Serum platelet-derived growth factor is a novel biomarker for the prediction of vulnerable plaque in patients with NSTE-ACS

Objective: To investigate the relationship between serum platelet-derived growth factor (PDGF) and vulnerable plaque in patients with non-ST-elevation acute coronary syndrome (NSTE-ACS). Methods: A total of 65 patients with NSTE-ACS were divided into vulnerable plaque group (n=46) and vulnerable plaque and stable plaque group (n=19) according to intravascular ultrasound (IVUS) examinations. Total cholesterol (TC), triglycerides (TG), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C) and serum PDGF were measured. Plaque characteristics and components were analyzed using gray-scale and iMap-IVUS. Correlation was performed between plaque characteristics and ACS markers. Logistic regression analysis was applied to determine risk factors. Receiver operating characteristic (ROC) curve was used to evaluate the predictive value. Results: Patients in vulnerable plaque group had visible higher levels of TG, LDL-C and PDGF (P < 0.05). There were significant differences in minimal lumen area (MLA), plaque area, plaque burden, fibrotic (FI), clipidic (LI) and necrotic core (NC) between the two groups (P < 0.05). PDGF was weakly correlated with plaque burden (R=0.428, P < 0.05), as well as moderately correlated with NC (R=0.669, P < 0.05). Multivariate analysis showed that serum PDGF (OR 4.751, [95% CI 1.534-29.543], P=0.05) was an independent risk factor of vulnerable plaque. The area under the curve (AUC) was 0.876 (95% CI 0.804-0.948, P=0.001). Conclusion: Serum PDGF could be used as a novel biomarker for the prediction of vulnerable plaque in patients with NSTE-ACS. study to find a novel biomarker to noninvasively predict vulnerable plaque in patients with NSTE-ACS. The results showed a higher level of serum PDGF in vulnerable plaque group; correlation analysis showed that PDGF was weakly correlated with plaque burden and moderately correlated with NC; Logistic regression analysis revealed that serum PDGF was independent risk factor of vulnerable plaque; ROS analysis demonstrated that PDGF could be used as a predictor to indicate plaque instability.


Introduction
Non-ST-elevation acute coronary syndrome (NSTE-ACS) is a major life-threatening disease worldwide, and the rupture of vulnerable plaque were demonstrated to be associated with ACS in clinic [1]. IVUS could be feasible to characterize the morphology of culprit coronary lesions or satellite lesions in patients undergoing coronary angiography, which provided clues to identify potentially vulnerable plaques [2]. However, it has not been widely used in clinic due to its invasiveness and high expenses.
Identifying a novel biomarker may aid detection of vulnerable coronary plaques [3,4], and such symptomatic and asymptomatic patients would dramatically benefit from intensive antithrombotic therapy or stenting in patients with NSTE-ACS [5].
As one of growth factors, platelet-derived growth factor (PDGF) was originally considered to be platelet and serum mitogen that regulated the growth and division of glial cells, smooth muscle cells (SMC) and fibroblasts [6]. Previous researches further unveiled that PDGF was involved in atherosclerosis [7] and tumor growth and metastasis [8]. In a heart-specific transgenic mice model, PDGF-D was demonstrated to contribute to the development of cardiac fibrosis and proliferation of vascular smooth muscle cells (vSMCs) [9]. On the contrary, there were some studies demonstrated that PDGF accelerated the generation of bone marrow-derived cardiac myocytes after acute coronary occlusion [10] and improved ventricular function after infarction [11]. Moreover, the relationship between PDGF and vulnerable plaque remains poorly elucidated.
Therefore, it is of great significance to identify the relationship between the level of serum PDGF and plaque characteristics in patients with NSTE-ACS from the perspective of clinical trials. In this study, the correlation analysis was performed between vulnerable coronary plaques characterized using IVUS in conjunction with the detection of serum PDGF, as well as the predictive value of PDGF for vulnerable plaques was also investigated in NSTE-ACS patients.

Participants
From May 2016 to February 2018, 65 patients diagnosed with NSTE-ACS were continuously selected from coronary angiography (CAG) examination in Tianjin Chest Hospital, and the degree of coronary stenosis was between 50% and 90%. Intravascular ultrasound (IVUS) examinations were performed, and the participants were divided into vulnerable plaque group (n = 46) and vulnerable plaque and stable plaque group (n = 19) according to the definition of vulnerable plaque [12]. Patients with bivascular or multivascular disease were divided into vulnerable plaque groups as long as they had vulnerable plaques in any major vessel. NSTE-ACS diagnosis was based on the guidelines for the diagnosis and treatment of NSTE-ACS [13]. Diabetes mellitus was defined as active use of an antidiabetic agent, or fasting plasma glucose level ≥7.0mmol/L or casual plasma glucose level ≥11.1mmol/L. All participants gave written informed consent and the study was approved by Tianjin Chest Hospital ethics committee. NSTE-ACS patients aged > 18 years old, who were first diagnosed with NSTE-ACS; who had diameter stenosis > 50% in two or more major vessels and had not recently been heavily exposed to antiplatelet and anticoagulant drugs were eligible for enrollment. Exclusion criteria: (1) Acute STEMI or previous history of old myocardial infarction; (2) Serum creatinine > 2.

CAG and IVUS procedure and analysis
Coronary angiography was performed using single-arm digital subtraction angiogram (DSA, Boston Scientific, Natick, MA, USA) and digital imaging system together with standard Judkins method. Visual measurement was used to estimate the stenosis degree of reference vessel diameter and lesion location diameter, and the angiographic results were evaluated by at least two experienced cardiologists.
IVUS imaging was then performed in patients with NSTE-ACS with a degree of coronary stenosis between 50% and 90% (Fig 1A).. The catheter was advanced sufficiently distal to the culprit lesion, and then automated pullback was performed at a speed of 0.5mm/s. Real-time image acquisition and continuous input and storage, followed by localization analysis were further applied using QIVUS iMap BasicViewer 3.0 (Medis medical imaging systems, Leiden, the Netherlands). After identifying the external elastic membrane (EEM) and the boundary of each layer (0.4mm thickness), the following parameters were measured: EEM cross-sectional area (EEM CSA), lumen area and minimal lumen area (MLA). Plaque area = EEM CSA -lumen area, and plaque burden = plaque area / EEM CSA. The plaque components were quantitatively analyzed by using different colors to represent different plaque components, and the plaque components were fibrotic (FI), clipidic (LI), necrotic core (NC) and dense calcium (DC) (Fig 1B and 1C)..

Statistical analysis
SPSS 20.0 statistical software (SPSS Inc, Chicago, IL) was utilized for all statistical analysis. The continuous variables were exhibited as mean ± SD. ANOVA was utilized for comparison between multiple groups, following Tukey's multiple comparison tests. Data were expressed as frequencies and percentages for category variables, using Chi-square test. Correlation was evaluated by using Pearson analysis. Logistic regression analysis was applied to determine risk factors. ROC curve was used to evaluate the predictive value. The level of statistical significance was 0.05.

Patient characteristics
A total of 65 consecutive patients with NSTE-ACS were enrolled in the study. The participants for inclusion were divided into stable plaque group (n = 19) and vulnerable plaque group (n = 46). As shown in Table 1, there were significant differences in sex, smoking history and diabetes mellitus history between the two groups (P < 0.05). Patients in vulnerable plaque group had visible higher levels of TG and LDL-C than those in stable plaque group (P < 0.05). In addition, elevated level of PDGF was observed in vulnerable plaque group (P < 0.05).
Plaque characteristics and components analysis Plaque characteristicsand components in vulnerable plaque group and stable plaque group were then examined using gray-scale and iMap-IVUS. The findings showed that there were significant differences in MLA, plaque area, plaque burden, FI, LI and NC between the vulnerable plaque and stable plaque groups (P < 0.05), while no statistic differences were found in EEM CSA and DC ( Table 2)..

Correlation analysis between plaque characteristics and ACS markers
Variables with P < 0.05 including TG, LDL-C and PDGF in Table 1 were selected to analyze the correlation with plaque characteristics. The results showed that TG and LDL-C were not correlated with plaque burden. In addition, PDGF was weakly correlated with plaque burden (R = 0.428, P < 0.05), as well as moderately correlated with NC (R = 0.669, P < 0.05) ( Table 3)..

Univariate logistic regression analysis for vulnerable plaque
Univariate logistic regression analysis was established with vulnerable plaque as dependent variables and variables with P < 0.05 in Table 1 as independent variables in this study, yielding that there were significant differences in smoking, diabetes mellitus and PDGF ( positive predictive value (PPV) and 59.4% negative predictive value (NPV) for prediction of vulnerable plaque. The area under the curve (AUC) was 0.876 (95% CI 0.804-0.948, P = 0.001) (Fig 2)..

Discussion
The proportion of patients manifesting with NSTE-ACS is increasing and remain a major cause of death worldwide [14]. Plaque instability and subsequent thrombus formation are the leading causes of ACS. Although IVUS has ability to characterize the culprit lesions, it is attractive to noninvasively characterize morphology and composition of vulnerable plaques in culprit and stable lesions in the future, especially in patients with ACS at low risk or in patients with suspected ACS. Therefore, the aim of this study was to find a novel biomarker to noninvasively predict vulnerable plaque in patients with NSTE-ACS. The results showed a higher level of serum PDGF in vulnerable plaque group; correlation analysis showed that PDGF was weakly correlated with plaque burden and moderately correlated with NC; Logistic regression analysis revealed that serum PDGF was independent risk factor of vulnerable plaque; ROS analysis demonstrated that PDGF could be used as a predictor to indicate plaque instability.
Gray-scale IVUS is commonly used to characterize coronary plaques in the clinical setting; however, it is insufficient for accurate assessments of plaque characteristics due to its lower resolution. overexpression of PDGF-D in transgenic mice promoted proliferation of cardiac interstitial fibroblasts and vSMCs, which led to cardiac fibrosis and dilated cardiomyopathy followed by subsequent cardiac failure. In this study, elevated level of serum PDGF was observed in patients with vulnerable plaques, and PDGF was proved to be an independent risk factor of vulnerable plaque. Further analysis suggested that PDGF could be used to predict vulnerable plaque in patients with NSTE-ACS.
In this study, in order to eliminate confounding factors caused by platelets, the patients who were first diagnosed with CHD, had not recently been heavily exposed to antiplatelet and anticoagulant drugs, and had no blood diseases were selected in this study. There are some limitations in this study.
Firstly, the present study comprised a relatively small sample size which may introduce selection bias.
Secondly, this single-center study lacks external validation. Finally, the follow-up time was missing.
Therefore, more multi-center, longer term follow-up and large-sample clinical studies are urgent to be performed in the future.
In conclusion, we demonstrated that the serum PDGF was independent risk factor of vulnerable plaque in patients with NSTE-ACS. Our findings suggested that serum PDGF might become a novel