Genetic Risk For Coronary Artery Disease Research Articles

Abstract 3074: RNA Editing Enzyme ADAR1 In Vascular Smooth Muscle Controls Vascular Integrity And Progression Of Atherosclerosis

Adenosine-to-inosine (A-to-I) RNA editing is a common RNA modification catalyzed by ADAR enzymes. ADAR1 serves a crucial function to edit specific immunogenic double stranded RNA (dsRNA) to prevent the dsRNA sensor, MDA5 ( IFIH1 ), activating an interferon stimulated gene (ISG) response. Discoveries in human genetics now implicate deficient ADAR1 RNA editing as a causal mechanism for coronary artery disease (CAD) risk. Here, we show that human atherosclerotic smooth muscle cells (SMCs) express immunogenic RNA at markedly higher levels compared to all other cell types. We demonstrate that ADAR1 is the master regulator of RNA editing in primary human coronary artery SMC (HCASMC) and loss of RNA editing causes activation of MDA5, characterized by upregulation of ISG genes (i.e. ISG15 ) and others with known roles in SMC phenotypic transition and CAD risk (i.e. KLF4 ). ADAR1 and MDA5 regulate SMC phenotypic modulation and calcification in vitro . In a high fat diet atherosclerosis model, scRNAseq analysis revealed SMC to have increasing ISG expression with phenotypic modulation. To evaluate SMC RNA editing in vivo , we generated a conditional SMC specific Adar1 KD mouse ( Adar1 flox/flox , Myh11 CreERT2 , ROSA tdT/+ , ApoE -/- ). At 2 weeks following SMC- Adar1 KD, we observed 50% mortality with a severe phenotype of elastin degradation and disarray, intravascular hemorrhage, and inflammatory cell infiltration in the aortic wall. scRNAseq of aortic tissue showed that SMC- Adar1 KD causes ISG activation and coordinates a complex transcriptional response throughout the vessel wall with robust macrophage cell infiltration. Through evaluation of ligand-receptor interaction, we show that Ccl5:Ccr5 is the principal regulator of cell infiltration following SMC- Adar1 KD. Through study of SMC- Adar1 haploinsufficiency in the atherosclerosis mouse model, we further reveal that MDA5 activation occurs in SMC phenotypic modulation and accelerates formation of chondromyocytes — indicating a cell type and context specific requirement of RNA editing. Through this work, we describe a fundamental new mechanism of CAD, where RNA editing and sensing of dsRNA mediates disease progression, promoting our understanding of disease causality and CAD genetic risk.

Abstract 2072: The 9p21.3 Coronary Artery Disease Risk Locus Induces Cell State Transitions In Vascular Smooth Muscle Cells

Background: Genome-wide association studies have identified common genetic variants (Single Nucleotide Polymorphisms - SNPs) at ~300 human genomic loci linked to coronary artery disease (CAD) susceptibility. Among these genomic regions, the most impactful is the 9p21.3 CAD risk locus, which spans a 60 kb gene desert and encompasses ~80 SNPs in high linkage disequilibrium. We have previously generated isogenic induced Pluripotent Stem Cells (iPSCs) lines from risk and non-risk donors at 9p21.3 and performed a complete deletion of the locus. By using iPSC-derived vascular smooth muscle cells (iPSC-VSMCs) we demonstrated that the risk haplotype at 9p21.3 causes altered expression of numerous genes essential for muscle function, including contraction and adhesion. Notably, deletion of the risk haplotype restores the non-risk phenotype, suggesting a gain of function effect. Hypothesis: The 9p21.3 risk haplotype alters VSMCs plasticity and induces dedifferentiation and alternative lineage acquisition. Aims: Identify transcriptomic signatures and trajectories of VSMCs induced by the 9p21.3 risk haplotype Methods: We have used iPSCs from individuals carrying the risk and non-risk haplotype at 9p21.3 and isogenic knockout lines. After VSMCs differentiation induction, mature VSMCs were used for single cell RNA sequencing using 10X Chromium platform. Results: Our analysis revealed that the 9p21.3 risk haplotype prompts VSMCs to acquire a novel cellular state showing reduced plasticity and divergent from other VSMC states previously linked to CAD. We found markers of alternative lineages, including osteogenic and neuronal, coupled with altered expression of genes within other CAD loci. We identified a set of new molecular markers crucial to define risk-associated VSMCs and uncovered their upstream regulators. Conclusions: Our study provides insights into CAD pathogenesis driven by the 9p21.3 risk locus and identified new gene regulatory networks essential for maintaining the normal functionality of the muscle layer of the arteries. Leveraging the power of iPSCs we present novel concepts about the biology of VSMCs and shed light on the impact of CAD genetic risk factors in preserving a healthy cellular state within the vasculature.

Abstract 16384: Interaction of Socioeconomic Status With Polygenic Risk for Coronary Artery Disease in a Novel Biobank

Introduction: Socioeconomic status (SES) has a measurable and significant impact on cardiovascular disease incidence and outcomes. Study of its potential interactions with other known risk factors, such as genetics, remains an area for further research. Aim: To examine the interaction of polygenic risk for coronary artery disease (CAD) with socioeconomic status, as represented by income bracket. Methods: We utilized the NIH All of Us biobank, which integrates whole genome sequencing, electronic health record, and survey data. A previously-validated PRS for CAD (Inouye et al., 2018) was calculated for each patient. We subsetted the sample into high, medium, and low genetic risk groups per z-scores for the PRS: z-score >1, between 1 and -1, and <-1 respectively. Income information was derived from the survey data. Multivariable logistic regression was used to assess the performance of the PRS in this sample, with outcome defined as presence of CAD. The model was adjusted for PRS, age, gender, and the first 4 principal components of ancestry. Both income and the interaction of PRS by income were included as covariates. Results: In this sample of 97,405 individuals, this PRS is strongly associated with CAD prevalence: adjusted odds ratio for CAD per SD of the PRS was 1.96 (95% CI, 1.83-2.10). We then observed a differential effect of income on CAD. In the low and medium genetic risk groups, increasing income was associated with lower CAD prevalence. In the high genetic risk group, this protective effect was lost and CAD prevalence was overall similar across income brackets. Conclusions: These findings suggest a complex interaction between genetic risk of CAD and SES. High risk individuals may not derive the same benefit of improved SES and its material benefits. This unique subgroup will require uniquely targeted interventions, highlighting the need for deeper study of genetic risk in order to actualize the promises of precision medicine.

Genetic Regulation of SMC Gene Expression and Splicing Predict Causal CAD Genes.

Coronary artery disease (CAD) is the leading cause of death worldwide. Recent meta-analyses of genome-wide association studies have identified over 175 loci associated with CAD. The majority of these loci are in noncoding regions and are predicted to regulate gene expression. Given that vascular smooth muscle cells (SMCs) play critical roles in the development and progression of CAD, we aimed to identify the subset of the CAD loci associated with the regulation of transcription in distinct SMC phenotypes. We measured gene expression in SMCs isolated from the ascending aortas of 151 heart transplant donors of various genetic ancestries in quiescent or proliferative conditions and calculated the association of their expression and splicing with ~6.3 million imputed single-nucleotide polymorphism markers across the genome. We identified 4910 expression and 4412 splicing quantitative trait loci (sQTLs) representing regions of the genome associated with transcript abundance and splicing. A total of 3660 expression quantitative trait loci (eQTLs) had not been observed in the publicly available Genotype-Tissue Expression dataset. Further, 29 and 880 eQTLs were SMC-specific and sex-biased, respectively. We made these results available for public query on a user-friendly website. To identify the effector transcript(s) regulated by CAD loci, we used 4 distinct colocalization approaches. We identified 84 eQTL and 164 sQTL that colocalized with CAD loci, highlighting the importance of genetic regulation of mRNA splicing as a molecular mechanism for CAD genetic risk. Notably, 20% and 35% of the eQTLs were unique to quiescent or proliferative SMCs, respectively. One CAD locus colocalized with a sex-specific eQTL (TERF2IP), and another locus colocalized with SMC-specific eQTL (ALKBH8). The most significantly associated CAD locus, 9p21, was an sQTL for the long noncoding RNA CDKN2B-AS1, also known as ANRIL, in proliferative SMCs. Collectively, our results provide evidence for the molecular mechanisms of genetic susceptibility to CAD in distinct SMC phenotypes.

Abstract 11954: Metabolomic Profiles Differentiate Arterial and Venous Polygenic Risk

Introduction : Genome-wide association studies reveal genetic correlations between venous and arterial cardiovascular phenotypes, such as venous thromboembolism (VTE), coronary artery disease (CAD), and peripheral arterial disease (PAD). The extent to which these diseases share metabolomic signatures has not been established. Methods : Polygenic risk scores (PRS) for CAD, PAD, and VTE were constructed using PRS-CS with external summary statistics and tested on ~120,000 UK Biobank participants with metabolomic profiling. Linear regression adjusted for age, sex, genotyping array, the first five principal components of ancestry, and statin prescription estimated the effect of polygenic disease risk on 249 metabolites among participants free of prevalent cardiovascular disease. For all metabolomic associations common to at least two diseases, bidirectional two-sample Mendelian randomization tested the causal relationship between metabolite and phenotype. Results : We found 206, 201, and 116 metabolites were associated with genetic risk for CAD, PAD, and VTE, respectively, at an adjusted p-value of 0.001 ( Figure ). Metabolomic profiles were similar between CAD and PAD (R = 0.75, P = 1.25E-46) and CAD and VTE (R = 0.76, P = 8.27E-49), but not between VTE and PAD (R = 0.24, P = 1.21E-4). Mendelian randomization revealed 28 metabolites that increased risk for both CAD and PAD and 2 metabolites that increased risk for CAD but decreased risk for VTE. There were no metabolites with shared causal relationship between PAD and VTE. Conclusions : Polygenic risks for arterial and venous disease are associated with unique metabolomic profiles. Mendelian randomization reveals overlap in causal metabolites between CAD and PAD but little overlap between VTE and either CAD or PAD despite moderate genetic correlation.

Prediction of coronary artery disease and major adverse cardiovascular events using clinical and genetic risk scores for cardiovascular risk factors

Background and Aims : Risk stratification of coronary artery disease (CAD) and major adverse cardiovascular events (MACE) remains suboptimal. CAD genetic risk scores (GRSs) predict risk independently from the QRISK3 score. We assessed the added value of GRSs for cardiovascular traits (CV GRSs).

Prediction of Coronary Artery Disease and Major Adverse Cardiovascular Events Using Clinical and Genetic Risk Scores for Cardiovascular Risk Factors.

Background:Coronary artery disease (CAD) and major adverse cardiovascular events (MACE) are the leading causes of death in the general population, but risk stratification remains suboptimal. CAD genetic risk scores (GRSs) predict risk independently from clinical tools, like QRISK3. We assessed the added value of GRSs for a variety of cardiovascular traits (CV GRSs) for predicting CAD and MACE and tested their early-life screening potential by comparing against the CAD GRS only.Methods:We used data from 379 581 participants in the UK Biobank without known cardiovascular conditions (follow-up, 11.3 years; 3.3% CAD cases and 5.2% MACE cases). In a training subset (50%) we built 3 scores: QRISK3; QRISK3 and an established CAD GRS; and QRISK3, the CAD GRS and the CV GRSs. In an independent subset (50%), we evaluated each score’s performance using the concordance index, odds ratio and net reclassification index. We then repeated the analyses without considering QRISK3.Results:For CAD, the combination of QRISK3 and the CAD GRS had a better performance than QRISK3 alone (concordance index, 0.766 versus 0.753; odds ratio, 5.47 versus 4.82; net reclassification index, 7.7%). Adding the CV GRSs did not significantly improve risk stratification. When only looking at genetic information, the combination of CV GRSs and the CAD GRS had a better performance than the CAD GRS alone (concordance index, 0.637 versus 0.625; odds ratio, 2.17 versus 2.07; net reclassification index, 3.3%). Similar results were obtained for MACE.Conclusions:In individuals without known cardiovascular disease, the inclusion of CV GRSs to a clinical tool and an established CAD GRS does not improve CAD or MACE risk stratification. However, their combination only with the CAD GRS increases prediction performance indicating potential use in early-life screening before the advanced development of conventional cardiovascular risk factors.

Abstract 12985: Genetic Susceptibility for Coronary Artery Disease Accentuates the Cardiovascular Disease Risk Associated With Air Pollution

Introduction: Air pollution (particulate matter of diameter <2.5 μm, PM 2.5 ) and genetic risk for coronary artery disease (CAD) each associate with cardiovascular disease (CVD). It is unknown whether genetic risk for CAD enhances CVD risk associated with air pollution exposure. Hypothesis: We tested the hypothesis that high genetic risk for CAD accentuates the adverse impact of air pollution on CVD, (as coronary artery calcium [CAC]) and major adverse cardiovascular events [MACE]). Methods: Participants (N=9340, median age 66 yrs, 49% male) in the Mass General Brigham Biobank with genetic and air pollution data were studied. Genome wide polygenic risk score for CAD (PRS CAD ) and principal components of ancestry (PCI) were determined. Average PM 2.5 exposure one year before each person’s enrollment was derived using home addresses and Environmental Protection Agency data. CAC was assessed on those with available CT images (N=548). MACE and CVD risk factors were evaluated via International Classification of Diseases codes. Socioeconomic status (SES) was assessed as residential zip code level income using US Census data. The presence of an interaction between high PM 2.5 and PRS CAD vs. MACE was tested. Results: The Combination of high PRS CAD and PM 2.5 (both ≥ median vs. others) associated with greater CAC [standardized β (95% CI): 0.13 (0.06, 0.19), p=0.0002, Fig.1A ] after adjusting for age, sex, and PCI and with further adjustment for CVD risk factors (p=0.007). Over 5 years median follow up, 1139 subjects experienced first time MACE. Combined high PRS CAD and PM 2.5 also associated with higher MACE risk [OR (95% CI): 1.20 (1.15, 1.25), p<0.001, p (interaction) <0.001, Fig.1B ] after adjusting for age, sex, and PCI and with further adjustment for CVD risk factors (p<0.001) and SES (p<0.001). Conclusions: Among individuals exposed to air pollution, genetic risk for CAD accentuates CAC and MACE risk in a graded manner. Genetic context may impact the health effects of air pollution.

Monogenic and Polygenic Models of Coronary Artery Disease.

Coronary artery disease (CAD) is a common disease globally attributable to the interplay of complex genetic and lifestyle factors. Here, we review how genomic sequencing advances have broadened the fundamental understanding of the monogenic and polygenic contributions to CAD and how these insights can be utilized, in part by creating polygenic risk estimates, for improved disease risk stratification at the individual patient level. Initial studies linking premature CAD with rare familial cases of elevated blood lipids highlighted high-risk monogenic contributions, predominantly presenting as familial hypercholesterolemia (FH). More commonly CAD genetic risk is a function of multiple, higher frequency variants each imparting lower magnitude of risk, which can be combined to form polygenic risk scores (PRS) conveying significant risk to individuals at the extremes. However, gaps remain in clinical validation of PRSs, most notably in non-European populations. With improved and more broadly utilized genomic sequencing technologies, the genetic underpinnings of coronary artery disease are being unraveled. As a result, polygenic risk estimation is poised to become a widely used and powerful tool in the clinical setting. While the use of PRSs to augment current clinical risk stratification for optimization of cardiovascular disease risk by lifestyle change or therapeutic targeting is promising, we await adequately powered, prospective studies, demonstrating the clinical utility of polygenic risk estimation in practice.

A Less than Provocative Approach for the Primary Prevention of CAD.

Coronary artery disease (CAD) risk increases in proportion to the magnitude and duration of exposure to plasma low-density lipoprotein cholesterol (LDL-C), doubling every additional decade of exposure. Early primary prevention is three times more effective than initiated later. Several clinical trials show plasma LDL-C of 15-40 mg/dL is more effective and equally safe as the Current Cardiovascular Clinical Practice Guidelines (CCCPG) recommended target of 70mg/dL. The cholesterol in the blood is the excess synthesized by the cells and secreted into the blood for disposal in the liver. The CCCPG is inadequate since traditional risk factors (TRF) are not detectable until the sixth and seventh decade. The genetic risk score (GRS) evaluated in 1 million individuals as a risk stratifier for CAD is superior to TRF. Genetic risk for CAD was reduced by 30-50% by statin therapy, PCSK9 inhibitors, and lifestyle changes. The GRS does not change during one's lifetime and is inexpensive. Incorporating genetic risk stratification into CCCPG would induce a paradigm shift in the primary prevention of CAD.

Risk prediction for coronary heart disease by a genetic risk score - results from the Heinz Nixdorf Recall study

BackgroundA Genetic risk score for coronary artery disease (CAD) improves the ability of predicting coronary heart disease (CHD). It is unclear whether i) the use of a CAD genetic risk score is superior to the measurement of coronary artery calcification (CAC) for CHD risk assessment and ii) the CHD risk assessment using a CAD genetic risk score differs between men and women.MethodsWe included 4041 participants (age-range: 45–76 years, 1919 men) of the Heinz Nixdorf Recall study without CHD or stroke at baseline. A standardized weighted CAD genetic risk score was constructed using 70 known genetic variants. The risk score was divided into quintiles (Q1-Q5). We specified low (Q1), intermediate (Q2-Q4) and high (Q5) genetic risk groups. Incident CHD was defined as fatal and non-fatal myocardial infarction, stroke and coronary death. The association between the genetic risk score and genetic risk groups with incident CHD was assessed using Cox models to estimate hazard ratios (HR) and 95%-confidence intervals (CI). The models were adjusted by age and sex (Model1), as well as by established CHD risk factors (RF) and CAC (Model2). The analyses were further stratified by sex and controlled for multiple testing.ResultsDuring a median follow-up time of 11.6 ± 3.7 years, 343 participants experienced CHD events (219 men). Per-standard deviation (SD) increase in the genetic risk score was associated with 18% increased risk for incident CHD (Model1: p = 0.002) which did not change after full adjustment (Model2: HR = 1.18 per-SD (p = 0.003)). In Model2 we observed a 60% increased CHD risk in the high (p = 0.009) compared to the low genetic risk group. Stratifying by sex, only men showed statistically significantly higher risk for CHD (Model2: HR = 1.23 per-SD (p = 0.004); intermediate: HR = 1.52 (p = 0.04) and high: HR = 1.88 (p = 0.008)) with no statistically significant risk observed in women.ConclusionOur results suggest that the CAD genetic risk score could be useful for CHD risk prediction, at least in men belonging to the higher genetic risk group, but it does not outbalance the value of CT-based quantification of CAC which works independently on both men and women and allows better risk stratification in both the genders.

Poly(ADP-Ribose) Polymerase Activity and Coronary Artery Disease in Type 2 Diabetes Mellitus: An Observational and Bidirectional Mendelian Randomization Study.

Experimental evidence suggests a close link between PARP (poly[ADP-ribose] polymerase) activation and diabetic endothelial dysfunction. Here, we tested whether PARP activity in circulating leukocytes was associated with coronary artery disease (CAD) among patients with type 2 diabetes mellitus (T2DM). Approach and Results: We performed observational and bidirectional Mendelian randomization studies of 3149 Chinese individuals with T2DM who underwent coronary angiography, with leukocyte PARP activity, 16 tag single-nucleotide polymorphisms in PARP1 and PARP2, and 17 CAD risk single-nucleotide polymorphisms analyzed. Of 3149 participants, 1180 who further received percutaneous coronary intervention were prospectively followed for 1 year to track major adverse cardiovascular and cerebrovascular events. Overall, greater PARP activity was cross-sectionally associated with an odds ratio of 1.23 for obstructive CAD, and prospectively with a hazard ratio of 1.34 for 1-year major adverse cardiovascular and cerebrovascular events after percutaneous coronary intervention (both P<0.001). Using a genetic score of 5 screened single-nucleotide polymorphisms in PARP1 and PARP2 as the instrumental variable, genetically predicted elevation in PARP activity showed a causal association with obstructive CAD (odds ratio=1.35, P<0.001). In contrast, the genetic risk of CAD had no significant effect on PARP activity. Ex vivo and in vitro cultures of human monocytes showed that rs747657, as the lead single-nucleotide polymorphism strongly associated with PARP activity, caused the differential binding of transcription factor GATA2 (GATA-binding protein 2) to an intronic regulatory region in PARP1, thus modulating PARP1 expression and PARP activity. Greater PARP activity may have causal roles in the development of obstructive CAD among patients with diabetes mellitus.

Genetic risk scores for coronary artery disease and its traditional risk factors: Their role in the progression of coronary artery calcification-Results of the Heinz Nixdorf Recall study.

BackgroundAtherosclerosis is the primary cause of coronary artery disease (CAD). Several observational studies have examined the association of traditional CAD risk factors with the progression of coronary artery calcification (CAC). In our study we investigated the effect of 11 different genetic risk scores associated with CAD and CAD risk factors on the progression of CAC.Methods and resultsWe included 3097 participants from the Heinz Nixdorf Recall study who had available CAC measurements at baseline (CACb) and at the 5-year follow-up (CAC5y). A weighted genetic risk score for CAD and each of the CAD-associated risk factors was constructed. Multiple regression analyses were applied to i) the difference between the observed log(CAC5y+1) (log(obs)) and expected log(CAC5y+1) (log(exp)) at the 5-year follow-up following the individual’s log(CACb+1) percentile for the time between scans (log(obs)–log(exp)) and ii) the 5-year CAC progression, defined as 5*(log(CAC5y+1)–log(CACb+1))/time between the scans, adjusted for age, sex, and log(CACb+1) as well as for risk factors. The median percent deviation from the expected (CAC5y+1) and the 5-year progression of (CAC+1) in our study were 0 (first quartile: Q1; third quartile: Q3: -0.32; 0.48) and 45.4% (0%; 171.0%) respectively. In the age-, sex- and log(CACb+1)-adjusted model, the per-standard deviation (SD) increase in CAD genetic risk score was associated with the percent deviation from the expected (CAC5y+1) (9.7% (95% confidence interval: 5.2%; 14.5%), p = 1.6x10-5) and the 5-year progression of CAC (7.1% (3.0%; 11.4%), p = 0.0005). The CAD genetic risk score explains an additional 0.6% of the observed phenotypic variance for “log(obs)–log(exp)” and 0.4% for 5-year progression of CAC. Additionally, the per-SD increase in the CAC genetic risk score was associated with the percent deviation from the expected (CAC5y+1) (6.2% (1.9%; 10.8%, p = 0.005)) explaining an additional 0.2% of the observed phenotypic variance. However, the per-SD increase in the CAC genetic risk score was not associated with the 5-year progression of CAC (4.4% (0.4%; 8.5%), p = 0.03) after multiple testing. Adjusting for risk factors did not change the results. None of the other genetic risk scores showed an association with the percent deviation from the expected (CAC5y+1) or with the 5-year progression of CAC.ConclusionsThe association of the CAC genetic risk score and the CAD genetic risk score provides evidence that genetic determinants for CAC and CAD influence the progression of CAC.

Polygenic Scores to Assess Atherosclerotic Cardiovascular Disease Risk: Clinical Perspectives and Basic Implications.

An individual's susceptibility to atherosclerotic cardiovascular disease is influenced by numerous clinical and lifestyle factors, motivating the multifaceted approaches currently endorsed for primary and secondary cardiovascular disease prevention. With growing knowledge of the genetic basis of atherosclerotic cardiovascular disease-in particular, coronary artery disease-and its contribution to disease pathogenesis, there is increased interest in understanding the potential clinical utility of a genetic predictor that might further refine the assessment and management of atherosclerotic cardiovascular disease risk. Rapid scientific and technological advances have enabled widespread genotyping efforts and dynamic research in the field of coronary artery disease genetic risk prediction. In this review, we describe how genomic analyses of coronary artery disease have been leveraged to create polygenic risk scores. We then discuss evaluations of the clinical utility of these scores, pertinent mechanistic insights gleaned, and practical considerations relevant to the implementation of polygenic risk scores in the health care setting.

Abstract P457: Age Dependence of Genetic Risk Scores in Relation to Coronary Artery Disease

Background: Coronary artery disease (CAD) is an age-related multi-factorial disease with a high public health burden. Genetic loci for CAD have been discovered and collectively predict CAD risk independent of traditional risk factors. It remains unclear whether age modifies the genetic predisposition to CAD. Methods: A total of 173,315 Caucasian subjects from the UK Biobank were included in this study, including 26,472 CAD patients with age of onset ranging from 32 to 78 years old, and 146,843 control subjects &gt; 60 years old without CAD. CAD diagnosis was extracted from in-patient Hospital Episode Statistics (HES) and National Death Registry based on ICD-10 and OPCS-4 codes. CAD genetic risk score (GRS) was calculated based on the 161 known CAD loci. Sub-GRSs for body mass index (BMI, 16 loci), blood pressure (BP, 26 loci) and lipids (17 loci) were constructed to examine the genetic risk for CAD from certain intermediate trait. The CAD GRSs among onset age groups were compared to the controls using linear regression models, controlling for sex, BMI, creatinine, c-reactive protein, high density lipoprotein, low density lipoprotein, triglyceride, diabetes, heart failure history, hypertension and smoking status. The sub-GRSs were compared using the same models, excluding sub-GRS specific intermediate traits. Interactions of age groups and GRSs were tested using logistic regression models for CAD status, controlling for the same variables described above. Association analyses of each GRS were performed within each category of CAD age at onset comparing to controls. Results: Compared to the controls, the mean CAD GRS in CAD patients with age of onset ≤ 50 years was 0.58 standard deviations (SD) higher (95% CI 0.53-0.63), followed by CAD age of onset between 50 and 60 years (0.50 SD higher, 95% CI 0.48-0.52), between 60 and 70 years (0.35 SD higher, 95% CI 0.33-0.37), and &gt; 70 years (0.25 SD higher, 95% CI 0.21-0.29). The sub-GRSs showed the similar trend. Age of onset stratified analysis revealed a significant CAD GRS-age interaction (p&lt;0.001) on CAD risk where genetic effect is stronger among those early onset CAD patients - the odds ratios (OR) per CAD GRS SD among subjects with CAD onset age ≤ 50, 50 to 60, 60 to 70, &gt; 70 years were 1.80 (95% CI 1.71-1.89), 1.66 (95% CI 1.62-1.70), 1.44 (1.41-1.47), 1.29 (95%CI 1.24-1.35), respectively. Among the three sub-GRSs, significant GRS-age interactions were observed for BP sub-GRS and lipids sub-GRS but not BMI sub-GRS. Conclusion: The observed gene-age interaction indicates that the genetic susceptibility for CAD decreases for older subjects. The contribution of the CAD GRS to CAD risk is almost 64% lower in those developing CAD at age &gt; 70 years compared to those ≤ 50 years. Future considerations and applications of CAD genetic risk in precision medicine needs to include age of onset to accurately evaluate the genetic contribution.

Abstract P148: Sex Differences in the Genetic Predisposition of Coronary Artery Disease

While coronary artery disease (CAD) is the leading cause of death for both men and women, sex differences in CAD risk are widely observed. Genetic susceptibility for CAD may be modified by sex and contribute to sex disparities in CAD. However, whether such modification exists is not clear given the limited sample size of previous studies. Using the UK Biobank, we constructed a CAD genetic risk score (CAD-GRS) based on 161 known loci, and assessed the sex-specific effect of the CAD-GRS on CAD incidence in ~307,147 men and women of European ancestry without previous history of CAD. During an average follow-up of 6.1 years 9,847 incident CAD events were identified using hospitalization data and the national death registry. To examine if certain intermediate trait contributes more predominantly on sex differences in CAD risk, we constructed three risk factor-based CAD-sub-GRSs after removing pleiotropic SNPs: 1) 26 loci associated with blood pressure (BP) and CAD (BP-sub-GRS); 2) 17 loci associated with lipids and CAD (lipids-sub-GRS); and 3) 16 loci associated with body mass index (BMI) and CAD (BMI-sub-GRS). Sex-specific effects of CAD-sub-GRSs were examined. Cox models were implemented to assess sex-specific genetic effects of CAD adjusted for age, smoking, alcohol drinking, history of hypertension and diabetes, BMI, cholesterol medication, education and SES, and population stratification. To avoid over-adjustment, history of hypertension, cholesterol medication and BMI were not controlled for in the analyses of BP-sub-GRS, lipids-sub-GRS and BMI-sub-GRS, respectively. Both CAD-GRS and CAD-sub-GRSs were categorized using quintiles into “low” (quintile 1), “intermediate” (quintile 2 - 4), and “high” (quintile 5) groups. We observed a strong sex discrepancy in the effect of the overall CAD-GRS on CAD incidence (HR high GRS vs. low GRS = 2.4 (95% CI: 2.2, 2.6) for males vs. HR high GRS vs. low GRS = 1.8 (95% CI: 1.6, 2.0) for females, interaction p-value = 0.002). A similar gene-sex interaction was observed for CAD-sub-GRS driven by BP-associated loci (HR high GRS vs. low GRS = 1.5 (95% CI: 1.4, 1.7) for males vs. HR high GRS vs. low GRS = 1.2 (95% CI: 1.1, 1.3) for females, interaction p-value = 0.0005) but not lipids or BMI-associated loci. The SNP-based analysis revealed significant gene-sex interaction (p=0.0001) at one CAD locus (21q22.11). The lead SNP (rs28451064) is a cis -eQTL for several neighboring genes’ expression ( MRPS6 , SLC5A3 , KCNE2 and AP000318.2 ) in aorta and tibial artery tissues. Previous genome-wide association studies reported significant associations between rs28451064 and bone mineral density, waist-hip ratio as well as pulse pressure. These results suggest that genetic predisposition of CAD based on known loci is stronger in men than in women. Future investigations of genetic risk and mechanism for CAD need to consider sex-specific effects.

Interaction Between Type 2 Diabetes Prevention Strategies and Genetic Determinants of Coronary Artery Disease on Cardiometabolic Risk Factors.

Coronary artery disease (CAD) is more frequent among individuals with dysglycemia. Preventive interventions for diabetes can improve cardiometabolic risk factors (CRFs), but it is unclear whether the benefits on CRFs are similar for individuals at different genetic risk for CAD. We built a 201-variant polygenic risk score (PRS) for CAD and tested for interaction with diabetes prevention strategies on 1-year changes in CRFs in 2,658 Diabetes Prevention Program (DPP) participants. We also examined whether separate lifestyle behaviors interact with PRS and affect changes in CRFs in each intervention group. Participants in both the lifestyle and metformin interventions had greater improvement in the majority of recognized CRFs compared with placebo (P < 0.001) irrespective of CAD genetic risk (Pinteraction > 0.05). We detected nominal significant interactions between PRS and dietary quality and physical activity on 1-year change in BMI, fasting glucose, triglycerides, and HDL cholesterol in individuals randomized to metformin or placebo, but none of them achieved the multiple-testing correction for significance. This study confirms that diabetes preventive interventions improve CRFs regardless of CAD genetic risk and delivers hypothesis-generating data on the varying benefit of increasing physical activity and improving diet on intermediate cardiovascular risk factors depending on individual CAD genetic risk profile.

P708Identification of 26 novel loci that confer susceptibility to early-onset coronary artery disease in a Japanese population

Abstract Background Early-onset coronary artery disease (CAD) has a strong genetic component. Although genome-wide association studies have identified various genes and loci significantly associated with CAD mainly in European ancestry populations, genetic variants that contribute to susceptibility to this condition in Japanese individuals remain to be identified definitively. Purpose The purpose of the study was to identify genetic variants that confer susceptibility to early-onset CAD in Japanese. We have now performed exome-wide association studies (EWASs) in subjects with early-onset CAD and controls. Methods A total of 7256 individuals aged ≤65 years was enrolled in the study. The EWAS was conducted with 1482 subjects with CAD and 5774 controls. Genotyping of single nucleotide polymorphisms (SNPs) was performed with Illumina Human Exome-12 DNA Analysis BeadChip or Infinium Exome-24 BeadChip arrays. The relation of allele frequencies for 31,465 SNPs that passed quality control to CAD was examined with Fisher's exact test. To compensate for multiple comparisons of allele frequencies with CAD, we applied a false discovery rate (FDR) of &lt;0.05 for statistical significance of association. Results The relation of allele frequencies for 31,465 SNPs to CAD with the use of Fisher's exact test showed that 170 SNPs were significantly (FDR &lt;0.05) associated with CAD. Multivariable logistic regression analysis with adjustment for age, sex, and the prevalence of hypertension, diabetes mellitus, and dyslipidemia revealed that 162 SNPs were significantly (P&lt;0.05) related to CAD. A stepwise forward selection procedure was performed to examine the effects of genotypes for the 162 SNPs on CAD. The 54 SNPs were significant (P&lt;0.05) and independent [coefficient of determination (R2), 0.0008 to 0.0297] determinants of CAD. These SNPs together accounted for 15.5% of the cause of CAD. After examination of results from previous genome-wide association studies and linkage disequilibrium of the identified SNPs, we newly identified 21 genes (RNF2, YEATS2, USP45, ITGB8, TNS3, FAM170B-AS1, PRKG1, BTRC, MKI67, STIM1, OR52E4, KIAA1551, MON2, PLUT, LINC00354, TRPM1, ADAT1, KRT27, LIPE, GFY, EIF3L) and five chromosomal regions (2p13, 4q31.2, 5q12, 13q34, 20q13.2) that were significantly associated with CAD. Gene ontology analysis showed that various biological functions were predicted in the 18 genes identified in the present study. The network analysis revealed that the 18 genes had potential direct or indirect interactions with the 30 genes previously shown to be associated with CAD or with the 228 genes identified in previous genome-wide association studies of CAD. Conclusion We have newly identified 26 loci that confer susceptibility to CAD. Determination of genotypes for the SNPs at these loci may prove informative for assessment of the genetic risk for CAD in Japanese.

Sleep Duration and Myocardial Infarction

BackgroundObservational studies suggest associations between extremes of sleep duration and myocardial infarction (MI), but the causal contribution of sleep to MI and its potential to mitigate genetic predisposition to coronary disease is unclear. ObjectivesThis study sought to investigate associations between sleep duration and incident MI, accounting for joint effects with other sleep traits and genetic risk of coronary artery disease, and to assess causality using Mendelian randomization (MR). MethodsIn 461,347 UK Biobank (UKB) participants free of relevant cardiovascular disease, the authors estimated multivariable adjusted hazard ratios (HR) for MI (5,128 incident cases) across habitual self-reported short (<6 h) and long (>9 h) sleep duration, and examined joint effects with sleep disturbance traits and a coronary artery disease genetic risk score. The authors conducted 2-sample MR for short (24 single nucleotide polymorphisms) and continuous (71 single nucleotide polymorphisms) sleep duration with MI (n = 43,676 cases/128,199 controls), and replicated results in UKB (n = 12,111/325,421). ResultsCompared with sleeping 6 to 9 h/night, short sleepers had a 20% higher multivariable-adjusted risk of incident MI (HR: 1.20; 95% confidence interval [CI]: 1.07 to 1.33), and long sleepers had a 34% higher risk (HR: 1.34; 95% CI: 1.13 to 1.58); associations were independent of other sleep traits. Healthy sleep duration mitigated MI risk even among individuals with high genetic liability (HR: 0.82; 95% CI: 0.68 to 0.998). MR was consistent with a causal effect of short sleep duration on MI in CARDIoGRAMplusC4D (Coronary ARtery DIsease Genome wide Replication and Meta-analysis plus Coronary Artery Disease Genetics Consortium) (HR: 1.19; 95% CI: 1.09 to 1.29) and UKB (HR: 1.21; 95% CI: 1.08 to 1.37). ConclusionsProspective observational and MR analyses support short sleep duration as a potentially causal risk factor for MI. Investigation of sleep extension to prevent MI may be warranted.

Clinical Utility and Practical Considerations of a Coronary Artery Disease Genetic Risk Score

BackgroundCoronary artery disease (CAD) risk traditionally has been assessed using clinical risk factors. We evaluated whether molecular genetic markers for CAD risk could add information to traditional variables. MethodsWe developed a false discovery rate 267-marker genetic risk score (FDR267) from markers that were significantly associated with CAD in the UK Biobank cohort meta-analysis. FDR267 was tested in the Atherosclerosis Risk in Communities cohort using logistic regression and Cox proportional hazards analyses in the European and African American groups. ResultsOur genetic risk score (FDR267) was associated with a 1.45 (95% confidence interval, 1.39-1.51) increase in odds ratio and a 1.32 (95% confidence interval, 1.26-1.38) increase in hazard ratio per standard deviation of the score. The score modestly improved the area under the curve (AUC) statistic when added to a clinical model (ΔAUC = 0.0112, P = 0.0002). FDR267 predicted incident CAD (C-index = 0.60), although it did not improve on clinical risk factors (ΔAUC = 0.0159, P = 0.0965). Individuals in the top quintile of FDR267 genetic risk were at approximately 2-fold increased risk compared with the bottom quintile, which is comparable to risk associated with self-reported family history. The performance of FDR267 was less robust in the African American sample. ConclusionsFDR267 is significantly associated with CAD in the European sample, with an effect size comparable to self-reported family history. FDR267 discriminated between individuals with and without CAD, but did not improve CAD risk prediction over clinical variables. FDR267 was less predictive of CAD risk in African Americans.