Use of a protein biomarker in the preparation of a product for predicting the future risk of coronary heart disease in a subject

Abstract

The present invention relates to the field of biomedicine, and particularly to the application of protein markers in the preparation of products for predicting the future incidence risk of coronary heart disease in subjects. The present invention provides an application of using 11 proteins as markers in the preparation of products for effectively distinguishing and predicting the future incidence risk of coronary heart disease in people without traditional risk factors. After this model is extended to the entire population, it can further improve the predictive ability of coronary heart disease on the basis of traditional risk factors or genetic risks, effectively identify and stratify the incidence risk of coronary heart disease, thereby improving primary prevention. The present invention can truly achieve individual coronary heart disease risk assessment, and thus can achieve better risk prediction or assessment in both the entire population and those who have been misjudged as healthy or at low risk in the past.

Description

Technical Field

[0001] The present invention relates to the field of biomedicine, and particularly relates to a risk prediction model for the onset of coronary atherosclerotic heart disease in patients without traditional risk factors based on proteomics. Background Art

[0002] Coronary atherosclerotic heart disease, commonly abbreviated as coronary heart disease, is an ischemic heart disease. It is caused by atherosclerotic lesions in the coronary artery blood vessels, resulting in stenosis or obstruction of the blood vessel lumen, including subtypes such as myocardial infarction, angina pectoris, and sudden coronary death. Coronary heart disease is the main cause of death from cardiovascular diseases globally and has a serious disease burden in countries around the world. In coronary heart disease, targeted strategies for recognized modifiable risk factors (referred to as standard modifiable cardiovascular risk factors, i.e., traditional risk factors) have made great progress in the prevention and treatment of coronary heart disease.

[0003] Clinically, a considerable number of patients develop coronary heart disease without any previous symptoms and without any traditional risk factors reaching or exceeding the diagnostic threshold. In recent years, international clinical practice has found that approximately 15% - 27% of coronary heart disease patients do not have the four traditional risk factors of hyperlipidemia, hypertension, diabetes, or smoking, and this proportion has been increasing year by year. Since in the subclinical stage, people without traditional risk factors are often misjudged as "healthy" or at low risk, missing the best window period for early prevention and treatment. Compared with patients with at least one traditional risk factor, coronary heart disease patients without traditional risk factors have a higher risk of in-hospital death, all-cause death after discharge, or death attributed to cardiovascular diseases, have a poorer prognosis, and have a heavier disease burden.

[0004] Currently, the Pooled cohort equation (PCE) score is commonly used in clinical practice to predict the 10-year risk of atherosclerotic cardiovascular disease. It mainly relies on traditional risk factors, but clinical practice has found that 15% - 27% of coronary heart disease patients do not have any traditional risk factors reaching or exceeding the diagnostic threshold. The PCE model for clinically evaluating the 10-year risk of atherosclerotic cardiovascular disease mainly focuses on the quantity and intensity of traditional risk factors, so it cannot effectively and accurately identify high-risk individuals in people without traditional risk factors.

[0005] In addition, existing polygenic risk scores for coronary heart disease, such as those for early risk prediction of coronary heart disease based on markers such as N-terminal pro-brain natriuretic peptide (NT-proBNP), high-sensitivity cardiac troponin T (hs-cTNT), or low-density lipoprotein cholesterol (LDL-c), have mostly been carried out in the general population or relying on case-control study designs, with poor screening effectiveness and low prediction accuracy for patients with coronary heart disease without traditional risk factors. Therefore, for the population without traditional risk factors (the population previously misjudged as "healthy"), exploring the key biomarkers for future coronary heart disease occurrence helps identify high-risk groups and then carry out targeted primary prevention measures to effectively reduce the disease burden of coronary heart disease.

[0006] Proteomics can provide valuable clues to potential pathological mechanisms at the molecular level. Since blood proteomes are easily accessible and close to the cardiovascular system, they have been widely characterized in cardiovascular diseases. Multiple studies have used mass spectrometry to quantitatively detect plasma proteins in small samples and identified proteins associated with coronary heart disease. However, the measurement of a single type of protein is one-sided and prone to overlooking other key markers that affect disease occurrence or progression. One study used an aptamer-based platform capable of analyzing more than a thousand proteins in plasma and detected many protein changes in the context of myocardial injury in the Framingham Heart Study cohort, thus highlighting the potential of proteomics tools for application in large human cohorts. However, current early risk prediction studies of coronary heart disease based on high-throughput proteomics only target the general population and do not pay attention to this special population without traditional risk factors. There is no study exploring the proteomic characteristics of coronary heart disease onset in people without traditional risk factors. Moreover, prediction models developed based on the general population have poor early risk prediction ability in people without traditional risk factors.

[0007] Therefore, there is an urgent need to invent a protein risk score model that can effectively distinguish and predict the future risk of coronary heart disease onset in the population without traditional risk factors (previously considered "healthy"). Summary of the Invention

[0008] In view of this, the purpose of the present invention is:

[0009] (1) In the population without traditional risk factors (i.e., the population previously considered "healthy"), construct a risk score based on protein markers (Protein risk score for CAD without traditional risk factor, PRS-NRCAD) model that can effectively distinguish and predict the future risk of coronary heart disease onset in the population previously considered "healthy";

[0010] (2) The constructed PRS-NRCAD score can effectively predict the future risk of coronary heart disease onset in the general population;

[0011] (3) The constructed PRS-NRCAD score can further improve the prediction accuracy compared with the Pooled cohort equation (PCE) or polygenic risk score for predicting the 10-year risk of atherosclerotic cardiovascular disease;

[0012] (4) The constructed PRS-NRCAD score can effectively stratify the future coronary heart disease risk in populations without traditional risk factors and the general population.

[0013] To achieve the above-mentioned invention objectives, the present invention provides the following technical solutions:

[0014] The present invention provides protein markers, including:

[0015] (1) The protein with UniProt ID Q99988; and

[0016] (2) The protein with UniProt ID O14763; and

[0017] (3) The protein with UniProt ID Q9Y286; and

[0018] (4) The protein with UniProt ID P48745; and

[0019] (5) The protein with UniProt ID P35318; and

[0020] (6) The protein with UniProt ID P24387; and

[0021] (7) The protein with UniProt ID P13236; and

[0022] (8) The protein with UniProt ID Q14767; and

[0023] (9) The protein with UniProt ID P00750; and

[0024] (10) The protein with UniProt ID P02760; and

[0025] (11) The protein with UniProt ID Q9HAV5.

[0026] The present invention also provides the application of the above protein markers in predicting the risk of coronary heart disease.

[0027] In some specific embodiments of the present invention, the above-mentioned risk of coronary heart disease is predicted based on PRS-NRCAD, and the determination rule is: if PRS-NRCAD ≥ -1.35, it is determined that the future risk of coronary heart disease is high risk; if PRS-NRCAD < -1.35, it is determined that the future risk of coronary heart disease is low risk;

[0028] The PRS-NRCAD is obtained according to Formula I, and the Formula I is:

[0029] PRS-NRCAD = 0.36 × c1 + 0.18 × c2 + 0.31 × c3 + 0.31 × c4 + 0.38× c5 + 0.34 × c6 + 0.2 × c7 + 0.28 × c8 + 0.32 × c9 + 0.3 × c 10 + 0.3 ×c 11 ;

[0030] wherein, c1 to c 11 are successively the normalized protein expression values (Normalized Protein eXpression, NPX) of the proteins with UniProt IDs of Q99988, O14763, Q9Y286, P48745, P35318, P24387, P13236, Q14767, P00750, P02760, and Q9HAV5, hereinafter referred to as "NPX values" for short.

[0031] In some specific embodiments of the present invention, the above-mentioned NPX values are obtained based on plasma samples or serum samples.

[0032] In some specific embodiments of the present invention, the above-mentioned prediction is based on PRS-NRCAD and PCE prediction.

[0033] In some specific embodiments of the present invention, the above-mentioned prediction is based on PRS-NRCAD and any one of the following predictions:

[0034] (i), a coronary heart disease polygenic risk score model;

[0035] (ii), a cardiovascular disease polygenic risk score model.

[0036] In some specific embodiments of the present invention, the above-mentioned risk of coronary heart disease is the risk of coronary heart disease in individuals without traditional risk factors.

[0037] The present invention also provides a prediction model for the risk of coronary heart disease, including:

[0038] Calculate the PRS-NRCAD of an individual. If the PRS-NRCAD ≥ -1.35, it is determined that the future risk of coronary heart disease is high risk. If the PRS-NRCAD < -1.35, it is determined that the future risk of coronary heart disease is low risk;

[0039] The PRS-NRCAD is obtained according to Formula I, and the Formula I is:

[0040] PRS-NRCAD = 0.36 × c1 + 0.18 × c2 + 0.31 × c3 + 0.31 × c4 + 0.38× c5 + 0.34 × c6 + 0.2 × c7 + 0.28 × c8 + 0.32 × c9 + 0.3 × c 10 + 0.3 ×c 11 ;

[0041] wherein, c1 to c 11 are successively the NPX values of the proteins with UniProt IDs Q99988, O14763, Q9Y286, P48745, P35318, P24387, P13236, Q14767, P00750, P02760, and Q9HAV5.

[0042] A method for predicting the future risk of coronary heart disease, comprising the following steps:

[0043] S1. Collect a blood sample of an individual and prepare a plasma sample or a serum sample;

[0044] S2. Quantify the proteins in the obtained plasma or serum sample by Olink Explore Proximity Extension Assay (PEA) to obtain the protein level of the sample, and convert the sample protein level into an NPX value;

[0045] S3. Substitute the NPX value into Formula I in the above prediction model to obtain PRS-NRCAD. If the PRS-NRCAD ≥ -1.35, it is determined that the future risk of coronary heart disease is high risk. If the PRS-NRCAD < -1.35, it is determined that the future risk of coronary heart disease is low risk.

[0046] In some specific embodiments of the present invention, there is no special limitation on the calculation method of the NPX value for the above-mentioned application and method, and a method well-known in the art can be adopted, such as the method described in https: / / biobank.ndph.ox.ac.uk / ukb / ukb / docs / Olink_1536_B0_to_B7_Normalization.pdf (UKB - Olink Explore 1536 - Data Normalization Strategy).

[0047] The present invention also provides the application of the above-mentioned protein biomarker in the preparation of a product for predicting the risk of coronary heart disease;

[0048] The product includes a reagent, a kit, a device or a system.

[0049] In some specific embodiments of the present invention, the reagent, the kit or the device in the above-mentioned application contains a specific antibody against the protein biomarker.

[0050] The present invention also provides a reagent for predicting the risk of coronary heart disease, and the detection target is the protein biomarker in the above-mentioned application.

[0051] The present invention also provides a kit for predicting the risk of coronary heart disease, and the detection target is the protein biomarker in the above-mentioned application.

[0052] The present invention also provides a device for predicting the risk of coronary heart disease, and the detection target is the protein biomarker in the above-mentioned application.

[0053] The present invention also provides a system for predicting the risk of coronary heart disease, including:

[0054] An acquisition module, configured to acquire the concentration of the protein biomarker in the plasma or serum of an individual and convert the concentration into an NPX value, and the protein biomarker is the protein biomarker in the above-mentioned application;

[0055] A processing module, configured to predict the risk of coronary heart disease of the individual, and the prediction is based on the PRS - NRCAD prediction in the above-mentioned application.

[0056] The present invention has the following effects:

[0057] 1. In the past, people without traditional risk factors were considered "healthy", that is, they would not develop coronary heart disease or had only a very low probability of developing it. However, clinical practice has found that a considerable proportion of this group of people still develop coronary heart disease in the future. The previous PCE score used to predict the risk of atherosclerotic cardiovascular disease mainly relied on traditional risk factors. Therefore, its risk stratification and prediction effect are poor in this group of people. The simple protein score PRS-NRCAD constructed by the present invention starts from this special group of people without traditional risk factors, can effectively screen the onset of coronary heart disease, and identify high-risk groups for future coronary heart disease occurrence, thereby effectively preventing the occurrence of coronary heart disease and reducing the disease burden of this part of the population;

[0058] 2. After applying the simple protein score PRS-NRCAD constructed by the present invention to the general population, it is found that PRS-NRCAD can further improve the prediction ability of the traditional PCE model and / or gene risk score for the risk of coronary heart disease onset, improve the prediction accuracy, and thus improve the effectiveness of primary prevention of coronary heart disease. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art.

[0060] Figure 1 Show the lasso regression penalty model to screen protein markers related to the risk of coronary heart disease onset in people without traditional risk factors;

[0061] Figure 2 Show the discrimination ability of PRS-NRCAD for future coronary heart disease onset in people without traditional risk factors. Different color lines represent the discrimination ability of each individual protein marker, and the red line represents the discrimination value of the integration of 11 proteins into PRS-NRCAD;

[0062] Figure 3 Show the prediction ability of PRS-NRCAD for the future risk of coronary heart disease onset in people without traditional risk factors. A is the training set and B is the validation set;

[0063] Figure 4 Show the prediction ability of PRS-NRCAD for the future risk of coronary heart disease onset in the general population;

[0064] Figure 5 Show the stratification ability of PRS-NRCAD for the risk of coronary heart disease onset in people without traditional risk factors;

[0065] Figure 6 Show the stratification ability of PRS-NRCAD for the risk of coronary heart disease onset in the general population. DETAILED DESCRIPTION OF THE INVENTION

[0066] The present invention discloses a prediction model for the onset risk of coronary atherosclerotic heart disease in patients without traditional risk factors based on proteomics. Those skilled in the art can draw on the content of this article and appropriately improve the process parameters to achieve it. It should be particularly noted that all similar substitutions and modifications are obvious to those skilled in the art and are all considered to be included in the present invention. The methods and applications of the present invention have been described through preferred embodiments, and it is obvious that relevant personnel can make changes or appropriate alterations and combinations to the methods and applications described herein without departing from the content, spirit, and scope of the present invention to implement and apply the technology of the present invention.

[0067] Based on the limitations of the existing research progress, the present invention provides a protein risk scoring model that can effectively distinguish and predict the future onset risk of coronary heart disease for populations without traditional risk factors who were previously considered "healthy" or at low risk. After promoting this risk scoring model to the entire population, it can further improve the predictive ability of coronary heart disease on the basis of traditional risk factors or genetic risks, effectively identify and stratify the onset risk of coronary heart disease, thereby improving primary prevention. The present invention can truly achieve individual coronary heart disease risk assessment, so that better risk prediction / assessment can be achieved in the entire population and in populations that were previously misjudged as healthy / low risk.

[0068] Unless otherwise specified, the raw materials, reagents, consumables, and instruments involved in the present invention are all ordinary commercially available products and can be purchased from the market.

[0069] The following further elaborates the present invention in conjunction with embodiments.

[0070] Embodiment

[0071] (1) Collection of patients with coronary heart disease without traditional risk factors and their data

[0072] The UK Biobank Pharma Proteomics Project (UKB-PPP) performed plasma proteomics measurements on baseline samples of more than 50,000 UKB participants. The Olink Explore 3072 proximity extension assay (PEA) technology based on antibodies was used on the Olink platform to quantify proteins in plasma samples, including 1,463 proteins in cardiac metabolism, inflammation, neurology, and oncology modules, and then the protein levels in the samples were converted into NPX values.

[0073] Clinical risk factor information, such as age, gender, education, smoking, etc., was collected using a standardized questionnaire; physical measurements such as height, weight, and blood pressure were evaluated by trained nurses; hypertension, diabetes, hypercholesterolemia, etc. were diagnosed based on the baseline self-reported medical history, health record diagnosis history, medication history, or baseline measurement values. The diagnosis of coronary heart disease was defined according to the International Classification of Diseases, Ninth Revision (ICD-9) or Tenth Revision (ICD-10) codes, or hospital records or self-reported diagnosis history. Traditional risk factors included smoking (currently smoking or having smoked more than 100 cigarettes in a lifetime), hypercholesterolemia, diabetes, and hypertension.

[0074] Participants with any form of coronary heart disease at baseline or those whose genetic data did not meet the quality control requirements were excluded. Finally, 49,530 subjects with genetic, plasma proteomic, and clinical data were included. Among them, 80% of the population without traditional risk factors at baseline was randomly selected as the test cohort (N = 5,772) to screen proteomic characteristics related to the risk of coronary heart disease onset and construct a protein prediction model, and the remaining 20% of the population was used as the validation cohort (N = 1,442) to verify the model effect. Then, the protein model was applied to the entire population with PCE scores (N = 47,755) to verify its generality.

[0075] In the discovery cohort, patients who developed coronary heart disease in the future were older at baseline, had a higher proportion of males, a higher proportion of obesity, higher average blood pressure, worse liver function, and worse lipid and glucose profiles. However, there was no statistical difference in the levels of polygenic risk scores predicting the risk of coronary heart disease or cardiovascular disease onset between the two groups of people, as shown in Table 1.

[0076] Table 1

[0077]

[0078] (2) Protein biomarker screening

[0079] Through a two-stage screening method, proteomic characteristics related to the risk of coronary heart disease onset in the population without traditional risk factors were screened.

[0080] First stage: The LASSO regression penalty model was used to include clinical characteristic factors (age, gender, and race) and protein biomarkers related to the risk of coronary heart disease onset in the population without traditional risk factors for screening, and 10-fold cross-validation was carried out to select the best penalty parameter. The coefficients of all three clinical characteristic factors (λ) were non-zero, so they were selected into the training model. Among 1463 proteins, 32 protein coefficients were non-zero and were selected into the further training model, as shown in Figure 1 . Then, the clinical characteristics and proteins with non-zero coefficients were uniformly included in the LASSO regression model to try to retain proteomic characteristics that may contribute less to disease classification. Subsequently, 31 protein biomarkers were screened out.

[0081] Phase 2: Use the Cox proportional hazards regression model to screen for protein biomarkers that can prospectively predict coronary heart disease in a population without traditional risk factors at baseline. For each of the 31 protein biomarkers screened above, a Cox proportional hazards regression model was constructed. The outcome of the Cox proportional hazards regression model was a binary variable: whether coronary heart disease occurred in the future. The variables were the NPX values of the 31 proteins. The hazard ratio (coef), hazard ratio (HR), and P-value of each individual protein were obtained. The P-values were multiplicatively adjusted using the Benjamini & Hochberg method, and a total of 11 protein biomarkers were screened (P < 0.05 / 31), as shown in Table 2.

[0082] Table 2

[0083]

[0084] (3) Construction of the protein prediction model

[0085] Based on the 11 protein biomarkers discovered through the above multi-stage screening, an additive model was used to construct a simple protein score for the risk of coronary heart disease in a population without traditional risk factors, PRS-NRCAD:

[0086]

[0087] By measuring the concentrations of the above proteins in plasma and calibrating and transforming the protein concentrations into protein NPX values using the NPX method, the simple protein score for the future risk of coronary heart disease in a population without traditional risk factors, PRS-NRCAD, can be calculated according to the formula.

[0088] Risk judgment criterion: When the individual PRS-NRCAD score ≥ -1.35, the risk of future coronary heart disease is high; when the individual PRS-NRCAD score < -1.35, the risk of future coronary heart disease is low.

[0089] (4) Analysis of the discriminatory ability of the protein prediction model for the occurrence of coronary heart disease events

[0090] First, the discriminatory ability of PRS-NRCAD and 11 individual protein markers for coronary heart disease (CHD) events in people without traditional risk factors was compared. The receiver operating characteristic curve (ROC) and the area under the curve (AUC) were used to evaluate the classification performance of PRS-NRCAD or 11 individual protein markers (presented by UniProt ID) between patients with future CHD events and normal controls without CHD events. The AUC ranges from 0 to 1, and can intuitively evaluate the prediction accuracy of the model. The larger the AUC value, the higher the prediction accuracy. The analysis found that the ROC curves of PRS-NRCAD were all higher than those of the other 11 individual protein markers, and PRS-NRCAD had the highest AUC value, indicating that the discriminatory ability of PRS-NRCAD for CHD onset was better than that of using 11 protein markers alone. In addition, the AUC value of PRS-NRCAD reached 0.767, indicating that PRS-NRCAD had good discriminatory ability for future CHD onset, as shown in Figure 2 (The lines of different colors represent the discriminatory ability of each individual protein marker, and the red line represents the discriminatory value of the 11 proteins integrated together).

[0091] (5) Analysis of the predictive ability of the protein prediction model for the risk of future CHD onset

[0092] Next, the predictive ability of the protein prediction model PRS-NRCAD for the risk of CHD onset in people without traditional risk factors was prospectively evaluated. First, in the training set, using the Cox regression model, it was found that PRS-NRCAD was positively correlated with the risk of future CHD onset in people without traditional risk factors. After adjusting for clinical factors (gender, age, and race) and genetic risks (polygenic risk score for CHD and polygenic risk score for cardiovascular disease) respectively, PRS-NRCAD was still associated with the risk of CHD onset in people without traditional risk factors (the HR and its 95% confidence interval CI did not include 1, indicating P < 0.05, as shown in Figure 3 A in).

[0093] Similarly, in the validation set, the Cox regression model found that the positive association between PRS-NRCAD and the risk of CHD onset in people without traditional risk factors remained significant after adjusting for clinical factors or the two polygenic risk scores (the HR and its 95% CI did not include 1, indicating P < 0.05, Figure 3 B in).

[0094] (6) Application and generalization of the predictive ability of the protein prediction model for the risk of CHD onset to the general population

[0095] Furthermore, the protein prediction model PRS-NRCAD was further applied to the entire population, and the predictive ability of PRS-NRCAD for the risk of coronary heart disease was evaluated. Currently, the PCE model is commonly used internationally to predict the 10-year risk of atherosclerotic cardiovascular disease in the general population. This model has a high degree of recognition and is widely used, including several clinical indicators such as age, gender, race, systolic blood pressure, hypertension treatment, diabetes, total cholesterol, and high-density lipoprotein cholesterol.

[0096] According to the data availability of PCE scores and genetic scores, a total of 47,755 people from the entire population were included. In the entire population, through the Cox regression model, it was found that after adjusting for the 10-year atherosclerotic cardiovascular disease risk and polygenic risk score, the association between PRS-NRCAD and the risk of coronary heart disease remained significant (the HR and its 95% confidence interval did not include 1, indicating P < 0.05, as Figure 4 shown).

[0097] As shown in Table 3, in the Cox regression model, the predictive abilities of PRS-NRCAD, PCE, and genetic risk score for the future risk of coronary heart disease were compared. The improvement in predictive ability was evaluated using the concordance index (C-index), and the improvement in disease classification, that is, risk stratification ability, was evaluated using the integrated discrimination improvement index (IDI) and net reclassification improvement index (NRI). Compared with using only the PCE model for prediction, combining the PCE model with the polygenic risk score for coronary heart disease or the polygenic risk score for cardiovascular disease could further improve the predictive ability for the risk of coronary heart disease (the C-index increased by 0.006 and 0.003 respectively, P < 0.001). On the basis of the PCE model, further adding PRS-NRCAD could increase the C-index by 0.016 (P < 0.001), and both the 10-year integrated discrimination improvement index (IDI) and net reclassification improvement index (NRI) were significantly improved. This indicates that adding the detection of 11 protein markers to the existing clinical risk prediction model can further improve the prediction ability for the onset of coronary heart disease.

[0098] As shown in Table 3, on the basis of the PCE and polygenic risk score model for coronary heart disease, further adding PRS-NRCAD could still further improve the predictive ability for the onset of coronary heart disease (the C-index increased by 0.023, the IDI increased by 0.013, the NRI increased by 0.180, P < 0.001). On the basis of the PCE and polygenic risk score model for coronary heart disease, further adding PRS-NRCAD could also improve the predictive ability for the onset of coronary heart disease (the C-index increased by 0.023, the IDI increased by 0.013, the NRI increased by 0.180, P < 0.001).

[0099] Table 3

[0100]

[0101] (7)Analysis of the effect of the protein prediction model on the risk stratification of coronary heart disease

[0102] Among the population without traditional risk factors, the population was divided into two groups: low-risk (PRS-NRCAD < -1.35, n = 3,600) and high-risk (PRS-NRCAD ≥ -1.35, n = 3,614) according to the PRS-NRCAD level. Kaplan-Meier survival analysis was used to compare the cumulative incidence risks of coronary heart disease in the low-risk and high-risk PRS-NRCAD populations during the median follow-up period of 13.9 years, and the log-rank test was used to determine that the high-risk population had a higher risk of coronary heart disease (P < 0.001). See Figure 5 , the two risk curves and their 95% confidence interval bands were separated from each other, without overlap, and the curve of the high-risk group was higher than that of the low-risk group at each time point, indicating that the high-risk PRS-NRCAD group had a higher risk of coronary heart disease.

[0103] Among the entire population, the population was divided into two groups: low-risk (PRS-NRCAD < -1.35) and high-risk (PRS-NRCAD ≥ -1.35) according to the PRS-NRCAD level. Kaplan-Meier survival analysis was used to compare the cumulative incidence risks of coronary heart disease in the low-risk and high-risk PRS-NRCAD populations during the median follow-up period of 13.9 years, and the log-rank test was used to determine that even in the entire population, the high-risk population still had a higher risk of coronary heart disease (P < 0.001). See Figure 6 , the two risk curves and their 95% confidence interval bands were separated from each other, without overlap, and the curve of the high-risk group was higher than that of the low-risk group at each time point, indicating that the high-risk PRS-NRCAD group had a higher risk of coronary heart disease.

[0104] (8)Clinical application value of the model

[0105] Among the population without traditional risk factors (the population previously misjudged as "healthy"), the ROC curve and AUC value were used to evaluate the screening ability of the protein risk score model PRS-NRCAD or 11 individual protein markers for future coronary heart disease. It was found that PRS-NRCAD had the strongest discrimination ability, with an AUC value of 0.767 and good discrimination effect. See Figure 2Furthermore, the prospective risk prediction ability of PRS-NRCAD was evaluated using a multivariable Cox regression analysis model. After adjusting for the confounding interference of demographic characteristics and genetic risk, PRS-NRCAD could still effectively predict the risk of coronary heart disease in the population without traditional risk factors, who are often regarded as "healthy" (P<0.05), as shown in Figure 3 .

[0106] When PRS-NRCAD was extended to the entire population, it was found that even after adjusting for the PCE score and genetic risk, which are widely used internationally to predict the 10-year risk of atherosclerotic cardiovascular disease in the general population, PRS-NRCAD could still effectively predict the risk of coronary heart disease in the general population (P<0.05), as shown in Figure 4 .

[0107] When compared with the existing 10-year atherosclerotic cardiovascular disease risk prediction score, PCE, adding PRS-NRCAD could further improve the ability to predict the risk of coronary heart disease, as shown in Table 3. Compared with the single PCE score, after adding PRS-NRCAD, the prediction accuracy (reflected by the C-index) increased by 1.6%, and both the overall prediction effect (represented by IDI) and the proportion of successful predictions (represented by NRI) were significantly improved, and the improvement effect was better than that of the genetic risk score. When compared with the combination of the PCE score and the genetic risk score, further adding the PRS-NRCAD variable could still improve the prediction accuracy, overall prediction effect, and proportion of successful predictions of coronary heart disease.

[0108] Using PRS-NRCAD to stratify the risk of the population without traditional risk factors, it was found that the high-risk population had a higher risk of coronary heart disease during the 13.9-year follow-up period (P<0.001, as shown in Figure 5 ), indicating that using PRS-NRCAD to stratify the future risk of coronary heart disease in the population without traditional risk factors can effectively identify high-risk populations. In addition, the ability of PRS-NRCAD to stratify the risk of coronary heart disease was further verified in the entire population, as shown in Figure 6 .

[0109] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. Use of a protein biomarker in the preparation of a product for predicting the future risk of coronary heart disease in a subject, characterized in that, The protein markers are as follows: (1) The protein with UniProt ID Q99988; and (2) The protein with UniProt ID O14763; and (3) The protein with UniProt ID Q9Y286; and (4) The protein with UniProt ID P48745; and (5) The protein with UniProt ID P35318; and (6) The protein with UniProt ID P24387; and (7) The protein with UniProt ID P13236; and (8) The protein with UniProt ID Q14767; and (9) The protein with UniProt ID P00750; and (10) The protein with UniProt ID P02760; and (11) The protein with UniProt ID Q9HAV5; The product includes reagents, kits, devices or systems.

2. The application according to claim 1, characterized in that, The future coronary heart disease (CHD) risk of the subject is predicted based on PRS-NRCAD, and the determination rule is: if PRS-NRCAD ≥ -1.35, it is determined that the subject has a high risk of future CHD; if PRS-NRCAD < -1.35, it is determined that the subject has a low risk of future CHD; PRS-NRCAD is obtained according to formula I, and formula I is: PRS-NRCAD = 0.36×c1 + 0.18×c2 + 0.31×c3 + 0.31×c4 + 0.38×c5 + 0.34×c6 + 0.2×c7 + 0.28×c8 + 0.32×c9 + 0.3×c 10 + 0.3×c 11 ; Among them, c1 to c 11 are the NPX values of the proteins with UniProt IDs Q99988, O14763, Q9Y286, P48745, P35318, P24387, P13236, Q14767, P00750, P02760, and Q9HAV5 in sequence.

3. The application according to claim 2, characterized in that, The prediction is based on PRS-NRCAD and PCE prediction.

4. The application according to claim 2, wherein The prediction is based on PRS-NRCAD and any one of the following predictions: (i) A polygenic risk score model for coronary heart disease; (ii) A polygenic risk score model for cardiovascular diseases.

5. The application according to claim 1 or 2, characterized in that, The CHD risk is the CHD risk of individuals without traditional risk factors.

6. The application according to claim 1 or 2, characterized in that The reagent, the kit or the device contains a specific antibody against the protein marker.

7. A reagent for predicting the future risk of coronary heart disease in a subject, characterized in that, The detection target is the protein marker; The protein markers are as follows: (1) The protein with UniProt ID Q99988; and (2) The protein with UniProt ID O14763; and (3) The protein with UniProt ID Q9Y286; and (4) The protein with UniProt ID P48745; and (5) The protein with UniProt ID P35318; and (6) The protein with UniProt ID P24387; and (7) The protein with UniProt ID P13236; and (8) The protein with UniProt ID Q14767; and (9) The protein with UniProt ID P00750; and (10) The protein with UniProt ID P02760; and (11) The protein with UniProt ID Q9HAV5.

8. A kit for predicting the future risk of coronary heart disease in a subject, characterized in that, The detection target is the protein marker; The protein markers are as follows: (1) The protein with UniProt ID Q99988; and (2) The protein with UniProt ID O14763; and (3) The protein with UniProt ID Q9Y286; and (4) The protein with UniProt ID P48745; and (5) The protein with UniProt ID P35318; and (6) The protein with UniProt ID P24387; and (7) The protein with UniProt ID P13236; and (8) The protein with UniProt ID Q14767; and (9) The protein with UniProt ID P00750; and (10) The protein with UniProt ID P02760; and (11) The protein with UniProt ID Q9HAV5.

9. A device for predicting the future risk of coronary heart disease in a subject, characterized in that, The detection target is a protein biomarker; The protein biomarker is: (1) The protein with UniProt ID Q99988; and (2) The protein with UniProt ID O14763; and (3) The protein with UniProt ID Q9Y286; and (4) The protein with UniProt ID P48745; and (5) The protein with UniProt ID P35318; and (6) The protein with UniProt ID P24387; and (7) The protein with UniProt ID P13236; and (8) The protein with UniProt ID Q14767; and (9) The protein with UniProt ID P00750; and (10) The protein with UniProt ID P02760; and (11) The protein with UniProt ID Q9HAV5.

10. A system for predicting the future risk of coronary heart disease in a subject, characterized in that, It includes: An acquisition module, which is used to acquire the concentration of the protein biomarker in the plasma or serum of the subject and convert the concentration into an NPX value; A processing module, which is used to predict the coronary heart disease onset risk of the subject. The prediction is based on the PRS-NRCAD prediction, and the determination rule is: if the PRS-NRCAD ≥ -1.35, it is determined that the subject has a high risk of future coronary heart disease onset; if the PRS-NRCAD < -1.35, it is determined that the subject has a low risk of future coronary heart disease onset; The protein biomarker is: (1) The protein with UniProt ID Q99988; and (2) The protein with UniProt ID O14763; and (3) The protein with UniProt ID Q9Y286; and (4) The protein with UniProt ID P48745; and (5) The protein with UniProt ID P35318; and (6) The protein with UniProt ID P24387; and (7) The protein with UniProt ID P13236; and (8) The protein with UniProt ID Q14767; and (9) The protein with UniProt ID P00750; and (10) The protein with UniProt ID P02760; and (11) The protein with UniProt ID Q9HAV5; The PRS-NRCAD is obtained according to Formula I, and the Formula I is: PRS-NRCAD = 0.36×c1+0.18×c2+0.31×c3+0.31×c4+0.38×c5+0.34×c6+0.2×c7+0.28×c8+0.32×c9+0.3×c 10 +0.3×c 11 ; Among them, c1 to c 11 are successively the NPX values of the proteins with UniProt IDs Q99988, O14763, Q9Y286, P48745, P35318, P24387, P13236, Q14767, P00750, P02760, and Q9HAV5, respectively.

Images