If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
CAD is a complex disease due to an interplay between genetic and lifestyle factors.
•
The genetic basis of CAD has been elusive until a decade ago.
•
GWAS detected >60 common variants associated with CAD risk.
•
Genomic techniques have provided useful insights into CAD pathophysiology.
•
Rare protective mutations paved the way to the development of new biologic drugs.
Abstract
Clinicians are well aware of the importance of a positive family history for coronary artery disease (CAD). Nonetheless, elucidation of the genetic basis of CAD has long proven difficult. The scenario changed in the last decade through the application of modern genomic technologies, like genome-wide association studies (GWAS) and next generation sequencing (NGS). GWAS have discovered over 60 common variants highly associated with CAD. For predictive purposes, such variants have been used to build up Genetic Risk Scores (GRSs), but their incorporation into classical prediction models does not appear substantially outperform the simple addition of family history. To date, the only strong case for the utility of incorporating genetic testing into clinical practice is represented by the diagnosis of Familial Hypercholesterolemia (FH). On the other hand, utilization of genomic techniques has driven formidable advances into the knowledge of CAD pathophysiology, particularly by addressing controversies on the causality of some lipid fractions that had long remained unsolved because of limitations of observational epidemiology. For example, NGS-derived rare variants with strong functional effects on key-genes like ANGPTL4, APOA5, APOC3, LPL, and SCARB1, have proven useful as proxies to demonstrate the causality of triglyceride-rich lipoproteins (TRLs) at variance with HDL-cholesterol concentration, thus contributing to tear down a dogma from classical epidemiology. Moreover, such variants have paved the way for the development of new biologic drugs (i.e. monoclonal antibodies or antisense oligonucleotides) targeting key proteins like PCSK9, Lipoprotein(a), and apolipoprotein C3. Such drugs are currently under active investigation, with first results being extremely promising.
Coronary Artery Disease (CAD), along with its main complication Myocardial Infarction (MI), remain the leading cause of death and disability worldwide (http://www.who.int/mediacentre/factsheets/fs317/en/), including in the developing countries where both lifespan and adoption of Western-like bad lifestyles are rising [
DALYs GBD, Collaborators H Global, regional, and national disability-adjusted life years (DALYs) for 306 diseases and injuries and healthy life expectancy (HALE) for 188 countries, 1990–2013: quantifying the epidemiological transition.
Clinicians approaching patients with suspected CAD are well aware of the importance of a positive family history for the disease, and hence of its genetic component. Indeed, classical epidemiological surveys like the Framingham Offspring Study [
], pointed out that subjects having a parent with premature CAD have a nearly two-fold increased risk, independent of classical risk factors. Studies on monozygotic twins, known as the best “natural experiments” to evaluate the fraction of phenotype variability attributable to genetic factors [
]), CAD is a typical example of “complex disease”, with a polygenic architecture interacting with a myriad of environmental and lifestyle risk factors [
]. Elucidation of the genetic basis of CAD has long been elusive until the recent decade, which has witnessed unprecedented progress in genomic medicine driven by both high throughput technologies and increased computational power (Fig. 1). Such advances hold promises of improving three major areas of practical medicine: i) prediction of CAD risk at individual levels; ii) therapeutic approaches based on novel pathophysiological insights; and iii) better use of existing therapies through pharmacogenetic (Fig. 1). Here we briefly review the practical implications of results obtained until now in the first two areas through major technological breakthroughs in genomics, i.e. genome wide association studies (GWAS) [
]. A further tool, i.e. the so-called Mendelian randomization (MR) approach that uses naturally occurring genetic variation to assess causality of risk factors from observational studies [
], has also substantially contributed to major advances. Relevant applications of the advances in cardiovascular pharmacogenomics are reviewed in detail elsewhere [
Fig. 1CAD is a complex disease in which lifestyle factors play a major role (thickened red arrow). Nonetheless, genetic factors also contribute, and CAD heritability has been estimated as high as 40–60% through twin studies in the pre-genomic era. The genetic basis of CAD has been elusive until a decade ago. Starting from 2007, a number of Genome Wide Association Studies (GWAS) have detected more than 60 common variants associated with CAD risk, with only a minor fraction of them being linked to known pathways and/or classical risk factors. By definition, such common alleles are not useful per se as predictors. They have been used to build up genetic risk scores that have been proved to be strongly associated with CAD risk, but still marginally useful (dotted gray arrows) in adding predictive value to classical models including family history. On the other hand, modern genetic approaches including next-generation sequencing (NGS) and Mendelian randomization studies have provided strong pathophysiological insights (thickened red arrows). For example, the LPL pathway is now recognized to play a role in atherogenesis, and HDL-cholesterol concentration is no longer viewed as protective per se, but rather a biomarker of impaired clearance of triglyceride-rich lipoproteins (TRLs), which in turn contain atherogenic, “non-HDL” cholesterol. Moreover, the identification of rare protective mutations has paved the way to the development of new biologic drugs mimicking their effects. Such drugs look very promising to further reduce CAD burden at population level in the coming years.
This narrative review is based on articles retrieved by the Authors through automatic weekly e-mail alerts from the National Center for Biotechnology Information (NCBI) at the U.S. National Library of Medicine (NLM), using the research string “coronary artery disease” AND “genetics”, from December 31, 2005 to March 6, 2017. Complete results were viewed in PubMed. The references of the retrieved papers were used to find more literature.
3. CAD prediction
3.1 Results from GWAS
Clinicians traditionally think about CAD genetics as the potential “Holy Grail” that should allow to identify individuals carrying risky mutations, in order to get an early diagnosis and to prescribe treatments able to prevent clinical manifestations [
]. In its shortest definition, a GWAS is an unbiased approach that uses chip arrays, containing up to >1,000,000 common (minor allele frequency – MAF > 5%, i.e. 1:20) Single Nucleotide Polymorphisms (SNPs) spanning throughout the human genome, to detect regions (loci) associated with a given phenotype [
]. An updated catalog of the impressive number of published GWAS on CAD and related traits can be found at http://www.ebi.ac.uk/gwas. Overall, >60 SNPs have shown robust association with CAD, the top twenty of which are reported in Table 1. A detailed list including all the remaining loci is reported elsewhere in a recent comprehensive review [
]. Noteworthy, only a minority (~40%) of all such risk variants appears to modulate CAD risk by influencing classical risk factors like plasma lipids, diabetes, and hypertension [
]. This is not surprising, considering that, as mentioned above, a positive family history remains a significant predictor of CAD even after full adjustment for traditional risk factors [
], but rather has biochemical functions and/or is transcribed into several regulatory elements, including long noncoding RNAs, able to modulate the expression of canonical genes [
]. A paradigm in this sense is represented by the 9p21.3 locus, which was the first one to emerge by three independent landmark GWAS published in 2007 [
]), the 9p21.3 locus remains the strongest genetic risk factor for CAD, with an impressive statistical power (P = 2 × 10−98). Indeed, the 9p21.3 genomic region appears to encode for a long noncoding RNA designated as ANRIL (antisense noncoding RNA in the INK4 locus) [
]. While the exact role of 9p21.3 risk variant(s) in promoting CAD is still not completely understood, the most recent data suggest that ANRIL influence the expression of neighboring genes (CDKN2B and/or 2A) encoding cell cycle inhibitor proteins (cyclin-dependent kinase inhibitor 2B/2A, cdkn2b, cdkn2A) expressed by several tissues including vascular smooth muscle cells [
]. Inadequate modulation of these inhibitory proteins is likely to promote cell proliferation within the arterial walls, which in turn would contribute to the development of the atheroma [
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.
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.
Because of intrinsic study methodology (see above), SNPs at 9p21.3 locus, like all GWAS-derived risk variants, are quite common in the general population. In this particular case, the MAF for the risk allele rs2891168 is 0.49, which means that near one fourth of the world population is homozygous [
]. Thus, it is not surprising that, notwithstanding the strength of the statistical association with CAD, the ensuing odds ratio (OR) for CAD of the individual 9p21.3 risk allele is low (1.21, with 95% CI 1.19–1.24 [
]), making it not at all useful in clinical practice when considered alone. This is true for all the GWAS-derived risk variants, whose individual ORs for CAD are in the range between 1.03 and 1.10 [
]. On the other hand, their frequencies make unavoidable that any given individual could carry simultaneously several risk variants. This represents the basis for calculating the genetic risk scores (GRSs) (Fig. 1), in the effort to eventually provide an effective tool for improving CAD prediction. In its simplest definition, a GRS is calculated by counting the number of “risk” alleles inherited by a given individual, each corrected for its own effect size, according to the following formula:
where n is the individual's genotype (0 = normal, 1 = heterozygous, 2 = homozygous) for the risk allele, and β is the effect size of the same allele (i.e. the log of the OR) derived from largest available GWAS. The resulting GRS is then used as a single variable added to classical prediction models. Several studies utilizing different GRSs (i.e. built up with different numbers and combinations of risk alleles) have shown that they are robustly associated with incident CAD, independent of classical risk factors [
]. However, this does not appear to translate into an actual usefulness in clinical practice. To this end, what is needed is the demonstration that incorporation of GRS into an established clinical prediction model (i.e. the Framingham Risk Score–FRS–, or ACC/AHA13 calculator [
2013 ACC/AHA guideline on the assessment of cardiovascular risk: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines.
]. With relatively few exceptions, the GRSs did not perform better than simple family history in improving individual's risk stratification after incorporation into existing clinical models [
]. Recently, Iribarren and colleagues found that application of GRSs including 8 to 51 SNPs to a cohort of 51,954 subjects of European-ancestry improved prediction of the FRS, although the net reclassification rate was modest (5–9%) [
Clinical utility of multimarker genetic risk scores for prediction of incident coronary heart disease: a cohort study among over 51 thousand individuals of European ancestry.
]. Tada and colleagues evaluated an extended GRS including 50 SNPs (“GRS-50”) in 23,595 Swedish, pointing out an effect independent of family history [
]. However, this is not surprising since self-reported family history in large population surveys is known to have a low accuracy and to reflect both genetic and non-genetic risk factors. A major general limitation of GWAS is that the risk variants explain only a minor fraction (no >20%) of total estimated CAD heritability [
] is hidden amongst the thousands of SNPs that did not reach genome-wide significance. Accordingly, Abraham and colleagues built up a “massive” GRS with 49,310 SNPs (“49K-GRS”), possibly reflecting the maximum available amount of genomic information, and successfully showed that it increased CAD prediction in FINRISK (n = 12,676) and Framingham Heart Study (n = 3406) prospective cohorts [
]. Again, however, the overall incremental predictive power of the 49K-GRS after incorporation into FRS and ACC/AHA13 models was very modest (C-index = +1.5%), except in certain subpopulations like individuals ≥ 60 years old (C-index = +4.6–5.1%). Arguments in favor of the possible use of the GRSs come from two studies that analyzed retrospectively DNA samples from patients participating to five major clinical trials on statin therapy (JUPITER, ASCOT, WOSCOPS, CARE and PROVE-IT-TIMI 22) [
Polygenic risk score identifies subgroup with higher burden of atherosclerosis and greater relative benefit from statin therapy in the primary prevention setting.
Polygenic risk score identifies subgroup with higher burden of atherosclerosis and greater relative benefit from statin therapy in the primary prevention setting.
] effectively discriminated patients with the highest genetic risk, who in turn derived the maximal benefit from statins. However, these intriguing results need validation in prospective trials. Moreover, a recent survey on 55,685 subjects from four studies still confirmed the association between GRSs (including up to 50 risk variants) and incident CAD, with subjects in the top GRS quintiles having a 91% higher relative risk as compared to those in the bottom quintiles [
]. Noteworthy, in any genetic risk category, the adherence to a healthy lifestyle was associated with a substantial risk reduction, up to nearly 50% in subjects with the highest genetic risk. Taken together, the results on GRS suggest at least two important practical lessons. First, the dream of a “perfect” prediction of CAD through the incorporation of genetic risk markers into more traditional models is likely unrealistic. The incremental benefit, if any, appears dull even with the addition of apparently robust and maximally informative markers integrating the most recent technological advances. Even though costs are consistently declining so that simultaneous genotyping of thousands of SNPs (i.e. for calculating the 49K-GRS) could be obtained with <100 €, the ensuing level of complexity makes unreliable the use of such information in daily clinical practice, particularly considering that an accurate family history (simple and inexpensive at same time) represents an excellent alternative. Second, widespread application of complex genetic tests at population level is unlikely to be cost-effective even in the future, since it is becoming evident that lifestyle factors can powerfully modify the risk of CAD regardless of individual's genetic risk profile. Thus, from a public health point of view, investing resources to promote adherence to healthy lifestyle should still remain a priority. Accordingly, there is no reason to disagree with recently updated guidelines by the European Society of Cardiology on cardiovascular disease prevention, which recommend against the use of genetic testing in clinical practice [
2016 European guidelines on cardiovascular disease prevention in clinical practice: the sixth joint task force of the European society of cardiology and other societies on cardiovascular disease prevention in clinical practice (constituted by representatives of 10 societies and by invited experts) developed with the special contribution of the European Association for Cardiovascular Prevention & Rehabilitation (EACPR).
The above mentioned rule has, however, a notable exception represented by Mendelian disorders strongly associated to premature CAD, mainly Familial Hypercholesterolemia (FH). FH is a disorder due to genetic defects in LDL-cholesterol (LDL-c) metabolism resulting in unusually high LDL-c and a markedly increased risk of early-onset CAD (for detailed reviews see [
Familial hypercholesterolaemia is underdiagnosed and undertreated in the general population: guidance for clinicians to prevent coronary heart disease: consensus statement of the European Atherosclerosis Society.
Homozygous familial hypercholesterolaemia: new insights and guidance for clinicians to improve detection and clinical management. A position paper from the Consensus Panel on Familial Hypercholesterolaemia of the European Atherosclerosis Society.
]. Homozygous FH is a rare disease with frequency variable from 1 in 1,000,000 (historical estimates) to 1 in 160,000 (most recent estimates in the Danish population) [
Familial hypercholesterolaemia is underdiagnosed and undertreated in the general population: guidance for clinicians to prevent coronary heart disease: consensus statement of the European Atherosclerosis Society.
Homozygous familial hypercholesterolaemia: new insights and guidance for clinicians to improve detection and clinical management. A position paper from the Consensus Panel on Familial Hypercholesterolaemia of the European Atherosclerosis Society.
]. Such extreme phenotype makes the diagnosis relatively easy. From a practical point of view, suspicion and diagnosis of heterozygous FH (which can be as common as 1 in 200 in general populations of European ancestry) is much more difficult, since LDL-c levels (generally >190 mg/dl) can overlap with other types of dyslipidemia, while retaining a substantially higher risk of developing CAD [
Familial hypercholesterolaemia is underdiagnosed and undertreated in the general population: guidance for clinicians to prevent coronary heart disease: consensus statement of the European Atherosclerosis Society.
Development of sensitive and specific age- and gender-specific low-density lipoprotein cholesterol cutoffs for diagnosis of first-degree relatives with familial hypercholesterolaemia in cascade testing.
Familial hypercholesterolaemia is underdiagnosed and undertreated in the general population: guidance for clinicians to prevent coronary heart disease: consensus statement of the European Atherosclerosis Society.
], with important public health implications. Indeed, FH has been designated by the U.S. Centers for Disease Control Office of Public Health Genomics as one of three top level genetic disorders (the others being BRCA1/2 related hereditary breast/ovarian cancer, and Lynch syndrome) with sufficient evidence for recommend implementation of case finding via all possible strategies, including genetic testing, to significantly improve public health [
]. Simultaneous, NGS-based, targeted sequencing of LDLR, APOB, and PCSK9, i.e. the three genes causally linked to the vast majority of FH, is now becoming widely available and cost-effective [
], so that national guidelines in Netherlands, Norway, and the U.K. currently recommend cascade screening using genetics and lipid profile after the identification of an index case [
]. The incremental values of genetic testing in FH over family history and clinical scores include certainty of the diagnosis in many cases, reduction of burden of undiagnosed patients, and improved outcomes in patients properly recognized and treated, altogether finally resulting absolute lowering of FH-related costs [
]. Noteworthy, Abul-Husn and colleagues recently applied exome sequencing to 50,726 individuals from a single U.S. health care system, showing that such genomic screening was effective in diagnosing FH patients (with an estimated prevalence of 1 in 256), most of whom were receiving inadequate lipid lowering therapy [
4. Pathophysiological insights and therapeutic implications
At variance with modest results in CAD prediction, genomic techniques have substantially contributed to new acquisitions in CAD pathophysiology, with relevant practical implications in terms of drug therapy. The identification of new targets is instrumental to the development of modern “biologic” or “intelligent” drugs, through either inactivating monoclonal antibodies (mAb) [
4.1 Proprotein Convertase Subtilisin/Kexin type 9 (PCSK9)
The most known and compelling example in this sense is Proprotein Convertase Subtilisin/Kexin type 9 (PCSK9), a liver-produced serin protease that accelerates the degradation of the LDL receptor [
], illustrates as only near a dozen years passed from protein/gene discovery (in 2003) to clinical trials showing the efficacy of target therapy through humanized monoclonal antibodies able to substantially lower LDL-cholesterol [
], an unprecedented fast-tracking process in cardiovascular drug development.
4.2 Lipoprotein(a)
A similar story can be depicted for Lipoprotein(a) (Lp(a), encoded by the LPA gene), although this LDL-like particle composed by apolipoprotein-B100 covalently bound to a distinct plasminogen-derived apolipoprotein (apolipoprotein(a)) [
], true causality remained unproven because of the lack of availability of drugs or other interventional strategies able to reduce Lp(a), and hence CAD risk. The scenario has recently clarified by the identification of LPA genetic variants associated with both elevated circulating Lp(a) and increased CAD risk [
]. A recent comprehensive analysis of such variants using data from both large-scale GWAS and NGS-based studies has consistently shown that one standard deviation (SD) genetically lowered is associated with a 29% lower risk of CAD [
Antisense oligonucleotides targeting apolipoprotein(a) in people with raised lipoprotein(a): two randomised, double-blind, placebo-controlled, dose-ranging trials.
A third, impressive example of genomic-driven pathophysiological insight is represented by recent downsizing of the causal role of HDL cholesterol (HDL-C) and, on the contrary, revitalized interest in triglyceride-rich lipoproteins (TRLs) [
]. The term TRLs designates very low-density lipoproteins (VLDL), chylomicrons, and their remnants. They contain both triglycerides and cholesterol (near 60% and 20% by mass in VLDL, respectively), as well as phospholipids and a number of proteins, including apolipoproteins C3 (ApoC3) and A5 (ApoA5) [
]. The major determinant of TRLs metabolism is represented by endothelial lipoprotein lipase (LPL), whose activity, in turn, is increased by ApoA5 and inhibited by ApoC3, as well as by circulating angiopoietin-like proteins 3 and 4 (Angptl3 and Angptl4) [
] (Fig. 1). Plasma triglycerides concentration represents a proxy of TRLs, and, at population level, it is known to be inversely related with HDL-C levels [
]. According to classic observational epidemiology, both high triglycerides and low HDL-C level were associated to CAD, but while the effect of triglycerides was substantially attenuated by adjusting for numerous covariates (including HDL-C), that of HDL-C appeared independent [
]. This led to the commonly accepted “dogma” that HDL-C levels were causally related with CAD, with high levels being protective, while triglycerides were considered no more than bystanders [
]. Functional and pathological studies supported this view, showing that HDL particles are involved in reverse cholesterol transport from atherosclerotic plaques, where, in turn, triglycerides do not accumulate, unlike cholesterol [
], at least with respect to its concentration. A first Mendelian randomization (MR) study used a variant (Asn396Ser) on the LIPG gene significantly associated with increased HDL-C levels (but without effects on other lipid fractions) as a probe to test causality, but, indeed, such variant was not associated with the expected reduced risk of MI in a meta-analysis of 20 studies including >116,000 participants [
] applied to HDL-C a very useful and innovative approach, namely targeted NGS of subjects with “extreme” phenotypes, where the probability to find mutations with large functional effects is substantially increased as compared to the general population [
]. In a relatively small group (n = 328) of subjects selected for having HDL-C levels > 95th percentile (mean level = 106.8 mg/dl) they detected a “human knock-out” homozygous for a loss of function [
DALYs GBD, Collaborators H Global, regional, and national disability-adjusted life years (DALYs) for 306 diseases and injuries and healthy life expectancy (HALE) for 188 countries, 1990–2013: quantifying the epidemiological transition.
]. A number of P376L heterozygous carriers were then found in a large population including 137,995 individuals (49,846 CAD cases and 88,149 controls), that in spite of having elevated HDL-C had increased (not decreased) CAD risk [
]. Finally, sequencing of the Icelandic population allowed the detection of rare mutations in the ASGR1 gene (encoding a subunit of the asialoglycoprotein receptor, involved in the homeostasis of circulating glycoproteins), which were “paradoxically” associated with both reduced HDL-C levels and reduced CAD risk [
]. A further argument seriously questioning the “dogma” of inverse causality of HDL-C in CAD/MI is the proven futility of a drug increasing HDL-cholesterol (the CETP inhibitor dalcetrapib) in a clinical trial evaluating the risk of recurrent CVD [
]. On the other hand, a number of NGS-based studies focusing on the main genes involving in modulation of the LPL pathway (APOA5, APOC3, ANGPTL4, and the LPL gene itself), often using the same strategy as above, i.e. initially focusing on individuals with extremely high or low plasma triglycerides levels, have yielded results univocally in favor of a causal role of TRLs in CAD (see Fig. 2 for a synopsis). Indeed, they found that carriers of mutations leading to increased LPL function (i.e. LoF mutations in APOC3 [
]) had both low plasma triglycerides and decreased risk of CAD, while carriers of mutations leading to decreased LPL function (i.e. LoF mutations in APOA5 [
]) had both high plasma triglycerides and increased risk of CAD (Fig. 2). Helgadottir and colleagues took advantage from the reduced allele diversity of the Icelandic population for a comprehensive study combining whole genome sequencing, GRS, and MR approaches to infer causality of all major plasma lipid fractions commonly measured in clinical practice (LDL-C, HDL-C, triglycerides, and non-HDL-C) [
]. They found that non-HDL-C GRS associated most strongly with CAD (P = 2.7 × 10−28), and no other GRS associated with CAD after accounting for non-HDL-C. In summary, according to the recent studies using sequence variants as proxies to infer causality, the long-lasting controversy on the role of HDL-C or triglycerides in CAD appears to be eventually solved as follows: i) HDL-C concentration per se is not causally linked to CAD (although concentration does not necessarily reflect the flux of HDL particles and their potential function in promoting reverse cholesterol transport [
]); ii) TRLs are causally linked to CAD; iii) however, the atherogenic role of TRLs is not due to triglycerides, but rather to their cholesterol content. The first practical insight is that clinicians should pay even more attention to non-HDL-C than to LDL-C [
Triglyceride-rich lipoproteins and high-density lipoprotein cholesterol in patients at high risk of cardiovascular disease: evidence and guidance for management.
], since this measure accounts for all atherogenic cholesterol-containing lipoproteins, including LDL, VLDL, intermediate-density lipoprotein, chylomicrons and their remnants, and Lp(a) [
Remnant cholesterol as a cause of ischemic heart disease: evidence, definition, measurement, atherogenicity, high risk patients, and present and future treatment.
], particularly considering that plasma APOC3 levels are strong and independent predictors of total and cardiovascular mortality in patients with established CAD [
Apolipoprotein C-III predicts cardiovascular mortality in severe coronary artery disease and is associated with an enhanced plasma thrombin generation.
Fig. 2The LPL-APOC3-APOA5-ANGPTL4 pathway. The lipoprotein lipase (LPL), a key enzyme in reducing plasma triglycerides, is attached to the capillary endothelium by the protein GPIHBP1. LPL acts by hydrolyzing the triglycerides present in the triglyceride-rich lipoproteins (VLDL and chylomicrons), and is critically controlled by the apolipoproteins APOC3 and APOA5, as well as by circulating angiopoetin like protein 4 (ANGPTL4). APOC3 and ANGPTL4 inhibit LPL, while APOA5 has a stimulatory effect. LoF mutations in APOC3 and in ANGPTL4 are therefore associated with low plasma triglycerides and protection from CAD, while LoF mutations in APOA5 and LPL are associated with high plasma triglycerides and increased CAD risk. Such mutations, discovered through NGS, have been used as proxies to infer causality of triglyceride-rich lipoproteins in CAD, which appears mediated by their cholesterol content (see text).
Of note, the three examples outlined above (anti-PCSK9, anti-Lp(a), and anti-APOC3 novel biologic therapies) well illustrate the extraordinary potential of genomics to guide efficient drug development [
], by discerning causal pathways, as well as by identifying genetic variations that can anticipate potential efficacy and safety of targeted inhibitory compounds [
], that not only are protected against CAD, but also live well without a given protein.
5. Conclusions and future perspectives
The last decade has witnessed unprecedented advances in understanding of the genetic architecture of CAD, which appears far more complex than expected. Regarding prediction, although a number of genetic markers have been identified through GWAS and incorporated into promising genetic risk scores, for the moment their use in clinical practice cannot be recommended. On the other hand, the impressive technological progress in human genomics has already provided important insights into disease pathobiology and novel therapeutic approaches. A lot of work remains to be done, particularly regarding the loci whose mechanistic link to CAD is still unknown. These loci are of paramount interest for future studies aimed to dissect new pathways, and, possibly, new therapeutic targets. For example, virtually nothing is known about the possible mechanism underlying the strong association (P = 1.8 × 10−42) between the 6p24.1 locus (i.e. the second most often identified after the 9p21.3, containing the PHACTR1 gene encoding for phosphatase and actin regulator 1) and CAD [
], requiring not only extremely large sample size (hundred of thousands, at minimum) and “deep” sequencing, but also “deep phenotyping”, i.e. meticulous gathering of clinical details at individual level [
]. Indeed, for example, most studies published so far have merged “CAD” and “MI” phenotypes. While instrumental to increase sample size, this approximation is intrinsically unable to dissect pathways involved in atheroma development (CAD) rather than in thrombotic complications (MI), which are expected to overlap only partially. To this end, an increasing participation of clinicians in large genomic studies, along with molecular biologists, epidemiologists, and bioinformatics, should be encouraged.
Acknowledgments
This work was performed (in part) in the LURM (Laboratorio Universitario Ricerca Medica) Research Center, University of Verona. This work was partially supported by the Cariverona Foundation (n. 2014.0851 and 2015.0872).
References
Wong N.D.
Epidemiological studies of CHD and the evolution of preventive cardiology.
Global, regional, and national disability-adjusted life years (DALYs) for 306 diseases and injuries and healthy life expectancy (HALE) for 188 countries, 1990–2013: quantifying the epidemiological transition.
2013 ACC/AHA guideline on the assessment of cardiovascular risk: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines.
Clinical utility of multimarker genetic risk scores for prediction of incident coronary heart disease: a cohort study among over 51 thousand individuals of European ancestry.
Polygenic risk score identifies subgroup with higher burden of atherosclerosis and greater relative benefit from statin therapy in the primary prevention setting.
2016 European guidelines on cardiovascular disease prevention in clinical practice: the sixth joint task force of the European society of cardiology and other societies on cardiovascular disease prevention in clinical practice (constituted by representatives of 10 societies and by invited experts) developed with the special contribution of the European Association for Cardiovascular Prevention & Rehabilitation (EACPR).
Familial hypercholesterolaemia is underdiagnosed and undertreated in the general population: guidance for clinicians to prevent coronary heart disease: consensus statement of the European Atherosclerosis Society.
Homozygous familial hypercholesterolaemia: new insights and guidance for clinicians to improve detection and clinical management. A position paper from the Consensus Panel on Familial Hypercholesterolaemia of the European Atherosclerosis Society.
Development of sensitive and specific age- and gender-specific low-density lipoprotein cholesterol cutoffs for diagnosis of first-degree relatives with familial hypercholesterolaemia in cascade testing.
Antisense oligonucleotides targeting apolipoprotein(a) in people with raised lipoprotein(a): two randomised, double-blind, placebo-controlled, dose-ranging trials.
Triglyceride-rich lipoproteins and high-density lipoprotein cholesterol in patients at high risk of cardiovascular disease: evidence and guidance for management.
Remnant cholesterol as a cause of ischemic heart disease: evidence, definition, measurement, atherogenicity, high risk patients, and present and future treatment.
Apolipoprotein C-III predicts cardiovascular mortality in severe coronary artery disease and is associated with an enhanced plasma thrombin generation.
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.
30123-1&id=fx1.jpg){kind=link}
30123-1&id=gr1.jpg){kind=link}
30123-1&id=gr2.jpg){kind=link}