The most award winning
healthcare information source.
TRUSTED FOR FOUR DECADES.
Genetic Predictors of Cardiovascular Disease
Abstract & Commentary
By Jonathan Abrams, MD, Professor of Medicine, Division of Cardiology, University of New Mexico, Albuquerque Dr. Abrams serves on the speaker's bureau for Merck, Pfizer, and Parke-Davis.
Sources: Paynter NP, et al. Cardiovascular disease risk prediction with and without knowledge of genetic variation at chromosome 9p21.3. Ann Intern Med. 2009:150:65-72.
Genetic research and results are playing an increasingly large role in clinical medicine. I suspect that most physicians are not adequately knowledgeable about genetics; nor am I. At present, the genetics of clopidogrel and warfarin are front and center, as well as an increasing number of genetic studies dealing with normal metabolism, as well as dyslipidemia.
This study utilized variability at chromosome 9p21.3 to assess whether cardiovascular (CV) risk factors can be identified in large cohorts of Caucasian women (white healthcare professionals). The cohort included 22,129 women, free of CV disease, who participated in the Women's Genome Healthy Study. They found that genetic variation at the chromosome 9p21.3 site "consistently was...associated with CAD and diabetes." This risk allele is common, carried by 75% of the white population, and has been correlated with CVD. In these studies, the authors asked whether a single nucleotide polymorphism (SNP) can influence long-term CV risk in these apparently healthy women followed for CV events.
Results: Healthy white women older than 45-years-old were studied and followed for more than 10 years. The subjects were all evaluated by two risk prediction models (NCEP-3, Reynolds Risk Score), as well as additional biomarkers and family history information. Extensive statistical methodology was employed, including the Hannelle index and the Hismer Lemeshow test. This report evaluated the 22,129 women who were genotyped for the rs10757274 SNP. Baseline CV risk information and extensive lipid panels, Lp(a), CRP, and hemoglobin A1C were measured. Genotypes were determined for the rs10757274 SNP using "an oligonucleotide ligation procedure."
Among these women 26% had no risk alleles (AA), 50% had one risk allele (AG), and 24% had two risk alleles (GG). There was only a minor increase in risk related to the number of risk alleles, and there were no associations between genotype, lipid levels, or other standard risk factors. SNP rs757274 was associated with an adjusted hazard ratio of 1.25 for the AG genotype, and even higher (1.63) with the GG genotype. Equivalent associations were positive for several standard risk factors.
The authors point out that the prediction models and performance measures in this study were similar to a previously reported study in white males, with similar outcomes. The association of the 9p21.3 SNP with CV risk was only modest, and not powerful enough to significantly affect standard and familiar CV risk factors; "knowledge of this genetic variation only marginally improved the classification of risk prediction," with only modest improvement in the ATP III status, family history, CRP, or the c-index. "Our data provide evidence that the effect of this genetic variation is unlikely to reflect genetic differences in lipid levels or biomarkers of homeostasis and inflammation." Importantly, they conclude that SNP identification is not useful in screening in daily practice. They do not rule out future studies with non-white populations or use of multi-gene panels that might have better predictive results. While this research "confirmed the association of chromosome 9p21.3 variation with total CV disease: CAD, MI, or stroke, traditional risk factors, CRP, and family history, all remained robust in predicting outcomes, with too small a contribution from SNPs alone, which are unlikely to improve prediction results."
This fascinating report should shake the trees a bit (ie, to see a genetic study as the lead article in an excellent general internal medicine journal). An accompanying editorial by Ioannis states that "expectations for personalized genetic prescriptions may be exaggerated and premature."1 Many data sets confirm the association of rs10757274 with CV disease; clinicians cannot use these data for decision making or as adding to traditional risk factors. Ioannis states, "We are still far from personalized medicine." He concludes, "We need to learn more about what our genome can tell us, and more important, what it cannot tell us.
While this interesting report is essentially negative regarding the usefulness is genomic CV risk screening of rs10757274, it can serve as a bridge from the usual discussions of traditional risk factors to an expanded view in a much larger diagnostic tent. In this case, the genomic exploration fell short in changing our approach to CV risk factors and their treatment. Nevertheless, pay attention to subsequent reports, which should help all of us to understand new terminology, concepts, and knowledge.
1. Ioannidis JPA. Personalized genetic prediction: too limited, too expensive, or too soon? Ann Intern Med. 2009;150:139-141.