Dangerous stratification model for hypertrophic cardiomyopathy and preparation method and application thereof

Abstract

The invention discloses a dangerous stratification model for hypertrophic cardiomyopathy and a preparation method and application thereof, and belongs to the technical field of biological medicine. The invention provides application of a reagent for detecting the expression level of PDCD5 protein. Researches find that expression increase of the serum PDCD5 protein is not only a risk factor of bad outcome of HCM patients, but also is independent of other clinical factors, and a survival curve of risk stratification according to the serum PDCD5 protein also has a good discrimination degree. The PDCD5 protein improves a risk stratification model of HCM established based on clinical factors. The non-synonymous variation of the VWCE, SPAG5, PDGFRB and PITX3 genes is independently related to the increase of the adverse outcome risk of HCM patients; on the basis of a score 2 established based on serum PDCD5 protein and clinical factors, the non-synonymous variation of the genes is combined, and the established multi-dimensional risk stratification model has good prediction efficiency on the adverse outcome of the HCM patient.

Description

Technical Field

[0001] This invention belongs to the field of biomedical technology, and in particular relates to a risk stratification model for hypertrophic cardiomyopathy, its preparation method and application. Background Technology

[0002] Hypertrophic cardiomyopathy (HCM) is characterized by unexplained left ventricular hypertrophy. Due to the significant heterogeneity in its clinical manifestations and natural course, the prognosis of HCM patients is highly uncertain. Mild cases may be asymptomatic, while severe cases can lead to serious complications such as sudden cardiac death (SCD), heart failure (HF), and malignant arrhythmias, posing a significant public health problem. Risk stratification of HCM patients to identify high-risk individuals with adverse outcomes is crucial. However, current risk stratification methods for HCM patients focus primarily on sudden cardiac death, with less attention paid to other adverse outcomes such as heart failure. Furthermore, existing models for assessing the risk of sudden cardiac death in HCM patients only show moderate C-index when applied to the HCM population. Therefore, developing a suitable risk stratification method for HCM patients has significant clinical implications.

[0003] Programmed cell death molecule 5 (PDCD5) is an apoptosis-related gene cloned from the human leukemia cell line TF-1. Besides its role in regulating apoptosis, PDCD5 plays a crucial role in the pathophysiology of various diseases, including tumors, autoimmune diseases, and atherosclerosis. However, no studies have yet explored whether PDCD5 can serve as a novel biomarker for the diagnosis and assessment of the severity of hepatocellular carcinoma (HCM).

[0004] Furthermore, as the most common hereditary cardiomyopathy, hepatocellular carcinoma (HCM) relies heavily on genetic factors in its progression. HCM patients exhibit gene variants with clear or potential pathogenic significance. In addition to being a major cause of HCM, gene variants encoding sarcomere proteins or sarcomere-related structural proteins are increasingly suggested by research to be valuable in assessing the prognosis of HCM patients. Given that HCM is a disease closely related to genetic factors, utilizing gene testing for precise risk stratification at the gene level provides a new approach and opportunity for HCM patients. Combining genetic information with traditional risk factors for risk stratification is of great significance for developing new concepts, technologies, and models for the diagnosis and treatment of HCM, and even for future personalized precision medicine. Summary of the Invention

[0005] Therefore, the purpose of this invention is to provide a risk stratification model for hypertrophic cardiomyopathy, its preparation method and application. This invention is the first to discover that serum PDCD5 protein is a biomarker for assessing the risk of having hypertrophic cardiomyopathy or the degree of risk of hypertrophic cardiomyopathy.

[0006] To achieve the above-mentioned objectives, the present invention provides the following technical solution:

[0007] This invention provides an application of a reagent for detecting PDCD5 protein expression levels, comprising at least one of the following:

[0008] (1) Application of reagents for detecting PDCD5 protein expression levels in the preparation of products for assessing the risk of hypertrophic cardiomyopathy;

[0009] (2) Application of reagents for detecting PDCD5 protein expression levels in the preparation of products for assessing the disease risk of patients with hypertrophic cardiomyopathy.

[0010] Preferably, for the application of (1), compared with the serum sample of the negative control, the expression of PDCD5 protein in the serum sample to be tested is significantly increased, the P of the serum sample to be tested vs the serum sample of the negative control is <0.05, and the PDCD5 protein in the serum sample to be tested is >0.95ng / mL, which indicates that the subject to be tested has a high risk of hypertrophic cardiomyopathy.

[0011] Regarding the application of (2), compared with the control serum sample group, the expression of PDCD5 protein in the serum sample of the hypertrophic cardiomyopathy patient to be tested was significantly increased. The serum sample group of the hypertrophic cardiomyopathy patient to be tested vs. the control serum sample group P < 0.05, and the PDCD5 protein in the serum sample to be tested was > 4.09 ng / mL, which indicates that the disease risk of hypertrophic cardiomyopathy patients is high.

[0012] This invention also provides an application of PDCD5 protein expression level combined with clinical factors in constructing a risk stratification model for hypertrophic cardiomyopathy, wherein the clinical factors are history of syncope / presyncope, NYHA classification, and echocardiographic findings of left ventricular enlargement.

[0013] This invention also provides an application of PDCD5 protein expression level, clinical factors, and nonsynonymous gene variants in constructing a risk stratification model for hypertrophic cardiomyopathy. The clinical factors are a history of syncope / presyncope, NYHA classification, and echocardiographic evidence of left ventricular enlargement. The nonsynonymous gene variants are nonsynonymous variants carrying von Willebrand factor C and EGF domain genes, nonsynonymous variants of sperm-associated antigen 5 genes, nonsynonymous variants of platelet-derived growth factor receptor β genes, and nonsynonymous variants of paired homeoheterodomain protein transcription factor 3 genes.

[0014] This invention also provides a method for constructing a risk stratification model for hypertrophic cardiomyopathy, comprising the following steps:

[0015] Candidate variables were included in multivariate Cox regression analysis and stepwise regression was used. Then, variables with p < 0.05 in the multivariate Cox regression analysis were used to build a risk stratification model. Each variable was weighted and scored according to the regression coefficients. The scores of each variable were summed to obtain the prediction score for adverse outcomes. The optimal cutpoint for risk stratification of the prediction score was calculated using survival analysis of R 4.2.3 and the surv_cutpoint function in its visualization toolkit. Patients with hypertrophic cardiomyopathy were stratified according to the optimal cutpoint, thus obtaining the risk stratification model for hypertrophic cardiomyopathy.

[0016] The candidate variables include clinical factors and serum PDCD5 protein expression levels; the clinical factors are age, sex, whether it is obstructive HCM, history of syncope / presyncope, NYHA classification, body mass index, history of hypertension, myoglobin, atrial flutter / fibrillation, non-sustained ventricular tachycardia, echocardiographic left ventricular enlargement, echocardiographic left ventricular ejection fraction, and echocardiographic wall motion abnormalities.

[0017] Preferably, the variables with p<0.05 are history of syncope / presyncope, NYHA classification, echocardiographic evidence of left ventricular enlargement, and serum PDCD5 protein expression level;

[0018] The calculation formula for the risk stratification model of hypertrophic cardiomyopathy is: Score 2 = Syncope / History of Syncope + NYHA Classification + Left Ventricular Enlargement + High Expression of Serum PDCD5 Protein. Among them, the presence of syncope / history of syncope is worth 1 point, NYHA Class I is worth 1 point, NYHA Class II is worth 2 points, NYHA Class III is worth 3 points, NYHA Class IV is worth 4 points, echocardiography showing left ventricular enlargement is worth 1 point, and serum PDCD5 protein >4.09 ng / mL is worth 1 point.

[0019] When stratifying patients with hypertrophic cardiomyopathy, they are divided into a high-risk group and a low-risk group. The high-risk group is defined as those with a predicted score > the optimal cutoff value, and the low-risk group is defined as those with a predicted score ≤ the optimal cutoff value.

[0020] Preferably, the candidate variables also include nonsynonymous variants of genes carrying von Willebrand factor C and EGF domains, nonsynonymous variants of genes carrying tissue factor pathway inhibitors, nonsynonymous variants of genes carrying mercaptopyruvate thiotransferase, nonsynonymous variants of genes carrying melanoma cell adhesion molecules, nonsynonymous variants of genes carrying paired homeotypical domain protein transcription factor 3, nonsynonymous variants of genes carrying p21 activated kinase 2, nonsynonymous variants of genes carrying sperm-associated antigen 5, and nonsynonymous variants of genes carrying platelet-derived growth factor receptor β.

[0021] Preferably, the variables with p < 0.05 are: history of syncope / presyncope, NYHA classification, left ventricular enlargement as indicated by echocardiography, serum PDCD5 protein expression level, carrying nonsynonymous variants of gene containing von Willebrand factor C and EGF domain, nonsynonymous variants of sperm-associated antigen 5 gene, nonsynonymous variants of platelet-derived growth factor receptor β gene, or nonsynonymous variants of gene containing paired homeotypical domain protein transcription factor 3.

[0022] When stratifying patients with hypertrophic cardiomyopathy, they are divided into high-risk, low-risk, and intermediate-risk groups.

[0023] Patients with a score 2 > the optimal cutoff value and carrying at least one of the following: nonsynonymous variants of von Willebrand factor C and EGF domain genes, nonsynonymous variants of sperm-associated antigen 5 gene, nonsynonymous variants of platelet-derived growth factor receptor β gene, and nonsynonymous variants of paired homeomorphic domain protein transcription factor 3 gene are stratified as high-risk group.

[0024] Patients with a score 2 ≤ the optimal cutoff value and who do not carry nonsynonymous variants of von Willebrand factor C and EGF domain genes, nonsynonymous variants of sperm-associated antigen 5 genes, nonsynonymous variants of platelet-derived growth factor receptor β genes, and nonsynonymous variants of paired homeomorphic domain protein transcription factor 3 genes are stratified into the low-risk group.

[0025] The intermediate-risk group is any one of the following:

[0026] Patients with a score of 2 > the optimal cutoff value and who do not carry nonsynonymous variants of von Willebrand factor C and EGF domain genes, nonsynonymous variants of sperm-associated antigen 5 genes, nonsynonymous variants of platelet-derived growth factor receptor β genes, or nonsynonymous variants of paired homeomorphic domain protein transcription factor 3 genes are stratified into the intermediate-risk group.

[0027] Patients with a score of 2 ≤ the optimal cutoff value and carrying at least one of the following: nonsynonymous variants of von Willebrand factor C and EGF domain genes, nonsynonymous variants of sperm-associated antigen 5 gene, nonsynonymous variants of platelet-derived growth factor receptor β gene, and nonsynonymous variants of paired homeomorphic domain protein transcription factor 3 gene are stratified into the intermediate-risk group.

[0028] Preferably, the optimal critical value is 2.

[0029] The present invention also provides a risk stratification model for hypertrophic cardiomyopathy prepared by the above-described construction method.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] This invention provides a risk stratification model for hypertrophic cardiomyopathy (HCM), its preparation method, and its application. Univariate and multivariate Cox proportional hazards regression models revealed that elevated serum PDCD5 protein expression is not only a risk factor for adverse outcomes in HCM patients but also independent of other clinical factors. Survival curves based on serum PDCD5 protein risk stratification also showed good discriminative power. These results suggest that patients with higher serum PDCD5 protein expression have a higher risk of future adverse outcomes. Based on a multivariate Cox proportional hazards regression model for adverse outcomes in HCM patients, this invention established a score 2 that includes serum PDCD5 protein and clinical factors. Compared to score 1 based on clinical factors, score 2 has a better C-index. Therefore, PDCD5 protein improves the risk stratification model for HCM based on clinical factors. Nonsynonymous variants of VWCE, SPAG5, PDGFRB, and PITX3 genes are independently associated with an increased risk of adverse outcomes in HCM patients. The multidimensional risk stratification model established by combining the above-mentioned nonsynonymous gene variants with score 2 based on serum PDCD5 protein and clinical factors has good predictive power for adverse outcomes in HCM patients. Attached Figure Description

[0032] Figure 1 The results show the comparison of serum PDCD5 protein expression in the population to be tested;

[0033] Figure 2 ROC curve for serum PDCD5 protein in the diagnosis of HCM;

[0034] Figure 3 Comparison of PDCD5 protein levels in serum of HCM patients with overall composite outcome and in serum of control group;

[0035] Figure 4 Forest plot of univariate Cox proportional hazards regression model for serum PDCD5 protein and various outcomes in HCM patients;

[0036] Figure 5 The hazard ratios and 95% confidence intervals for serum PDCD5 protein and the study endpoint in HCM patients are as follows: (a) hazard ratios and 95% confidence intervals for serum PDCD5 protein and the overall composite outcome; (b) hazard ratios and 95% confidence intervals for serum PDCD5 protein and the composite outcome of HF.

[0037] Figure 6 Survival curves for the primary endpoint in HCM patients with different serum PDCD5 protein expression levels;

[0038] Figure 7Survival curves for secondary endpoints in HCM patients with different serum PDCD5 protein expression levels: (a) Survival curves for the SCD composite outcome in HCM patients with different serum PDCD5 protein expression levels; (b) Survival curves for the HF composite outcome in HCM patients with different serum PDCD5 protein expression levels.

[0039] Figure 8 Forest plot of a multivariate Cox proportional hazards regression model for the overall composite outcome of HCM patients, based on a risk stratification model of HCM patients using PDCD5 combined with clinical factors;

[0040] Figure 9 Survival curves for HCM patients after risk stratification based on score 2 and the overall composite outcome;

[0041] Figure 10 Forest plot of a multivariate Cox proportional hazards regression model for the overall composite outcome of HCM patients, based on a clinical factor-based risk stratification model for HCM patients;

[0042] Figure 11 Predictive scores for overall composite outcomes and stratified survival curves for HCM patients;

[0043] Figure 12 The frequency of variant alleles representing nonsynonymous variants of pathogenic genes in HCM patients;

[0044] Figure 13 Survival curves for HCM patients carrying nonsynonymous gene variants and overall adverse outcomes are shown below: (a) Survival curves for HCM patients carrying VWCE gene nonsynonymous variants; (b) Survival curves for HCM patients carrying TFPI gene nonsynonymous variants; (c) Survival curves for HCM patients carrying MPST gene nonsynonymous variants; (d) Survival curves for HCM patients carrying MCAM gene nonsynonymous variants; (e) Survival curves for HCM patients carrying PITX3 gene nonsynonymous variants; (f) Survival curves for HCM patients carrying PAK2 gene nonsynonymous variants; (g) Survival curves for HCM patients carrying SPAG5 gene nonsynonymous variants; (h) Survival curves for HCM patients carrying PDGFRB gene nonsynonymous variants.

[0045] Figure 14 Multivariate Cox regression analysis of overall composite outcome and nonsynonymous gene variants in HCM patients;

[0046] Figure 15 Survival curves for HCM patients with nonsynonymous variants of any of the VWCE, SPAG5, PDGFRB, or PITX3 genes and their overall composite outcome;

[0047] Figure 16To assess the predictive efficacy of the multidimensional risk stratification model and score 2 on the overall composite outcome of HCM patients, where Model 1... a Representative score 2, Model 2 b This represents a multi-dimensional hazard stratification model;

[0048] Figure 17 Survival curves for the overall composite outcome of HCM patients stratified according to a multidimensional risk stratification model. Detailed Implementation

[0049] This invention has discovered that serum PDCD5 protein may be a novel prognostic biomarker for hepatocellular carcinoma (HCM). First, serum PDCD5 protein levels were significantly elevated in HCM patients compared to healthy individuals without cardiovascular disease, indicating that PDCD5 protein may be generated during the development and progression of HCM. Detection of serum PDCD5 protein expression levels in the subjects showed that the level of PDCD5 protein expression can, to some extent, diagnose HCM, with an optimal cutoff value of 0.95 ng / mL. Second, high serum PDCD5 protein expression can independently predict the risk of adverse outcomes in HCM patients. Patients with high PDCD5 expression have a more than 5 times higher risk of adverse outcomes than those with low expression. Furthermore, the optimal cutoff value for serum PDCD5 protein in predicting adverse outcomes (overall composite outcome) in HCM patients was determined to be 4.09 ng / mL.

[0050] The adverse outcomes of this invention are categorized into overall composite outcome, HF composite outcome, and SCD composite outcome. This invention considers the overall composite outcome as the primary adverse outcome, and the HF composite outcome and SCD composite outcome as secondary adverse outcomes. The overall composite outcome is defined as cardiac death, SCD survival, fatal ventricular arrhythmia, appropriate implantable cardioverter-defibrillator (ICD) intervention, heart transplantation or implantation of a left ventricular assist device, and uncontrollable NYHA Class IV symptoms. The SCD composite outcome is defined as the composite outcome of SCD, SCD survival, fatal ventricular arrhythmia, and appropriate ICD intervention. The HF composite outcome is defined as the composite outcome of HF-related death, heart transplantation or implantation of a left ventricular assist device, and uncontrollable NYHA Class IV symptoms. NYHA Class I / II / III / IV are assigned scores of 1 / 2 / 3 / 4, respectively.

[0051] Based on this, the present invention provides an application of a reagent for detecting PDCD5 protein expression levels, comprising at least one of the following:

[0052] (1) Application of reagents for detecting PDCD5 protein expression levels in the preparation of products for assessing the risk of hypertrophic cardiomyopathy;

[0053] (2) Application of reagents for detecting PDCD5 protein expression levels in the preparation of products for assessing the disease risk of patients with hypertrophic cardiomyopathy.

[0054] In this invention, the reagents for detecting PDCD5 protein expression levels include antibodies such as ELISA kits, probes, or fluorescent reagents.

[0055] In this invention, for the application of (1), compared with the negative control serum sample, the expression of PDCD5 protein in the serum sample to be tested is significantly increased, with P < 0.05 for the serum sample to be tested vs. the negative control serum sample, and the PDCD5 protein in the serum sample to be tested is > 0.95 ng / mL, indicating that the subject to be tested has a high risk of hypertrophic cardiomyopathy. The negative control serum sample is peripheral venous blood serum from healthy individuals without cardiovascular disease.

[0056] For the application of (2), compared with the control serum sample group, the expression of PDCD5 protein in the serum samples of the patients with hypertrophic cardiomyopathy (HCM) to be tested was significantly increased, P < 0.05, and the PDCD5 protein in the serum samples to be tested was > 4.09 ng / mL, indicating that the disease risk of HCM patients was high. The control serum sample was the peripheral venous blood serum of HCM patients who did not experience the overall composite outcome, the peripheral venous blood serum of HCM patients who did not experience the HF composite outcome, or the peripheral venous blood serum of HCM patients who did not experience the SCD composite outcome. The disease risk refers to the risk of adverse outcomes.

[0057] Following the discovery that serum PDCD5 protein independently predicts the risk of adverse outcomes in HCM patients, this invention further provides an application of PDCD5 protein expression levels combined with clinical factors in constructing a risk stratification model for hypertrophic cardiomyopathy. The clinical factors include a history of syncope / presyncope, NYHA classification, and echocardiographic evidence of left ventricular enlargement. The risk stratification model for hypertrophic cardiomyopathy established by this invention using serum PDCD5 protein and clinical factors demonstrates good predictive efficacy, superior to risk stratification models based solely on clinical factors.

[0058] This invention also provides the application of PDCD5 protein expression level, clinical factors, and gene nonsynonymous variants in constructing a risk stratification model for hypertrophic cardiomyopathy, wherein the clinical factors are history of syncope / syncope precursor, NYHA classification, and echocardiographic evidence of left ventricular enlargement.

[0059] The nonsynonymous gene variants are nonsynonymous variants carrying the von Willebrand factor C and EGF domain gene, nonsynonymous variants of the sperm-associated antigen 5 gene, nonsynonymous variants of the platelet-derived growth factor receptor β gene, or nonsynonymous variants of the paired homeotypical domain protein transcription factor 3 gene.

[0060] This invention constructs a multidimensional risk stratification model based on PDCD5 protein expression levels, clinical factors, and gene nonsynonymous variants. In essence, it is a tandem model consisting of two risk stratification models: a risk stratification model established by clinical factors and serum PDCD5 protein, and a risk stratification model based on gene nonsynonymous variants that are independently associated with poor prognosis in HCM.

[0061] Based on this, the present invention also provides a method for constructing a risk stratification model for hypertrophic cardiomyopathy, comprising the following steps:

[0062] Candidate variables were included in multivariate Cox regression analysis and stepwise regression was used. Then, variables with p < 0.05 in the multivariate Cox regression analysis were used to build a risk stratification model. Each variable was weighted and scored according to the regression coefficients. The scores of each variable were summed to obtain the prediction score for adverse outcomes. The optimal cutpoint for risk stratification of the prediction score was calculated using survival analysis of R 4.2.3 and the surv_cutpoint function in its visualization toolkit. Patients with hypertrophic cardiomyopathy were stratified according to the optimal cutpoint, thus obtaining the risk stratification model for hypertrophic cardiomyopathy.

[0063] The candidate variables include clinical factors and serum PDCD5 protein expression levels; the clinical factors are age, sex, whether it is obstructive HCM, history of syncope / presyncope, NYHA classification, body mass index, history of hypertension, myoglobin, atrial flutter / fibrillation, non-sustained ventricular tachycardia, echocardiographic left ventricular enlargement, echocardiographic left ventricular ejection fraction, and echocardiographic wall motion abnormalities.

[0064] In this invention, the variables with p<0.05 are history of syncope / presyncope, NYHA classification, echocardiographic findings of left ventricular enlargement, and serum PDCD5 protein expression level; the calculation formula for the risk stratification model of hypertrophic cardiomyopathy is Score2 = History of syncope / presyncope + NYHA classification + left ventricular enlargement + high serum PDCD5 protein expression, where the presence of history of syncope / presyncope is worth 1 point, NYHA classification I is worth 1 point, NYHA classification II is worth 2 points, NYHA classification III is worth 3 points, NYHA classification IV is worth 4 points, echocardiographic findings of left ventricular enlargement are worth 1 point, and serum PDCD5 protein >4.09 ng / mL is worth 1 point;

[0065] When stratifying patients with hypertrophic cardiomyopathy, they are divided into a high-risk group and a low-risk group. The high-risk group is defined as those with a predicted score > the optimal cutoff value, and the low-risk group is defined as those with a predicted score ≤ the optimal cutoff value.

[0066] In this invention, the candidate variables also include nonsynonymous variants of genes carrying von Willebrand factor C and EGF domains, nonsynonymous variants of genes carrying tissue factor pathway inhibitors, nonsynonymous variants of genes carrying mercaptopyruvate thiotransferase, nonsynonymous variants of genes carrying melanoma cell adhesion molecules, nonsynonymous variants of genes carrying paired homeotypical domain protein transcription factor 3, nonsynonymous variants of genes carrying p21 activated kinase 2, nonsynonymous variants of genes carrying sperm-associated antigen 5, and nonsynonymous variants of genes carrying platelet-derived growth factor receptor β.

[0067] The variables with p < 0.05 were: history of syncope / presyncope, NYHA classification, left ventricular enlargement as indicated by echocardiography, serum PDCD5 protein expression level, carrying nonsynonymous variants of gene containing von Willebrand factor C and EGF domain, nonsynonymous variants of gene containing sperm-associated antigen 5, nonsynonymous variants of gene containing platelet-derived growth factor receptor β, and nonsynonymous variants of gene containing paired homeotypical domain protein transcription factor 3.

[0068] When stratifying patients with hypertrophic cardiomyopathy, they are divided into high-risk, intermediate-risk, and low-risk groups.

[0069] Patients with a score 2 > the optimal cutoff value and carrying at least one of the following: nonsynonymous variants of von Willebrand factor C and EGF domain genes, nonsynonymous variants of sperm-associated antigen 5 gene, nonsynonymous variants of platelet-derived growth factor receptor β gene, and nonsynonymous variants of paired homeomorphic domain protein transcription factor 3 gene are stratified as high-risk group.

[0070] Patients with a score 2 ≤ the optimal cutoff value and who do not carry nonsynonymous variants of von Willebrand factor C and EGF domain genes, nonsynonymous variants of sperm-associated antigen 5 genes, nonsynonymous variants of platelet-derived growth factor receptor β genes, and nonsynonymous variants of paired homeomorphic domain protein transcription factor 3 genes are stratified into the low-risk group.

[0071] The intermediate-risk group is any one of the following:

[0072] (S1) Patients with a score 2 > the optimal cutoff value and who do not carry nonsynonymous variants of von Willebrand factor C and EGF domain genes, nonsynonymous variants of sperm-associated antigen 5 genes, nonsynonymous variants of platelet-derived growth factor receptor β genes, or nonsynonymous variants of paired homeomorphic domain protein transcription factor 3 genes are stratified.

[0073] (S2) Patients with a score 2 ≤ the optimal cutoff value and carrying at least one of the following: nonsynonymous variants of von Willebrand factor C and EGF domain genes, nonsynonymous variants of sperm-associated antigen 5 gene, nonsynonymous variants of platelet-derived growth factor receptor β gene, and nonsynonymous variants of paired homeomorphic domain protein transcription factor 3 gene.

[0074] In this invention, the optimal critical value is 2.

[0075] The present invention also provides a risk stratification model for hypertrophic cardiomyopathy prepared by the above-described construction method.

[0076] This invention employs the classic indicator for evaluating the effectiveness of Cox proportional hazards regression models: the C-index and its 95% CI. The merits of different models are assessed by comparing whether there are statistically significant differences in their C-indexes. The C-index represents the probability that the predicted result matches the actual result; a C-index of 1 indicates perfect agreement, a C-index > 0.90 indicates high accuracy, a C-index between 0.71 and 0.90 indicates moderate accuracy, and a C-index between 0.50 and 0.70 indicates low accuracy. The study found that Score 1 has strong clinical applicability, but its C-index is only around 0.7, indicating moderate accuracy. Score 2, based on the optimal critical value of the univariate Cox proportional hazards regression model of serum PDCD5 protein and adverse outcomes, transforms it from a continuous variable into a categorical variable, dividing it into high and low PDCD5 expression groups. Furthermore, high PDCD5 expression and other clinical factors are then combined using a multivariate Cox proportional hazards regression model to construct Score 2. The C-index of score 2 as a continuous variable for predicting adverse outcomes in HCM patients was 0.866, and its C-index as a categorical variable for risk stratification reached 0.774. Statistical comparison of C-index among different models suggests that score 2 has better predictive power for adverse outcomes in HCM patients than score 1. This invention combines a risk stratification model based on clinical factors, including high PDCD5 protein expression, with a risk stratification model based on non-synonymous gene variant information to obtain a final multidimensional risk stratification model. This model has a C-index of 0.941, which is superior to the risk stratification model based on PDCD5 protein and clinical factors. The survival curves of the three groups of HCM patients stratified according to this multidimensional model showed good pairwise discrimination. These results suggest that the multidimensional model is a strategy for improving risk stratification in HCM patients.

[0077] The technical solutions provided by the present invention will be described in detail below with reference to the embodiments, but they should not be construed as limiting the scope of protection of the present invention.

[0078] In the following embodiments, clinical data were collected as follows:

[0079] Baseline clinical data of all enrolled HCM patients were collected and recorded through the electronic inpatient medical record system, including demographic characteristics, medical history, and laboratory test information. Twelve-lead electrocardiograms, Holter monitors, and echocardiograms of all enrolled HCM patients during their hospitalization were reviewed through the electronic inpatient medical record system, and their arrhythmia types, electrocardiographic features, and echocardiographic features were collected and recorded.

[0080] Demographic characteristics included sex, age, height, and weight. Height and weight were measured on the second day of hospitalization, under conditions of wearing only a hospital gown and fasting. Body mass index (BMI) was calculated using the following formula: BMI = weight (kg) / height 2 (m 2 ).

[0081] Medical history information includes present illness, past medical history, personal history, family history, physical examination, etc. Specifically, it includes NYHA functional classification, symptom information such as history of syncope / prodromal syncope, information on common comorbidities such as coronary heart disease (CAD), hypertension, diabetes, smoking history, drinking history, family history information such as HCM family history and SCD family history of first-degree relatives, as well as vital signs such as systolic blood pressure, diastolic blood pressure, and heart rate at the time of admission.

[0082] Laboratory test information includes complete blood count, liver and kidney function tests, blood glucose and lipids, coagulation function, myocardial injury markers, and brain natriuretic peptide (BNP) levels. Among these, the highest values ​​of myocardial injury markers and BNP levels were recorded during hospitalization, while the remaining information was recorded based on fasting test results on the second day of admission.

[0083] By reviewing 12-lead electrocardiograms, ECG monitoring, and Holter monitoring images collected and recorded in the electronic inpatient medical record system, the following arrhythmia types were identified: sick sinus syndrome, atrial flutter / fibrillation, non-sustained ventricular tachycardia (NSVT), left bundle branch block, right bundle branch block, and atrioventricular block. NSVT was defined as at least three consecutive ventricular tachycardias recorded by ECG monitoring or Holter monitoring, with a frequency ≥120 bpm, a duration <30 seconds, and spontaneous termination within 30 seconds. In addition to arrhythmia type, ECG characteristics were recorded, including: biphasic P waves, tall peaked P waves, pathological q waves, ST segment elevation, ST segment depression, T wave inversion, and left ventricular hypertrophy [defined as RV5+SV1>4.0mV (male) / 3.5mV (female)].

[0084] Echocardiographic features include left ventricular maximal wall thickness (LVMWT), interventricular septal end-diastolic thickness (IVSTd), left ventricular posterior wall end-diastolic thickness (LVPWTd), left atrial diameter (LAD), left ventricular end-diastolic diameter (LVEDd), left ventricular end-systolic diameter (LVESd), maximal left ventricular outflow tract pressure gradient (LVOTmaxPG), LVEF, peak velocity of the early diastolic mitral valve filling wave (E), and peak velocity of the late diastolic mitral valve filling wave. The left ventricular wall is defined as follows: diastolic inflow velocity (A) ratio E / A; early transmitral flow velocity (E) to early diastolic annular velocity (e') ratio E / e'; and the presence of abnormal wall motion. Severe left ventricular wall hypertrophy is defined as LVMWT ≥ 3.0 cm; left atrial enlargement is defined as LAD > 4.0 cm (male) / 3.8 cm (female); left ventricular enlargement is defined as LVEDd > 5.5 cm (male) / 5.0 cm (female); and left ventricular systolic dysfunction is defined as left ventricular ejection fraction (LVEF) ≤ 50%.The left ventricular mass (LVM) and left ventricular mass index (LVMI) are calculated according to the following formulas: (1) LVM = 0.8 × 1.04 × [(IVSTd + LVEDd + LVPWTd)3 – LVEDd3] + 0.6g; (2) First calculate the body surface area (BSA), then calculate LVMI: male BSA = 0.00607 × height (cm) + 0.0127 × weight (kg) - 0.0698; female BSA = 0.00586 × height (cm) + 0.0126 × weight (kg) - 0.0461; LVMI = LVM / BSA.

[0085] Example 1: Application of reagents for detecting PDCD5 protein expression levels in the preparation of products for assessing the risk of hypertrophic cardiomyopathy.

[0086] (1) Sample grouping

[0087] The samples are divided into serum samples to be tested and serum samples for negative control. The serum samples for negative control are peripheral venous blood serum from healthy individuals without cardiovascular disease.

[0088] (2) Sample collection:

[0089] 4 mL of peripheral venous blood was collected from both the subjects to be tested and healthy individuals using disposable vacuum blood collection tubes (red cap) without additives. After collection, the blood was centrifuged at 3000 rpm for 10 min using an SL02 low-speed centrifuge to collect serum. Serum samples for testing and negative control were obtained separately, aliquoted, and stored at -80℃.

[0090] (3) Serum samples were used to detect serum PDCD5 protein.

[0091] The PDCD5 protein in the serum samples to be tested and the negative control serum samples was detected by ELISA using a human PDCD5 enzyme-linked immunosorbent assay (ELISA) kit. The absorbance was measured by an ELISA reader, and a standard protein curve was plotted using Excel software to calculate the PDCD5 protein concentration of each serum sample.

[0092] (4) Evaluation methods

[0093] Compared with the negative control serum sample, the expression of PDCD5 protein in the serum sample to be tested was significantly increased. The P value of the serum sample to be tested versus the negative control serum sample was 0.004, and the PDCD5 protein in the serum sample to be tested was >0.95 ng / mL, which indicates that the subjects to be tested had a high risk of hypertrophic cardiomyopathy.

[0094] Example 2: Reagents for detecting PDCD5 protein expression levels were used to assess the risk of hypertrophic cardiomyopathy.

[0095] 1.1 Research Methods

[0096] 1.1.1 Research Subjects

[0097] This embodiment was approved by the Ethics Committees of Peking University People's Hospital and Fuwai Hospital of the Chinese Academy of Medical Sciences, and all enrolled patients signed informed consent forms.

[0098] Case group: Patients diagnosed with HCM who were hospitalized in the Department of Cardiology at Peking University People's Hospital from April 2021 to April 2023, and patients diagnosed with HCM who were hospitalized in the Department of Cardiomyopathy at Fuwai Hospital, Chinese Academy of Medical Sciences in June 2023 were consecutively enrolled.

[0099] The inclusion criteria for the case group were: (1) age between 18 and 80 years; and (2) meeting the diagnostic criteria for HCM in the 2020 American Heart Association (AHA) / American College of Cardiology (ACC) HCM guidelines.

[0100] The exclusion criteria for the case group were: (1) age <18 years or age >80 years; (2) left ventricular hypertrophy caused by other reasons such as amyloidosis, Fabry disease, hypertensive heart disease, aortic stenosis, etc. could not be ruled out; (3) refusal to participate in this study.

[0101] Based on the inclusion and exclusion criteria for the above case groups, a total of 177 HCM patients were included in this study. Two patients were excluded because their blood samples were of poor quality and serum PDCD5 protein could not be detected or whole exome sequencing (WES) was not possible. In the end, a total of 175 HCM patients were analyzed, of which 164 HCM patients also completed WES testing.

[0102] Control group: Healthy individuals without cardiovascular disease were recruited. Inclusion criteria were: (1) age between 18 and 80 years; (2) no cardiovascular diseases such as coronary heart disease, hypertension, or cardiomyopathy. Exclusion criteria were: (1) age <18 years or age >80 years; (2) pregnancy or lactation; (3) presence of underlying cardiovascular disease; (4) refusal to participate in this study. A total of 194 subjects were recruited as the control group.

[0103] 1.1.2 PDCD5 expression level is used to assess the risk of hypertrophic cardiomyopathy.

[0104] The method described in Example 1 was used to detect PDCD5 in the serum of peripheral venous blood samples from 175 HCM patients (subjects to be tested) and 194 control subjects (negative control subjects).

[0105] Figure 1 The results showed that, compared with 194 healthy individuals without cardiovascular disease, serum PDCD5 protein expression was elevated in 175 HCM patients, and this was statistically significant [HCM group vs. control group: 1.74 (1.13–2.69) vs. 1.51 (0.84–2.43) ng / mL, P = 0.004].

[0106] The diagnostic efficacy of serum PDCD5 protein for hepatocellular carcinoma (HCM) was evaluated using receiver operating characteristic (ROC) curves. The area under the curve (AUC) and its 95% confidence interval (CI) were calculated. The optimal cutoff value for serum PDCD5 protein in diagnosing HCM was calculated using the Youden index, and its sensitivity, specificity, positive predictive value, and negative predictive value were also calculated.

[0107] Figure 2 The results showed that serum PDCD5 protein has certain diagnostic value for hepatocellular carcinoma (HCM), with an AUC and 95% CI of 0.587 (0.530–0.645) (p = 0.004). The optimal cutoff value for serum PDCD5 protein in diagnosing HCM was 0.95 ng / mL. A serum PDCD5 protein level >0.95 ng / mL had a sensitivity of 83.1%, a specificity of 32.0%, a positive predictive value of 52.7%, a negative predictive value of 67.4%, and a diagnostic accuracy of 56.3% in diagnosing HCM.

[0108] Example 3: Application of reagents for detecting PDCD5 protein expression levels in the preparation of products for assessing disease risk in HCM patients.

[0109] (1) Adverse outcomes in HCM patients

[0110] All enrolled HCM patients were followed up via semi-structured interviews by telephone or outpatient visits. The primary endpoint of the study was a composite endpoint, defined as the overall composite outcome of cardiac death, survival of SCD, fatal ventricular arrhythmia, appropriate implantable cardioverter-defibrillator (ICD) intervention, heart transplantation or implantation of left ventricular assist device, and uncontrollable NYHA class IV symptoms.

[0111] Cardiac death was defined as death related to SCD, HF, and operation. Secondary endpoints were defined as SCD composite outcome and HF composite outcome: (1) SCD composite outcome was defined as the composite outcome of SCD, SCD survival, fatal ventricular arrhythmia, and appropriate ICD intervention; (2) HF composite outcome was defined as the composite outcome of HF-related death, heart transplantation or implantation of left ventricular assist device, and uncontrollable NYHA class IV symptoms.

[0112] (2) Serum samples from HCM patients

[0113] HCM patient serum samples were divided into a test HCM patient serum sample group and a control serum sample group. The control serum samples were peripheral venous blood serum from HCM patients who did not have the overall composite outcome, peripheral venous blood serum from HCM patients who did not have the HF composite outcome, or peripheral venous blood serum from HCM patients who did not have the SCD composite outcome.

[0114] (3) Detection of PDCD5 protein in serum of HCM patients

[0115] The method for detecting PDCD5 protein in the serum of HCM patients is described in step (3) of Example 1.

[0116] (4) The relationship between serum PDCD5 protein and adverse outcomes in HCM patients was assessed using a Cox proportional hazards regression model, and the hazard ratio (HR) and 95% CI were calculated. Patients were grouped according to the optimal cutoff value for predicting adverse outcomes based on serum PDCD5 protein, and the relationship between different serum PDCD5 protein expression levels and adverse outcomes was analyzed using Kaplan-Meier curves. The relationship between PDCD5 gene variants and adverse outcomes was also analyzed using Kaplan-Meier curves.

[0117] Univariate Cox proportional hazards regression analysis was used to screen all variables. Variables with p < 0.10 in the univariate Cox regression analysis (the variables being gender, body mass index, history of syncope / presyncope, NYHA classification, history of hypertension, myoglobin, atrial flutter / fibrillation, non-sustained ventricular tachycardia, echocardiographic left ventricular enlargement, echocardiographic left ventricular ejection fraction, and echocardiographic wall motion abnormalities), as well as other clinically significant variables (the significant variables being age and whether obstructive HCM is present), were used as candidate variables. Pearson or Spearman correlation analysis was used to assess the correlation between candidate variables. Variables with strong correlations were combined, and variables with too many missing values ​​were removed. The variables were weighted and scored according to the regression coefficients. The scores of each variable were summed to obtain the predicted score for adverse outcomes. The optimal cutpoint value for risk stratification of the predicted score was calculated using the surv_cutpoint function (Kassambara) in the survival analysis and visualization (survminer) toolkit of R 4.2.3.

[0118] (5) Evaluation methods

[0119] Compared with the control group, HCM patients who experienced the overall composite outcome had significantly higher baseline serum PDCD5 protein levels, and the PDCD5 protein in the serum sample to be tested was >4.09 ng / mL, which indicates that HCM patients have a high risk of adverse outcomes or a high disease risk.

[0120] Example 4: Reagents for detecting PDCD5 expression levels were used to assess the disease risk in HCM patients.

[0121] The study subjects used in this embodiment were the same as those described in Example 2. A total of 166 (94.9%) HCM patients completed follow-up, with a loss to follow-up rate of 5.1%. The follow-up time was 5.03 (4.67–7.78) months, with a potential median follow-up time and 95% CI of 5.03 (4.91–5.15) months. During the follow-up period, 14 patients (8.4%) experienced an overall composite outcome, 10 patients experienced a SCD composite outcome, 6 patients experienced a HF composite outcome, 3 patients experienced cardiac death, 3 patients experienced fatal ventricular arrhythmias, and 3 patients experienced uncontrollable NYHA class IV symptoms.

[0122] The method described in Example 3 was used to detect PDCD5 in the serum of peripheral venous blood samples from 14 HCM patients (subjects to be tested) with overall composite outcome and 152 control subjects.

[0123] Figure 3The results showed that HCM patients who experienced the overall composite outcome had higher baseline serum PDCD5 protein levels compared with those who did not [overall composite outcome group: 3.04 (1.38–4.78) vs. 1.67 (1.06–2.60) ng / mL, p = 0.025].

[0124] Univariate Cox proportional hazards regression analysis indicated that higher serum PDCD5 protein levels in HCM patients were associated with a higher risk of overall composite outcome [HR and 95% CI: 1.263 (1.068–1.493), p = 0.006], HF composite outcome [HR and 95% CI: 1.392 (1.088–1.780), p = 0.008], or cardiac death [HR and 95% CI: 1.550 (1.164–2.064), p = 0.003] (see [link to relevant data]). Figure 4 Furthermore, serum PDCD5 protein levels showed a linear relationship with the risk of the overall composite outcome (P = 0.036, nonlinear P = 0.412) and the HF composite outcome (P = 0.028, nonlinear P = 0.927) (see [link to relevant documentation]). Figure 5 ).

[0125] The optimal cutoff value for serum PDCD5 protein in predicting the overall composite outcome was 4.09 ng / mL. HCM patients were divided into a high-expression group (serum PDCD5 protein > 4.09 ng / mL) and a low-expression group (serum PDCD5 protein ≤ 4.09 ng / mL) based on serum PDCD5 protein levels.

[0126] Compared with HCM patients with low serum PDCD5 protein expression, patients with high PDCD5 expression had a more than 5-fold increased risk of the primary endpoint [HR and 95% CI: 6.722 (2.317-19.504), p<0.001], and the survival curves of the two groups were well-discriminatory (log rank test p<0.001) (see Figure 6 The C-index and 95% CI for predicting the overall composite outcome by serum PDCD5 protein expression level were 0.697 (0.560-0.834).

[0127] Serum PDCD5 protein expression levels also showed good predictive value for secondary endpoints. Compared with HCM patients with low serum PDCD5 protein expression, patients with high serum PDCD5 expression had a higher risk of the SCD composite outcome [HR and 95% CI: 5.742 (1.605-20.540), p = 0.007] or the HF composite outcome [HR and 95% CI: 7.128 (1.433-35.463), p = 0.016]. Survival curves for both the SCD composite outcome (log rank test p = 0.002) and the HF composite outcome (log rank test p = 0.005) showed good discriminatory power between the high and low serum PDCD5 protein expression groups (see...). Figure 7 The C-index and 95% CI of serum PDCD5 protein expression level for predicting the SCD and HF composite outcomes were 0.673 (0.511-0.834) and 0.738 (0.501-0.976), respectively.

[0128] Example 5: Application of PDCD5 expression level combined with clinical factors in the preparation of products for assessing disease risk in HCM patients

[0129] Based on Example 4, the disease risk of HCM patients was assessed by combining PDCD5 with clinical factors.

[0130] The study subjects used in this embodiment were the same as those described in Example 2. A total of 166 (94.9%) HCM patients completed follow-up, with a loss to follow-up rate of 5.1%. The follow-up time was 5.03 (4.67–7.78) months, with a potential median follow-up time and 95% CI of 5.03 (4.91–5.15) months. During the follow-up period, 14 patients (8.4%) experienced an overall composite outcome, 10 patients experienced a SCD composite outcome, 6 patients experienced a HF composite outcome, 3 patients experienced cardiac death, 3 patients experienced fatal ventricular arrhythmias, and 3 patients experienced uncontrollable NYHA class IV symptoms.

[0131] The relationship between serum PDCD5 protein and adverse outcomes in HCM patients was assessed using a Cox proportional hazards regression model, calculating the hazard ratio (HR) and 95% confidence interval (CI). Patients were grouped according to the optimal cutoff value for predicting adverse outcomes based on serum PDCD5 protein expression, and the relationship between different serum PDCD5 protein expression levels and adverse outcomes was analyzed using Kaplan-Meier curves. The relationship between PDCD5 gene variants and adverse outcomes was also analyzed using Kaplan-Meier curves.

[0132] A hazard stratification model was established using univariate and multivariate Cox proportional hazards regression models: Univariate Cox proportional hazards regression was used to screen all variables. Variables with p < 0.10 in the univariate Cox regression analysis (the variables being gender, body mass index, history of syncope / presyncope, NYHA classification, history of hypertension, myoglobin, atrial flutter / fibrillation, non-sustained ventricular tachycardia, echocardiographic left ventricular enlargement, echocardiographic left ventricular ejection fraction, and echocardiographic wall motion abnormalities), as well as other clinically significant variables such as age and gender (the significant variables being age and whether obstructive HCM exists), were used as candidate variables. Pearson or Spearman correlation analysis was used to assess the correlation between candidate variables. Variables with strong correlations were combined, and variables with too many missing values ​​were removed. The processed candidate variables were then included in the multivariate Cox regression analysis using stepwise regression. In multivariate Cox regression analysis, variables with p < 0.05 were ultimately used to establish a risk stratification model. Each variable was weighted and scored according to its regression coefficients, and the scores of all variables were summed to obtain a predictive score for adverse outcomes. The C-index and its 95% confidence interval (CI) were calculated to evaluate the predictive efficacy of the established predictive score for adverse outcomes. Survival analysis using R4.2.3 and the surv_cutpoint function (Kassambara) in the visualization toolkit were used to calculate the optimal cutoff value for risk stratification based on the predictive score. Based on this cutoff value, HCM patients were stratified into high-risk (predictive score > cutoff value) and low-risk (predictive score ≤ cutoff value) groups. Kaplan-Meier curves were used to calculate the time to reach the study endpoint for patients in the high-risk and low-risk groups, and the log-rank test was used to compare the differences between the groups. The more obvious the separation of the curves between the two groups, the better the efficacy of the risk stratification method in identifying high-risk patients.

[0133] Statistical analyses were performed using IBM SPSS 25.0 and R 4.2.3 software. All statistical analyses employed two-tailed tests, with p < 0.05 considered statistically significant.

[0134] In this embodiment, age, sex, presence of obstructive HCM, history of syncope / prodromal syncope, NYHA classification, body mass index (BMI), history of hypertension, myoglobin, atrial flutter / fibrillation, non-sustained ventricular tachycardia, echocardiographic findings of left ventricular enlargement, left ventricular ejection fraction (LVEF), abnormal wall motion, and serum PDCD5 protein expression level were included in the multivariate Cox proportional hazards regression model for the overall composite outcome.

[0135] Figure 8 The results showed that a history of syncope / presyncope [HR and 95% CI: 4.825 (1.562-14.906), P = 0.006], NYHA classification [HR and 95% CI: 3.599 (1.353-9.577), P = 0.010], left ventricular enlargement [HR and 95% CI: 3.975 (1.174-13.454), P = 0.027], and serum PDCD5 protein expression level [HR and 95% CI: 5.676 (1.775-18.148), P = 0.003] were independent risk factors for the overall composite outcome.

[0136] Based on a multivariate Cox proportional hazards regression model, this embodiment establishes a new predictive score (risk stratification model) for adverse outcomes in HCM patients by combining serum PDCD5 protein expression levels and clinical factors. This score includes four variables: history of syncope / presyncope, NYHA classification, echocardiographic (UCG) findings of left ventricular enlargement, and serum PDCD5 protein expression level. The calculation formula is: Score2 = History of syncope / presyncope + NYHA classification + Left ventricular enlargement + High serum PDCD5 protein expression (i.e., 1 point for the presence of syncope / presyncope, 1 / 2 / 3 / 4 points for NYHA classification I / II / III / IV respectively, 1 point for echocardiographic findings of left ventricular enlargement, and 1 point for serum PDCD5 protein >4.09 ng / mL). For every 1-point increase in this score, the risk of adverse outcomes in HCM patients increases more than fourfold [HR and 95% CI: 4.347 (2.494-7.576), P<0.001].

[0137] Figure 9 The results showed that the optimal cutoff value for score 2 in predicting the overall composite outcome in HCM patients was 2. HCM patients with scores 2 > 2 and scores 2 ≤ 2 were stratified into high-risk and low-risk groups, respectively. The risk of overall composite outcome in high-risk HCM patients was more than 10 times that in low-risk patients [HR and 95% CI: 10.406 (2.324-46.591), p = 0.002]. The survival curves of the two groups and the overall composite outcome showed good discrimination (log rank test p < 0.001).

[0138] Comparative Example 1: Clinical Factors Used to Assess Disease Risk in HCM Patients

[0139] (1) Adverse outcomes in HCM patients

[0140] All enrolled HCM patients were followed up via semi-structured interviews by telephone or outpatient visits. The primary endpoint of the study was a composite endpoint, defined as the overall composite outcome of cardiac death, survival of SCD, fatal ventricular arrhythmia, appropriate implantable cardioverter-defibrillator (ICD) intervention, heart transplantation or implantation of left ventricular assist device, and uncontrollable NYHA class IV symptoms.

[0141] Cardiac death was defined as death related to SCD, HF, and operation. Secondary endpoints were defined as SCD composite outcome and HF composite outcome: (1) SCD composite outcome was defined as the composite outcome of SCD, SCD survival, fatal ventricular arrhythmia, and appropriate ICD intervention; (2) HF composite outcome was defined as the composite outcome of HF-related death, heart transplantation or implantation of left ventricular assist device, and uncontrollable NYHA class IV symptoms.

[0142] (2) Research subjects

[0143] The diagnostic criteria for hepatocellular carcinoma (HCM) are defined as a maximum end-diastolic ventricular wall thickness ≥15 mm at any location of the ventricle as measured by echocardiography (UCG) / cardiac magnetic resonance (CMR), which cannot be explained by other cardiovascular diseases, metabolic diseases, or systemic diseases. For patients with a clear family history of HCM or whose gene testing reveals gene mutations of clear pathogenic significance, this diagnostic threshold is relaxed to 13 mm. Exclusion criteria are defined as: (1) age >80 years; (2) malignant tumor; (3) severe liver or kidney dysfunction; (4) no UCG examination performed during hospitalization.

[0144] This study included 618 patients with hepatocellular carcinoma (HCM). 52 patients who did not undergo UCG examination during hospitalization and 5 patients whose UCG characteristics did not meet the diagnostic criteria for HCM were excluded. In addition, 40 patients aged >80 years, 36 patients with malignant tumors, and 4 patients with severe liver and kidney dysfunction were excluded. A total of 481 patients were included in the analysis.

[0145] A total of 481 HCM patients were included in the analysis, including 111 (23.1%) obstructive HCM patients (LVOTmaxPG ≥ 30 mmHg) and 370 (76.9%) non-obstructive HCM patients (LVOTmaxPG < 30 mmHg). The study population was aged 61.0 (51.0–69.5) years, predominantly male (60.3%), with a higher proportion of males among non-obstructive HCM patients than among obstructive HCM patients (65.7% vs 42.3%, p < 0.001). A history of syncope / prodromal syncope (36.8% vs 25.2%, p = 0.020) and NYHA grade III / IV symptoms (14.4% vs 6.5%, p = 0.014) were more common in obstructive HCM patients. The proportions of common comorbidities and a positive family history of HCM did not differ significantly between patients with obstructive HCM and those with non-obstructive HCM. Regarding treatment, angiotensin-converting enzyme inhibitors (ACEIs) or angiotensin receptor blockers (ARBs) were used more frequently in patients with non-obstructive HCM (47.8% vs 27.9%, p<0.001). The HCM-risk-SCD score, recommended for SCD risk stratification in HCM patients by the 2014 European Society of Cardiology (ESC) guidelines for the diagnosis and treatment of HCM, was also higher in patients with obstructive HCM [2.06 (1.44–3.43) vs 1.61 (1.27–2.19), p<0.001].

[0146] A total of 380 HCM patients were followed up for 5.58 years (range 2.27–10.29), with a potential median follow-up time and 95% CI of 6.38 years (range 5.62–7.15). During the follow-up period, 89 patients (23.4%) experienced an overall composite outcome, 41 patients (10.8%) experienced a SCD composite outcome, 60 patients (15.8%) experienced a HF composite outcome, 79 patients experienced all-cause mortality (38 of which were cardiac deaths), 12 patients survived SCD / SCD, 20 patients experienced fatal ventricular arrhythmias, 29 patients underwent appropriate ICD intervention, and 35 patients developed NYHA class IV symptoms that were difficult to control with medication.

[0147] (3) Establish a hazard stratification model using a multifactor Cox proportional hazards regression model.

[0148] The specific method for establishing a risk stratification model using a multivariate Cox proportional hazards regression model is described in Example 5. The difference between this method and Example 5 is that, in this example, when establishing a risk stratification model for HCM patients based on clinical factors, the survival data of 380 HCM patients with a median follow-up time of 6.38 years were analyzed. Age, sex, NYHA classification, family history of HCM, atrial flutter / fibrillation, non-sustained ventricular tachycardia, left bundle branch block, echocardiographic findings of severe left ventricular wall hypertrophy, left atrial enlargement, left ventricular enlargement, and left ventricular systolic dysfunction were included in the multivariate Cox proportional hazards regression analysis.

[0149] Multivariate Cox proportional hazards regression analysis of the overall composite outcome indicated that NYHA class [hazard ratio (HR) and 95% confidence interval (CI): 1.830 (1.399-2.393), P<0.001], NSVT [HR and 95% CI: 2.861 (1.577-5.191), P=0.001], echocardiographic evidence of severe left ventricular wall hypertrophy [HR and 95% CI: 2.750 (1.251-6.041), P=0.012], and left atrial enlargement [HR and 95% CI: 1.701 (1.087-2.661), P=0.020] were independent risk factors for the overall composite outcome in HCM patients (see [link to relevant data]). Figure 10 ).

[0150] Based on the multivariate Cox proportional hazards regression model of the above overall composite outcome, a predictive score for adverse outcomes in HCM patients was established. The score includes four variables: NYHA class, non-sustained ventricular tachycardia (NSVT), severe left ventricular wall hypertrophy and left atrial enlargement as indicated by echocardiography (UCG). The calculation formula is: Score 1 = NYHA class × 1 + NSVT × 2 + severe left ventricular wall hypertrophy × 2 + left atrial enlargement × 1 (that is, NYHA class I / II / III / IV are scored as 1 / 2 / 3 / 4 points respectively, non-sustained ventricular tachycardia is scored as 2 points, severe left ventricular wall hypertrophy is scored as 2 points, and left atrial enlargement is scored as 1 point).

[0151] Figure 11The results showed that the C-index and 95% CI of the predictive score for adverse outcomes in HCM patients were 0.699 (0.635-0.762) (P<0.001). For every 1-point increase in the predictive score, the risk of adverse outcomes in HCM patients increased 1.7-fold, with a HR and 95% CI of 1.737 (1.481-2.036) (P<0.001). The optimal cutoff value for the predictive score in HCM patients was 2 points. HCM patients with a predictive score >2 points were stratified as high-risk, and those with a predictive score ≤2 points were stratified as low-risk. The risk of adverse outcomes in the high-risk group was nearly 4 times that in the low-risk group [HR and 95% CI: 3.894 (2.530-5.992), P<0.001], and the survival curves of the two groups were well separated (logrank test, P<0.001).

[0152] Example 6

[0153] The subjects studied in this embodiment were the same as in Example 5. The C-index and 95% CI for predicting adverse outcomes in HCM patients were calculated for both the score 1 established in Comparative Example 1 and the score 2 established in Example 5, and compared to evaluate the superiority of the risk stratification model established by combining PDCD5 and clinical factors versus the risk stratification model based on clinical factors. The results are shown in Table 1.

[0154] Table 1. Predictive efficacy of different risk stratification models for overall composite outcomes in HCM patients.

[0155] Hazard stratification model C Index (95% CI) p-value HR (95% CI) p-value Rating 1 0.707(0.569-0.844) / 1.667(1.160-2.397) 0.006 Rating 2 0.866(0.773-0.960) <0.001 4.347(2.494-7.576) <0.001

[0156] The results in Table 1 show that, compared with the risk stratification model based on clinical factors (score 1) established in Comparative Example 1, the risk stratification model established in Example 5, which combines serum PDCD5 protein levels and clinical factors (score 2), has better efficacy in predicting the risk of adverse outcomes in HCM patients [score 2 vs score 1, C index and 95% CI: 0.866 (0.773-0.960) vs 0.707 (0.569-0.844); p<0.001].

[0157] Example 7: Using whole-exome sequencing (WES) results combined with serum PDCD5 levels to evaluate the disease risk in HCM patients.

[0158] 1.1 Follow-up and outcomes

[0159] All enrolled HCM patients in the case group were followed up via semi-structured interviews by telephone or outpatient visits. The primary endpoint of the study was a composite endpoint, defined as the overall composite outcome of cardiac death, survival of SCD, fatal ventricular arrhythmia, appropriate implantable cardioverter-defibrillator (ICD) intervention, heart transplantation or implantation of left ventricular assist device, and uncontrollable NYHA class IV symptoms.

[0160] Cardiac death was defined as death related to SCD, HF, and operation. Secondary endpoints were defined as SCD composite outcome and HF composite outcome: (1) SCD composite outcome was defined as the composite outcome of SCD, SCD survival, fatal ventricular arrhythmia, and appropriate ICD intervention; (2) HF composite outcome was defined as the composite outcome of HF-related death, heart transplantation or implantation of left ventricular assist device, and uncontrollable NYHA class IV symptoms.

[0161] 1.2 Whole exome sequencing (WES)

[0162] All enrolled HCM patients underwent WES testing. Frozen blood samples were transported in batches via cold chain to Shanghai Meiji Biotechnology Co., Ltd., where professional technicians performed WES testing. The control group consisted of East Asian populations selected from the Genome Aggregation Database (gnomAD).

[0163] First, DNA extraction is performed: (1) Pretreatment: the frozen blood sample is thawed on ice, the washing solution is added with the corresponding anhydrous ethanol according to the packaging label, and the elution solution is preheated on a 65°C metal bath.

[0164] (2) Grinding and lysis: Add 50-200 mg of sample to the grinding tube and grind at 45-65 Hz for 15-25 s; after grinding, add 20 μL proteinase K and 500 μL cell lysis buffer in sequence and mix thoroughly for 10 s; place on a 65 ℃ metal bath and heat at 500 rpm for 3 h, and take it out and mix it every hour.

[0165] (3) Column chromatography: After the polished tube has been allowed to stand at room temperature, centrifuge at 13000 rcf for 3 min; take the supernatant, add 250 μL of binding solution, mix well, centrifuge again at 13000 rcf for 1 min, and discard the supernatant; add 500 μL of washing solution A, centrifuge at 13000 rcf for 1 min, and discard the supernatant; add 500 μL of washing solution B, centrifuge at 13000 rcf for 1 min, and discard the supernatant; add 700 μL of washing solution C, centrifuge at 13000 rcf for 1 min, and discard the supernatant; centrifuge at 13000 rcf for 3 min, spin dry thoroughly, and air dry for 5 min;

[0166] (4) Elution: Transfer the centrifuge column to a new centrifuge tube, add 50–100 μL of preheated elution buffer to the center of the column, incubate at room temperature for 10 min, centrifuge at 13000 rcf for 3 min to obtain extracted DNA, and add 1–2 μL of RNase. Next, DNA quality control is performed. The DNA purity of the fully dissolved sample is detected using an anoDrop2000 spectrophotometer, the DNA concentration is detected using a Quantus Flurometer (Picogreen), and the DNA integrity is detected using agarose gel electrophoresis. Samples with a concentration ≥10 ng / μL, a total sample volume ≥0.8 μg, and a DNA main band >10 kb or diffuse gel electrophoresis with a DNA main band ≥3 kb, a non-viscous DNA solution, no pigments, suspended matter, and no severe contamination from RNA, protein, sugars, or other impurities are considered to have passed quality control. Finally, whole-exome sequencing is performed. After the genomic DNA passed quality control, the DNA sequence was randomly fragmented into 180–280 bp fragments using a Covaris disruptor with ultrasound. Library construction and capture experiments were performed using the Agilent SureSelect HumanAll ExonV6 kit. The randomly fragmented DNA fragments underwent end repair, phosphorylation, and 3' A-tailing. Adapters were then ligated to both ends of the fragments to prepare DNA libraries. Next, the DNA library with a specific index was hybridized to biotin-labeled probes in liquid chromatography, and exon sequences were captured by elution with streptomycin-containing magnetic beads. Linear PCR amplification was then performed to construct sequencing libraries. Finally, the constructed sequencing libraries underwent quality control, and those that passed quality control were sequenced using the Illumina Novoseq™ platform.

[0167] After obtaining the raw sequencing data, quality control was first performed to filter out low-quality sequencing data. Next, the data was aligned with the reference genome sequence (GRCh38), and the BWA-MEME software was used to obtain the sequence location file. Then, the location file was corrected using the Best Practices workflow of GATK software, and single nucleotide polymorphisms (SNPs) and small insertion and deletion (Indel) variants were detected. Finally, VEP software was used for functional annotation of the variant information. Manta software was used to detect chromosomal structural variations in the sample compared to the reference genome sequence, and CNVkit software was used to detect copy number variations.

[0168] 1.3 Analysis of whole exome sequencing data

[0169] The analysis methods for the above-mentioned variant information are as follows. First, variants with sequencing depth <10X were filtered out. Nonsynonymous variants were extracted and analyzed using the R 4.2.3 toolkit maftools, where nonsynonymous variants were defined as: nonsense variants, stop codon variants, start codon variants, splice site variants, frameshift variants, missense variants, and in-frame deletion or insertion variants. The distribution of variant information across chromosomes was visualized using Circos software. The relationship between gene nonsynonymous variants and the study endpoint in HCM patients was analyzed using a univariate Cox proportional hazards regression model, and the Benjamini-Hochberg method was used to correct for false discovery rate of p-values. The impact of whether or not a gene carries nonsynonymous variants on the study endpoint was analyzed using Kaplan-Meier curve analysis. The R 4.2.3 toolkit clusterProfier was used to perform functional predictions on multiple gene nonsynonymous variants that were screened and might predict adverse outcomes in HCM patients. Multivariate Cox regression analysis was performed on nonsynonymous variants with corrected p-values ​​<0.05 in univariate Cox proportional hazards regression analysis to screen for nonsynonymous variants that may be independently associated with an increased risk of adverse outcomes in HCM patients. HCM patients carrying at least one of these nonsynonymous variants were classified as high-risk, and others as low-risk. Patients with a predictive HCM adverse outcome score >2 established in Example 5 were classified as high-risk, and others as low-risk. In the final multidimensional risk stratification model, patients classified as high-risk by both stratification methods were ultimately classified as high-risk, patients classified as low-risk by both methods were ultimately classified as low-risk, and other patients were ultimately classified as intermediate-risk. Statistical analysis was performed using IBM SPSS 25.0 and R 4.2.3 software. All statistical analyses used two-tailed tests, and p <0.05 was considered statistically significant.

[0170] This embodiment uses the same research subjects as described in Example 2. Among them, 164 HCM patients underwent WES testing, and 155 patients completed follow-up, with a loss to follow-up rate of 5.5%. The follow-up time was 4.93 (4.67–6.47) months, with a potential median follow-up time and 95% CI of 4.93 (4.84–5.02) days. A total of 12 patients experienced an overall composite outcome, 9 experienced a SCD composite outcome, 5 experienced a HF composite outcome, 2 experienced cardiac death, 2 experienced fatal ventricular arrhythmias, and 3 experienced uncontrollable New York Heart Association Class III / IV symptoms.

[0171] A total of 139,449 single nucleotide polymorphisms (SNPs) and 5,543 small fragment insertions and deletions (Indels) were obtained from 164 HCM patients examined by WES. Among the nonsynonymous variants, SNPs were the most prevalent, while missense variants were the most prevalent in terms of functional type.

[0172] like Figure 12 As shown, among the 20 pathogenic genes for hypertrophic cardiomyopathy summarized in the 2023 Chinese Guidelines for the Diagnosis and Treatment of Hypertrophic Cardiomyopathy in Adults, non-synonymous variants of 14 genes were detected in HCM patients in this embodiment. The frequencies from high to low are as follows: titin (TTN) gene, ALPK3 gene, calcium voltage-gated channel alpha subunit 1C (CACNA1C), JPH2 gene, FLNC gene, MYBPC3 gene, LIM domain binding 3 (LDB3) gene, MYH7 gene, ACTN2 gene, TNNI3 gene, TNNT2 gene, CSRP3 gene, MYL2 gene, and TPM1 gene.

[0173] Univariate Cox proportional hazards regression analysis indicated the presence of nonsynonymous variants in the following genes: von Willebrand factor C and EGF domains (VWCE), tissue factor pathway inhibitor (TFPI), mecaptopyruvate sulfurtransferase (MPST), melanoma cell adhesion molecule (MCAM), paired-like homeodomain 3 (PITX3), p21-activated kinase 2 (PAK2), sperm-associated antigen 5 (SPAG5), and platelet-derived growth factor receptor β (PDG5). Patients with nonsynonymous variants of the factor receptor beta (PDGFRB) gene had a higher risk of overall adverse outcomes (adjusted P < 0.05 for all) (see [link to relevant documentation]). Figure 13 ).

[0174] Figure 14 The results showed that, in multivariate Cox proportional hazards regression analysis of the eight gene nonsynonymous variants included in the univariate Cox proportional hazards regression analysis for the overall composite outcome of HCM patients, the results suggested that carrying the VWCE gene nonsynonymous variant [hazard ratio ( hazard ratio ) ] was associated with a risk ratio of 0.05. The following gene variants were independently associated with an increased risk of overall composite outcome in HCM patients: SPAG5 gene nonsynonymous variants [HR and 95% CI: 10.529 (2.002-55.379), P = 5.45E-03], SPAG5 gene nonsynonymous variants [HR and 95% CI: 32.245 (6.747-154.102), P = 1.35E-05], PDGFRB gene nonsynonymous variants [HR and 95% CI: 24.394 (5.861-101.524), P = 1.13E-05], and PITX3 gene nonsynonymous variants [HR and 95% CI: 203.394 (25.651-1612.79), P = 4.87E-07].

[0175] Figure 15 The results showed that compared with patients who did not carry nonsynonymous variants of VWCE, SPAG5, PDGFRB, and PITX3, patients carrying any one of these four genes had a nearly 15-fold increased risk of overall composite outcome [HR and 95% CI: 14.960 (6.137-36.468), P = 4.16E-10]. Patients carrying at least one nonsynonymous variant of VWCE, SPAG5, PDGFRB, or PITX3 had a 73-fold increased risk of overall composite outcome compared with wild-type HCM patients [HR and 95% CI: 73.335 (9.378-573.457), P = 3.16E-05]. The C-index and 95% CI for predicting overall composite outcome were 0.897 (0.812-0.982), and the survival curves showed good discrimination (logrank test P < 0.0001).

[0176] In this embodiment, based on the score 2 established in Example 5 using serum PDCD5 protein and clinical factors, a new multidimensional risk stratification model was established using four gene nonsynonymous variants independently associated with the overall composite outcome of HCM patients: patients with a score 2 > 2 and carrying at least one gene nonsynonymous variant of VWCE, SPAG5, PDGFRB, or PITX3 were stratified as high-risk; patients with a score 2 ≤ 2 and not carrying any gene nonsynonymous variant of VWCE, SPAG5, PDGFRB, or PITX3 were stratified as low-risk; and the remaining patients (patients with a score 2 > 2 and not carrying any gene nonsynonymous variant of VWCE, SPAG5, PDGFRB, or PITX3; and patients with a score 2 ≤ 2 and carrying at least one gene nonsynonymous variant of VWCE, SPAG5, PDGFRB, or PITX3) were stratified as intermediate-risk.

[0177] Example 8

[0178] The research object used in this embodiment is the same as that described in Embodiment 2.

[0179] The method described in Example 7 was used to assess the risk stratification of HCM patients as described in Example 2. The Kaplan-Meier curve was used to evaluate the discriminative power of the multidimensional risk stratification model. The C-index and 95% confidence interval (CI) of the model were calculated to assess its predictive efficacy. The C-index and 95% CI were compared with those obtained using the optimal cutoff value 2 of score 2 in Example 5 to evaluate whether the final multidimensional risk stratification model was superior to the risk stratification model established by comprehensively considering serum PDCD5 protein and clinical factors. Statistical analysis was performed using IBM SPSS 25.0 and R 4.2.3 software. All statistical analyses used two-tailed tests, and p < 0.05 was considered statistically significant.

[0180] Figure 16 The results showed that the multidimensional risk stratification model of Example 6 was superior to the C-index and 95% CI of the score 2>2 used in Example 5 to predict the overall composite outcome of HCM patients [C-index and 95% CI: 0.774 (0.685-0.862), P<0.001].

[0181] Figure 17The results showed that the multidimensional risk stratification model established in Example 6 could predict the overall composite outcome of HCM patients, with a HR and 95% CI of 33.740 (7.677-148.282) (P = 1.24E-06). Based on the established multidimensional risk stratification model, HCM patients were stratified into high-risk, intermediate-risk, and low-risk groups. The survival curves of the three groups showed good differentiation (overall log rank test P < 0.001; intermediate-risk vs. low-risk group, log rank test P = 0.037; high-risk vs. intermediate-risk group, log rank test P < 0.001; high-risk vs. low-risk group, log rank test P < 0.001).

[0182] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. Use of a reagent for detecting the expression level of PDCD5 protein, characterized in that, The application comprises at least one of the following: (1) use of a reagent for detecting PDCD5 protein expression level in the preparation of a product for evaluating the risk of suffering from hypertrophic cardiomyopathy; (2) use of a reagent for detecting PDCD5 protein expression level in the preparation of a product for evaluating the risk degree of a patient suffering from hypertrophic cardiomyopathy.

2. Use according to claim 1, characterized in that, For the application of (1), if the PDCD5 protein expression in the serum sample to be detected is significantly higher than that in the negative control serum sample, P < 0.05, and the PDCD5 protein in the serum sample to be detected is > 0.95 ng / mL, it indicates that the subject to be detected has a high risk of suffering from hypertrophic cardiomyopathy; For the application of (2), if the PDCD5 protein expression in the serum sample of the patient to be detected is significantly higher than that in the control serum sample group, P < 0.05, and the PDCD5 protein in the serum sample to be detected is > 4.09 ng / mL, it indicates that the patient has a high risk degree of suffering from hypertrophic cardiomyopathy.

3. Application of PDCD5 protein expression level combined with clinical factors in construction of a risk stratification model of hypertrophic cardiomyopathy, characterized in that, The clinical factors are syncope / syncope precursor history, NYHA classification, and left ventricular enlargement indicated by echocardiography.

4. Use of PDCD5 protein expression level, clinical factors and gene non-synonymous variation in constructing a risk stratification model of hypertrophic cardiomyopathy, characterized in that, The clinical factors are syncope / syncope precursor history, NYHA classification, and left ventricular enlargement indicated by echocardiography; and the non-synonymous variations are non-synonymous variations in the von Willebrand factor C and EGF domain gene, non-synonymous variations in the sperm-associated antigen 5 gene, non-synonymous variations in the platelet-derived growth factor receptor beta gene, or non-synonymous variations in the paired homeodomain protein transcription factor 3 gene.

5. A method for constructing a risk stratification model for hypertrophic cardiomyopathy, characterized by, The application comprises the following steps: The candidate variables are included in the multivariate Cox regression analysis and the stepwise regression method is used, then the variables with p < 0.05 in the multivariate Cox regression analysis are used to establish a risk stratification model, the regression coefficients are used to weight and score each variable, the scores of each variable are added to obtain a prediction score for adverse outcomes, the survival analysis and its visualization toolkit in R 4.2.3 are used to calculate the best critical value of the prediction score for risk stratification, and the hypertrophic cardiomyopathy patients are stratified according to the best critical value, thereby obtaining a risk stratification model for hypertrophic cardiomyopathy; The candidate variables include clinical factors and serum PDCD5 protein expression level; the clinical factors are age, gender, whether obstructive HCM, syncope / syncope precursor history, NYHA classification, body mass index, hypertension history, myoglobin, atrial flutter / fibrillation, non-sustained ventricular tachycardia, left ventricular enlargement indicated by echocardiography, left ventricular ejection fraction indicated by echocardiography, and abnormal wall motion indicated by echocardiography.

6. The method of building a risk stratification model according to claim 5, characterized in that, The variables with p < 0.05 are syncope / syncope precursor history, NYHA classification, left ventricular enlargement indicated by echocardiography, and serum PDCD5 protein expression level. The calculation formula of the risk stratification model of the hypertrophic cardiomyopathy is score 2 = syncope / syncope precursor history + NYHA classification + left ventricular enlargement + high expression of serum PDCD5 protein, wherein the presence of syncope / syncope precursor history is valued at 1 point, NYHA classification of grade I is valued at 1 point, NYHA classification of grade II is valued at 2 points, NYHA classification of grade III is valued at 3 points, NYHA classification of grade IV is valued at 4 points, ultrasonic cardiogram indicating left ventricular enlargement is valued at 1 point, and serum PDCD5 protein > 4.09 ng / mL is valued at 1 point; When the hypertrophic cardiomyopathy patients are stratified, the hypertrophic cardiomyopathy patients are divided into a high-risk group and a low-risk group, the high-risk group is predicted score > optimal critical value, and the low-risk group is predicted score ≤ optimal critical value.

7. The construction method of claim 5, wherein, The candidate variables further include non-synonymous variation of von Willebrand factor C and EGF domain gene, non-synonymous variation of tissue factor pathway inhibitor gene, non-synonymous variation of mercaptopyruvate sulfurtransferase gene, non-synonymous variation of melanoma cell adhesion molecule gene, non-synonymous variation of paired-like homeodomain transcription factor 3 gene, non-synonymous variation of p21-activated kinase 2 gene, non-synonymous variation of sperm associated antigen 5 gene, and non-synonymous variation of platelet-derived growth factor receptor beta gene.

8. The construction method of claim 5, wherein, The variables with p < 0.05 are syncope / syncope precursor history, NYHA classification, left ventricular enlargement indicated by ultrasonic cardiogram, serum PDCD5 protein expression level, non-synonymous variation of von Willebrand factor C and EGF domain gene, non-synonymous variation of sperm associated antigen 5 gene, non-synonymous variation of platelet-derived growth factor receptor beta gene, and non-synonymous variation of paired-like homeodomain transcription factor 3 gene; When the hypertrophic cardiomyopathy patients are stratified, the hypertrophic cardiomyopathy patients are divided into a high-risk group, a low-risk group and a medium-risk group; The patients with score 2 > optimal critical value and at least one of non-synonymous variation of von Willebrand factor C and EGF domain gene, non-synonymous variation of sperm associated antigen 5 gene, non-synonymous variation of platelet-derived growth factor receptor beta gene and non-synonymous variation of paired-like homeodomain transcription factor 3 gene are stratified into the high-risk group; The patients with score 2 ≤ optimal critical value and without non-synonymous variation of von Willebrand factor C and EGF domain gene, non-synonymous variation of sperm associated antigen 5 gene, non-synonymous variation of platelet-derived growth factor receptor beta gene or non-synonymous variation of paired-like homeodomain transcription factor 3 gene are stratified into the low-risk group; The medium-risk group is any one of the following: (S1) the patients with score 2 > optimal critical value and without non-synonymous variation of von Willebrand factor C and EGF domain gene, non-synonymous variation of sperm associated antigen 5 gene, non-synonymous variation of platelet-derived growth factor receptor beta gene or non-synonymous variation of paired-like homeodomain transcription factor 3 gene are stratified into the medium-risk group; (S2) Patients carrying at least one of Factor C and EGF domains gene non-synonymous variation, Sperm associated antigen 5 gene non-synonymous variation, Platelet-derived growth factor receptor beta gene non-synonymous variation and Paired Hox-like homeodomain transcription factor 3 gene non-synonymous variation are stratified into the intermediate risk group when the score is 2.

9. The construction method according to any one of claims 5 to 8, characterized in that, The optimal critical value is 2.

10. The risk stratification model of hypertrophic cardiomyopathy prepared by the construction method according to any one of claims 5-9.

Images