Unraveling the Complex Genetics of Coronary Artery Disease

Artificial Intelligence for Risk Assessment on Primary Prevention of Coronary Artery Disease

Purpose of Review

Coronary artery disease (CAD) is a common and etiologically complex disease worldwide. Current guidelines for primary prevention, or the prevention of a first acute event, include relatively simple risk assessment and leave substantial room for improvement both for risk ascertainment and selection of prevention strategies. Here, we review how advances in big data and predictive modeling foreshadow a promising future of improved risk assessment and precision medicine for CAD.

Recent Findings

Artificial intelligence (AI) has improved the utility of high dimensional data, providing an opportunity to better understand the interplay between numerous CAD risk factors. Beyond applications of AI in cardiac imaging, the vanguard application of AI in healthcare, recent translational research is also revealing a promising path for AI in multi-modal risk prediction using standard biomarkers, genetic and other omics technologies, a variety of biosensors, and unstructured data from electronic health records (EHRs). However, gaps remain in clinical validation of AI models, most notably in the actionability of complex risk prediction for more precise therapeutic interventions.

Summary

The recent availability of nation-scale biobank datasets has provided a tremendous opportunity to richly characterize longitudinal health trajectories using health data collected at home, at laboratories, and through clinic visits. The ever-growing availability of deep genotype-phenotype data is poised to drive a transition from simple risk prediction algorithms to complex, “data-hungry,” AI models in clinical decision-making. While AI models provide the means to incorporate essentially all risk factors into comprehensive risk prediction frameworks, there remains a need to wrap these predictions in interpretable frameworks that map to our understanding of underlying biological mechanisms and associated personalized intervention. This review explores recent advances in the role of machine learning and AI in CAD primary prevention and highlights current strengths as well as limitations mediating potential future applications.

Examining the Association between Mitochondrial Genome Variation and Coronary Artery Disease

Large-scale genome-wide association studies have identified hundreds of single-nucleotide variants (SNVs) significantly associated with coronary artery disease (CAD). However, collectively, these explain <20% of the heritability. Hypothesis: Here, we hypothesize that mitochondrial (MT)-SNVs might present one potential source of this “missing heritability”. Methods: We analyzed 265 MT-SNVs in ~500,000 UK Biobank individuals, exploring two different CAD definitions: a more stringent (myocardial infarction and/or revascularization; HARD = 20,405), and a more inclusive (angina and chronic ischemic heart disease; SOFT = 34,782). Results: In HARD cases, the most significant (p < 0.05) associations were for m.295C>T (control region) and m.12612A>G (ND5), found more frequently in cases (OR = 1.05), potentially related to reduced cardiorespiratory fitness in response to exercise, as well as for m.12372G>A (ND5) and m.11467A>G (ND4), present more frequently in controls (OR = 0.97), previously associated with lower ROS production rate. In SOFT cases, four MT-SNVs survived multiple testing corrections (at FDR < 5%), all potentially conferring increased CAD risk. Of those, m.11251A>G (ND4) and m.15452C>A (CYB) have previously shown significant associations with body height. In line with this, we observed that CAD cases were slightly less physically active, and their average body height was ~2.00 cm lower compared to controls; both traits are known to be related to increased CAD risk. Gene-based tests identified CO2 associated with HARD/SOFT CAD, whereas ND3 and CYB associated with SOFT cases (p < 0.05), dysfunction of which has been related to MT oxidative stress, obesity/T2D (CO2), BMI (ND3), and angina/exercise intolerance (CYB). Finally, we observed that macro-haplogroup I was significantly (p < 0.05) more frequent in HARD cases vs. controls (3.35% vs. 3.08%), potentially associated with response to exercise. Conclusions: We found only spurious associations between MT genome variation and HARD/SOFT CAD and conclude that more MT-SNV data in even larger study cohorts may be needed to conclusively determine the role of MT DNA in CAD.

[Molecular genetics of human hypertension]

A genetic influence on blood pressure was demonstrated more than 100 years ago and a simple Mendelian inheritance was initially presumed. Platt and Pickering conducted a lively debate on this topic. Platt favored the idea that a single gene or only a few genes were responsible for high blood pressure. Pickering presented research results, which supported the assumption that many genes exerted an influence on blood pressure. This was all in a period when it was not even known what genes were. Genome-wide association studies (GWAS) according to the Pickering model have identified > 500 blood pressure relevant gene loci, which are distributed over the whole genome. Each individual gene exerts only a small effect on blood pressure. The dark horses of hypertension research are the secondary causes. In pheochromocytoma, primary aldosteronism, Cushing's syndrome and even fibromuscular dysplasia (renovascular hypertension) the results indicate that a genetic cause regularly underlies secondary hypertension. This would therefore also partially confirm Platt's theory. In the meantime, a multitude of forms of hypertension have been described with a genetic inheritance according to Mendel. Each of these genetic variants exerts a considerable influence on blood pressure. A multitude of novel physiological mechanisms were explained by this. These findings will become therapeutically important. Therefore, it is incumbent upon clinicians to be optimally informed about these research results.

Molecular genetics of human hypertension

PURPOSE OF REVIEW

Genetic variance on blood pressure was shown about 100 years ago; a Mendelian inheritance was initially presumed. Platt and Pickering conducted a lively debate, whether blood pressure was inherited in a Mendelian fashion or whether the condition was polygenic. Genetic-hypertension research has appropriately followed both pathways.

RECENT FINDINGS

Genome-wide association studies, Pickering model, have identified more than 500 blood-pressure loci, the targets of which are waiting to be evaluated. Then, come the 'dark-horses' of hypertension, namely 'secondary' causes. These conditions have been remarkably elucidative including pheochromocytoma, primary aldosteronism, Cushing's syndrome, and even renovascular hypertension. All these conditions feature genetic causes. Finally, arrive the Platt followers. A plethora of Mendelian conditions located within the kidney are established. These syndromes involve increased sodium (as chloride) absorption in the distal nephron. Finally, nonsalt-dependent Mendelian forms involving the vascular directly have been described. Mechanistically, Mendelian forms have large effects on blood pressure and offer effective treatment targets.

SUMMARY

Which genetic models will bring us improved therapies? Ongoing studies will answer that question. It behooves the clinician to follow this dynamic area of research.

Multiple independent mechanisms link gene polymorphisms in the region of ZEB2 with risk of coronary artery disease

BACKGROUND AND AIMS

Coronary artery disease (CAD) arises from the interaction of genetic and environmental factors. Although genome-wide association studies (GWAS) have identified multiple risk loci and single nucleotide polymorphisms (SNPs) associated with risk of CAD, they are predominantly located in non-coding or intergenic regions and their mechanisms of effect are largely unknown. Accordingly, our objective was to develop a data-driven informatics pipeline to understand complex CAD risk loci, and to apply this to a poorly understood cluster of SNPs in the vicinity of ZEB2.

METHODS

We developed a unique informatics pipeline leveraging a multi-tissue CAD genetics-of-gene-expression dataset, GWAS datasets, and other resources. The pipeline first dissected SNP locations and their linkage disequilibrium relationships, and progressed through analyses of tissue-specific expression quantitative trait loci, and then gene-gene, gene-phenotype, SNP-phenotype relationships. The pipeline concluded by exploring CAD-relevant gene regulatory networks (GRNs).

RESULTS

We identified three independent CAD risk SNPs in close proximity to the ZEB2 coding region (rs6740731, rs17678683 and rs2252641/rs1830321). Our pipeline determined that these SNPs likely act in concert via the atherosclerotic arterial wall and adipose tissues, by governing metabolic and lipid functions. In addition, ZEB2 is the top key driver of a liver-specific GRN that is related to lipid levels, metabolic and anthropometric measures, and CAD severity.

CONCLUSIONS

Using a novel informatics pipeline, we disclosed the multi-faceted mechanisms of action of the ZEB2-associated CAD risk SNPs. This pipeline can serve as a roadmap to dissect complex SNP-gene-tissue-phenotype relationships and to reveal targets for tissue- and gene-specific therapeutic interventions.

Hsa-miR-584-5p as a novel candidate biomarker in Turkish men with severe coronary artery disease

Coronary artery disease (CAD) is still the preliminary cause of mortality and morbidity in the developed world. Identification of novel predictive and therapeutic biomarkers is crucial for accurate diagnosis, prognosis and treatment of the CAD. The aim of this study was to detect novel candidate miRNA biomarker that may be used in the management of CAD. We performed miRNA profiling in whole blood samples of angiographically confirmed Turkish men with CAD and non-CAD controls with insignificant coronary stenosis. Validation of microarray results was performed by qRT-PCR in a larger cohort of 62 samples. We subsequently assessed the diagnostic value of the miRNA and correlations of miRNA with clinical parameters. miRNA-target identification and network analyses were conducted by Ingenuity Pathway Analysis (IPA) software. Hsa-miR-584-5p was one of the top significantly dysregulated miRNA observed in miRNA microarray. Men-specific down-regulation (p = 0.040) of hsa-miR-584-5p was confirmed by qRT-PCR. ROC curve analysis highlighted the potential diagnostic value of hsa-miR-584-5p with a power area under the curve (AUC) of 0.714 and 0.643 in men and in total sample, respectively. The expression levels of hsa-miR-584-5p showed inverse correlation with stenosis and Gensini scores. IPA revealed CDH13 as the only CAD related predicted target for the miRNA with biological evidence of its involvement in CAD. This study suggests that hsa-miR-584-5p, known to be tumor suppressor miRNA, as a candidate biomarker for CAD and highlighted its putative role in the CAD pathogenesis. The validation of results in larger samples incorporating functional studies warrant further research.