Aortic dissection related biomarker, kit, system and application thereof

Abstract

The invention discloses a biomarker related to aortic dissection, a kit, a system and application thereof, and relates to the technical field of biological medicine. The biological marker related to the aortic dissection comprises a second biological marker; the second biomarker is used for predicting the clinical death risk of aortic dissection patients, and the second biomarker comprises SYTL1 protein, B2M protein, LYVE1 protein, TTR protein, PRSS2 protein, CSTF1 protein and H2BC5 protein. The plasma-based protein biomarker capable of predicting the mortality risk of the aortic dissection patient is successfully screened out, the pathogenesis of the disease can be explored easily, and the method has important significance on efficient, accurate and low-cost diagnosis of the aortic dissection patient.

Description

Technical Field

[0001] This case is a divisional application of patent application number 202411816983.6, filed on December 11, 2024, entitled "Biomarkers, Reagent Kits, Systems and Applications Related to Aortic Dissection". Specifically, this invention relates to the field of biomedical technology and pertains to a biomarker, reagent kit, system and application related to aortic dissection. Background Technology

[0002] Acute aortic dissection is a dangerous cardiovascular disease caused by damage to the innermost lining of the aorta, resulting in a catastrophic separation of the aortic wall due to blood flowing through it. Treatment for aortic dissection is relatively well-established. Stanford type A aortic dissection is primarily treated surgically, while type B is mainly treated endovascularly. Although treatment guidelines are well-established, the mortality rate for untreated aortic dissection is very high, increasing by 1-3% per hour, reaching as high as 21% within 24 hours and as high as 74% within one week. For Stanford type A aortic dissection, the mortality rate after surgery reaches 25% due to various complications, while Stanford type B also carries the risk of death due to postoperative retrograde rupture or distal dissection. Early identification of aortic dissection and high-risk patients, minimizing the time from symptom onset to appropriate treatment, is crucial for improving patient survival rates.

[0003] The gold standard for diagnosing aortic dissection is computed tomography angiography (CTA). Combining imaging techniques with highly sensitive and specific biomarkers to rapidly diagnose aortic dissection before patients reach hospitals with surgical capabilities will significantly improve survival rates. Furthermore, pre-hospital prognostic assessment, early identification of patients with serious consequences, and attention during diagnosis and treatment will also improve patient outcomes. However, most primary care hospitals lack the equipment to perform CTA, magnetic resonance angiography (MRA), and transthoracic ultrasound, increasing the difficulty and complexity of diagnosing aortic dissection.

[0004] Several potential biomarkers exist in existing research on acute aortic dissection, including CTRP-9, C-reactive protein, and D-dimer. For example, Chinese invention patent application CN201910839515.3 screens for acute aortic dissection using plasma CTRP-9 and D-dimer concentrations. However, the reproducibility, specificity, and sensitivity of these potential biomarkers in expanded populations remain unclear, and no specific biomarker is found that is associated with the prognosis of patients with aortic dissection.

[0005] Therefore, there is an urgent need to identify reliable early biomarkers for aortic dissection to aid in the diagnosis of whether patients with chest pain have aortic dissection, as well as predictive techniques for the clinical prognosis of confirmed patients, in order to guide the clinical diagnosis and treatment process. Summary of the Invention

[0006] To overcome the shortcomings of the above-mentioned technologies, the purpose of this invention is to provide a biomarker, reagent kit, system and its application related to aortic dissection, to discover biomarkers related to aortic dissection and to construct models and systems for the diagnosis of aortic dissection, so as to achieve accurate and rapid diagnosis of aortic dissection.

[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A biomarker associated with aortic dissection, the biomarker comprising a first biomarker and / or a second biomarker; the first biomarker for diagnosing aortic dissection, the first biomarker comprising RCN1, SERPINA3, and CPN1; the second biomarker for predicting clinical mortality risk in patients with aortic dissection, the second biomarker comprising SYTL1, B2M, LYVE1, TTR, PRSS2, CSTF1, and H2BC5.

[0008] Information on the above biomarkers can be found in the Uniprot database (https: / / www.uniprot.org).

[0009] The present invention also provides the use of a first biomarker or a detection reagent thereof in the preparation of a product for diagnosing aortic dissection, wherein the first biomarker includes RCN1, SERPINA3 and CPN1.

[0010] The present invention also provides a kit for diagnosing aortic dissection, the kit comprising reagents for detecting the expression level of a first biomarker in a sample; the first biomarker includes RCN1, SERPINA3 and CPN1; the sample is blood or plasma of a subject.

[0011] The present invention also provides the use of a second biomarker or its detection reagent in the preparation of a product for predicting the clinical mortality risk of patients with aortic dissection, the second biomarker including SYTL1, B2M, LYVE1, TTR, PRSS2, CSTF1 and H2BC5.

[0012] The present invention also provides a kit for predicting the clinical mortality risk of patients with aortic dissection, the kit comprising reagents for detecting the expression level of a second biomarker in a sample; the second biomarker includes SYTL1, B2M, LYVE1, TTR, PRSS2, CSTF1 and H2BC5; the sample is the blood or plasma of the subject.

[0013] This invention also provides a method for constructing a diagnostic model for aortic dissection, comprising the following steps: Plasma protein data from healthy individuals were used to construct a healthy sample group, and plasma protein data from patients with aortic dissection were used to construct a disease sample group. The disease sample group included plasma protein data from patients with aortic dissection who died clinically and data from a sample group with good prognosis. Each sample in the healthy sample group and the disease sample group included the expression level of each protein feature in the plasma protein and the annotation information of the corresponding auxiliary examinations for whether the subject was diagnosed. The healthy sample group and the disease sample group are randomly divided into a training set and a test set; Differential protein analysis was performed on the training set, and differential proteins were screened using t-tests and P-values. Proteins whose expression levels differed between healthy individuals and patients with aortic dissection according to P < 0.05 were identified as the optimal combination of biomarkers for the diagnosis of aortic dissection. The optimal combination of biomarkers for the diagnosis of aortic dissection included RCN1, SERPINA3, and CPN1. Using the expression level of the optimal biomarker for the diagnosis of aortic dissection in each sample as feature information, and utilizing training and test set data, an aortic dissection diagnostic model is constructed based on machine learning algorithms. The aortic dissection diagnostic model is used to obtain risk parameters for aortic dissection in subjects based on the expression levels of the optimal combination of biomarkers for aortic dissection diagnosis.

[0014] This invention also provides a diagnostic model for aortic dissection, constructed using the method described above. By inputting the optimal biomarker combination into the aortic dissection diagnostic model and performing analysis, risk parameters for aortic dissection in the subject are obtained.

[0015] This invention also provides a method for constructing a clinical mortality risk prediction model for patients with aortic dissection, comprising the following steps: Plasma proteomic data of aortic dissection patients with good clinical outcomes were obtained as the good prognostic sample group, and plasma proteomic data of aortic dissection patients with clinical outcomes of death were obtained as the poor prognostic sample group. Each sample includes the corresponding patient's clinical characteristics, the expression level of each protein feature in the plasma proteomic sample, and the annotation information of whether the corresponding patient died. The good prognosis sample group and the poor prognosis sample group are randomly divided into a training set and a test set; Differential protein analysis was performed on the training set, and differentially expressed proteins were screened using t-tests and p-values. Proteins with expression level differences in the plasma proteome that met the p-value < 0.05 were selected as the optimal biomarker combination for predicting clinical mortality risk in patients with aortic dissection. The optimal biomarker combination for predicting clinical mortality risk in patients with aortic dissection included SYTL1, B2M, LYVE1, TTR, PRSS2, CSTF1, and H2BC5. Using the expression levels and clinical characteristics of the optimal biomarkers for predicting the clinical mortality risk of aortic dissection patients in each sample as feature information, a clinical mortality risk prediction model for aortic dissection patients is constructed using machine learning algorithms based on the training set. The optimal combination of clinical features is determined based on the feature importance of each clinical feature in the predictive performance of the clinical mortality risk prediction model for aortic dissection patients. The predictive performance of the aortic dissection patient clinical mortality risk prediction model was evaluated using information on the expression levels of the optimal biomarkers and the optimal clinical characteristics of each sample in the test set data. The clinical mortality risk prediction model for patients with aortic dissection is used to obtain risk parameters for clinical mortality of patients with aortic dissection based on the expression levels of the optimal combination of biomarkers for predicting clinical mortality risk of patients with aortic dissection and the optimal combination of clinical features.

[0016] Furthermore, the clinical characteristics information includes gender, age, CTA, BMI, BaPWV, and clinical outcome.

[0017] As a preferred option, the optimal combination of clinical characteristics includes gender, age, CTA, BMI, and BaPWV. Combining the optimal combination of biomarkers for predicting clinical mortality risk in aortic dissection patients with the optimal combination of clinical characteristics yields more accurate results than using the optimal combination of biomarkers for predicting clinical mortality risk in aortic dissection patients alone.

[0018] Furthermore, the machine learning algorithm is selected from any of the following algorithms: logistic regression, linear regression, random forest, neural network, support vector machine, Bayesian classification, gradient boosting, K-nearest neighbors, and decision tree.

[0019] This invention also provides a clinical mortality risk prediction model for patients with aortic dissection, constructed using the method described above. Clinical mortality risk parameters for patients with aortic dissection are obtained by inputting the optimal biomarker for predicting clinical mortality risk of patients with aortic dissection into the model and performing analysis.

[0020] The present invention also provides a system for diagnosing aortic dissection or predicting the clinical mortality risk of patients with aortic dissection, including a processor and a display; The processor is configured to predict the risk of a subject having aortic dissection or the clinical mortality risk of a clinical aortic dissection patient based on the expression levels of the biomarkers associated with aortic dissection using an aortic dissection diagnostic model or an aortic dissection patient clinical mortality risk prediction model; the display is used to present the risk parameters of aortic dissection or clinical mortality risk parameters predicted by the processor for the subject. The biomarkers associated with aortic dissection are those mentioned above.

[0021] Furthermore, the display also presents treatment recommendations corresponding to the risk parameters.

[0022] Furthermore, when the system uses the aortic dissection diagnostic model to predict the risk of a subject having aortic dissection, the prediction is based on the expression level of the first biomarker.

[0023] Furthermore, when the system uses the clinical mortality risk prediction model for patients with aortic dissection to predict the clinical mortality risk of patients with aortic dissection, it does so based on the expression level of a second biomarker and clinical characteristic information; the clinical characteristic information includes computed tomography angiography (CTA), body mass index (BMI), brachial-ankle pulse wave velocity (BaPWV), gender, and age.

[0024] Furthermore, the system includes a liquid chromatography device and a mass spectrometry analysis device, wherein the liquid chromatography device is used for protein extraction from the collected plasma samples of the subjects, and the mass spectrometry analysis device is used for quantitative analysis of the extracted proteins.

[0025] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention provides biomarkers, reagent kits, systems, and applications related to aortic dissection. It mines biomarkers related to aortic dissection and constructs models and systems for the diagnosis of aortic dissection. By screening 10 protein biomarkers, it achieves accurate and rapid diagnosis of aortic dissection and accurate prediction of the clinical mortality risk of patients with aortic dissection. It has the advantages of low cost, simple method, high accuracy, and high efficiency.

[0026] This invention establishes a correlation between 10 protein features and the diagnosis and clinical outcome prediction of subjects with aortic dissection, along with a corresponding predictive model. Obtaining these protein and clinical features is minimally invasive for the subjects. The protein features are obtained using plasma samples (usually collected and prepared during other aortic dissection procedures) through proteomics and mass spectrometry analysis. The clinical features should also have been obtained during the subjects' standardized treatment and follow-up. Therefore, the features used in this invention have excellent compatibility with existing aortic dissection-related medical procedures. Furthermore, the negative predictive value of both the aortic dissection diagnostic model and the aortic dissection patient clinical mortality risk prediction model of this invention are close to 95%, and the area under the AUC curve is close to or exceeds 0.9. They also exhibit high specificity, sensitivity, and accuracy, enabling efficient and accurate screening of aortic dissection patients and those with a high clinical mortality risk. In particular, it can accurately exclude non-dissection patients, reducing unnecessary treatment and waste of medical resources, and alleviating the psychological stress of these patients. In addition, this invention can enable patients with aortic dissection to predict clinical outcomes as early as possible with high efficiency and accuracy without the need for additional invasive examinations. It allows for more rigorous, regular, and high-frequency monitoring of various indicators during treatment, preventing serious cardiovascular events or even death, and guiding clinical decision-making to form individualized and precise treatment plans, thereby improving prognosis.

[0027] This invention uses only blood biomarkers, or blood biomarkers combined with clinical indicators, to screen for aortic dissection patients and those with poor clinical outcomes in a relatively non-invasive manner. This allows for more precise treatment policies, including the accurate exclusion of healthy individuals or patients with good prognoses from the disease group. For patients with aortic dissection at high risk of postoperative death or other adverse outcomes, less invasive surgical methods can be carefully selected during clinical decision-making. More frequent and regular monitoring of the predictive indicators and other clinical follow-up items provided by this invention can also be conducted to avoid these patients suffering unnecessary cardiovascular events or even death.

[0028] This invention successfully screened plasma-based protein biomarkers that can accurately screen patients with aortic dissection and predict the risk of death in patients with aortic dissection. This will also help to explore the pathogenesis of this disease and is of great significance for the efficient, accurate and low-cost diagnosis of patients with aortic dissection. Attached Figure Description

[0029] Figure 1 This is a schematic diagram of some components of the aortic dissection diagnostic system and the aortic dissection patient clinical mortality risk prediction system according to an embodiment of the present invention; Figure 2The diagram shows the importance of the three protein features selected as the optimal biomarker combination for the diagnosis of aortic dissection (part A in the figure) and the performance results of the aortic dissection diagnosis model on the test set (parts B and C in the figure). Figure 3 The diagram illustrates the importance of the seven protein features selected by the embodiments of the present invention for predicting the clinical mortality risk of patients with aortic dissection (part D in the diagram) and the performance results of the aortic dissection patient clinical mortality risk prediction model on the test set (parts E and F in the diagram). Detailed Implementation

[0030] To better explain the present invention, the main contents of the present invention are further illustrated below with reference to specific embodiments, but the contents of the present invention are not limited to the following embodiments.

[0031] Example 1: Biomarkers associated with aortic dissection and their screening process Step 1: Experimental batch design and quality control sample preparation Experimental designs for clinical cohort samples comprehensively consider the diverse clinical characteristics of the population, employing any applicable proteomics big data design and quality control tools to ensure the samples are distributed as uniformly as possible, minimizing batch effects. Furthermore, QC (quality control) samples—equal-volume pools of experimental samples—can be prepared concurrently with sample processing to evaluate system stability throughout the experiment.

[0032] Step 2: Protein extraction and peptide enzymatic hydrolysis The plasma sample to be tested was thawed at 4°C. An appropriate amount of sample was added to a new 1.5 mL EP tube, and 50 mM ammonium bicarbonate solution was added to bring the total volume to 100 μL. The tube was heated at 95°C for 3 min to denature proteins, then cooled to room temperature and enzymatically digested with trypsin at 37°C for 16 h. The digestion product was extracted and lyophilized. Desalting was then performed. After lyophilization, the peptides were reconstituted in 100 μL of 0.1% formic acid solution and injected at a 2% concentration for analysis.

[0033] Samples were separated using an EASY-nLC 1200 high-performance liquid chromatography system with a flow rate of nanoliters. Mobile phase A was a 0.1% formic acid aqueous solution, and mobile phase B was a 0.1% formic acid-acetonitrile aqueous solution (acetonitrile to formic acid volume ratio of 4:1). The loading and analytical columns were first equilibrated with 100% mobile phase A. Then, the enzymatically digested peptides of the sample were delivered to the loading column (2 cm, ID 100 μm, 3 μm, C18) via an autosampler, followed by separation on the analytical column (15 cm, ID 150 μm, 1.9 μm, C18) at a flow rate of 600 nL / min.

[0034] Step 3: Mass Spectrometry Data Acquisition and Analysis 1. Mass Spectrometry Analysis: After separation by liquid chromatography in step two, the sample was analyzed by mass spectrometry using an HFX mass spectrometer. The detection mode was positive ion, with a precursor ion scan range of 300–1400 m / z. The primary mass spectrometry resolution was 60,000 at 200 m / z, the AGC (Automatic Gain Control) target was 3e6, and the maximum IT was 20 ms. The mass-charge ratio of the peptide and peptide fragments was acquired using the following method: 30 DIA scans were acquired after each full scan, using HCD fragmentation mode, with a Normalized Collision Energy of 27%, and the isolation window was adjusted according to the isolation window. The secondary mass spectrometry resolution was 15,000 at 200 m / z.

[0035] 2. Quality control analysis: The correlation data of plasma proteome QC samples showed that the median correlation of plasma proteome QC samples was 98%, indicating that the experimental data had high consistency and reproducibility.

[0036] 3. Protein differential analysis: With a p-value of less than 0.05, differentially expressed proteins were screened among healthy subjects and patients with aortic dissection with good and poor prognoses. The screened proteins were then subjected to further targeted validation.

[0037] Step 4: Targeted Proteomics Analysis First, the quantitative information of the target protein set is normalized (z-score). Then, the pheatmap R package is used to classify both the sample and protein expression levels (distance algorithm: Euclidean, connection method: Average linkage).

[0038] The raw data from mass spectrometry analysis were RAW files. Qualitative and quantitative analysis was performed using the iProteome one-stop data analysis cloud platform. The search parameters were set as follows: Enzyme: Trypsin; Fixed modifications: Urea methylation (C); Variable modifications: Methionine oxidation, Acetyl (protein N-terminus); Missed cleavages: 2 Peptide; Mass Tolerance: 20 ppm; Mass Tolerance: 0.05 Da.

[0039] Furthermore, the use of a high-precision, high-resolution liquid chromatography-mass spectrometry (LC-MS / MS) instrument ensured good mass deviation during data acquisition, ultimately yielding high-quality MS1 and MS2 spectra. The mass deviations of all identified peptides were primarily within 10 ppm, indicating accurate and reliable identification results. The spectral data were then analyzed using the Mascot search engine to obtain a score for each MS2 spectrum. The distribution of scores for excellent peptides further demonstrates that the mass spectrometry instrument can produce high-quality experimental data. Additionally, in the qualitative analysis of DIA data, a peptide FDR ≤ 0.05 was used as the screening criterion.

[0040] DIA (data-independent acquisition) refers to a data-independent scanning mode, which is a holographic mass spectrometry data acquisition mode based on the electrostatic field orbital trap (Orbitrap). After primary mass spectrometry detection, DIA fragments the precursor ions within a specific mass-to-charge ratio range, collects the corresponding fragment ions, and rapidly scans all fragment ions within adjacent precursor ion windows sequentially, thereby enabling qualitative and quantitative analysis of proteins.

[0041] Based on steps one through four above, the expression levels of various proteins that serve as biomarkers can be detected from the plasma samples of the subjects, and differential proteomes between healthy individuals and patients with aortic dissection can be screened, as well as differential proteomes between patients with good and poor prognosis in aortic dissection. The expression levels of each protein characteristic in the differential proteome can also be obtained.

[0042] The differentially expressed proteome between healthy individuals and patients with aortic dissection served as the first biomarker, including RCN1, SERPINA3, and CPN1; the differentially expressed proteome between patients with good and poor prognoses served as the second biomarker, including SYTL1, B2M, LYVE1, TTR, PRSS2, CSTF1, and H2BC5.

[0043] All ten biomarkers mentioned above are proteins, and the main functions of each protein are currently considered to be as follows in clinical practice.

[0044] RCN1: Human reticulocyte cadherin 1, as a potential regulator, plays a role in regulating calcium-dependent activity in the endoplasmic reticulum (ER) lumen or post-ER compartment. The RCN1 protein, Human (P. pastoris, His), is a recombinant RCN1 protein expressed by P. pastoris and tagged with N-His. The specific mechanisms by which it exerts its regulatory influence on calcium-dependent processes remain to be fully elucidated, prompting further investigation into its functional significance in these cellular compartments.

[0045] SERPINA3: α-1 antichymotrypsin, a cytochrome P450 2E1 (CYP2E1) enzyme, can be induced by compounds such as ethanol and participates in the metabolic activation of various procarcinogens (N-nitrosamines, aniline, vinyl chloride, urethane), inducing tumorigenesis. Receptor tyrosine kinases participate in the development and maturation of the central and peripheral nervous systems by regulating the proliferation, differentiation, and survival of sympathetic nerves and nerve cells.

[0046] CPN1: Copine protein 1, a calcium-dependent membrane-bound protein, is a calcium-dependent phospholipid-binding protein that actively participates in calcium-mediated intracellular processes. Copine-1 participates in the TNF-α receptor signaling pathway, exhibiting calcium-dependent phospholipid-binding properties. Furthermore, it acts in a calcium-independent manner through an AKT-dependent signaling cascade, inducing neurite growth and thus playing a crucial role in neural progenitor cell differentiation. Copine-1 may also promote the recruitment of target proteins to the cell membrane in a calcium-dependent manner and participate in membrane transport.

[0047] SYTL1: Synaptic binding protein 1 (SYT1) is an intact membrane protein of synaptic vesicles, believed to act as a Ca(2+) sensor during vesicle transport and exocytosis. Its binding to synaptotagmin-1 with calcium participates in triggering neurotransmitter release at the synapse. Its role in the nervous system is crucial, especially during the docking and fusion of synaptic vesicles with the presynaptic membrane. This role is essential for maintaining mature neurons and neurotransmitter release.

[0048] B2M: β2-microglobulin (β2-MG) is a light chain of human leukocyte antigen (HLA) class I antigen. It is a low molecular weight serum globulin produced by lymphocytes, platelets, and polymorphonuclear leukocytes, and is present on the surface of various cells outside mature erythrocytes and placental trophoblast cells. β2-MG is widely found in plasma, urine, cerebrospinal fluid, saliva, and colostrum. Clinically, blood and urine β2-MG measurements can be used to evaluate renal tubular function.

[0049] LYVE1: Hyaluronic acid receptor 1 on lymphatic endothelial cells. Lymphatic endothelial receptor-1 is a close relative of the leukocyte receptor CD44 and is the main receptor for hyaluronic acid (HA) on lymphatic endothelial cells. It is a commonly used marker for distinguishing blood vessels from lymphatic vessels. It is a lymphoapeptide receptor on dendritic cells, selectively binding to their surface HA glycocalyx, regulating the entry of peripheral lymphatic vessels and migration to downstream lymph nodes, thereby activating the immune system.

[0050] TTR: Transthyretin, is a homotetrameric protein primarily synthesized by the liver and choroid plexus. Its function is to transport thyroxine and retinol-bound proteins in plasma and cerebrospinal fluid. It is present in plasma, serum, and cerebrospinal fluid, and is mainly synthesized by the liver and choroid plexus.

[0051] PRSS2: Serine protease 2, is a serine protease belonging to the protease family. This enzyme plays a crucial role in mammals, particularly in digestion, coagulation, and the complement system. Activation of serine proteases is achieved through changes in a set of amino acid residues at their active site, one of which must be serine, hence its name.

[0052] CSTF1: Cleavage-stimulating factor subunit 1. This gene encodes one of three subunits that bind to form cleavage-stimulating factor (CSTF). CSTF is involved in the polyadenylation and 3' end cleavage of precursor mRNA. Similar to the β subunit of mammalian G proteins, this protein contains transduction protein-like repetitive sequences. Furthermore, transcript variants of this gene with different 5' UTRs but encoding the same protein have been found. The CSTF1 gene is widely expressed in various tissues, including the testes and lymph nodes, and is also expressed in 25 other tissues.

[0053] H2BC5: H2B aggregate histone 5. Histones are basic nucleoproteins that make up the structure of nucleosomes on eukaryotic chromosomes. A nucleosome consists of approximately 146 bp of DNA wrapped around a histone octamer, which is composed of each of the four core histones (H2A, H2B, H3, and H4). This gene has no introns and encodes a replication-dependent histone, a member of the histone H2B family.

[0054] Example 2: Diagnostic model for aortic dissection and predictive model for clinical mortality risk in patients with aortic dissection, and their construction methods I. Diagnostic Model for Aortic Dissection The aortic dissection diagnostic model of this embodiment uses the expression level of the first biomarker in Example 1 as its input parameter and the risk parameter of the subject's aortic dissection as its output parameter. The construction method of this model includes the following steps: 1) Obtain plasma protein data from a group of healthy individuals as a healthy sample group, and obtain plasma protein data from a group of patients with aortic dissection as a disease sample group, including sample groups with good and poor prognoses in clinical outcomes. Each sample in the healthy sample group and the disease sample group includes the expression level of each protein feature in the plasma protein and the corresponding annotation information of the auxiliary examination for patient diagnosis.

[0055] 2) The healthy group and the aortic dissection (good prognosis + poor prognosis) patient sample groups were divided into training set and test set.

[0056] 3) Use t-test and P-value to screen features and determine the protein features in the plasma proteome that meet the expression level difference requirement P<0.05 as the optimal biomarker combination for predicting whether the subject has aortic dissection; this step is the screening process in Example 1, and the optimal biomarker combination for predicting the risk of aortic dissection in the subject is the first biomarker in Example 1.

[0057] 4) Based on the expression levels of the optimal biomarker combination for each sample in the training set, a random forest algorithm is used to construct and train an application prediction model. The application prediction model is then tested on the test set to evaluate its prediction performance. The application prediction model whose prediction performance exceeds the performance index threshold is used as the aortic dissection diagnostic model to provide a prediction result on the risk of aortic dissection based on the expression levels of the optimal biomarker combination in the subject's plasma.

[0058] II. Clinical Mortality Risk Prediction Model for Patients with Aortic Dissection The clinical mortality risk prediction model for patients with aortic dissection in this embodiment uses the expression level of the second biomarker in Example 1 as the input parameter and the clinical mortality risk parameter for patients with aortic dissection as the output parameter. The construction method includes the following steps: 1) Obtain plasma proteomic data from a group of aortic dissection patients with good clinical outcomes as the good prognosis sample group, and obtain plasma proteomic data from a group of aortic dissection patients with poor clinical outcomes (death) as the poor prognosis sample group. Each sample in the good prognosis sample group and the poor prognosis sample group includes the expression level of each protein feature in the plasma proteomic data and the corresponding patient's mortality information. Obtain the clinical characteristic information of the patients corresponding to each sample. The clinical characteristic information includes at least gender, age, CTA, BMI, BaPWV, and clinical outcome. A good clinical outcome refers to a patient who is discharged after improvement in condition during clinical treatment.

[0059] 2) Divide the aortic dissection samples with good prognosis and the aortic dissection samples with poor prognosis into training set and test set.

[0060] 3) Based on the training set, features are screened using t-tests and P-values, and each protein feature in the plasma proteome that meets the expression level difference requirement of P<0.05 is identified as the optimal biomarker combination for predicting the clinical mortality risk of patients with aortic dissection. This step is the screening process in Example 1, and the optimal biomarker combination for predicting the clinical mortality risk of patients with aortic dissection is the second biomarker in Example 1.

[0061] 4) Based on the expression levels and clinical characteristics of the optimal biomarker combination for predicting the clinical mortality risk of aortic dissection patients in each sample of the training set, an application prediction model is constructed and trained using the random forest algorithm. The optimal combination of clinical features is determined based on the feature importance of each clinical feature in the predictive performance of the application prediction model. The application prediction model is tested to evaluate its predictive performance, and the application prediction model whose predictive performance exceeds the performance index threshold is used as the clinical mortality risk prediction model for aortic dissection patients to provide prediction results regarding the mortality risk of aortic dissection. The prediction results are represented by risk parameters, including but not limited to percentages, such as "probability of having aortic dissection: 90%", etc. The actual risk parameters can also be other numerical ranges, and the content and method of presentation can be set as needed, which will not be elaborated here.

[0062] The optimal combination of clinical features is determined by ranking the features by importance and selecting the top five features based on clinical experience.

[0063] In the above modeling method, the preset condition for expressing the degree of difference at different levels is that the P-value is less than 0.05, that is, P-value < 0.05, where P-value is the P-value. The specific calculation method is known to those skilled in the art and will not be elaborated here.

[0064] The above method for constructing and training an application prediction model using the random forest algorithm includes the following steps: 1) Data cleaning and population; 2) The data is randomly divided into training set and test set.

[0065] 3) Model training and parameter optimization were performed on the training set. First, a random search was conducted to explore various parameter combinations, then the model with the best performance was selected. ntree = 500, and the minimum size of the terminal node was 1. Triple-fold cross-validation was used to ensure the model had good generalization ability.

[0066] In the optimal parameters, the number of trees, ntree, is set to 500. Each tree is randomly sampled for learning and construction. Due to the randomness of samples and features, the random forest is less prone to overfitting, and this randomness helps improve the model's generalization ability. Each internal node represents a "test" for an attribute (e.g., aortic dissection or no), each branch represents the result of the test, and each leaf node represents a class label (made after calculating all attributes). Nodes without children are leaves. The minimum size of a terminating node is 1, meaning each terminating node has at least one sample, allowing the model to explore patterns in the data more comprehensively.

[0067] After the above application prediction model based on the random forest algorithm is trained, it is further necessary to test the application prediction model based on the test set to evaluate the prediction performance, and use the application prediction model whose prediction performance exceeds the performance index threshold to give a prediction result on the risk of having aortic dissection or the clinical death risk of aortic dissection patients based on the expression level of the above biomarkers.

[0068] Among them, performance indicators such as AUC can be used, and other performance indicators and their corresponding thresholds can also be selected according to the actual situation.

[0069] The performance indicators of the above aortic dissection diagnosis model (Model 1) and the clinical death risk prediction model for aortic dissection patients (Model 2) on the test set are shown in Table 1, Figure 2 (B, C) and Figure 3 (E, F), where the aortic dissection diagnosis model only uses 3 protein features of the first biomarker, while the clinical death risk prediction model for aortic dissection patients uses the second biomarker and optimal clinical feature information to predict the risk of a subject having aortic dissection and whether the affected limb of aortic dissection will die.

[0070] Table 1: Performance indicators of Model 1 and Model 2 on the test set The area under the curve (AUC) of the AUC curve is the area covered under the curve of the receiver operating characteristic curve. The AUC value is not greater than 1. The larger the AUC value, the better the model performance. When 0.5 < AUC < 1, it indicates that the model prediction performance is better than random prediction and the performance is good.

[0071] It can be seen from Table 1 that for both the aortic dissection diagnosis model and the clinical death risk prediction model for aortic dissection patients, the area under the AUC curve is close to or exceeds 0.9, and the specificity, sensitivity, and accuracy are all relatively high, indicating that the prediction performances of both models are very good and the prediction results are highly credible. Thus, unnecessary waste of medical resources and side effects of patients and anxiety about the disease caused by over-treatment can be avoided.

[0072] In addition, through the random forest algorithm, feature importance analysis is carried out on each protein in the plasma protein data, and the analysis results are respectively as Figure 2 (A) and Figure 3 (D) shown. In the figure, the importance of protein features decreases in order from top to bottom. From Figure 2 it can be seen that the importance ranking of the three protein features as the optimal biomarker combination for diagnosing aortic dissection is the most forward. From Figure 3It can be seen that the seven protein features in the optimal biomarker combination for predicting clinical mortality risk in patients with aortic dissection are ranked highest in importance. The feature importance analysis results of the random forest algorithm further validate the biomarker screening results in Example 1.

[0073] Example 3: Aortic dissection diagnostic system and aortic dissection patient clinical mortality risk prediction system I. Aortic Dissection Diagnostic System like Figure 1 As shown, the aortic dissection diagnostic system 001 includes a first processor 002 and a first display 003. The first processor 002 obtains the expression levels of RCN1, SERPINA3 and CPN1 of the subject, uses the aortic dissection diagnostic model in Example 2 to predict the risk parameters of the subject having aortic dissection, and causes the first display 003 to display the predicted risk parameters of the subject having aortic dissection and the corresponding diagnosis and treatment suggestions.

[0074] Risk parameters can be expressed as percentages, such as "probability of having aortic dissection: 90%". Specifically, based on the clinical experience of medical professionals, two risk parameter thresholds can be set for subjects to have aortic dissection: a low threshold of 20% and a high threshold of 50%. Treatment recommendations are as follows: If the predicted risk parameter is below the low threshold, the risk of aortic dissection in the subject is considered to be very low. The subject can be treated as other clinical chest pain conditions and some patients (e.g., young patients with no history of smoking, normal BaPWV, normal BMI and no family history of genetic diseases) are exempt from invasive imaging examinations. If the low threshold < the predicted risk parameter ≤ the high threshold, further investigation using other medical methods is necessary. When the predicted risk parameter exceeds the high threshold, it indicates an extremely high risk of the subject having aortic dissection. It is recommended to conduct detailed, high-risk procedures such as CTA, echocardiography, DSA, or MRI in advance. For abnormalities detected during imaging, such as double-lumen changes, aortic dilation, or intramural hematoma, a value exceeding the high threshold can also help clinicians make a predisposing diagnosis of aortic dissection. This invention does not specifically limit the specific setting of the threshold or the specific application of the prediction results by medical professionals.

[0075] Clinical mortality risk prediction system for patients with aortic dissection The aortic dissection patient clinical mortality risk prediction system 004 includes a second processor 005 and a second display 006. Before imaging reveals aortic dissection rupture or poor clinical outcome in the subject, the aortic dissection patient clinical mortality risk prediction model in Example 2 is used to predict the risk parameters of clinical mortality for the aortic dissection patient. Based on the predicted risk parameters of the subject's mortality, the second display 006 presents the predicted risk parameters and corresponding treatment recommendations.

[0076] For subjects diagnosed with aortic dissection, two risk parameter thresholds were set: low and high.

[0077] If the predicted risk parameter is below a low threshold (the specific value can be set to 20%), the risk of death of the subject can be considered very low, and treatment can be carried out according to the standard treatment procedure. Repeated imaging examinations can be waived for some patients (e.g., young patients with no history of smoking, normal BaPWV, normal BMI and no family history of genetic diseases). If the predicted risk parameters are between the low and high thresholds, further investigation using other medical methods is necessary. When the predicted risk parameter exceeds a high threshold (for example, 50%), it signifies an extremely high risk of death for the subject. The prediction results are provided to physicians to help them determine whether further detailed, high-risk procedures (e.g., CTA, echocardiography, DSA, or MRI) are necessary, and to recommend close monitoring for serious cardiovascular events such as death. For abnormalities detected during imaging, such as severe lung infection, neurological complications, double-lumen changes in other parts of the aorta, aortic dilation, or surgical anastomotic hematoma, a threshold above the high threshold can also help clinicians make a propensity prediction of extremely poor prognosis or even death for patients with aortic dissection. This invention does not specifically limit the specific setting of the threshold or the specific application of the prediction results by medical professionals.

[0078] Close monitoring can further include: 1) shortening the time interval between postoperative follow-up examinations for aortic dissection patients who have not yet experienced severe pain in other parts of the body, in order to detect potential aortic dissection rupture early; 2) completely eliminating the entire aortic tear and false lumen during aortic dissection surgery to avoid serious cardiovascular events or even death; 3) for patients with unsatisfactory clinical prognosis, if the predicted risk parameters are higher than the first threshold, it is recommended that clinicians consider the necessity of more rigorous monitoring of postoperative biochemical indicators and cardiovascular function.

[0079] The aforementioned second display 006 can be the first display 003. When the first processor 002 processes the risk of the subject's illness and gives the subject's disease risk parameter as higher than a high threshold, the second processor 004 is used to predict the subject's mortality risk, so that the first display 003 simultaneously displays the disease risk parameter and the patient's clinical mortality risk parameter and the corresponding diagnosis and treatment opinions.

[0080] Aortic dissection diagnostic systems and clinical mortality risk prediction systems for aortic dissection patients may also include liquid chromatography and mass spectrometry equipment to acquire plasma samples collected from subjects, analyze and quantify them through targeted proteomics (MRMHR), and detect the expression levels of biomarkers.

[0081] In some implementations, the plasma used to obtain the expression levels of biomarkers in subjects can be obtained simultaneously with routine blood tests, requiring a small amount of plasma, without increasing blood loss or blood collection trauma, and without the need for additional blood collection. This reduces unnecessary CTA and MRI examinations for low-risk patients and the number of invasive examinations for some low-risk patients, thereby reducing the psychological stress, contrast agent risks, radiation exposure, and bodily invasion of subjects.

[0082] Example 4 The use of first biomarkers or their detection reagents in the preparation of products for diagnosing aortic dissection, including RCN1, SERPINA3 and CPN1.

[0083] Example 5 A kit for diagnosing aortic dissection, comprising reagents for detecting the expression levels of a first biomarker in plasma samples from subjects; the first biomarker includes RCN1, SERPINA3, and CPN1.

[0084] Example 6 The use of a second biomarker or its detection reagent in the preparation of products that predict the risk of clinical death in patients with aortic dissection, wherein the first biomarker includes RCN1, SERPINA3 and CPN1.

[0085] Example 7 A kit for predicting clinical mortality risk in patients with aortic dissection, comprising reagents for detecting the expression levels of a second biomarker in plasma samples from subjects; the second biomarker includes SYTL1, B2M, LYVE1, TTR, PRSS2, CSTF1, and H2BC5.

[0086] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A biomarker associated with aortic dissection, characterized in that: The biomarkers include a second biomarker; the second biomarker is used to predict the clinical mortality risk of patients with aortic dissection, and the second biomarker includes SYTL1 protein, B2M protein, LYVE1 protein, TTR protein, PRSS2 protein, CSTF1 protein and H2BC5 protein.

2. The use of a second biomarker or its detection reagent in the preparation of products for predicting the clinical mortality risk in patients with aortic dissection, characterized in that: The second biomarker includes SYTL1 protein, B2M protein, LYVE1 protein, TTR protein, PRSS2 protein, CSTF1 protein, and H2BC5 protein.

3. A kit for predicting the clinical mortality risk in patients with aortic dissection, characterized in that: The kit contains reagents for detecting the expression level of a second biomarker in a sample; the second biomarker includes SYTL1 protein, B2M protein, LYVE1 protein, TTR protein, PRSS2 protein, CSTF1 protein, and H2BC5 protein; the sample is the subject's blood or plasma.

4. A method for constructing a clinical mortality risk prediction model for patients with aortic dissection, characterized in that: Includes the following steps: Plasma proteomic data of aortic dissection patients with good clinical outcomes were obtained as the good prognostic sample group, and plasma proteomic data of aortic dissection patients with clinical outcomes of death were obtained as the poor prognostic sample group. Each sample includes the corresponding patient's clinical characteristics, the expression level of each protein feature in the plasma proteomic sample, and the annotation information of whether the corresponding patient died. The good prognosis sample group and the poor prognosis sample group are randomly divided into a training set and a test set; Differential protein analysis was performed on the training set, and differentially expressed proteins were screened using t-tests and p-values. Proteins with expression level differences in the plasma proteome that met the p-value < 0.05 were selected as the optimal biomarker combination for predicting clinical mortality risk in patients with aortic dissection. The optimal biomarker combination for predicting clinical mortality risk in patients with aortic dissection included SYTL1 protein, B2M protein, LYVE1 protein, TTR protein, PRSS2 protein, CSTF1 protein, and H2BC5 protein. Using the expression levels and clinical characteristics of the optimal biomarkers for predicting the clinical mortality risk of aortic dissection patients in each sample as feature information, a clinical mortality risk prediction model for aortic dissection patients is constructed using machine learning algorithms based on the training set. The optimal combination of clinical features is determined based on the feature importance of each clinical feature in the predictive performance of the clinical mortality risk prediction model for aortic dissection patients. The predictive performance of the aortic dissection patient clinical mortality risk prediction model was evaluated using information on the expression levels of the optimal biomarkers and the optimal clinical characteristics of each sample in the test set data. The clinical mortality risk prediction model for patients with aortic dissection is used to obtain risk parameters for clinical mortality of patients with aortic dissection based on the expression levels of the optimal combination of biomarkers for predicting clinical mortality risk of patients with aortic dissection and the optimal combination of clinical features.

5. A clinical mortality risk prediction model for patients with aortic dissection, characterized in that: It is constructed using the method described in claim 4 for constructing a clinical mortality risk prediction model for patients with aortic dissection.

6. A system for predicting the clinical mortality risk in patients with aortic dissection, characterized in that: Including the processor and display; The processor is configured to predict the clinical mortality risk of aortic dissection patients using a clinical mortality risk prediction model based on the expression levels of biomarkers used to predict the clinical mortality risk of aortic dissection patients. The display is used to present the clinical mortality risk parameters predicted by the processor; The biomarkers used to predict the clinical mortality risk in patients with aortic dissection include SYTL1 protein, B2M protein, LYVE1 protein, TTR protein, PRSS2 protein, CSTF1 protein, and H2BC5 protein.

Images