An integrated approach to coronary heart disease diagnosis and clinical management

Abstract

The major issue in coronary heart disease (CHD) diagnosis and management is that symptoms onset in an advanced state of disease. Despite the availability of several clinical risk scores, the prediction of cardiovascular events is lacking, and many patients at risk are not well stratified according to the canonical risk factors alone. Therefore, adequate risk assessment remains the most challenging issue. Recently, the integration of imaging data with biochemical markers in a radiogenomic framework has been proposed in many fields of medicine as well as in cardiology. Multimodal imaging and advanced processing techniques can provide both direct (e.g., remodeling index, calcium score, total plaque volume, plaque burden) and indirect (e.g., myocardial perfusion index, coronary flow reserve) imaging features of CHD. Furthermore, the identification of novel non-invasive biochemical markers, mainly focused on plasma and/or serum samples, has increased the specificity of findings, reflecting several pathophysiological pathways of atherosclerosis, the principal actor in CHD. In this context, a multifaced approach, derived from the strengths of all these modalities, appears promising for finer risk stratification and treatment strategies, facilitating the decision-making and clinical management of patients. This review underlines the role of different imaging modalities in the quantification of coronary atherosclerosis and describes novel blood-based markers that could improve diagnosis and have a better predictive value in CHD.

Introduction

Cardiovascular diseases (CVDs) are the primary cause of mortality worldwide with 17.3 million deaths per year, and an estimation of 23.6 million in 2030, placing it as a relevant issue for the public health system [1,2]. Coronary heart disease (CHD) is the largest contributor of CVDs and mortality rate is due in prevalence to atherosclerosis, a chronic inflammatory condition of the arterial wall. Unfortunately, myocardial infarction (MI) is still a first common manifestation of CHD and, in about 50% of patients, angina pectoris is the first symptom of the pathology [1]. For this reason, an accurate and prompt diagnosis in CHD patients could improve prognosis and/or quality of life and allow timely and adequate therapeutic treatments (percutaneous or surgical myocardial revascularization, pharmacological therapy). Furthermore, efforts should be focused on primary prevention or early detection of subjects suffering from coronary atherosclerosis, in order to implement therapeutic strategies, so reducing morbidity, health expenditure, and mortality [3-6]. The risk of clinical manifestations of CHD is currently estimated according to multifactorial integrated prediction models developed on the basis of population studies that have allowed to evaluate the likelihood of cardiac events, even though some of them have a poor predictive value [7-13]. Imaging techniques have deeply increased early detection of CHD, although the invasive approach restricts their feasibility mainly to symptomatic patients. Among them, for their higher spatial resolution, intravascular ultrasound (IVUS), X-ray angiography (XRA), and computed tomography coronary angiography (CTCA) provide a direct evaluation and quantification of coronary artery alteration, while cardiac magnetic resonance (CMR) and nuclear medicine techniques (single photon computed tomography (SPECT), and positron emission tomography (PET)) provide indirect information of CHD, estimating myocardial perfusion and metabolism abnormalities that are consequent to coronary artery disease [14]. In addition, serum/plasma biomarkers can be mini-invasively extracted also in asymptomatic patients in order to analyze at different levels (e.g., cellular, biochemical, epigenetic and/or transcriptional) atherosclerosis and CHD development.

This review underlines the role of different imaging modalities in the setting of coronary atherosclerosis and describes novel blood-based markers that could improve diagnosis and have a better predictive value in CHD.

Imaging markers for direct CHD diagnosis

Several imaging techniques have been developed and used extensively over the last years in order to exclude and/or detect CHD with the aim to guide optimal patient management.

IVUS employs a miniature ultrasound probe placed at the tip of a coronary catheter; the signals received according to the different acoustic impedance are reconstructed into a real-time tomographic gray-scale image [15]. In coronary arteries, two borders are well defined: the blood-intimal border and the external elastic membrane (EEM). Some morphologic features of atherosclerotic coronary plaques can be invasively assessed by this tool such as soft, fibrous, calcified, necrotic and lipid components. The reliability of ultrasound imaging in predicting the composition of atherosclerotic plaque components has been demonstrated in histological comparative studies [16]. Indeed, IVUS allows to clearly depict the vascular wall and accurately calculate the remodeling index.

Many clinical trials have disclosed significant improvements in patient outcomes and reduced complications [17,18]. Meanwhile, plenty of clinical pharmacological trials employs IVUS to demonstrate its beneficial effects because this technique offers great value for the precise quantification of atherosclerosis progression or regression [19].

Differently, XRA is a procedure that uses a special dye (contrast medium) and x-rays to selectively highlight the arterial lumen (lumenogram), allowing the identification of caliber alterations, like stenoses/occlusions or aneurysms, and, at the same time, to treat these conditions. This technique is a reliable reference standard for the estimation of the real coronary lumen and intracoronary pressure measurement for the detection of functionally relevant stenoses and fractional flow reserve (FFR) calculation. However, XRA is an invasive technique like IVUS, but unlike it, it is not convincing with regard to prognostic value [20]. Moreover, conversely to other imaging techniques for direct CHD diagnosis, XRA does not allow the evaluation of vascular wall, and CHD is largely not a disease of the lumen itself but an abnormality of the vessel wall [21].

CTCA has overcome several of previous technique drawbacks, allowing an accurate quantitative evaluation of the arterial lumen, vessel wall and the extension, severity and composition of the atherosclerotic plaque [14,22]. A first level of evidence that can be extracted by CTCA is represented by the detection of coronary calcium deposits and the assessment of lumen stenosis [23]. Calcium burden is usually quantified using the ‘Agatston score’, an aggregate score across entire vessel territories rather than across specific atherosclerotic plaques, and mainly related to risk stratification of asymptomatic patients [24], than of symptomatic ones [25]. The amount of calcium correlates roughly to the total amount of coronary atherosclerosis but the correlation with the degree of luminal narrowing is poor. Even with severe calcifications, luminal stenosis is not necessarily present and a ‘zero’ calcium score cannot be used to rule out coronary artery stenosis in symptomatic individuals, especially in young subjects and in individuals with acute symptoms [24,26]. This evidence can be described by the remodeling index, a CTCA parameter used to define the main grow direction of plaque toward the inner or outer layers of the vascular wall and not evaluable by XRA technique [27]. Moreover, atherosclerotic plaques are often composed by both calcific and non-calcific components (mixed plaque), a feature highly predictive for CHD complications and easily quantifiable (e.g., total plaque volume, calcific plaque volume, non-calcific plaque volume) by advanced CTCA scanners, like dual energy CT or spectral CT [28,29]. These state-of-art computed tomographies, thanks to improved spatial and temporal resolution, better tissue characterization, and lowering of dose and iodine load, have a wide range of applications also in patients with high or irregular heart rate, encompassing from myocardial perfusion, coronary blood flow velocity and pressure and non-invasive FFR assessment by the application of the computational fluid dynamics and mathematical models [30]. In conclusion, among the imaging techniques for direct CHD diagnosis, CTCA is the unique that allows a “simultaneous” analysis of the entire coronary tree at a low dose of contrast medium load, proposing this tool as suboptimal for early diagnosis and CHD risk assessment (Figure 1).

Figure 1.

Integrated diagnostic approach in coronary heart disease. The figure shows the imaging modalities for direct assessment of coronaries and the diversity of circulating biomarkers that could be associated with the diagnosis, outcome prediction and risk assessment in coronary heart disease (CHD). The left panel represents X-ray angiography (XRA), coronary computed tomography angiography (CTCA) and intravascular ultrasound (IVUS) in coronary imaging and respective parameters. For XRA: luminal narrowing, blood flow and fractional flow reserve (FFR); for CTCA: coronary artery calcium (CAC), remodeling index (RI), plaque features, plaque burden and FFR CT; for IVUS: lumen, RI, vessel structure, external elastic membrane (EEM) and plaque features. The right panel depicts markers that can be analyzed in serum/plasma or whole blood and reflect coronary artery alterations: cellular: peripheral blood mononuclear cells (PBMCs), neutrophils, lymphocytes, endothelial progenitors cells (EPCs), CD31+ cells; biochemical markers: inflammatory pathways, cytokines, adhesion molecules, enzymes and metabolites; epigenetic markers: DNA methylation levels and differential expression of circulating non-coding RNAs (ncRNAs); and transcriptional markers: genes coding for pro-and antioxidant molecules, cell motility proteins, signaling receptors, transcription factors, inflammatory and metabolic mediators.

Imaging markers for indirect CHD diagnosis

CMR provides excellent soft tissue contrast, allowing functional, perfusional, morphological and anatomical evaluations, without exposing the patient to ionizing radiation, like both XRA and CTCA [14]. Recently, CMR has been applying also to direct anatomical coronary visualization, although it is still limited in the visualization of distal coronary segments due to inferior spatial resolution, compared to CTCA [31]. Several applications for the assessment and follow-up of patients with CHD include the estimation of myocardial viability thanks to T₁ mapping, T₂ mapping and the monitoring of left ventricle remodeling after acute myocardial infarct through diffusor tensor imaging besides first pass perfusion CMR and delayed contrast enhancement [32].

Nuclear medicine imaging by SPECT and PET, instead, provide, through the injection of a radiotracer, functional and perfusional analysis, the evaluation of ischemia and blood flow quantification [33,34]. Myocardial perfusion imaging triggered by ECG monitoring allows also consideration of wall motion and thickening; moreover, changes in blood flow are considered positive for suspected CHD. Cardiac PET is useful for perfusion imaging, functional evaluation and assessment of myocardial viability. For perfusion imaging, commonly used PET radiotracers are ⁸²Rubidium, ¹³Nitrogen-ammonia and ¹⁵Oxygen-water. ¹⁸F-FDG is used for diagnosis of myocardial viability and it is considered the most sensitive modality for predicting left ventricular functional recovery post-coronary revascularization. In particular, the integration of perfusion and viability, detected by PET, allows the identification of myocardial stunning (normal perfusion but reduced metabolism) and myocardial hibernation (reduced perfusion but preserved metabolism) [35].

In vitro biomarkers for CHD diagnosis and risk prediction

Several studies focused on the introduction of multiplex biomarker assays analyzing different circulating molecules and their integration with current clinical scores in multivariable prediction model for a better diagnosis and risk stratification [36,37]. Concerning in vitro biomarkers, many molecules can be extracted from serum and/or plasma of asymptomatic subjects and CHD patients that could reflect at different levels (e.g., cellular, biochemical, epigenetic and/or transcriptional) atherosclerosis and coronary alterations (Figure 1).

Cellular markers

Circulating cells have been proposed as blood-derived biomarkers in CHD in the pathogenesis of CHD, secreting a wide plethora of specific biomarkers [38]. Several studies showed an increased number of monocytes in the blood of patients with atherosclerosis and in subjects with cardiovascular risk factors [38,39]. The multiple roles of monocytes in the atherogenic process are attributed to the existence of sub-populations characterized by the expression of different surface markers with consequent functional changes and response to stimuli as well as alterations in gene expression [40-42]. Circulating neutrophil count and neutrophil/lymphocyte ratio are also emerging markers of the presence and severity of CHD, correlating with plaque vulnerability [43,44]. In peripheral blood mononuclear cell (PBMCs) of CHD patients, were found a set of 190 genes differentially expressed compared to healthy controls [45]. Another heterogeneous class of cells involved in CVDs is represented by the endothelial progenitor cells (EPCs). Several studies have reported a variation of cell number and a functional impairment in patients with CHD strictly dependent on the presence of cardiovascular risk factors and associated with the severity of coronary lesions and clinical outcome [46-48]. Finally, Kim et al. performed a flow cytometry analysis of CD31(+) cells from the peripheral blood of CHD patients showing a significant correlation between CD31(+) cell count and the number of atherosclerotic vessels [49] (Table 1).

Table 1.

Cellular markers associated to CHD diagnosis and prognosis

Biomarker	Source	Regulation	Clinical role	Ref.
CD14++CD16-CCR2+; CD14++CD16+CCR2+; CD14+CD16++CCR2- monocytes	PBMCs	Reduced expression of CD14 and CD14+CD16++CCR2- subpopulation in CHD	Diagnostic	[40]
CD4+CD28null T cells	PBMCs	CD4+CD28null population from RF and ACS groups had higher expression levels of cytotoxic molecules	Predictive for treatment	[41]
CD3+/CD31+ T cells	PBMCs	Increased levels in ACS patients	Predictive after PCI	[42]
Neutrophil/Lymphocyte ratio	Leucocytes	Increased N/L ratio associated with severity of CHD and plaque vulnerability	Diagnostic/Predictive of long term outcome	[43,44]
EPCs	PBMCs	Number reduction and functional impairment in CHD patients; cell count dependent on the number of RF; associated with the severity of coronary lesions and less sub-stent plaque burden	Diagnostic/Predictive of future CV events and PCI follow-up	[46-48]
CD31(+) cells	Blood	Increased CD31(+) cells in UA patients; significant correlations between cell count and the number of atherosclerotic coronaries	Diagnostic/Predictive of UA	[49]

Abbreviations: CHD = Coronary heart disease; RF = risk factor; ACS = Acute coronary syndrome; PBMC = Peripheral blood mononuclear cell; PCI = Percutaneous coronary intervention; UA = Unstable angina.

Biochemical biomarkers

Inflammatory biomarkers appear to have an important prognostic value in patients with CVDs and may be useful in the diagnosis of apparently healthy subjects without known CHD who cannot be assessed with conventional risk factors. In the last years, several biomarkers have been identified, although not for everyone there are studies that show a possible correlation with instrumental imaging parameters.

In particular, the group of biomarkers of interest includes: transforming growth factor beta (TGF-β1) [50]; cellular adhesion molecules (CAMs) [51,52]; monocyte chemoattractant protein-1 (MCP-1) [53-55]; stromal cell-derived factor-1α (SDF-1α) [56-58]; lectin-like oxidized low density lipoprotein receptor 1 (LOX-1) [59-61]; pentraxin 3 (PTX3) [62-64]; bilirubin [65-67] and haemoglobin A1c (HbA1c) [68,69], whose serum and/or plasmatic levels were associated with the presence and severity of CHD.

In addition, among emerging biomarkers, we selected: serum amyloid A (SAA) proteins [70,71]; fibrinogen [72,73]; myeloperoxidase (MPO) [74-77]; paraoxonase-1 (PON1) [78]; matrix metalloproteinases (MMPs) [79,80]; proprotein convertase subtilisin/kexin type 9 (PCSK9) [81-83]; lipoprotein-associated phospholipase A2 (Lp-PLA2) [84,85]; retinol-binding protein-4 (RBP4) [86,87]; angiopoietin-like 4 (ANGPTL4) [88] and sphingolipids [89-91] that could be potential biomarkers for the assessment of CHD severity. Indeed, these emerging biochemical markers could potentially find a useful application for the diagnosis and prognosis assessment in CHD patients. Furthermore, several works focused on the introduction of multiplex biomarker assay analyzing different circulating molecules and their integration in new grading score for a better diagnosis and risk stratification [55,92-96] (Table 2).

Table 2.

Biochemical markers associated to CHD diagnosis and prognosis

Biomarker	Source	Regulation	Clinical role	Ref.
TGF-β1	Serum	High levels in AMI patients compared to SA and UA	Diagnostic/Risk stratification	[50]
sVCAM-1/sICAM-1	Serum	Higher levels in UA patients	Predictive of ACS	[51,52]
MCP-1	Plasma	Positive correlation with the extent of coronary atherosclerosis in UA patients detected by RXA; high concentrations in ACS patients	Diagnostic/Predictive of increased risk for death or AMI	[53-55]
SDF-1α	Plasma	Reduction in STEMI patients compared to SA or to NSTEMI; association with disease severity	Diagnostic/Predictive of negative outcome	[56-58]
LOX-1	Serum	Increased concentrations in CHD patients in association with disease severity	Diagnostic/Predictive of increased risk of CHD	[59-61]
PTX3	Plasma	Biomarker of plaque vulnerability	Diagnostic	[62-64]
Bilirubin	Serum	Lower levels in patients with critical stenosis and non-calcified/mixed plaques detected by CTCA	Diagnostic/Predictive of risk stratification, CHD onset and long term-mortality	[65-67,162]
HbA1c	Plasma	High levels associated to CAC and CHD severity in non diabetic patients	Diagnostic/Predictive	[68,69]
SAA	Plasma	Elevated levels in CHD patients	Predictive of disease risk and worse prognosis	[70,71]
MPO	Plasma	Marker of plaque instability and disease severity	Diagnostic	[74-77]
PON1	Plasma	Reduced activity in CHD patients	Diagnostic/Predictive	[78]
MMP-9/MMP-8	Plasma/Serum	Biomarkers of plaque instability	Diagnostic	[79,80]
PCSK9	Serum	High concentrations are associated with CAC and disease severity	Diagnostic/Predictive of CV events	[81-83]
Lp-PLA2	Plasma	High levels associated with CHD severity	Diagnostic/Predictive of CHD and mortality	[84,85]
RBP4	Serum	High levels associated with coronary lesion severity in stable CHD and ACS	Diagnostic/Predictive of CV events	[86,87]
ANGPTL4	Plasma	High levels in patients at risk	Predictive of CV event occurrence	[88]
Sphingolipids	Plasma	Positively related to CHD and subclinical atherosclerosis	Predictive of CV events	[89-91]
hs-TnT/hs-CRP	Serum	Associated with CHD burden and change in plaque composition detected by CCTA	Diagnostic	[154]
IL-10	Serum	Reduced levels in patients with ACS	Diagnostic/Predictive of long-term adverse outcomes	[155,156]
IL-8	Serum	High levels in patients in CHD patients	Predictive of long-term outcome	[157]
IL-6	Plasma/Serum	High concentration in patients with multivessel atherosclerosis and calcified plaque assessed by CCTA	Diagnostic	[158]
Adiponectin	Serum	Low levels associated with lipid-rich, non-calcified plaques and multivessel CHD by CTCA	Diagnostic	[159-161]

Abbreviations: ACS = Acute coronary syndrome; CAC = Coronary Artery Calcium; CHD = Coronary heart disease; CTCA = Coronary Computed Tomography Angiography; SA = stable angina; STEMI = ST-elevation myocardial infarction; NSTEMI = non-ST-elevation myocardial infarction; AMI = acute myocardial infarction; UA = unstable angina.

Epigenetic markers

Epigenetic modifications are involved in a variety of pathological conditions as well as in CVDs and atherosclerosis [4-6,97-101]. Epigenetic mechanisms include DNA methylation, histone modifications, and regulation by non-coding RNA (ncRNA) [102,103]. Several studies evaluated the methylation status of genomic DNA from blood cells in CHD and ACS patients and in subjects with cardiovascular risk factors, observing a different methylation pattern, as well as a methylation signature predictive of increased risk of cardiac events, ischemic heart disease, stroke, and mortality [104-108].

Furthermore altered specific DNA methylation has been reported in several promoters, particularly in genes related to inflammation [109,110], coagulation [111], hypertension [112], glucose [113,114] and lipid metabolism [115,116].

In addition, regulatory T (Treg) cells have also been shown to play a protective role in atherosclerosis avoiding plaque rupture. DNA demethylation of the transcription factor Forkhead box P3 (FOXP3) gene was found to be essential for the maintenance of the suppressive properties of this cellular subtype. Indeed, an increase of FOXP3 methylation was observed in cultures of PBMCs obtained from ACS patients compared to controls [117] (Table 3).

Table 3.

Epigenetic markers associated to CHD diagnosis and prognosis

Biomarker	Source	Regulation	Clinical role	Ref.
Global DNA methylation	Lymphocytes	Hypermethylation in patients related to hyperhomocysteinemia	Diagnostic	[104]
LINE-1	Leucocytes	Lower methylation in CHD patients	Diagnostic/Predictive of increased risk of acute events and mortality	[105]
Alu/Sat2	Leucocytes	High methylation in CHD patients with cardiovascular risk factor	Diagnostic	[106]
Global DNA methylation	Leucocytes	Hypermethylated regions in patients with hyperhomocysteinemia	Diagnostic	[107]
Global DNA methylation	Leucocytes	Identification of 47 CpG islands associated with ACS	Diagnostic	[108]
PLA2G7	Leucocytes	High promoter methylation in CHD patients	Diagnostic/Predictive of CHD risk gender and age specific	[109]
PTX3	Leucocytes	Lower promoter methylation in CHD group	Diagnostic	[110]
Factor VII	PBMCs	Promoter hypomethylation in CHD patients	Diagnostic	[111]
HSD11B2	PBMCs	Promoter methylation associated with hypertension	Diagnostic	[112]
INS/GNASAS	Leucocytes	Locus hypermethylation in AMI patients	Predictive of risk of AMI in women	[113]
GCK	Leucocytes	Hypomethylation in CHD patients compared to controls	Predictive of risk of CHD onset	[114]
ABCA1	Leucocytes	Higher methylation in CHD patients associated with low HDL; association with aging and CHD in men	Diagnostic	[115,116]
FOXP3	Regulatory T cells	Increased methylation in ACS patients	Diagnostic	[117]
Microarray	Plasma/Serum	Downregulation: miR-17-92 cluster, -126, -145, -155; Upregulation: miR-133, -208a associated with CHD severity	Diagnostic	[121-123]
Microarray	PBMCs	Upregulation of miR-134, -135a, -147, -198, -370 in UA compared to SA patients	Diagnostic/Predictive of acute events	[124]
Microarray	Plasma	Upregulation of miR-1, -126, and -483-5p in SA patients vs. controls; upregulation miR-1, -126, and -133a in UA patients vs. controls	Diagnostic	[125]
Microarray	Plasma	Overexpression of miR-106b/25 cluster, miR-17/92a cluster, miR-21/590-5p family, miR-126* and miR-451 in patients with vulnerable CHD	Diagnostic	[126]
mir-197/mir-223	Serum	Elevated levels in CHD patients	Predictive of cardiovascular death	[127]
miR-31	Serum	Higher levels in CHD patients with restenosis compared to patients without restenosis and in healthy controls	Diagnostic	[128]
miR-214	Plasma	Circulating levels related to the severity of coronary stenosis	Diagnostic	[129,130]
Realtime PCR	Plasma	Upregulation of miR-122 and miR-370 in hyperlipidemic CHD patients; association of increased levels of miR-122 and miR-370 with disease severity	Diagnostic	[131]
Microarray	Plasma	Low levels of miR-145, miR-155 and let-7c levels in CHD patients compared to controls	Diagnostic	[132]
Realtime PCR	Plasma	High levels of miR-17-5p are associated with CHD severity	Diagnostic	[133]
Microarray	Platetets	Upregulation of miRNA340* and miRNA624 in patients with CHD	Diagnostic	[134]
Realtime PCR	Plasma/PBMCs	Downregulation of miR-155 and miR-146a/b during ACEI/ARB treatment	Predictive markers of CHD risk and treatment efficacy	[135,136]
EPCs	PBMCs	Upregulation of miR-221, miR-222 and miR-92a; increased EPC number and decreased miR-221/222 levels after atorvastatin therapy	Predictive markers of treatment efficacy	[137,138]

Abbreviations: ACEI = Angiotensin converting-enzyme inhibitor; ARB = Angiotensin II receptor blocker; CHD = Coronary heart disease; PBMC = Peripheral blood mononuclear cells; AMI = acute myocardial infarction; SA = stable angina; UA = unstable angina; EPC = endothelial progenitor cell.

MicroRNAs (MiRNAs) are short noncoding single-strand RNAs with the function of inhibiting gene expression through mRNA degradation or translational repression [118]. MiRNAs are stable in several bodily fluids, so several studies focused on the identification of these molecules in plasma specimens as well as in blood cells with the aim to correlate their expression with CHD [119,120].

Circulating levels of specific endothelial cell, smooth muscle cell and inflammation associated miRNAs, were significantly reduced in patients with stable CHD, while cardiac-specific miRNAs were upregulated [121-123].

Several studies also showed the important role of miRNA signature in vulnerable CHD, particularly for their ability to discriminate patients with UA from those with SA, suggesting that these circulating molecules could be useful to identify patients at risk for ACS and predict the clinical outcome [124-127].

Serum levels of miR-31 were higher in CHD patients with in stent restenosis compared to patients without stent restenosis and miR-214 downregulation was linked to disease severity [128-130]. In addition, other studies also showed that miRNA alterations are associated with disease severity [131,132].

Furthermore, miR-17-5p overexpression was reported to be an independent factor associated with the severity of atherosclerosis [133].

Platelets play an important role, both during plaque rupture and in the formation of atherosclerotic plaque. Indeed, miRNA expression profiles of platelets from patients with premature CHD revealed an upregulation of miR-340* and miR-624* in patients compared to healthy controls [134].

The pattern of miRNA expression is also influenced by therapeutic treatments, suggesting important implications for patient management. In whole blood samples of CHD patients and controls, 11 miRNAs were significantly down-regulated in CHD group. Particularly, miR-155, which is known to target the AT1 receptor, was found to be associated with ACEI/ARB use [135]. Furthermore, miR-146a/b were higher in PBMCs of CHD patients but under angiotensin II receptor blocker inhibitors and statin treatments, their expression levels decreased [136]. Interestingly, also lipid lowering therapy with statins is able to influence miRNA expression in CHD [137,138] (Table 3).

Transcriptional markers

Genome-wide gene expression profiling is a promising strategy for the identification of novel disease biomarkers [139]. Several studies on gene expression profiling of blood cells identified a differential transcriptional signature in CHD patients and healthy subjects. Particularly, the major alterations were discovered in genes codifying for pro- and antioxidant molecules, cell motility proteins, signaling receptors, transcription factors, inflammatory molecules and mediators. Interestingly, the expression pattern was found to correlate with the severity of CHD and gene expression in vascular tissues, indicating a mirroring between circulating cells and changes in the atherosclerotic vessel wall [140-146]. Transcriptional deregulations were also found in genes involved in DNA repair, as reported in a recent study by Ahmadi el al. [147].

Gene expression changes may reflect not only the presence and activity of disease but also environmental modifier effects, as well as treatment response. Indeed, in a study by Taurino et al. 365 differentially expressed genes were found in patients with CHD vs. healthy controls (175 genes were upregulated, and 190 genes were downregulated). Furthermore, in a group of patients was analyzed whole-blood gene expression before and after a cardiac rehabilitation following surgical coronary revascularization. In patients underwent rehabilitation, were found 645 genes differentially expressed at the beginning and the end of the program, with 196 genes upregulated and 449 genes downregulated. The expression levels of genes involved in oxidative phosphorylation and mitochondrial dysfunction were higher in CHD patients compared to control subjects. Completion of the rehabilitation treatment was characterized by a downregulation of the genic signature of oxidative stress and mitochondrial impairment [148]. Furthermore, a whole genome expression analysis identified six genes with a differential expression in monocytes of patients versus controls; ABCA1, ABCG1, and RGS1 were down-regulated in patients, whereas ADRB2 and FOLR3 were upregulated compared to matched controls and this expression pattern was influenced by aspirin and statin therapy [149] (Table 4).

Table 4.

Transcriptional markers associated to CHD diagnosis and prognosis

Biomarker	Source	Regulation	Clinical role	Ref.
Microrray	PBMCs	Upregulation of cytokine and EGR family genes; EGR1 levels able to discriminate ischemic and non-ischemic CHD patients	Diagnostic	[140]
Whole-genome microarray	PBMCs	Differential expression of 14 genes related to coronary stenosis	Diagnostic	[141]
Array	PBMCs	160 differently expressed genes in patients; expression signature correlated with the severity of CHD and gene expression in vascular tissues	Diagnostic	[142]
Myocardin/GATA4/Nkx2.5	PBMCs	Higher transcriptional levels in patients in relation to disease severity	Diagnostic	[143]
Nrf2	PBMCs	Lower gene expression in patients than control subjects	Diagnostic	[144]
Homer1/IL-1β/TNF-α	Leucocytes	Higher mRNA levels in CHD patients compared to controls	Diagnostic	[145]
MT-COI	Monocytes	Low mRNA levels in patients at risk	Predictive of CV events	[146]
MSH2/XRCC1/ATM	PBMCs	Upregulation in diabetic CHD patients	Diagnostic	[147]
Microarray	Leucocytes	Downregulation of COX7C, ATP5I NDUFA1 and CASP3	Predictive markers after cardiac rehabilitation	[148]
Microarray	Monocytes	Downregulation of ABCA1, ABCG1 and RGS1, upregulation of ADRB2 and FOLR3	Diagnostic	[149]
GES score	Whole blood	GES score is associated with plaque volume and phenotype by IVUS and atherosclerotic plaque burden and stenosis by CTCA	Diagnostic	[152,153]

Abbreviations: EGR = Early growth response; CHD = Coronary heart disease; PBMCs = Peripheral blood mononuclear cell.

Integration of biomarkers and direct imaging features in CHD

While several studies have tested circulating biomarkers in imaging-detected CHD (only recently by CTCA), some authors have tried to correlate imaging features to serum/plasma biomarkers, in order to increase diagnostic power and risk stratification of patients with coronary alterations.

In a multicenter study a gene expression score (GES) based on age, sex and the expression levels of 23 genes in peripheral blood cells was developed to assess the likelihood of CHD in non diabetic patients [150]. A subsequent study showed that GES could be a more accurate predictor of obstructive CHD compared to clinical estimation scores [151].

In the Atlanta study, Joshi et al. correlated the validated GES score and coronary plaque composition by IVUS with radiofrequency backscatter analysis (IVUS/VH). GES was significantly associated with plaque volume, necrotic core composition, and dense calcium. Data suggested that this composite gene expression score is not only predictive of obstructive coronary artery disease, as previously reported, but also predictive of larger atherosclerotic plaque burden with a more vulnerable phenotype [152]. GES was also significantly associated with plaque burden and luminal stenosis assessed by CTCA. Indeed, in a study on 610 patients, Voros et al. reported a significant association between GES score, plaque burden expressed by coronary artery calcium and stenosis severity reported as score index. Particularly, GES significantly correlated with maximum luminal stenosis and segment stenosis score index. A low score had a sensitivity of 0.90 and a high score a specificity of 0.87 for stenosis ≥ 70% [153].

C-reactive protein (CRP) is the most extensively studied systemic marker of inflammation. A recent study by Seifarth et al. has shown that plasmatic levels of high-sensitivity C-reactive protein (hs-CRP) and high-sensitivity troponin T (hs-TnT) are weakly associated with a significant increase in CHD burden and change in plaque composition detected by CTCA using a semiquantitative crosssection-based score in a 2-year follow-up [154].

Several studies showed also the involvement of interleukins in plaque instability and their possible predictive value of cardiovascular events in long-term outcome [155-157]. Indeed, in a recent study Harada K. et al. evaluated the association between inflammatory markers and coronary artery plaque parameters assessed by CTCA on 220 subjects with suspected CHD showing that circulating levels of hs-CRP and IL-6 were significantly higher in CHD patients. Particularly, plasma IL-6 concentration was significantly associated with 4-9 segment plaques compared to patients without or with 1-3 segments. On the other hand, plasma hs-CRP level was associated with the presence of calcified plaque, independently of traditional cardiovascular risk factors [158].

MCP-1 plays a key role in the recruitment of monocytes to sites of inflammation, promoting the initiation of the fatty streak, plaque instability, as well as remodeling after MI. A positive correlation between circulating MCP-1 levels and the extent of coronary atherosclerosis, expressed by an angiographic severity score, was found in patients with UA. These findings suggest that circulating MCP-1 concentration likely reflects the coronary wall damage [53].

Adiponectin is a protein hormone playing a protective role of vascular walls from atherosclerosis. Several studies showed a significant inverse correlation between multivessel CHD and lipid-rich non-calcified plaques assessed by CTCA and serum levels of adiponectin independently by other significant risk factors [159-161]. Although bilirubin has long been considered a waste product, it is currently recognized as an endogenous antioxidant molecule, involved in the attenuation of lipid peroxidation and playing an anti-atherogenic role. In a study on 1151 patients the association between total serum bilirubin levels and presence, severity and plaque composition evaluated by CTCA was determined. Data showed that subjects with primarily non-calcified plaques and mixed plaques had lower bilirubin levels compared to patients with calcified plaques and normal subjects. Furthermore, serum total bilirubin levels were found to be lower in patients with any coronary plaque [162].

Conclusions and future perspectives

The need to improve diagnosis and risk prediction has prompted the search for novel markers in cardiovascular medicine. Literature data suggest that CTCA could substantially reduce the number of invasive procedures, increasing the safety of patients, and allows a more precise planning of potential treatment options. Furthermore, strong evidence has also emerged on the usefulness of coronary calcium score assessed by CTCA. In association with imaging improvements, novel high-throughput platforms investigating proteomic, metabolomic, epigenomic, and transcriptomics profiles together with genome-wide association studies may generate “multimarker CHD scores” with a higher predictive power than the use of a single biomarker. Surrogate biomarkers of coronary atherosclerosis and advanced imaging techniques could represent important cornerstones to characterize sub-clinical and clinical atherosclerosis with a consequent facilitation in the decision-making and clinical management of patients.

Disclosure of conflict of interest

None.