cohort included 19,313 initially healthy white women (median age 52.8 years at baseline) in the Women’s Genome Health Study who gave blood for genotyping. During a median follow-up of 12.3 years, 777 incident cardiovascular disease events occurred.

The mean number of risk alleles was 102 in the 101-SNP risk score and 10.7 in the 12-SNP risk score.

The 101 SNP genetic risk score was positively correlated with total cholesterol, systolic blood pressure and C-reactive protein, and negatively associated with HDL-cholesterol. Both genetic risk scores were associated with increased risk of cardiovascular disease after adjusting for age.

However, neither score remained independently associated with disease risk after adjusting for other risk factors, including history of diabetes, hemoglobin A1c level, C-reactive protein and family history.

While the genetic risk scores do not help to distinguish between women at high and low risk, they do confirm the value of traditional risk factors, the authors point out. The risk scores also validate family history of early cardiovascular disease, “which integrates shared genetics, shared behaviors, and environmental factors.”

“I believe the reason for our results is two-fold,” Dr. Paynter said to Reuters Health. “First, many of the markers in the 101 marker score were associated with intermediate risk factors, and their effect beyond a current measurement of the risk factors was not enough to predict risk. Second, our understanding of the genetics of cardiovascular disease is still incomplete.”

Still, she added, “Our results in no way detract from the biologic importance of genetics or the potential for genetics to eventually be useful in cardiovascular risk prediction. However, they do suggest that at this point it may be more useful for predicting your cardiovascular disease risk to know your family history.”


JAMA 2010;303:631-637.