Coronary Artery Disease Risk Prediction Research Articles

Abstract 4136501: Machine Learning Uncovers the Contribution of Rare Genetic Variants and Enhances Risk Prediction for Coronary Artery Disease in the Japanese Population

Background: Genome-wide association studies (GWASs) have enhanced our understanding of the genetic basis of coronary artery disease (CAD), and polygenic risk scores (PRSs) have facilitated the assessment of genetic risk. However, these methods predominantly focus on common variants due to statistical power, potentially leaving rare variants insufficiently analyzed and thus limiting the predictive performance of PRS. Methods: We conducted whole genome sequencing (WGS) of 1,752 Japanese early-onset myocardial infarction (MI) patients and 3,019 controls from Biobank Japan (BBJ). We performed case-control association studies including GWAS and gene-based tests, as well as a novel machine learning-based framework. In this framework, we developed a penalized regression model to predict the CAD status from genome-wide rare nonsynonymous variants. The model identified the minimal set of most distinguishing features (genes) and generated a rare variant-based risk score (RVS). The RVS was evaluated on an independent validation WGS cohort of 200 cases and 824 controls. We also derived a PRS based on CAD-GWAS (25,668 CAD cases vs 141,667 controls from BBJ) to compare the properties and performance between the RVS and PRS. Results: In the case-control studies, only two common variants in chromosome 12 were identified in GWAS, with no genes in gene-based analysis (SKAT-O), suggesting the challenges in rare variant analysis. On the other hand, our machine-learning framework identified 59 CAD-related genes, including LDLR , a causal gene of familial hypercholesteremia. Functional analyses revealed that various biological pathways, including lipid metabolism, immune system, and vessel development, are involved in CAD. For the genetic risk prediction, RVS significantly predicted CAD (area under the curve [AUC], 0.58; p=0.001, pseudo-R 2 , 0.051; p=7.92*10 -9 ). RVS was significantly associated with LDL cholesterol levels and coagulation function (Pearson’s r, 0.21; p=4.5*10 -5 and 0.10; p=0.03, respectively) and MI patients with high RVS (top 5%) showed higher cardiovascular mortality rate (p=0.03, log-rank test), highlighting the clinical importance of RVS. Finally, the combined risk score (CRS) of RVS and PRS significantly improved CAD prediction compared to PRS (AUC, 0.66 (CRS) vs 0.61 (PRS); p=0.007, pseudo-R 2 , 0.093 (CRS) vs 0.040 (PRS); p=0.0018, Figure ). Conclusions: Our machine learning framework successfully characterized rare variants and enhanced genetic risk prediction in CAD.

Coronary artery disease risk prediction by combined stratification of low-density lipoprotein-cholesterol and high-density lipoprotein-cholesterol levels across different glucose statuses.

To investigate the association between low-density lipoprotein-cholesterol (LDL-C) levels and coronary artery disease (CAD) incidence based on combining high-density lipoprotein-cholesterol (HDL-C) levels and glucose status. In this retrospective cohort study, we used data from a nationwide claims database (1,524,289 individuals without a history of CAD or familial hypercholesterolaemia; 2008-2019). Cox proportional hazards modelling identified the risk of incident CAD by a novel combination of four HDL-C levels, seven LDL-C levels and glucose status. During the follow-up period (mean: 5.5 years), 8301 (0.99/1000 person-years) events occurred. The risk of CAD increased from lower LDL-C levels accompanied by lower HDL-C levels regardless of the glucose status. Using the most favourable levels of HDL-C and LDL-C (i.e. 60-99 mg/dL and <80 mg/dL, respectively) as references, the hazard ratios (95% confidence interval) for the group with HDL-C levels <40 mg/dL and LDL-C levels <80 mg/dL were 2.74 (1.47-5.11), 2.52 (1.30-4.91) and 2.85 (1.68-4.84) for normoglycaemia, borderline glycaemia and diabetes, respectively. Comparison of the most favourable levels of HDL-C and LDL-C with their least favourable levels (i.e. <40 mg/dL and 180-199 mg/dL, respectively) revealed that the risk of new-onset CAD exhibited a 19-, nine- and seven-fold increase in individuals with normoglycaemia, borderline glycaemia and diabetes, respectively. To prevent CAD, LDL-C levels should be strictly controlled in patients with low HDL-C levels regardless of glucose tolerance. Individualized treatment, which involves setting target LDL-C levels based on glucose tolerance and HDL-C values, is required.

Current and Emerging Approaches for Primary Prevention of Coronary Artery Disease Using Cardiac Computed Tomography.

To summarize the current use of cardiac computed tomography (CT) technologies as well as their pertinent evidence in regards to prevention of coronary artery disease (CAD). Cardiac CTA has now become a main non-invasive method for the evaluation of symptomatic CAD. In addition to coronary calcium score, other CT technologies such as atherosclerotic plaque analysis, fractional flow reserve estimation by CT, pericoronary fat attenuation, and endothelial wall shear stress have emerged. Whether the use of CT modalities can enhance risk prediction and prevention in CAD has not been fully answered. We discuss the evidence for coronary artery calcium scoring and coronary CT angiography in primary prevention and the current barriers to their use. We attempt to delineate what can be done to expand use and what studies are needed to broaden adoption in the future. We also examine the potential roles of emerging CT technologies. Finally, we describe potential clinical approaches to prevention that would incorporate cardiac CT technologies.

Coronary Artery Disease in Patients Undergoing Hemodialysis: A Problem that Sounds the Alarm.

Chronic kidney disease (CKD) is affecting more and more individuals over time. The importance of the increased prevalence is enhanced by the close association with the increased risk of poor individual outcomes such as death, fatal and non-fatal cardiovascular (CV) events and progression to end stage kidney disease (ESKD). ESKD requires replacement treatment such as hemodialysis (HD), a particular and complex context that unfortunately has been rarely considered in observational studies in the last few decades. The current perspective of HD as a bridge to kidney transplant requires greater attention from observational and experimental research both in the prevention and treatment of CV events in ESKD patients. We present a narrative review by performing a literature review to extrapolate the most significant articles exploring the CV risk, in particular coronary artery disease (CAD), in ESKD and evaluating possible innovative diagnostic and therapeutic tools in these patients. The risk of CAD increases linearly when the estimated glomerular filtration rate (eGFR) declines and reached the most significant level in ESKD patients. Several diagnostic techniques have been evaluated to predict CAD in ESKD such as laboratory tests (Troponin-T, N-terminal pro b-type natriuretic peptide, alkaline phosphatase), echocardiography and imaging techniques for vascular calcifications evaluation. Similarly, treatment is based on lifestyle changes, medical therapy and invasive techniques such as coronary artery bypass grafting (CABG) and percutaneous coronary intervention (PCI). Unfortunately in the literature there are no clear indications of the usefulness and validity of biomarkers and possible treatments in ESKD patients. Considering the ESKD weight in terms of prevalence and costs it is necessary to implement clinical research in order to develop prognostic reliable biomarkers for CV and CAD risk prediction, in patients with ESKD. It should be highlighted that HD is a peculiar setting that offers the opportunity to implement research and facilitates patient monitoring by favoring the design of clinical trials.

Evaluation of Polygenic Risk Scores for Prediction of Coronary Artery Disease in a Greek Case-Control Study.

Coronary artery disease (CAD) stands as the most predominant type of cardiovascular disease (CVD). Polygenic risk scores (PRSs) have become essential tools for quantifying genetic susceptibility, and researchers endeavor to improve their predictive precision. The aim of the present work is to assess the performance and the relative contribution of PRSs developed for CVD or CAD within a Greek population. The sample under study comprised 924 Greek individuals (390 cases with CAD and 534 controls) from the THISEAS study. Nine PRSs drawn from the PGS catalog were replicated and tested for CAD risk prediction. PRSs computations were performed in the R language, and snpStats was used to process genotypic data. Descriptive characteristics of the study were analyzed using the statistical software IBM SPSS Statistics v21.0. The effectiveness of each PRS was assessed using the PRS R2 metric provided by PRSice2. Among nine PRSs, PGS000747 greatly increased the predictive value of primary CAD risk factors by 21.6% (p-value = 2.63 × 10-25). PGS000012 was associated with a modest increase in CAD risk by 2.2% (p-value = 9.58 × 10-4). The remarkable risk discrimination capability of PGS000747 stands out as the most noteworthy outcome of our study.

Plasma Homocysteine Levels as Cardiovascular Disease Risk vis-a-vis Estrogen Levels in Pre and Postmenopausal Women

Objectives: The objective of the study is to evaluate plasma homocysteine level and its relation with serum estrogen in pre and postmenopausal women as a risk factor in coronary artery disease (CAD). Materials and Methods: A cross-sectional study was conducted among premenopausal (n = 50) and postmenopausal women (n = 50) with an estimation of plasma homocysteine (Hcy) and serum estradiol (E2). Postmenopausal women based on their angiographic status were sub-grouped into A (with normal coronaries) and B (with CAD). Results: The levels of Hcy, 21.98 (20.89–24.05) μmol/L, increased significantly (P &lt; 0.0001) and that of E2, 3.05 (2.32–3.6) pg/mL, decreased significantly (P = 0.0001) in the postmenopausal group when compared to Hcy, 6.11 (4.94–8.27) μmol/L and E2, 26.37 (25.3–29) pg/mL in premenopausal group. The levels of Hcy were elevated (24.7 ± 3.92 μmol/L) in postmenopausal Group B in comparison with postmenopausal Group A (21.37 ± 3.31 μmol/L), P = 0.002. Logistic regression analysis showed Hcy and body mass index to be independent predictors of CAD. Plasma Hcy showed an area under curve (AUC) of 1.00 with sensitivity and specificity of 100% at cutoff &gt;12.6 μmol/L between pre and postmenopausal groups and AUC of 0.722 with 60% sensitivity and 84% specificity at cutoff &gt;23.7 μmol/L between postmenopausal Groups A and B. Conclusion: We observed low E2 and high Hcy levels in postmenopausal women in comparison with premenopausal women. Among postmenopausal women, those with CAD had higher Hcy levels. Assessing Hcy levels routinely in postmenopausal women will help in better risk prediction of CAD and may also help in conjunction with other risk factors to decide the initiation of hormone replacement therapy.

Decreased plasma miR-140-3p is associated with coronary artery disease

BackgroundAlthough many circulating miRNAs (c-miRNAs) are associated with coronary artery disease (CAD), they are far from being the biomarker for CAD diagnosis or risk prediction. Therefore, novel c-miRNAs discovery and validation are still required, especially evaluating their prediction capacity. ObjectivesIdentify novel CAD-related c-miRNAs and evaluate its risk prediction capacity for CAD. Methods: miRNAs associated with CAD were preliminarily investigated in three paired samples representing pre-CAD stage and CAD stage of three female individuals using the Applied Biosystems miRNA TaqMan® Low-Density Array (TLDA). Then, the candidate miRNAs were further verified in an independent case-control study including 129 CAD patients and 76 controls, and their potential practical value in prediction for CAD was evaluated using a machine learning (ML) algorithm. The accuracy of classification and prediction was assessed with the area under the receiver operating characteristic curve (AUC). ResultsTLDA analysis shows that miR-140-3p decreased significantly in CAD-stage (FC = −3.01, P = 0.007). Further study shows that miR-140-3p was significantly lower in CAD group [1.26 (0.68, 2.01)] than in control group [2.07 (1.19, 3.21)] (P < 0.001) and independently associated with CAD (P < 0.001). The addition of miR-140-3p to the variables including smoking history, HDL-c, and APOA1 improved the accuracy of classification by logistic regression and of prediction for CAD by ML models. The ML models built with miR-140-3p and HDL-c, respectively, had a similar prediction accuracy. The feature importance of miR-140-3p and HDL-c in the ML models was also similar. Decision curve analysis showed that miR-140-3p and HDL-c had almost identical net benefits. ConclusionReduced levels of miR-140-3p is linked to CAD, and it is possible to use the plasma level of miR-140-3p as a means of evaluating the risk of CAD.

Association Between Serum Uric Acid Levels and Oxido-Inflammatory Biomarkers With Coronary Artery Disease in Type 2 Diabetic Patients.

Cardiovascular disease signifies a major cause of morbidity and mortality among patients with type 2 diabetes mellitus (T2DM). Serum uric acid (SUA) levels are elevated during the initial phases of impaired glucose metabolism. This work was designed to explore the association between SUA levels, serum oxido-inflammatory biomarkers, and the risk of coronary artery disease (CAD) in T2DM patients as the primary outcome. The secondary outcome was to assess the prognostic role of SUA in the prediction of the risk of CAD in T2DM patients. In this case-control study, we enrolled 110 patients with T2DM who were further divided into patients with CAD and without CAD. In addition, 55 control participants were stringently matched to cases by age. Diabetic patients with CAD had significantly higher serum levels of the inflammatory biomarkers and the oxidative malondialdehyde but significantly lower levels of serum total antioxidant capacity (TAC) compared with the controls and diabetic patients without CAD. Significant positive correlations existed between SUA levels and serum levels of the inflammatory biomarkers and malondialdehyde, while a significant negative correlation existed between SUA levels and serum TAC. SUA demonstrated an accepted discrimination ability. SUA can differentiate between T2DM patients with CAD and patients without CAD, an area under the curve of 0.759. Elevated serum levels of SUA and oxido-inflammatory biomarkers are associated with an increased risk of CAD in T2DM. SUA levels reflect the body's inflammatory status and oxidant injury in T2DM. SUA could be utilized as a simple biomarker in the prediction of CAD risk in T2DM.

Coronary Artery Disease Risk Prediction in Patients With Severe Aortic Stenosis: Development and Validation of the Aortic Stenosis-Coronary Artery Disease (AS-CAD) Score

Patients at a low risk of coronary artery disease (CAD) could be triaged to noninvasive coronary computed tomography angiogram instead of invasive coronary angiography, reducing health care costs and patient morbidity. Therefore, we aimed to develop a CAD risk prediction score to identify those who underwent transcatheter aortic valve implantation (TAVI) at a low risk of CAD. We enrolled 1,782 patients who underwent TAVI and randomized the patients to the derivation or validation cohort 2:1. The aortic stenosis-CAD (AS-CAD) score was developed using logistic regression, followed by separation into low- (score 0 to 5), intermediate- (6 to 10), or high-risk (>11) categories. The AS-CAD was validated initially through the k-fold cross-validation, followed by a separately held validation cohort. The average age of the cohort was 82 ± 7years, and 41% (730 of 1,782) were female; 35% (630) had CAD. The male sex, previous percutaneous coronary intervention, stroke, peripheral arterial disease, diabetes, smoking status, left ventricular ejection fraction <50%, and right ventricular systolic pressure >35mm Hg were all associated with an increased risk of CAD and were included in the final AS-CAD model (all p <0.03). Within the validation cohort, the AS-CAD score stratified those into low, intermediate, and high risk of CAD (p <0.001). Discrimination was good within the internal validation cohort, with a c-statistic of 0.79 (95% confidence interval 0.74 to 0.84), with similar power obtained using k-fold cross-validation (c-statistic 0.74 [95% confidence interval 0.70 to 0.77]). In conclusion, The AS-CAD score robustly identified those at a low risk of CAD in patients with severe AS. The use of AS-CAD in practice could avoid potential complications of invasive coronary angiogram by triaging low-risk patients to noninvasive coronary assessment using existing computed tomography data.

PREDICTION OF CORONARY ARTERY DISEASE USING URINARY PROTEOMICS

Objective: Coronary artery disease (CAD) is multifactorial, caused by complex pathophysiology, and contributes to a high burden of mortality worldwide. Urinary proteomic analyses may help to identify predictive biomarkers and provide insights into the pathogenesis of CAD. Design and method: Urinary proteome was analyzed in 965 participants using capillary electrophoresis coupled with mass spectrometry. A proteomic classifier was developed in a discovery cohort with 36 individuals with CAD and 36 matched controls using the support vector machine. The classifier was tested in a validation cohort with 115 individuals who progressed to CAD and 778 controls and compared with a previously developed CAD-associated classifier, ACSP75. The Framingham risk score was available in 737 participants. Bioinformatics analysis was performed based on the CAD-associated peptides. Results: The novel proteomic classifier was comprised of 160 urinary peptides, mainly related to collagen turnover, lipid metabolism, and inflammation. In the validation cohort, the classifier provided an AUC of 0.82 (95% CI: 0.78-0.87) for the CAD prediction in 8 years, superior to ACSP75 (AUC: 0.53, 95% CI: 0.47-0.60). On top of ACSP75, the addition of the novel classifier improved the AUC to 0.84 (95% CI: 0.80-0.89). In a multivariable Cox model, a 1-SD increment in the novel classifier was associated with CAD (HR: 1.54, 95% CI: 1.26-1.89, P&lt;0.0001) higher risk of CAD. The new classifier further improved the risk reclassification of CAD on top of the Framingham risk score (net reclassification index: 0.61, 95% CI: 0.25-0.95, P = 0.001). Conclusions: A novel urinary proteomic classifier related to collagen metabolism, lipids, and inflammation showed potential for the risk prediction of CAD. Urinary proteome provides an alternative approach to personalized prevention.

The association between circulating lipoprotein subfractions and lipid content in coronary atheromatous plaques assessed by near-infrared spectroscopy

Abstract Funding Acknowledgements Type of funding sources: Foundation. Main funding source(s): Norwegian Health Association The Liaison Committee for education, research and innovation in Central Norway. Background Lipid content in coronary atheromatous plaques, measured by near-infrared spectroscopy (NIRS), can predict the risk of future coronary events. Biomarkers that reflect lipid content in coronary plaques may therefore improve coronary artery disease (CAD) risk assessment, and lipoprotein subfractions could potentially add value beyond traditional lipid measurements. Purpose We aimed to investigate the association between circulating lipoprotein subfractions and lipid content in coronary atheromatous plaques in statin-treated patients with stable CAD undergoing percutaneous coronary intervention. Methods 56 patients with stable CAD underwent three vessel imaging with NIRS when feasible. The coronary artery segment with the highest lipid content, defined as the maximum lipid core burden index within any 4mm length across the entire lesion (maxLCBI4mm), was selected as target segment. Lipoprotein subfractions were analyzed in fasting serum samples by nuclear magnetic resonance spectroscopy. Penalized linear regression analyses (lasso) were used to identify the best predictors of maxLCBI4mm. The uncertainty of the lasso estimates was assessed as the percentage of the resampled datasets by bootstrapping. Results Only modest evidence was found for an association with lipoprotein subfractions and maxLCBI4mm. The lipoprotein subfractions with strongest potential as predictors according to the percentage presence in resampled datasets were lipoprotein a (Lp(a)) (reg.coeff = 57.0; 78.1 % presence) and free cholesterol in the smallest high-density lipoprotein (HDL) subfractions (reg.coeff = 36.5; 74.3 % presence). When including established cardiovascular risk factors in the regression model, none of the lipoprotein subfractions was considered potential predictors of maxLCBI4mm. Conclusion In this study, Lp(a) and free cholesterol in the smallest HDL subfractions in serum showed the strongest potential as predictors for lipid content in coronary atheromatous plaques. Although the evidence is modest, our study suggests that measurements lipoprotein subfractions may provide additional information with respect to coronary plaque composition compared to traditional lipid measurements, but not in addition to established risk factors. Future studies are needed to verify our findings, and to investigate the clinical implication of lipoprotein subfractions as meaningful biomarkers for lipid-rich plaques and enhanced risk prediction in CAD.

Coronary Artery Disease Risk Prediction in Young Adults: How Can We Overcome the Dominant Effect of Age?

We review the limitations of current approaches for predicting Coronary Artery Disease (CAD) in young adults and explore the alternative approaches identify high-risk individuals in this population. Atherosclerosis begins in childhood, and young individuals with genetic predisposition and individuals with early exposure to traditional and non-traditional risk factors have an increased lifetime risk of CAD. However, most risk prediction models have been developed and validated in middle and older age groups and focus on short-term risk. Therefore, alternative approaches are needed in younger individuals. Genetic scores, biomarkers, imaging studies, and multi-omics data all have the potential to be used and help identify high-risk individuals.

Minimal improvement in coronary artery disease risk prediction in Chinese population using polygenic risk scores: evidence from the China Kadoorie Biobank.

Several studies have reported that polygenic risk scores (PRSs) can enhance risk prediction of coronary artery disease (CAD) in European populations. However, research on this topic is far from sufficient in non-European countries, including China. We aimed to evaluate the potential of PRS for predicting CAD for primary prevention in the Chinese population. Participants with genome-wide genotypic data from the China Kadoorie Biobank were divided into training ( n = 28,490) and testing sets ( n = 72,150). Ten previously developed PRSs were evaluated, and new ones were developed using clumping and thresholding or LDpred method. The PRS showing the strongest association with CAD in the training set was selected to further evaluate its effects on improving the traditional CAD risk-prediction model in the testing set. Genetic risk was computed by summing the product of the weights and allele dosages across genome-wide single-nucleotide polymorphisms. Prediction of the 10-year first CAD events was assessed using hazard ratios (HRs) and measures of model discrimination, calibration, and net reclassification improvement (NRI). Hard CAD (nonfatal I21-I23 and fatal I20-I25) and soft CAD (all fatal or nonfatal I20-I25) were analyzed separately. In the testing set, 1214 hard and 7201 soft CAD cases were documented during a mean follow-up of 11.2 years. The HR per standard deviation of the optimal PRS was 1.26 (95% CI:1.19-1.33) for hard CAD. Based on a traditional CAD risk prediction model containing only non-laboratory-based information, the addition of PRS for hard CAD increased Harrell's C index by 0.001 (-0.001 to 0.003) in women and 0.003 (0.001 to 0.005) in men. Among the different high-risk thresholds ranging from 1% to 10%, the highest categorical NRI was 3.2% (95% CI: 0.4-6.0%) at a high-risk threshold of 10.0% in women. The association of the PRS with soft CAD was much weaker than with hard CAD, leading to minimal or no improvement in the soft CAD model. In this Chinese population sample, the current PRSs minimally changed risk discrimination and offered little improvement in risk stratification for soft CAD. Therefore, this may not be suitable for promoting genetic screening in the general Chinese population to improve CAD risk prediction.

Prediction of coronary artery disease using urinary proteomics.

Coronary artery disease (CAD) is multifactorial, caused by complex pathophysiology, and contributes to a high burden of mortality worldwide. Urinary proteomic analyses may help to identify predictive biomarkers and provide insights into the pathogenesis of CAD. Urinary proteome was analysed in 965 participants using capillary electrophoresis coupled with mass spectrometry. A proteomic classifier was developed in a discovery cohort with 36 individuals with CAD and 36 matched controls using the support vector machine. The classifier was tested in a validation cohort with 115 individuals who progressed to CAD and 778 controls and compared with two previously developed CAD-associated classifiers, CAD238 and ACSP75. The Framingham and SCORE2 risk scores were available in 737 participants. Bioinformatic analysis was performed based on the CAD-associated peptides. The novel proteomic classifier was comprised of 160 urinary peptides, mainly related to collagen turnover, lipid metabolism, and inflammation. In the validation cohort, the classifier provided an area under the receiver operating characteristic curve (AUC) of 0.82 [95% confidence interval (CI): 0.78-0.87] for the CAD prediction in 8 years, superior to CAD238 (AUC: 0.71, 95% CI: 0.66-0.77) and ACSP75 (AUC: 0.53 and 95% CI: 0.47-0.60). On top of CAD238 and ACSP75, the addition of the novel classifier improved the AUC to 0.84 (95% CI: 0.80-0.89). In a multivariable Cox model, a 1-SD increment in the novel classifier was associated with a higher risk of CAD (HR: 1.54, 95% CI: 1.26-1.89, P < 0.0001). The new classifier further improved the risk reclassification of CAD on top of the Framingham or SCORE2 risk scores (net reclassification index: 0.61, 95% CI: 0.25-0.95, P = 0.001; 0.64, 95% CI: 0.28-0.98, P = 0.001, correspondingly). A novel urinary proteomic classifier related to collagen metabolism, lipids, and inflammation showed potential for the risk prediction of CAD. Urinary proteome provides an alternative approach to personalized prevention.

Screening for Coronary Artery Disease inCancer Survivors: JACC: CardioOncology State-of-the-ArtReview.

Coronary artery disease (CAD) is an important contributor to the cardiovascular burden in cancer survivors. This review identifies features that could help guide decisions about the benefit of screening to assess the risk or presence of subclinical CAD. Screening may be appropriate in selected survivors based on risk factors and inflammatory burden. In cancer survivors who have undergone genetic testing, polygenic risk scores and clonal hematopoiesis markers may become useful CAD risk prediction tools in the future. The type of cancer (especially breast, hematological, gastrointestinal, and genitourinary) and the nature of treatment (radiotherapy, platinum agents, fluorouracil, hormonal therapy, tyrosine kinase inhibitors, endothelial growth factor inhibitors, and immune checkpoint inhibitors) are also important in determining risk. Therapeutic implications of positive screening include lifestyle and atherosclerosis interventions, and in specific instances, revascularization may be indicated.

Long-Term Coronary Artery Disease Risk Prediction with Machine Learning Models.

The heart is the most vital organ of the human body; thus, its improper functioning has a significant impact on human life. Coronary artery disease (CAD) is a disease of the coronary arteries through which the heart is nourished and oxygenated. It is due to the formation of atherosclerotic plaques on the wall of the epicardial coronary arteries, resulting in the narrowing of their lumen and the obstruction of blood flow through them. Coronary artery disease can be delayed or even prevented with lifestyle changes and medical intervention. Long-term risk prediction of coronary artery disease will be the area of interest in this work. In this specific research paper, we experimented with various machine learning (ML) models after the use or non-use of the synthetic minority oversampling technique (SMOTE), evaluating and comparing them in terms of accuracy, precision, recall and an area under the curve (AUC). The results showed that the stacking ensemble model after the SMOTE with 10-fold cross-validation prevailed over the other models, achieving an accuracy of 90.9 %, a precision of 96.7%, a recall of 87.6% and an AUC equal to 96.1%.

Sex-Specific Survival Bias and Interaction Modeling in Coronary Artery Disease Risk Prediction.

The 10-year Atherosclerotic Cardiovascular Disease risk score is the standard approach to predict risk of incident cardiovascular events, and recently, addition of coronary artery disease (CAD) polygenic scores has been evaluated. Although age and sex strongly predict the risk of CAD, their interaction with genetic risk prediction has not been systematically examined. This study performed an extensive evaluation of age and sex effects in genetic CAD risk prediction. The population-based Norwegian HUNT2 (Trøndelag Health Study 2) cohort of 51 036 individuals was used as the primary dataset. Findings were replicated in the UK Biobank (372 410 individuals). Models for 10-year CAD risk were fitted using Cox proportional hazards, and Harrell concordance index, sensitivity, and specificity were compared. Inclusion of age and sex interactions of CAD polygenic score to the prediction models increased the C-index and sensitivity by accounting for nonadditive effects of CAD polygenic score and likely countering the observed survival bias in the baseline. The sensitivity for females was lower than males in all models including genetic information. We identified a total of 82.6% of incident CAD cases by using a 2-step approach: (1) Atherosclerotic Cardiovascular Disease risk score (74.1%) and (2) the CAD polygenic score interaction model for those in low clinical risk (additional 8.5%). These findings highlight the importance and complexity of genetic risk in predicting CAD. There is a need for modeling age- and sex-interaction terms with polygenic scores to optimize detection of individuals at high risk, those who warrant preventive interventions. Sex-specific studies are needed to understand and estimate CAD risk with genetic information.

Small LDL subfractions are associated with coronary atherosclerosis despite no differences in conventional lipids.

Lipoprotein subfractions currently represent a new source of cardiovascular disease (CVD) risk markers that may provide more information than conventional lipid measures. We aimed to investigate whether lipoprotein subfractions are associated with coronary atherosclerosis in patients without prior known CVD. Fasting serum samples from 60 patients with suspected coronary artery disease (CAD) were collected before coronary angiography and analyzed by nuclear magnetic resonance (NMR) spectroscopy. The severity of coronary atherosclerosis was quantified by the Gensini score (≤20.5 = nonsignificant coronary atherosclerosis, 20.6-30.0 = intermediate coronary atherosclerosis, ≥30.1 = significant CAD). Differences in lipoprotein subfractions between the three Gensini groups were assessed by two-way ANOVA, adjusted for statin use. Despite no differences in conventional lipid measures between the three Gensini groups, patients with significant CAD had higher apolipoprotein-B/apolipoprotein-A1 ratio, 30% more small and dense low-density lipoprotein 5 (LDL-5) particles, and increased levels of cholesterol, triglycerides, and phospholipids within LDL-5 compared with patients with nonsignificant coronary atherosclerosis and intermediate coronary atherosclerosis (P ≤ 0.001). In addition, the low-density lipoprotein (LDL) cholesterol/high-density lipoprotein cholesterol ratio, and triglyceride levels of LDL 4 were significantly increased in patients with significant CAD compared with patients with nonsignificant coronary atherosclerosis. In conclusion, small and dense lipoprotein subfractions were associated with coronary atherosclerosis in patients without prior CVD. Additional studies are needed to explore whether lipoprotein subfractions may represent biomarkers offering a clinically meaningful improvement in the risk prediction of CAD.

Polygenic risk score improves the accuracy of a clinical risk score for coronary artery disease

BackgroundThe value of polygenic risk scores (PRSs) towards improving guideline-recommended clinical risk models for coronary artery disease (CAD) prediction is controversial. Here we examine whether an integrated polygenic risk score improves the prediction of CAD beyond pooled cohort equations. MethodsAn observation study of 291,305 unrelated White British UK Biobank participants enrolled from 2006 to 2010 was conducted. A case–control sample of 9499 prevalent CAD cases and an equal number of randomly selected controls was used for tuning and integrating of the polygenic risk scores. A separate cohort of 272,307 individuals (with follow-up to 2020) was used to examine the risk prediction performance of pooled cohort equations, integrated polygenic risk score, and PRS-enhanced pooled cohort equation for incident CAD cases. The performance of each model was analyzed by discrimination and risk reclassification using a 7.5% threshold.ResultsIn the cohort of 272,307 individuals (mean age, 56.7 years) used to analyze predictive accuracy, there were 7036 incident CAD cases over a 12-year follow-up period. Model discrimination was tested for integrated polygenic risk score, pooled cohort equation, and PRS-enhanced pooled cohort equation with reported C-statistics of 0.640 (95% CI, 0.634–0.646), 0.718 (95% CI, 0.713–0.723), and 0.753 (95% CI, 0.748–0.758), respectively. Risk reclassification for the addition of the integrated polygenic risk score to the pooled cohort equation at a 7.5% risk threshold resulted in a net reclassification improvement of 0.117 (95% CI, 0.102 to 0.129) for cases and − 0.023 (95% CI, − 0.025 to − 0.022) for noncases [overall: 0.093 (95% CI, 0.08 to 0.104)]. For incident CAD cases, this represented 14.2% correctly reclassified to the higher-risk category and 2.6% incorrectly reclassified to the lower-risk category.ConclusionsAddition of the integrated polygenic risk score for CAD to the pooled cohort questions improves the predictive accuracy for incident CAD and clinical risk classification in the White British from the UK Biobank. These findings suggest that an integrated polygenic risk score may enhance CAD risk prediction and screening in the White British population.

Association between NMR metabolomic signatures of healthy lifestyle and incident coronary artery disease.

To identify metabolites associated with a healthy lifestyle and explore the possible mechanisms of lifestyle in coronary artery disease (CAD). The nuclear magnetic resonance metabolomics platform was applied to perform metabolomic profiling of baseline plasma samples from a randomly selected subset of 121 733 UK Biobank participants. Cox proportional hazards models with covariate adjustments were used to investigate the associations between validated lifestyle-associated metabolites and incident CAD and to estimate the accuracy of the inclusion of metabolites to predict CAD compared with traditional prediction models. The discriminatory ability of each model was evaluated using Harrell's C statistic, integrated discrimination improvement (IDI), and continuous net reclassification improvement (NRI) indexes. During a median of 8.6 years of follow-up, 5513 incident CAD cases were documented. Among the 111 lifestyle-associated metabolites, 65 were significantly associated with incident CAD after multivariate adjustment (Bonferroni P < 3.11 × 10-04). The addition of these metabolites to classic risk prediction models [Framingham Risk Score (FRS) using lipids; FRS using body mass index] improved CAD prediction accuracy as assessed by the C statistic (increasing to 0.739 [95% CI, 0.731-0.747] and 0.752 [95% CI, 0.746-0.758]), respectively; continuous NRI (0.274 [0.227-0.325] and 0.266 [0.223-0.317]) and IDI (0.003 [0.002-0.004] and 0.003 [0.002-0.004]). Healthy lifestyle-associated metabolites are associated with the incidence of CAD and may help improve the prediction of CAD risk. The use of metabolite information combined with the FRS model warrants further investigation before clinical implementation.