Cardiovascular genetics--two steps forward, one step back.

The recent publication of 2 studies of the genetics of coronary heart disease1,2 provides an opportune moment to reassess strategies and progress towards the time when the genetic investment in this common cause of death and disability will be repaid with better detection and prevention. These 2 studies are important because they are large and their findings appear to be consistent. The genetic literature has been plagued by small and discrepant reports of genetic discovery. So much so that journal editors are beginning to demand more stringent definitions of statistical significance3 to avoid publishing false positive results. However, one must not assume that inconsistency equates to error, as there is sufficient genetic heterogeneity in different populations around the world to explain why one study might be positive and another negative. Even within the recent reports, the authors found that the association of myocardial infarction with a sequence variant on chromosome 9p21 was evident in Caucasian but not in black subjects.2 The size of a study is important as it determines statistical power to detect genetic associations.4 The numbers needed to study depend on whether one takes a pessimistic or optimistic view of the likely frequency and magnitude of effect of influential genetic variants. The pessimistic view that subscribes to infrequent variants of small individual effect has led to predictions of sample sizes in the hundreds of thousands. This view no doubt was a stimulus for the studies of Helgadottir et al and McPherson et al to include a combined total of about 40,000 subjects. However, these studies revealed surprisingly that the genetic variants were relatively common, with homozygotes (carrying 2 copies of the variant) comprising about 20% of the Caucasian population. The results also showed that the effect size was not inconsequential, with these homozygotes displaying an apparent 30% to 100% increase in risk of myocardial infarction. This genetic risk is less of a risk than smoking, hypertension, diabetes or even a positive family history of coronary heart disease in a first-degree relative, which are associated individually with a 200% to 500% increase in risk, and much higher when combined. Nevertheless, the genetic variant might be something worth testing, and may Cardiovascular Genetics – Two Steps Forward, One Step Back Stephen B Harrap,1PhD, MBBS, FRACP