Alameda, CA - Four genes have been identified that are associated with an increased risk for myocardial infarction, none of which have been previously associated with MI risk [1]. The authors believe further evaluation of these genes, identified in three successive studies of MI survivors, could help illuminate new mechanisms for coronary heart disease and be used in the future for risk stratification.

The paper, by researchers led by Dr Dov Shiffman—a scientist at Celera Diagnostics—is available online August 29, 2005 in the American Journal of Human Genetics and will be published in the October 2005 issue [1]. As the authors explained to heartwire, unlike most previous studies of gene variability in heart-disease patients, Shiffman et al's paper looked at a large cross-section of genes, about one third of all those with a known or predicted function. By contrast, most genetic studies of heritable cardiovascular disease have looked at limited candidate gene scans, limiting the focus to genes associated with particular processes that could lead to heart disease, namely genes associated with lipid synthesis, lipid metabolism, hypertension, etc.

Shiffman et al first studied over 11 000 single-nucleotide polymorphisms (SNPs) in almost 7000 genes, paying particular attention to SNPs involved in gene function. In the first study, the authors identified 637 SNPs associated with MI in 340 MI survivors, but not in 346 controls. In the second study, investigators looked for the same SNPs in 445 cases and 606 controls, a process that narrowed the list of MI-associated SNPs to 31. In a third and final study, the 31 SNPs were reduced to four gene variants that appeared to be associated with MI and had the same risk allele as in the first two studies. Even after adjusting for known risk factors, the risk estimates associated with these variants remained largely unchanged.

The first of the gene variants, KIAA0992, codes for the cytoskeletal protein palladin. A second, ROS1, is a tyrosine kinase, and the two others, TAS2R50 and OR13G1, are G-protein-coupled receptors. The authors hypothesize that the presumed function of these genes may "give some hints as to how they may be associated with MI." Because palladin is a cytoskeletal protein, a change in palladin expression might affect endothelial cell junctions, mononuclear cell trafficking into arterial walls, or platelet adhesion, Shiffman et al hypothesize. Likewise, ROS1 plays a role in cell proliferation and differentiation and might, theoretically, affect arterial plaque formation and stability. The role of G-protein-coupled receptors are even less clear, they note, but since one is a bitter-taste-receptor homolog and the other is an olfactory-receptor homolog, both might play a role in influencing a person's dietary preferences, a factor that could affect eating habits and lifestyle, potentially increasing MI risk.

Expressed as odds ratios, the odds ratio for having two risk alleles in the case group compared with having no risk alleles ranged from approximately 1.4 to 1.8. In an interview with heartwire, Shiffman and Thomas White (chief scientific officer for Celera Diagnostics) explained that the odds ratios are on a par with other recognized risk factors for coronary disease.

"If you look at what the odds ratio would be for having an LDL level of greater than 160, the odds ratio would be about 1.6, and if you had type 2 diabetes, it would be about 1.9," White said. "So our odds ratios are in the same ballpark as the kinds of conventional risk factors that physicians currently use to make treatment decisions."

Shiffman and colleagues believe having diagnostic "gene panels" could be particularly useful in young patients defined as ATP III low or moderate risk. Seventy percent of cardiac events in younger patients occur in moderate- or low-risk groups not typically assigned to the types of treatments used in older, higher-risk patients, they note. "It may be that some of these genetic risk factors could be combined into something we call a genetic risk score, where you would look at all SNPs that convey risk. And those people who are otherwise at moderate risk according to conventional factors [who are also found to have a high genetic risk score] would then be considered to be at significantly elevated risk to warrant treatment and lifestyle recommendations," White commented.

The authors add the caveat that all of their analyses were conducted in MI survivors, so it is possible that gene variants associated with fatal MI might not have been picked up. As well, the first study in the series looked only at male MI survivors and controls, "so it is unlikely that genetic markers that are primarily associated with MI in females would have been discovered in this study."

Still, say the authors, the association of these four gene variants is striking. "If confirmed in prospective, population-based studies, these markers—individually or in combination—may be useful in identifying individuals at a risk for MI higher than that predicted by traditional risk factors alone."