Coronary artery disease (CAD) remains the most common disease and the leading cause of death, with a strong genetic risk. Clinical trials have documented that modification of known risk factors can prevent 30–40 % of CAD.1 However, epidemiologic studies have shown that genetic risk accounts for approximately 50 % of CAD.2 If the aim is to prevent CAD in this century, it will require comprehensive prevention modifying genetic, as well as known, risk factors. In 2006, we proposed that technological advances were ripe for a whole genome-based approach to the discovery of genetic risk factors for CAD.3 The hypothesis that common genetic variants would contribute to the risk of CAD was borne out by an explosion of gene discovery in the past five years made possible by high throughput genotyping arrays. Importantly, the formation of international consortia bringing together data from many genome-wide association studies (GWAS) has led to the discovery of 31 CAD risk loci. Although these loci account for only 10 % of the heritability of CAD, most loci do not associate with conventional risk factors for CAD, like cholesterol, demonstrating that much more work will be required to understand the mechanistic underpinnings of CAD.
Technological Advances in Whole-genome Genotyping
Genetic variants, most commonly seen as single nucleotide polymorphisms (SNPs), can now be genotyped across the entire genome thanks to two important technological advances. First, microlithography and nanotechnology have enabled the mass production of microarrays that interrogate between 500,000 to one million SNPs. Second, by genotyping more than 2,000 selected human genomes with over 13,000,000 SNPs the Thousand Genomes project has enabled the fine mapping of genetic haplotypes so that for any individual genotyped on a microarray, genotypes of as many as 11,000,000 SNPs can be imputed with accuracy to an allele frequency as low as 0.02.4 This provides an unprecedented degree of genomic coverage, with one SNP every 300 base pairs. The drawback to whole-genome genotyping is that any association, to be statistically valid, must overcome a stringent threshold of significance to account for multiple testing. In the case of arrays that genotype 1,000,000 SNPs, this threshold is p<5 x 10-8, or p<0.05 divided by 1,000,000 (assuming each SNP tested is independent). Thus, enormous sample sizes including well-defined cases and controls are required.
The First Common Genetic Risk Factor for Coronary Artery Disease
Early in 2007, we, and others, simultaneously reported on the discovery of the first common genetic variant to predict CAD risk.5–8 This finding was quickly confirmed by multiple other groups9–13 and has become a benchmark by which GWAS for CAD are measured. The 9p21.3 locus covers 53,000 base pairs of sequence approximately 130,000 base pairs downstream from the CDKN2B gene encoding a cyclin dependent kinase inhibitor. The locus overlaps the 3’ sequence of a long alternatively spliced non-coding RNA that initiates transcription upstream of CDKN2B and runs antisense to this gene. Initially called ANRIL for antisense non-coding RNA of the INK4 locus, this gene now bears the official name CDKN2BAS. The 9p21.3 locus contains cis-regulatory sequences that influence the expression of CDKN2B as well as the related flanking gene CDKN2A.14,15 The CAD risk allele is associated with reduced expression of CDKN2A14,15 and the release of cell cycle inhibition likely contributes to the cellular proliferation characteristic of atheromatous lesions.
9p21 occurs in 75 % of the population.5 If one is homozygous, the relative risk is increased to 50 % and if heterozygous, 25 %. 9p21 is now observed in all ethnic groups except Africans and African Americans.16 The very important observation that 9p21 risk is independent of all known risk factors such as cholesterol, blood pressure, or diabetes suggests a novel mechanism.5 While the molecular mechanism of why 9p21 mediates risk is unknown, we have shown it acts through atherosclerosis of the vessel wall.22 It is not related to thrombosis or plaque rupture and as such does not associate with increased risk for myocardial infarction. We showed a good correlation between the number of coronary vessels involved and the number of copies of 9p21 present; patients presenting with one vessel disease were more likely to be homozygous non-risk, while those presenting with triple vessel disease were more likely to have two copies of the risk allele. This finding has been confirmed by others.17
International Meta-analysis Consortia Augment the Power of Genome-wide Association Studies
Subsequent to the discovery of the 9p21.3 locus, nine other risk loci for CAD and/or myocardial infarction (MI) were identified in GWAS from several laboratories around the world.18–21 The frequency of discovery of novel loci has gradually tapered off with the realization that individual laboratories sampling regional populations do not have the resources or sample sizes to provide sufficient power for the discovery of loci with ever diminishing effect sizes.
To overcome this limitation, large international consortia were formed to meta-analyze data from individual GWAS samples. One such consortium is Coronary artery disease genome-wide replication and meta-analysis (CARDIoGRAM), bringing together 14 GWAS including 22,233 individuals with CAD (cases) and 64,762 controls of European descent followed by genotyping of top association signals in 56,682 additional individuals.22,23 CARDIoGRAM doubled the number of known CAD loci, identifying 13 novel ones. Another consortium combining four GWAS of European and South East Asian samples called C4D24 comprised 15,420 individuals with CAD and 15,062 controls with replication in an independent sample of 21,408 cases and 19,185 controls, and one combining data from three Asian studies25 have also reported the discovery of novel loci. Together, these consortia bring to 31 the number of loci associated with CAD and/or MI (see Table 1).
Overview of the Genetic Variants for Coronary Artery Disease
In Table 1 there are many features characterizing the genetic variants. The frequency of the risk alleles in the population varies from between 2 and 91 %. The odds ratio of increased risk is small, averaging 15 % with only two outliers, one of 65 % and the other 92 % with the remainder in the range of 10–25 %. The maximum number of risk alleles one could have is 62. The maximum number of alleles present per individual we observed in CARDIoGRAM was 37 and the minimum was nine. Individuals with early onset of CAD had the most risk alleles. Development of a weighted risk score showed the risk associated with the most alleles was threefold greater than individuals with the least number of alleles. If risk is assessed in percentiles, the top tenth percentile had an odds ratio for CAD of 1.88 versus 0.55 for the lowest tenth percentile.23
Refining Phenotypes—Distinguishing Coronary Artery Disease from Myocardial Infarction
Myocardial infarction is a phenotype associated with CAD resulting from the rupture of the vessel wall and formation of a thrombus over a substrate of coronary atherosclerosis. Most GWAS have not distinguished between the phenotype of coronary atherosclerosis and myocardial infarction and treated them as equals. However, recent evidence shows that these are genetically heterogeneous phenotypes. For example, we reported that each allele of the 9p21.3 locus is associated with the degree of severity of angiographically defined CAD, but does not predict myocardial infarction among cases of CAD.26
In another recent study, we showed that the ADAMTS7 and ABO loci discovered in the CARDIoGRAM consortium show different associations with CAD and MI; whereas the ADAMTS7 predicts CAD, it does not predict MI and conversely whereas the ABO locus predicts MI it is not associated with CAD.27 These results highlight the complexity of coronary artery disease and the necessity to refine phenotypic classification.
Previously, we emphasized the importance of having well-defined controls, as well as cases, in genetic association studies.3 We proposed that with the advent of non-invasive multi-slice computed tomography (CT) angiography, identifying controls free of CAD would improve the power of association studies. Unfortunately, a more expedient and less expensive approach has been used in most GWAS by comparing asymptomatic and age-matched controls to CAD cases to search for genetic loci associated with CAD risk. We estimate that as many as one in three controls are misclassified using this approach, strongly undermining the power of GWAS. Thus, a further refinement of the phenotype of controls as well as cases will be needed if additional loci are to be discovered.
The genetic loci are already used clinically to predict disease severity.28 Thanks to improved coverage from the 1,000 genomes project, imputation enables identification of loci with allele frequencies above 2 %. It is predicted that within the next five years the power to impute will be further improved to predict alleles with frequencies as low as 1 %. New methods of whole genome sequencing are rapidly coming online and should enable the discovery of rare alleles (with frequencies between 0.1 and 1 %) with strong effects, but validation of these rare alleles will require genotyping across very large samples of the population. In addition, confirmation by functional replication in suitable animal models will also be necessary.
International consortia are already joining forces and it is anticipated that sample sizes of up to one million genotyped individuals, with refined phenotyping, will lead to the discovery of even more loci. The statistical threshold of genome-wide significance remains an impediment to the discovery of loci that may have weak effects that are biologically important over a patient’s lifetime. Visscher’s group has shown that up to half of the heritability of common traits (including common diseases) can be accounted for when considering the complete genotypic profile of an individual, not just alleles that reach genome-wide levels of significance.29
Even though the GWAS approach has not discovered all loci associated with common diseases, it has set the stage for important new directions. First, it clearly emphasized the importance of careful phenotyping. Second, it has identified genetic risk loci that were previously unsuspected. Third, it has enabled the development of genetic risk assessment and disease stratification. The GWAS discoveries will launch an intensive research effort to elucidate novel biologic mechanisms contributing to the risk of CAD and MI and will ultimately lead to new therapies.
Implications for Prevention and Therapy
The prevention of CAD will require the full discovery of genetic risk, however, a major effort is in progress. While the relative risk of many of these risk variants are significant, it remains to be determined whether genetic testing should be part of clinical management. We do not know how to treat these risk factors so many would claim it is premature. It is important to recognize that with the microarray and high throughput technology, a few hundred risk factors can be assessed within an hour or less.
The most important finding to emerge from the genome wide associations studies for CAD is the observation that of 31 risk variants, only six mediate their risk through known risk factors. Thus, there are mechanisms contributing to atherosclerosis that remain unknown. We cannot be comprehensive in preventing CAD or treating its sequelae unless we understand these mechanisms. There is great potential for these new targets to facilitate development of drugs specifically inhibiting these genetic risk factors.
- Roger VL, Go AS, Lloyd-Jones DM, et al., Heart disease and stroke statistics--2011 update: a report from the American Heart Association, Circulation, 2011;123(4):e18–e209.
- Chan L, EB, Gene-environment interactions and gene therapy in atherosclerosis, Cardiology in Review, 1994;2(3):130–7.
- Roberts R, Stewart AFR, Personalized genomic medicine: a future prerequisite for the prevention of coronary artery disease, Am Heart Hosp J, 2006;4(3):222–7.
- Durbin RM, Abecasis GR, Altshuler DL, et al., A map of human genome variation from population-scale sequencing, Nature, 2010;467(7319):1061–73.
- Genome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controls, Nature, 2007;447(7145):661–78.
- Helgadottir A, Thorleifsson G, Manolescu A, et al., A common variant on chromosome 9p21 affects the risk of myocardial infarction, Science, 2007;316(5830):1491–3.
- McPherson R, Pertsemlidis A, Kavaslar N, et al., A common allele on chromosome 9 associated with coronary heart disease, Science, 2007;316(5830):1488–91.
- Samani NJ, Erdmann J, Hall AS, et al., Genome wide association analysis of coronary artery disease, N Engl J Med, 2007;357(5):443–53.
- Ding H, Xu Y, Wang X, et al., 9p21 is a shared susceptibility locus strongly for coronary artery disease and weakly for ischemic stroke in Chinese Han population, Circ Cardiovasc Genet, 2009;2(4):338–46.
- Shen GQ, Rao S, Martinelli N, et al., Association between four SNPs on chromosome 9p21 and myocardial infarction is replicated in an Italian population, J Hum Genet, 2008;53(2):144–50.
- Hiura Y, Fukushima Y, Yuno M, et al., Validation of the association of genetic variants on chromosome 9p21 and 1q41 with myocardial infarction in a Japanese population, Circ J, 2008;72(8):1213–7.
- Hinohara K, Nakajima T, Takahashi M, et al., Replication of the association between a chromosome 9p21 polymorphism and coronary artery disease in Japanese and Korean populations, J Hum Genet, 2008;53(4):357–9.
- Bown MJ, Braund PS, Thompson J, et al., Association between the coronary artery disease risk locus on chromosome 9p21.3 and abdominal aortic aneurysm, Circ Cardiovasc Genet, 2008;1(1):39–42.
- Jarinova O, Stewart AFR, Roberts R, et al., Functional analysis of the chromosome 9p21.3 coronary artery disease risk locus, Arterioscler Thromb Vasc Biol, 2009;29(10):1671–7.
- Harismendy O, Notani D, Song X, et al., 9p21 DNA variants associated with coronary artery disease impair interferon-gamma signalling response, Nature, 2011;470(7333):264–8.
- Assimes TL, Knowles JW, Basu A, et al., Susceptibility locus for clinical and subclinical coronary artery disease at chromosome 9p21 in the multi-ethnic ADVANCE study, Hum Mol Genet, 2008;17(15):2320–8.
- Patel RS, Su S, Neeland IJ, et al., The chromosome 9p21 risk locus is associated with angiographic severity and progression of coronary artery disease, Eur Heart J, 2010;31(24):3017–23.
- Kathiresan S, Voight BF, Purcell S, et al., Genome-wide association of early-onset myocardial infarction with single nucleotide polymorphisms and copy number variants, Nat Genet, 2009;41(3):334–41.
- Tregouet DA, Konig IR, Erdmann J, et al., Genome-wide haplotype association study identifies the SLC22A3-LPAL2-LPA gene cluster as a risk locus for coronary artery disease, Nat Genet, 2009;41(3):283–5.
- Samani NJ, Deloukas P, Erdmann J, et al., Large scale association analysis of novel genetic loci for coronary artery disease, Arterioscler Thromb Vasc Biol, 2009;29(5):774–80.
- Erdmann J, Grosshennig A, Braund PS, et al., New susceptibility locus for coronary artery disease on chromosome 3q22.3, Nat Genet, 2009;41(3):280–2.
- Preuss M, Konig IR, Thompson JR, et al., Design of the Coronary ARtery DIsease Genome-Wide Replication And Meta-Analysis (CARDIoGRAM) Study: A Genome-Wide Association Meta-analysis Involving More Than 22,000 Cases and 60,000 Controls, Circ Cardiovasc Genet, 2010;3(5):475–83.
- Schunkert H, Konig IR, Kathiresan S, et al., Large-scale association analysis identifies 13 new susceptibility loci for coronary artery disease, Nat Genet, 2011;43(4):333–8.
- A genome-wide association study in Europeans and South Asians identifies five new loci for coronary artery disease, Nat Genet, 2011;43(4):339–44.
- Wang F, Xu CQ, He Q, et al., Genome-wide association identifies a susceptibility locus for coronary artery disease in the Chinese Han population, Nat Genet, 2011;43(4):345–9.
- Dandona S, Stewart AFR, Chen L, et al., Gene dosage of the common variant 9p21 predicts severity of coronary artery disease, J Am Coll Cardiol, 2010;56(6):479–86.
- Reilly MP, Li M, He J, et al., Identification of ADAMTS7 as a novel locus for coronary atherosclerosis and association of ABO with myocardial infarction in the presence of coronary atherosclerosis: two genome-wide association studies, Lancet, 2011;377(9763):383–92.
- Davies RW, Dandona S, Stewart AFR, et al., Improved prediction of cardiovascular disease based on a panel of single nucleotide polymorphisms identified through genome-wide association studies, Circ Cardiovasc Genet, 2010;3(5):468–74.
- Yang J, Benyamin B, McEvoy BP, et al., Common SNPs explain a large proportion of the heritability for human height, Nat Genet, 2010;42(7):565–9.
- Gudbjartsson DF, Bjornsdottir US, Halapi E, et al., Sequence variants affecting eosinophil numbers associate with asthma and myocardial infarction, Nat Genet, 2009;41(3):342–7.
- Soranzo N, Spector TD, Mangino M, et al., A genome-wide meta-analysis identifies 22 loci associated with eight hematological parameters in the HaemGen consortium, Nat Genet, 2009;41(11):1182–90.