Molecular marker for noninvasive evaluation of severity of esophageal and gastric varicosity

Abstract

The invention relates to a molecular marker for noninvasive evaluation of severity of esophageal and gastric varicosity, and belongs to the technical field of molecular biology. The invention provides a molecular marker for noninvasive evaluation of severity of esophageal and gastric varicosity. The molecular marker comprises gender, platelet count, spleen thickness, spleen major diameter and spleen hardness. The molecular marker is used for evaluating the severity of esophageal varicose veins, is noninvasive, has higher detection accuracy (AUC is greater than 0.9), can more accurately and directly reflect the severity of esophageal varicose veins of liver cirrhosis patients, is more accurate in grading, and can be used for detecting the severity of esophageal varicose veins of liver cirrhosis patients. And the risk of the liver cirrhosis patient suffering from esophageal and gastric varicose veins can be accurately divided into low, medium and high levels. Moreover, when the molecular marker is used for evaluating the severity of esophageal and gastric varicose veins, the molecular marker is easy to realize clinically, economical and practical, is easy to accept by patients, can obtain a better clinical effect, and has a great application prospect.

Description

Technical Field

[0001] The invention relates to a molecular marker for non-invasively evaluating the severity of esophageal and gastric varices, and belongs to the technical field of molecular biology. Background Art

[0002] Cirrhosis is a chronic, progressive, diffuse liver disease caused by one or more etiologies. Diffuse hepatic fibrosis, regenerative nodules, and pseudolobule formation are its main pathological features. Gastroesophageal varices (GOV) are complications of portal hypertension in cirrhosis. About 3% to 12% of patients with cirrhosis develop esophageal varices each year, and about 8% to 12% develop from small varicose veins to large varicose veins. Esophageal variceal bleeding (EVB) is the main cause of death in patients with cirrhosis, with an annual incidence of 5% to 15%, and the mortality rate of acute bleeding is as high as 15% to 35%. In the case of large variceal bleeding, the 6-week mortality rate is about 11.1% to 40%.

[0003] The presence of red signs in the gastroesophageal veins, the size of the varices, and the severity of cirrhosis are considered to be the most important predictors of esophagogastric variceal bleeding. Since the first Baveno consensus workshop in 1992, the use of screening esophagogastroduodenoscopy (EGD) to detect high-risk varices (HRV) in all patients with cirrhosis has been recommended. It is recommended that patients with cirrhosis undergo screening endoscopic examinations once a year to screen for high-risk varices in the esophagogastric region, and timely intervention can be performed based on the screening results. Until now, screening endoscopic examinations are still the preferred method for identifying the presence and size of esophagogastric varices. However, a large proportion of patients with cirrhosis do not have varicose veins or high-risk varices, so endoscopy is a suboptimal screening test. In addition, screening endoscopic examinations have the disadvantages of being prone to invasive endoscopy-related complications, being relatively expensive, and causing patient discomfort, which have promoted the exploration of new screening methods for gastroesophageal varices.

[0004] At present, more and more studies have focused on finding new minimally invasive or non-invasive screening methods for gastroesophageal varices to replace screening gastroscopy to exclude and include high-risk varices, thereby reducing or avoiding the use of screening gastroscopy. However, these methods still have some shortcomings. For example, it has been reported that the three indicators of liver stiffness, platelet count and spleen thickness are combined to assess the severity of esophageal and gastric varices. Although these three indicators can be directly obtained by non-invasive methods, the combination of these three indicators can only classify the probability of esophageal and gastric varices in patients with cirrhosis into high risk and low risk. In addition, the combination of these three indicators has a very high misdiagnosis rate, which can easily classify some low-risk patients as high-risk and some high-risk patients as low-risk, thereby missing some patients. In general, the existing combination model around liver stiffness, platelet count and spleen thickness cannot accurately stratify the risk of esophageal and gastric varices, causing a large number of patients to still undergo gastroscopy screening, which brings certain physical pain and economic pressure to patients. Therefore, there is an urgent need to find a minimally invasive or non-invasive gastroesophageal varices screening method with higher detection accuracy and more precise risk stratification of esophageal and gastric varices in patients with cirrhosis. Summary of the invention

[0005] To solve the above problems, the present invention provides a molecular marker for non-invasively evaluating the severity of esophageal and gastric varices, wherein the molecular marker includes gender, platelet count, spleen thickness, spleen long diameter and spleen hardness.

[0006] In one embodiment of the present invention, the molecular markers consist of gender, platelet count, spleen thickness, spleen long diameter and spleen hardness.

[0007] In one embodiment of the present invention, the esophageal and gastric varices include esophageal and gastric varices caused by liver cirrhosis.

[0008] In one embodiment of the present invention, the esophageal and gastric varices are esophageal and gastric varices caused by liver cirrhosis.

[0009] The present invention also provides the use of a reagent for detecting the above molecular marker or a device for detecting the above molecular marker in the preparation of a product for non-invasively evaluating the severity of esophageal and gastric varices.

[0010] In one embodiment of the present invention, the esophageal and gastric varices include esophageal and gastric varices caused by liver cirrhosis.

[0011] In one embodiment of the present invention, the esophageal and gastric varices are esophageal and gastric varices caused by liver cirrhosis.

[0012] In one embodiment of the present invention, the application comprises the following steps:

[0013] Information collection step: detecting and collecting information related to the evaluation object and the above-mentioned molecular markers;

[0014] Data analysis step: Analyze and calculate the information obtained in the information collection step to obtain the risk prediction value related to the severity of esophageal and gastric varices;

[0015] Evaluation steps: The evaluation subjects are classified into high risk, medium risk or low risk of esophageal and gastric varices according to the risk prediction value obtained by the data analysis module. The evaluation subjects included in the high-risk level can start primary preventive drug treatment without gastroscopy in clinical diagnosis and treatment to prevent esophageal and gastric varices rupture and bleeding; the evaluation subjects included in the medium-risk level need to be closely followed up and undergo gold standard gastroscopy to clarify the situation of esophageal and gastric varices; the evaluation subjects included in the low-risk level can be exempted from gastroscopy in clinical diagnosis and treatment, and only need regular follow-up.

[0016] In one embodiment of the present invention, the data analysis step includes: first constructing a multivariate lossistics regression prediction model, and calculating a predicted probability value related to the severity of esophageal and gastric varices according to the multivariate lossistics regression prediction model, and then substituting the predicted probability value into a transformation equation to calculate a risk prediction value related to the severity of esophageal and gastric varices;

[0017] The multivariate lossistics regression prediction model is: P = e riskscore / (1+e riskscore );where P is the predicted probability value, ∈(0,1); risk score is the risk score; e is the base of the natural logarithm (where ∈(0,1) means that the predicted probability value P follows a uniform distribution in the interval (0,1));

[0018] The calculation formula of the risk score is: risk score = -11.246 - 0.013 × PLT + 1.673 × Sex (Male: 1,

[0019] Female: 0)+0.028×spleenD+0.067×spleenT+0.061×SSM630; where risk score refers to risk score; PLT refers to platelet count, unit is ×10 9 / L; Sex (Male: 1, Female: 0) refers to gender. When calculating the risk score, if the gender of the assessed subject is male, the value of this item is 1, and if the gender of the assessed subject is female, the value of this item is 0; spleenD refers to the thickness of the spleen, in mm; spleenT refers to the long diameter of the spleen, in mm; SSM630 refers to the stiffness of the spleen, in kPa (where spleenD and spleenT can be measured in vitro using a Doppler convenient ultrasound instrument; SSM630 can be measured in vitro using a liver transient elastic imaging instrument based on the principle of transient elastic imaging);

[0020] The transformation equation is: risk prediction value = [score - minimum score / (maximum score - minimum score)] × 100; where the score is the predicted probability value P, and the minimum score is 3.343713 × 10 -5 , the maximum score is 0.9917606.

[0021] In one embodiment of the present invention, the evaluation step includes: determining whether the risk prediction value obtained by the data analysis module is greater than the high-risk cutoff value or less than the low-risk cutoff value (double cut off); if the risk prediction value is greater than the high-risk cutoff value (>78.28±1), the evaluation object is classified as high risk of esophageal and gastric varices; if the risk prediction value is less than the low-risk cutoff value (<4.84±1), the evaluation object is classified as low risk of esophageal and gastric varices; if the risk prediction value is between the high-risk cutoff value and the low-risk cutoff value ([4.84±1, 78.28±1]), the evaluation object is classified as medium risk of esophageal and gastric varices; the high-risk cutoff value is 78.28±1; the low-risk cutoff value is 4.84±1.

[0022] In one embodiment of the present invention, the product is a detection kit or a detection device.

[0023] The present invention also provides a product for non-invasively evaluating the severity of esophageal and gastric varices, wherein the detection kit comprises a reagent for detecting the above-mentioned molecular marker and / or a device for detecting the above-mentioned molecular marker.

[0024] In one embodiment of the present invention, the esophageal and gastric varices include esophageal and gastric varices caused by liver cirrhosis.

[0025] In one embodiment of the present invention, the esophageal and gastric varices are esophageal and gastric varices caused by liver cirrhosis.

[0026] In one embodiment of the present invention, the product is a detection kit or a detection device.

[0027] The present invention also provides an evaluation model for non-invasively evaluating the severity of esophageal and gastric varices, the evaluation model comprising an information collection module, a data analysis module and an evaluation module;

[0028] The information collection module is used to detect and collect information related to the evaluation object and the above-mentioned molecular markers;

[0029] The data analysis module is used to analyze and calculate the information obtained in the information collection step to obtain a risk prediction value related to the severity of esophageal and gastric varices;

[0030] The evaluation module is used to classify the evaluation object into high risk, medium risk or low risk of esophageal and gastric varices according to the risk prediction value obtained by the data analysis module.

[0031] In one embodiment of the present invention, the esophageal and gastric varices include esophageal and gastric varices caused by liver cirrhosis.

[0032] In one embodiment of the present invention, the esophageal and gastric varices are esophageal and gastric varices caused by liver cirrhosis.

[0033] In one embodiment of the present invention, the data analysis step includes: first constructing a multivariate lossistics regression prediction model, and calculating a predicted probability value related to the severity of esophageal and gastric varices according to the multivariate lossistics regression prediction model, and then substituting the predicted probability value into a transformation equation to calculate a risk prediction value related to the severity of esophageal and gastric varices;

[0034] The multivariate lossistics regression prediction model is: P = e riskscore / (1+e riskscore );where P is the predicted probability value, ∈(0,1); risk score is the risk score; e is the base of the natural logarithm;

[0035] The calculation formula of the risk score is: risk score = -11.246 - 0.013 × PLT + 1.673 × Sex (Male: 1,

[0036] Female: 0)+0.028×spleenD+0.067×spleenT+0.061×SSM630; where risk score refers to risk score; PLT refers to platelet count, unit is ×10 9 / L; Sex (Male: 1, Female: 0) refers to gender. When calculating the risk score, if the gender of the assessed subject is male, the value of this item is 1, and if the gender of the assessed subject is female, the value of this item is 0; spleenD refers to the thickness of the spleen, in mm; spleenT refers to the long diameter of the spleen, in mm; SSM630 refers to the stiffness of the spleen, in kPa;

[0037] The transformation equation is: risk prediction value = [score - minimum score / (maximum score - minimum score)] × 100; where the score is the predicted probability value P, and the minimum score is 3.343713 × 10 -5 , the maximum score is 0.9917606.

[0038] In one embodiment of the present invention, the evaluation step includes: determining whether the risk prediction value obtained by the data analysis module is greater than the high-risk cutoff value or less than the low-risk cutoff value; if the risk prediction value is greater than the high-risk cutoff value (>78.28±1), the evaluation object is classified as a high risk of esophageal and gastric varices; if the risk prediction value is less than the low-risk cutoff value (<4.84±1), the evaluation object is classified as a low risk of esophageal and gastric varices; if the risk prediction value is between the high-risk cutoff value and the low-risk cutoff value ([4.84±1, 78.28±1]), the evaluation object is classified as a medium risk of esophageal and gastric varices; the high-risk cutoff value is 78.28±1; the low-risk cutoff value is 4.84±1.

[0039] The present invention also provides an evaluation method for non-invasively evaluating the severity of esophageal and gastric varices, which is not for the purpose of disease diagnosis and treatment. The evaluation method comprises: using the above-mentioned molecular marker or the above-mentioned evaluation model to non-invasively evaluate the severity of esophageal and gastric varices.

[0040] In one embodiment of the present invention, the esophageal and gastric varices include esophageal and gastric varices caused by liver cirrhosis.

[0041] In one embodiment of the present invention, the esophageal and gastric varices are esophageal and gastric varices caused by liver cirrhosis.

[0042] The present invention also provides an evaluation method for non-invasively evaluating the severity of esophageal and gastric varices, which is not for the purpose of disease diagnosis and treatment. The evaluation method comprises: using the above-mentioned molecular marker or the above-mentioned evaluation model to non-invasively evaluate the severity of esophageal and gastric varices.

[0043] The technical solution of the present invention has the following advantages:

[0044] 1. The present invention provides a molecular marker for non-invasively assessing the severity of esophageal and gastric varices, wherein the molecular marker includes gender, platelet count, spleen thickness, spleen long diameter and spleen hardness. The use of the molecular marker to assess the severity of esophageal and gastric varices has the advantages of being non-invasive, having a higher detection accuracy (AUC greater than 0.9), being able to more accurately and directly reflect the severity of esophageal and gastric varices in patients with cirrhosis, and being more accurate in grading, and being able to accurately divide the risk of esophageal and gastric varices in patients with cirrhosis into three levels: low, medium and high. Moreover, the use of the molecular marker to assess the severity of esophageal and gastric varices is easy to implement in the clinic, economical and affordable, easily accepted by patients, and can obtain good clinical effects, with great application prospects.

[0045] 2. The present invention provides an evaluation model for non-invasively evaluating the severity of esophageal and gastric varices, the evaluation model comprising an information collection module, a data analysis module and an evaluation module; the information collection module is used to detect and collect information related to the evaluation object and the above-mentioned molecular markers; the data analysis module is used to analyze and calculate the information obtained in the information collection step to obtain a risk prediction value related to the severity of esophageal and gastric varices; the evaluation module is used to classify the evaluation object into high risk, medium risk or low risk of esophageal and gastric varices according to the risk prediction value obtained by the data analysis module; the molecular markers include gender, platelet count, spleen thickness, spleen long diameter and spleen hardness. The evaluation model is used to evaluate the severity of esophageal and gastric varices, which has the advantages of being non-invasive, having a higher detection accuracy (AUC greater than 0.9), being able to more accurately and directly reflect the severity of esophageal and gastric varices in patients with liver cirrhosis, and being more accurately classified into three levels of low, medium and high risk of esophageal and gastric varices in patients with liver cirrhosis. Furthermore, the use of the evaluation model to assess the severity of esophageal and gastric varices is easy to implement in clinical practice, economical and affordable, easily accepted by patients, can achieve good clinical effects, and has great application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 : Patient selection process.

[0047] Figure 2 : To evaluate the diagnostic performance of the model based on a combination of five molecular markers, namely sex (Sex), platelet count (PLT), spleen thickness (spleenD), spleen long diameter (spleenT) and spleen stiffness (SSM630), for high-risk varicose veins in the validation cohort.

[0048] Figure 3: To evaluate the diagnostic performance of the model based on a combination of five molecular markers, namely sex (Sex), platelet count (PLT), spleen thickness (spleenD), spleen long diameter (spleenT) and spleen stiffness (SSM630), for high-risk varicose veins in the training cohort.

[0049] Figure 4 : To evaluate the diagnostic performance of a model based on a combination of three molecular markers, platelet count (PLT), spleen thickness (spleenD), and spleen stiffness (SSM630), for high-risk varicose veins in a validation cohort. DETAILED DESCRIPTION

[0050] The following examples are provided for a better understanding of the present invention, but are not intended to limit the best mode of implementation, nor to limit the content and protection scope of the present invention. Any product identical or similar to the present invention obtained by anyone under the inspiration of the present invention or by combining the features of the present invention with other prior arts shall fall within the protection scope of the present invention.

[0051] If no specific experimental steps or conditions are specified in the following examples, the conventional experimental steps or conditions described in the literature in the field can be used. If no manufacturer is specified for the reagents or instruments used, they are all conventional reagent products that can be purchased commercially.

[0052] Experimental Example 1: Screening of molecular markers and construction of evaluation model

[0053] The experimental process is as follows:

[0054] 1. Screening of molecular markers

[0055] This study was approved by the hospital ethics committee (Ethics Committee of Nanfang Hospital, Southern Medical University), and the patients signed informed consent. A total of 1289 patients with cirrhosis admitted to Nanfang Hospital from June 2021 to April 2024 were selected as the research subjects. Inclusion criteria: 1) Adult patients aged 18 to 75 years; 2) Imaging diagnosis of cirrhosis; 3) Patients who need to undergo gastroscopy screening according to the judgment of the treating physician; 4) Voluntary signing of informed consent. Exclusion criteria: 1) History of liver transplantation / liver cancer; 2) Acute or chronic portal vein thrombosis; 3) Previous transjugular intrahepatic portosystemic shunt (TIPS); 4) Splenectomy or splenic embolization; 5) Moderate to severe ascites; 6) Inappropriate for participation in this study as judged by the researchers. A total of 804 patients with cirrhosis were included in the analysis, including 183 patients with high-risk varicose veins and 621 patients with low-risk varicose veins (see the patient selection process for details). Figure 1The baseline data of the patients included in the analysis are shown in Table 1).

[0056] According to the ratio of 7:3, the analysis population was randomly divided into a training cohort and a validation cohort, of which 564 cases were in the learning cohort and 240 cases were in the validation cohort. The characteristics of the patients in the training cohort and the validation cohort were compared by paired sample t-test, and the comparison results are shown in Table 2. The comparison results in Table 2 suggest that the characteristics of the patients in the training cohort and the validation cohort are similar.

[0057] LASSO regression analysis was performed in the training cohort. The lambda value of 0.03015075 and the log(λ) value of -3.501545 (1-SE standard) were selected based on 10 cross-validations. Finally, seven variables were screened, including sex (Sex), red blood cell count (RBC), platelet count (PLT), serum albumin concentration (ALB), spleen thickness (spleenD), spleen long diameter (spleenT) and spleen stiffness (SSM630).

[0058] These seven variables were introduced into multivariate logistic regression analysis, and it was found that sex (Sex), platelet count (PLT), spleen thickness (spleenD), spleen long diameter (spleenT) and spleen stiffness (SSM630) were independent variables of high-risk varicose veins. These five independent variables of high-risk varicose veins were used as molecular markers for non-invasive assessment of the severity of esophageal and gastric varices, and these five molecular markers were combined for non-invasive assessment of the severity of esophageal and gastric varices. Among them, platelet count (PLT) was a protective factor, and sex (Sex), spleen thickness (spleenD), spleen long diameter (spleenT) and spleen stiffness (SSM630) were risk factors.

[0059] 2. Construction of evaluation model

[0060] A multivariate loss regression prediction model was constructed based on the five independent variables of high-risk varicose veins, namely sex, platelet count (PLT), spleen thickness (spleenD), spleen long diameter (spleenT) and spleen stiffness (SSM630). The constructed multivariate loss regression prediction model was: P = e risk score / (1+e riskscore); In the multivariate lossistics regression prediction model, P is the predicted probability value, ∈(0,1) (∈(0,1) means that the predicted probability value P follows a uniform distribution in the interval (0,1)); e is the base of the natural logarithm; risk score is the risk score, and the calculation formula is: risk score=-11.246-0.013×PLT+1.673×Sex(Male:1,Female:0)+0.028×spleenD+0.067×spleenT+0.061×SSM630; In the calculation formula of the risk score, Sex(Male:1,Female:0) refers to gender. When calculating the risk score, if the gender of the evaluated object is male, the value of this item is 1, and if the gender of the evaluated object is female, the value of this item is 0.

[0061] On the basis of the multivariate lossistics regression prediction model, a transformation equation was introduced to obtain the risk prediction value related to the severity of esophageal and gastric varices. The introduced transformation equation was: risk prediction value = [score-minimum score / (maximum score-minimum score)] × 100; in the transformation equation, the score is the predicted probability value P, and the minimum score is 3.343713 × 10 -5 , the maximum score is 0.9917606.

[0062] 3. Evaluation and validation of molecular marker combinations

[0063] According to the important indicators for evaluating the performance of the classification model, the receiver operating characteristic (ROC) curve was established to verify the effectiveness of the combination of five molecular markers: sex (Sex), platelet count (PLT), spleen thickness (spleenD), spleen long diameter (spleenT) and spleen stiffness (SSM630). The ROC curve is shown in Figure 2-Figure 3 (ROC value is used as a standard to measure the quality of classifiers. It is generally believed that values ​​above 0.7 have good discrimination ability, values ​​above 0.8 indicate that the model has strong discrimination ability, and values ​​close to 1 indicate that the model is very good). Figure 2 and Figure 3 The ROC curve showed that the area under the ROC curve of the model based on the combination of five molecular markers, namely sex (Sex), platelet count (PLT), spleen thickness (spleenD), spleen long diameter (spleenT) and spleen stiffness (SSM630), and the severity of esophageal and gastric varices was 0.94 (95% CI 0.92-0.97) in the learning cohort and 0.94 (95% CI 0.88-0.96) in the validation cohort. This result showed that the model had excellent discrimination ability.

[0064] According to the important indicators for evaluating the performance of the classification model, the receiver operating characteristic (ROC) curve was established to verify the effectiveness of the combination of three molecular markers: platelet count (PLT), spleen thickness (spleenD) and spleen stiffness (SSM630). The ROC curve is shown in Figure 4 . Figure 4 The ROC curve of the model based on the combination of three molecular markers, platelet count (PLT), spleen thickness (spleenD) and spleen stiffness (SSM630), and the severity of esophageal and gastric varices had an area under the ROC curve of 0.829 (95% CI 0.80-0.85) in the validation cohort. This result shows that the model has a strong ability to distinguish, but it is lower than the ability of the model based on the combination of five molecular markers, sex (Sex), platelet count (PLT), spleen thickness (spleenD), spleen long diameter (spleenT) and spleen stiffness (SSM630). It can be seen that the use of sex (Sex), platelet count (PLT), spleen thickness (spleenD), spleen long diameter (spleenT) and spleen stiffness (SSM630) to non-invasively assess the severity of esophageal and gastric varices has the advantage of higher detection accuracy.

[0065] 4. Confirmation of double cut-off values ​​of the evaluation model

[0066] According to the sensitivity and specificity of the model, the exclusion value was calculated (the exclusion value is Rule out, looking for the cut-off with the maximum Youden index and a missed diagnosis rate of <5%). The calculated exclusion value, that is, the optimal cut-off value for excluding patients with high-risk gastroesophageal varices (low-risk population) is 4.84. These patients can be exempted from gastroscopy in clinical practice and only need regular follow-up.

[0067] According to the model's ability to predict and distinguish patients with true high-risk gastroesophageal varices (that is, those who are identified as truly diseased by the model / determined to be diseased by the model), the inclusion value is calculated (the inclusion value is Rule in, looking for a cut-off with a positive predictive value PPV>90% and as large a true positive case as possible. The true positive case is TP, which refers to the number of samples that are actually positive and predicted to be positive). The inclusion value is calculated, that is, the optimal cut-off value for including patients with high-risk gastroesophageal varices (high-risk population) is 78.28. These patients can start primary preventive drug treatment without gastroscopy in clinical diagnosis and treatment to prevent gastroesophageal varicose bleeding.

[0068] Patients with a value between 4.84 and 78.28 ([4.84, 78.28]) are in the gray area. These patients are medium-risk patients and require close follow-up and gold standard gastroscopy to clarify the presence of gastroesophageal varices.

[0069] 5. Validation of the evaluation model

[0070] The test is based on sensitivity (the ability of the test to correctly identify people with the disease, that is, accuracy when the test result is positive. It measures the ability of the test to detect the disease. High sensitivity means few false negative results), specificity (the ability of the test to correctly identify people without the disease, that is, accuracy when the test result is negative. In other words, it measures the rate at which the test will not mistakenly diagnose healthy people as patients. High specificity means few false positive results), positive predictive value (the proportion of individuals with positive test results who actually have the disease. It reflects the possibility that the subject actually has the disease when the test result is positive), negative predictive value (the proportion of individuals with positive test results who actually have the disease among all individuals with positive test results. It reflects the possibility that the subject actually has the disease when the test result is positive), and negative predictive value (the proportion of individuals with positive test results who actually have the disease among all individuals with positive test results. The performance of the model was evaluated by the following indicators: the proportion of individuals with negative test results who are truly healthy. It reflects the possibility that the subject is actually healthy when the test result is negative), the positive likelihood ratio (the ratio of the proportion of individuals who actually have the disease to the proportion of individuals who do not have the disease but have a positive test result among all samples with positive test results. It reflects the possibility that the subject is actually sick when the test result is positive), and the negative likelihood ratio (the ratio of the proportion of individuals who are truly healthy to the proportion of individuals who have the disease but have a negative test result among all samples with negative test results. It reflects the possibility that the subject is actually not sick when the test result is negative). The results are shown in Table 3.

[0071] The results in Table 3 show that the model's exclusion value of 4.84 in the training set had a sensitivity of 98.89% and a negative predictive value of 99.57%, both higher than 95% and close to 100%, with a negative likelihood ratio of 0.02; the sensitivity and negative predictive value in the validation set were 98.89% and 98.98%, both higher than 90% and close to 100%, with a negative likelihood ratio of 0.04, indicating that the model can accurately exclude patients without high-risk gastroesophageal varices; at the same time, the model's inclusion value of 78 .28 The specificity and positive predictive value in the training set were 97.32% and 92.50%, both higher than 90%, and the positive likelihood ratio was 15.52; the specificity and positive predictive value in the validation set were 97.96% and 94.12%, both higher than 90%, and the positive likelihood ratio was 21.19, indicating that the model can accurately identify patients with high-risk gastroesophageal varices, thereby including patients with high-risk gastroesophageal varices. In summary, it can be concluded that the model has very good performance in exclusion and inclusion. It can be seen that the use of the evaluation model and the double cutoff value can indeed accurately classify the risk of esophageal and gastric varices in patients with cirrhosis into three levels: low, medium, and high, and the classification is more accurate.

[0072] Table 1. Baseline data of patients included in the analysis

[0073]

[0074] Table 2. Comparison of patient characteristics between the training cohort and the validation cohort

[0075]

[0076] Table 3. Accuracy results of the evaluation model and double cutoff values ​​for diagnosing high-risk varicose veins in the training cohort and validation cohort

[0077]

[0078] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

Claims

1. A molecular marker for non-invasively assessing the severity of esophageal and gastric varices, characterized in that: The molecular markers include gender, platelet count, spleen thickness, spleen long diameter and spleen hardness.

2. The molecular marker according to claim 1, characterized in that The molecular markers consist of gender, platelet count, spleen thickness, spleen long diameter and spleen hardness.

3. The molecular marker according to claim 1 or 2, characterized in that The esophageal and gastric varices include esophageal and gastric varices caused by liver cirrhosis.

4. Use of a reagent for detecting the molecular marker described in any one of claims 1 to 3 or a device for detecting the molecular marker described in any one of claims 1 to 3 in the preparation of a product for non-invasively assessing the severity of esophageal and gastric varices.

5. A product for non-invasively assessing the severity of esophageal and gastric varices, characterized in that: The detection kit comprises a reagent for detecting the molecular marker according to any one of claims 1 to 3 and / or a device for detecting the molecular marker according to any one of claims 1 to 3.

6. An evaluation model for non-invasively assessing the severity of esophageal and gastric varices, characterized in that: The evaluation model includes an information collection module, a data analysis module and an evaluation module; The information collection module is used to detect and collect information related to the evaluation object and the molecular marker according to any one of claims 1 to 3; The data analysis module is used to analyze and calculate the information obtained in the information collection step to obtain a risk prediction value related to the severity of esophageal and gastric varices; The evaluation module is used to classify the evaluation object into high risk, medium risk or low risk of esophageal and gastric varices according to the risk prediction value obtained by the data analysis module.

7. The evaluation model according to claim 6, characterized in that: The esophageal and gastric varices include esophageal and gastric varices caused by liver cirrhosis.

8. The evaluation model according to claim 6 or 7, characterized in that: The data analysis step includes: first constructing a multivariate lossistics regression prediction model, and calculating a predicted probability value related to the severity of esophageal and gastric varices according to the multivariate lossistics regression prediction model, and then substituting the predicted probability value into a transformation equation to calculate a risk prediction value related to the severity of esophageal and gastric varices; The multivariate lossistics regression prediction model is: P = e riskscore / (1+e riskscore );where P is the predicted probability value, ∈(0,1); risk score is the risk score; e is the base of the natural logarithm; The calculation formula of the risk score is: risk score = -11.246 - 0.013 × PLT + 1.673 × Sex (Male: 1, Female: 0)+0.028×spleenD+0.067×spleenT+0.061×SSM630; where risk score refers to risk score; PLT refers to platelet count, unit is ×10 9 / L; Sex (Male: 1, Female: 0) refers to gender. When calculating the risk score, if the gender of the subject is male, the value of this item is 1, and if the gender of the subject is female, the value of this item is 0; spleenD refers to the thickness of the spleen, in mm; spleenT refers to the long diameter of the spleen, in mm; SSM630 refers to the stiffness of the spleen, in kPa; The transformation equation is: risk prediction value = [score - minimum score / (maximum score - minimum score)] × 100; where the score is the predicted probability value P, and the minimum score is 3.343713 × 10 -5 , the maximum score is 0.9917606.

9. The evaluation model according to claim 6 or 7, characterized in that: The evaluation step includes: determining whether the risk prediction value obtained by the data analysis module is greater than the high-risk cutoff value or less than the low-risk cutoff value; if the risk prediction value is greater than the high-risk cutoff value, the evaluation object is classified as high risk for esophageal and gastric varices; if the risk prediction value is less than the low-risk cutoff value, the evaluation object is classified as low risk for esophageal and gastric varices; if the risk prediction value is between the high-risk cutoff value and the low-risk cutoff value, the evaluation object is classified as medium risk for esophageal and gastric varices; the high-risk cutoff value is 78.28±1; the low-risk cutoff value is 4.84±1.

10. A method for non-invasively evaluating the severity of esophageal and gastric varices, characterized in that: The method is not intended for the diagnosis and treatment of a disease, and the evaluation method comprises: using the molecular marker described in any one of claims 1 to 3 or the evaluation model described in any one of claims 6 to 9 to non-invasively evaluate the severity of esophageal and gastric varices.

Images