Serum Golgi protein 73-based metabolism-related fatty liver disease hepatic fibrosis evaluation method and system
By establishing a non-invasive diagnostic algorithm based on serum Golgi protein 73, and combining it with pathological evaluation of liver tissue and liver elastography, the GFA algorithm was constructed. This solved the problem of insufficient sensitivity and consistency of existing non-invasive diagnostic algorithms in metabolic-related fatty liver disease, and achieved a more efficient and accurate evaluation of liver fibrosis.
Patent Information
- Application Number
- CN202411113947.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-14
- Publication Date
- 2024-11-26
AI Technical Summary
Existing non-invasive diagnostic algorithms lack sufficient sensitivity and consistency in diagnosing liver fibrosis in metabolic-related fatty liver disease. Furthermore, traditional biomarkers such as FIB4, NFS, and Hepamet fibrosis scoring systems exhibit low consistency in application, making it difficult to effectively identify the risk of disease progression.
A non-invasive diagnostic algorithm based on serum Golgi protein 73 was developed. By combining the pathological evaluation mechanism of liver tissue, GP73 level detection and liver elastography, the GFA algorithm was constructed to evaluate the level of liver fibrosis and its performance was compared with that of traditional algorithms.
It improves the diagnostic sensitivity and consistency of liver fibrosis in metabolic-associated fatty liver disease, simplifies the diagnostic process, reduces reliance on invasive examinations, and provides higher diagnostic accuracy and cost-effectiveness.
Smart Images

Figure CN119028584A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the fields of artificial intelligence and medical treatment, and in particular to a method and system for evaluating liver fibrosis in metabolic-related fatty liver disease based on serum Golgi protein 73. Background Art
[0002] Currently, the development of drug treatments for metabolic-related fatty liver disease requires an invasive examination method - liver puncture pathology to determine the treatment endpoint. Biomarkers can not only achieve non-invasive diagnosis, but also reflect the treatment effect and predict the prognosis of the disease. Since fibrosis is the strongest predictor of mortality in patients with metabolic-related fatty liver disease, accurate staging of fibrosis and the ability to distinguish metabolic-related fatty liver disease from early fibrosis are essential for identifying patients at risk of disease progression. Simple and reliable biomarkers will enable medical professionals to determine which patients need close follow-up (patients with metabolic-related fatty liver disease with F2-F4 fibrosis) and which patients need less frequent follow-up (F0-F1 patients). According to the European Association for the Study of the Liver's Guidelines for Identification and Stratification of Chronic Liver Disease, the purpose of establishing a non-invasive diagnostic algorithm is to exclude low-prevalence populations and to simply achieve this through two unrelated non-invasive diagnostic methods. One of the common combinations is serological markers combined with liver elasticity measurement.
[0003] Traditional noninvasive diagnostic algorithms, such as FIB4 (or FI B-4), NFS, Hepamet Fibros is Score (HFS) and APRI, have been clinically used to monitor the risk of liver disease progression, but the sensitivity of diagnosing metabolic-related fatty hepatitis and early fibrosis is moderate. In addition, when applied to the same patient, there is low consistency between these fibrosis scoring systems, and it is necessary to develop biomarkers with stable diagnostic performance.
[0004] Golgi protein 73 (GP73) is a transmembrane protein found in the Golgi apparatus in recent years. Its relative molecular mass according to polyacrylamide gel electrophoresis is 7.3×10 4 , so it is called Golgi protein 73. It is mainly expressed in bile duct epithelial cells. Many studies have shown that it can be used as a serum marker for the early diagnosis of liver cancer. Its sensitivity and specificity are higher than those of the traditional liver cancer serum marker alpha-fetoprotein, and its role in various chronic liver diseases has gradually been revealed. Our research team has found that serum Golgi protein 73 levels in patients with chronic hepatitis B are helpful in determining the prognosis of the disease. Other studies have found that serum Golgi protein 73 is parallel to liver fibrosis, and compared with FI B-4, serum Golgi protein 73 has better diagnostic performance. However, there are very few studies at home and abroad on the risk stratification of metabolic-related fatty liver disease with Golgi protein 73 levels, and the sample size is small.
[0005] Based on the above background, the present invention discloses a method and system for evaluating liver fibrosis in metabolic-related fatty liver disease based on serum Golgi protein 73. Summary of the invention
[0006] The purpose of the present invention is to provide a method and system for evaluating liver fibrosis in metabolic-related fatty liver disease based on serum Golgi protein 73, aiming to solve the problems mentioned in the above-mentioned background technology; the present invention prospectively enrolls patients with metabolic-related fatty liver disease and measures their serum Golgi protein 73 levels, evaluates its correlation with liver fibrosis, and uses it in combination with transient elastography to explore a new diagnostic algorithm based on Golgi protein 73 (i.e., GP73) for identifying significant liver fibrosis.
[0007] The first aspect of the present invention is achieved by providing a method for evaluating liver fibrosis in metabolic-related fatty liver disease based on serum Golgi protein 73, the method comprising:
[0008] Establish a liver tissue pathological evaluation mechanism, the scoring criteria of which at least include fatty degeneration score, lobular inflammation score, ballooning degeneration score, and fibrosis stage;
[0009] Establishing detection conditions for GP73 levels, and obtaining the detected GP73 results as first detection data, wherein GP73 represents serum Golgi protein 73 of the subject;
[0010] Based on the evaluation mechanism, construct a non-invasive diagnostic model for the subject;
[0011] Obtaining basic information of the subject, and obtaining the test result of liver elasticity measurement of the subject as the second test data;
[0012] The basic information, the first detection data and the second detection data are input into the non-invasive diagnosis model to evaluate the liver fibrosis level of the subject.
[0013] The non-invasive diagnosis model is built based on the GFA algorithm, which satisfies:
[0014] GFA=-8.757+1.048×Age(y)+1.028×GP73(ng / mL)+1.588×LSM(kpa);
[0015] Among them, Age (y) is the age of the subject, which can be obtained through the basic information, GP73 (ng / mL) is the GP73 result, and LSM (kpa) is the test result of liver elasticity measurement.
[0016] Furthermore, the method further comprises:
[0017] Comparative test data of liver fibrosis evaluation of subjects using a selected reference non-invasive diagnostic algorithm;
[0018] Drawing a receiver operating characteristic curve based on the comparative detection data and the output of the non-invasive diagnostic model;
[0019] The difference between the diagnostic performance of the GFA algorithm and the reference non-invasive diagnostic algorithm in the area under the receiver operating characteristic curve was compared to perform a performance comparison between different algorithms.
[0020] Furthermore, after the step of drawing the receiver operating characteristic curve, data processing and analysis are performed, specifically including: continuous variable comparison; categorical variable comparison; correlation analysis; multivariate analysis.
[0021] Furthermore, the reference non-invasive diagnostic algorithm includes at least: FIB algorithm and NFS algorithm.
[0022] In a second aspect of the present invention, a system for evaluating liver fibrosis in metabolic-related fatty liver disease based on serum Golgi protein 73 is used in the method described, and the system comprises:
[0023] An evaluation mechanism setting module is used to set a liver tissue pathological evaluation mechanism. The scoring criteria of the evaluation mechanism at least include fatty degeneration score, lobular inflammation score, ballooning degeneration score, and fibrosis stage;
[0024] An evaluation data acquisition module, used to establish detection conditions for GP73 levels and obtain the detected GP73 results as first detection data, wherein GP73 represents serum Golgi protein 73 of the subject;
[0025] A model building module, used to build a non-invasive diagnostic model for the subject based on the evaluation mechanism;
[0026] An information acquisition module, used to acquire basic information of the subject and obtain the test result of liver elasticity measurement of the subject as the second test data;
[0027] The result output module is used to input the basic information, the first detection data and the second detection data into the non-invasive diagnosis model to evaluate the liver fibrosis level of the subject.
[0028] The non-invasive diagnosis model is built based on the GFA algorithm, which satisfies:
[0029] GFA=-8.757+1.048×Age(y)+1.028×GP73(ng / mL)+1.588×LSM(kpa);
[0030] Among them, Age (y) is the age of the subject, which can be obtained through the basic information, GP73 (ng / mL) is the GP73 result, and LSM (kpa) is the test result of liver elasticity measurement.
[0031] Furthermore, the system further comprises:
[0032] A comparative data acquisition module, used for selecting a reference non-invasive diagnostic algorithm to evaluate the comparative detection data of liver fibrosis in a subject;
[0033] A comparative data processing module, used for drawing a receiver operating characteristic curve according to the comparative detection data and the output of the non-invasive diagnosis model;
[0034] The algorithm performance evaluation module is used to compare the difference between the diagnostic performance of the GFA algorithm and the reference non-invasive diagnostic algorithm in the area under the receiver operating characteristic curve, so as to compare the performance between different algorithms.
[0035] The present invention provides a method for evaluating liver fibrosis in metabolic-related fatty liver disease based on serum Golgi protein 73. Compared with the prior art, the present method has the following effects: on the one hand, a non-invasive diagnostic algorithm based on Golgi protein 73 is prospectively established (implemented by a non-invasive diagnostic model); on the other hand, the correlation between the measured GP73 level and liver fibrosis can be evaluated, and on this basis, significant liver fibrosis can be identified or evaluated; finally, the clinical potential value of Golgi protein 73 in stratified analysis of the metabolic-related fatty liver disease population is revealed. BRIEF DESCRIPTION OF THE DRAWINGS
[0036] Figure 1 A flow chart of a method for evaluating liver fibrosis in metabolic-related fatty liver disease based on serum Golgi protein 73 provided in an embodiment of the present invention;
[0037] Figure 2 A structural block diagram of a system for evaluating liver fibrosis in metabolic-related fatty liver disease based on serum Golgi protein 73 provided in an embodiment of the present invention;
[0038] Figure 3 A flowchart of performing liver tissue pathological feature analysis in one embodiment;
[0039] Figure 4 A Venn diagram of the distribution of metabolic cardiovascular risk factors in one embodiment;
[0040] Figure 5 A comparison chart of GP73 levels in subjects with different fibrosis grades in one embodiment;
[0041] Figure 6The relationship between the number of combined different metabolic cardiovascular risk factors and the prevalence of significant fibrosis in one embodiment;
[0042] Figure 7 The relationship between the number of combined different metabolic cardiovascular risk factors and the prevalence of metabolic-related steatohepatitis in one embodiment;
[0043] Figure 8 is a receiver operating characteristic curve diagram of different non-invasive diagnostic algorithms for diagnosing significant liver fibrosis in the whole sample (n=228) in one embodiment;
[0044] Fig. 9 The figure is a receiver operating characteristic curve diagram of different non-invasive diagnostic algorithms for diagnosing significant liver fibrosis in a validation group (n=92) according to an embodiment. DETAILED DESCRIPTION
[0045] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.
[0046] like Figure 1 As shown, in one embodiment, a method for evaluating liver fibrosis in metabolic-related fatty liver disease based on serum Golgi protein 73 comprises steps S101 to S105;
[0047] S101. Establish a liver tissue pathological evaluation mechanism, and the scoring criteria of the evaluation mechanism at least include fatty degeneration score, lobular inflammation score, ballooning degeneration score, and fibrosis stage;
[0048] In this step, the fatty degeneration score is scored according to the proportion of liver tissue involved in the fatty degeneration of hepatocytes. No score is given for less than 5%, 5% to 33% is grade 1, 33% to 66% is grade 2, and more than 66% is grade 3. Lobular inflammation score: graded according to the degree of inflammatory cell infiltration, no inflammation is graded as grade 0, less than 2 inflammatory foci per 200 times field of view is grade 1, 2 to 4 inflammatory foci are grade 2, and more than 4 inflammatory foci are grade 3. Ballooning score: The presence and degree of ballooning are scored from 0 to 2, where grade 0 means no ballooning, and grades 1 and 2 are divided according to the number and distribution of ballooning cells. Fibrosis stage: No fibrosis is stage 0, perisinusoidal or periportal fibrosis is stage 1 (specifically divided into 1A, 1B, and 1C, but uniformly counted as grade 1 in the analysis), the coexistence of perisinusoidal and periportal fibrosis is stage 2, bridging fibrosis is stage 3, and cirrhosis is defined as stage 4.
[0049] In some scenarios, the pathological scoring process involved in the evaluation mechanism can be completed independently by three senior liver pathologists, and the final staging results can be determined by consensus to ensure the objectivity and accuracy of the evaluation.
[0050] S102, establishing detection conditions for GP73 levels, and obtaining the detected GP73 results as first detection data, wherein GP73 represents serum Golgi protein 73 of the subject;
[0051] Among them, the test conditions include the selection criteria of the subjects, such as:
[0052] The participants may include: i) being able to provide informed consent; ii) being between 18 and 75 years old; iii) having liver steatosis confirmed by abdominal ultrasound, computed tomography or other imaging techniques, and liver biopsy confirming that the proportion of hepatocyte steatosis is 5% or more; iv) meeting the diagnostic criteria of at least one of the following metabolic cardiovascular risk factors: body mass index ≥ 23 kg / m 2 Or waist circumference >90cm for men and >80cm for women; fasting blood glucose concentration ≥5.6mmo l / L, or blood glucose level 2 hours after loading ≥7.8mmo l / L, or glycosylated hemoglobin ≥5.7%, or has been diagnosed with type 2 diabetes or received diabetes treatment; blood pressure ≥130 / 85mmHg or received hypertension treatment; plasma triglyceride concentration ≥1.70mmol / L or received lipid-lowering treatment; plasma high-density lipoprotein cholesterol concentration ≤1.0mmo l / L for men and ≤1.3mmol / L for women or received lipid-lowering treatment.
[0053] The detection conditions include the following exclusion criteria: i) history of excessive drinking (average alcohol volume ≥30g / day for men and ≥20g / day for women) or history of other drugs that may cause hepatic fatty degeneration (such as corticosteroids, amiodarone and methotrexate, etc.); ii) known to have other liver diseases, including but not limited to viral hepatitis, drug-induced liver disease or autoimmune liver disease; iii) known or suspected to have malignant diseases or human immunodeficiency virus infection; iv) liver tissue specimens do not meet the requirements. Liver tissue specimens that meet the diagnostic criteria must be no less than 10 mm in length and have no less than 6 portal areas.
[0054] In some scenarios, the detection of GP73 levels can be carried out by using a double antibody sandwich enzyme-linked immunosorbent assay (ELISA) provided by a company to quantitatively detect Golgi protein 73 in serum; the kit uses a combination of anti-Golgi protein 73 monoclonal antibodies and polyclonal antibodies.
[0055] The following is a detailed description of the test process; Sample preparation: All serum samples were stored at -80°C before analysis. On the day of the test, the samples were thawed and equilibrated at room temperature for at least 30 minutes, and then gently mixed before analysis. Sample dilution: Serum samples were appropriately diluted using the sample dilution buffer provided in the kit according to the ratio recommended in the kit instructions. Mix 20 μL of serum with 50 μL of dilution buffer. Sample loading and incubation: The diluted samples were added to the microplate wells pre-coated with anti-Golgi protein 73 monoclonal antibodies. Incubate at 37°C for 30 minutes to allow Golgi protein 73 in the serum to bind to the antibodies fixed on the well wall. Washing: After the incubation, the microplate was thoroughly washed using an automatic plate washer to remove unbound substances and wash 5 times. Secondary antibody binding: 100 μL of horseradish peroxidase-labeled secondary antibody specifically targeting human Golgi protein 73 was added to each well. Incubate again at 37°C for 30 minutes to form an "antibody-Golgi protein 73-antibody" sandwich complex. Color development and reading: After thoroughly washing the wells with washing solution, add the substrate solution containing hydrogen peroxide and tetramethylbenzidine. Cover with aluminum foil and incubate for 15 minutes at room temperature to avoid light, then add the stop solution to stop the reaction. Use a microplate reader to measure the absorbance (Optical Density, OD) value of each well at a wavelength of 450nm. Calculation of results: According to the standard curve, the OD value is converted into the concentration value of Golgi protein 73 (in ng / mL). The standard curve is prepared by detecting different concentrations of recombinant Golgi protein 73 protein. At the same time, the purified recombinant Golgi protein 73 is detected as a calibration standard.
[0056] In this step, all samples can be tested in two batches, and each sample is tested twice to ensure the repeatability and reliability of the results; the coefficient of variation of the test process is kept less than 15% to ensure the accuracy of the experiment. All testing work is completed by the same experienced technician to avoid the impact of operational differences on the results, while ensuring that the clinical information of the subjects is kept confidential to the technician.
[0057] S103, constructing a non-invasive diagnostic model for the subject based on the evaluation mechanism;
[0058] In step S103, the non-invasive diagnosis model is built based on the GFA algorithm, which satisfies:
[0059] GFA=-8.757+1.048×Age(y)+1.028×GP73(ng / mL)+1.588×LSM(kpa);
[0060] Among them, Age (y) is the age of the subject, which can be obtained through the basic information, GP73 (ng / mL) is the GP73 result, and LSM (kpa) is the test result of liver elasticity measurement.
[0061] S104, obtaining basic information of the subject, and obtaining the test result of liver elasticity measurement of the subject as second test data;
[0062] The basic information of the subjects includes but is not limited to height, weight, waist circumference, blood pressure, current medications and records of any previous illnesses; based on the height and weight data, the body mass index of each participant (or subject) is calculated.
[0063] S105, inputting the basic information, the first detection data and the second detection data into the non-invasive diagnosis model to evaluate the liver fibrosis level of the subject.
[0064] In addition, in some scenarios, a series of standardized laboratory tests were performed on fresh samples from each participant to assess liver health and related disease risks. These tests included viral hepatitis indicators, autoimmune antibodies, comprehensive liver function tests (including alanine aminotransferase, total bilirubin, albumin, alkaline phosphatase, glutamyl transpeptidase, etc.), blood routine, coagulation function tests, blood lipid analysis (triglycerides, total cholesterol, high-density lipoprotein cholesterol and low-density lipoprotein), uric acid, blood sugar and glycosylated hemoglobin, etc.
[0065] For liver elasticity measurement, all participants underwent TE ( The examination was performed by an experienced operator who was blinded to the clinical data. To accommodate participants of different body sizes, the M-type probe was used for routine examinations, while for participants with a body mass index ≥ 30 kg / m 2 For participants with ≥ 100% sensitivity, the XL probe was used to ensure the accuracy of the test. The patient was in the supine position with the right arm fully abducted, and the right liver lobe was scanned through the intercostal space to obtain at least 10 valid measurements. The final result was the median of all valid measurements, and if the ratio of the interquartile range to the median was less than or equal to 30%, the result was considered as the median.
[0066] The calculation method of the traditional non-invasive diagnostic algorithm is as follows:
[0067] FIB-4 = age (years) × aspartate aminotransferase (U / L) / platelet count (10 9 / L)× alanine aminotransferase (U / L) 1 / 2 ;
[0068] NFS = -1.675 + 0.037 × age (years) + 0.094 × body mass index (kg / m 2)+1.13×DM(0 / 1)+0.99×AST / AST-0.013×platelet count(109 / L)-0.66×albumin(g / dL).
[0069] Among them, FIB-4 and NFS both represent a noninvasive diagnostic algorithm, and DM stands for diabetes mellitus.
[0070] In an example of this embodiment, the method further includes:
[0071] Comparative test data of liver fibrosis evaluation of subjects using a selected reference non-invasive diagnostic algorithm;
[0072] Drawing a receiver operating characteristic curve based on the comparative detection data and the output of the non-invasive diagnostic model;
[0073] The difference between the diagnostic performance of the GFA algorithm and the reference non-invasive diagnostic algorithm in the area under the receiver operating characteristic curve was compared to perform a performance comparison between different algorithms.
[0074] In an example of this embodiment, after the step of drawing the receiver operating characteristic curve, data processing and analysis are performed, specifically including: continuous variable comparison; categorical variable comparison; correlation analysis; multivariate analysis.
[0075] More specifically, continuous variables were expressed as mean ± SD or median and (IQR) according to whether they conformed to normal distribution; categorical variables were expressed as frequency (n) or percentage (%). Comparison of continuous variables: Student's t-test or Mann-Whitney U test was used for comparison between two groups, and analysis of variance or Kruskal-Wallis test was used for comparison between multiple groups. Comparison of categorical variables: Chi-square test or Fischer's exact test was used according to the data distribution. Correlation analysis: Pearson's correlation coefficient was used to evaluate the correlation between variables. Multivariate analysis: Based on the significant liver fibrosis suggested by liver puncture pathology, binary logistic regression analysis was performed, and variables with p < 0.10 in univariate analysis were selected to construct a multivariate algorithm by forward stepwise method. The samples were divided into a modeling group (60%, 136 cases) and a validation group (40%, 92 cases) using a random number generation method. The diagnostic cut-off value was determined by the maximum value of the Youden index. The performance characteristics of the noninvasive diagnostic algorithm were evaluated, including sensitivity, specificity, positive predictive value, and negative predictive value. The receiver operating characteristic curve was drawn using R language, and the area under the receiver operating characteristic curve and its 95% confidence interval were calculated. A two-tailed P value less than 0.05 was considered statistically significant. All analyses were completed using SPSS25.0, Medcalc for Windows version 22, and GraphPad Prism 9.3.1 software. Sample size calculation: This study is a single-sample diagnostic test. In order to reduce the number of patients undergoing invasive procedures, the expected specificity is 80%, the allowable error is 0.1, a two-sided test is required, α=0.05, and the confidence level is 1-β=0.9. The sample size formula for diagnostic studies is as follows:
[0076]
[0077] Among them, Z 1-α / 2 and Z 1-β The results in the table are 1.96 and 1.28 respectively. p represents the expected specificity, p 0 In this case, the expected specificity p is 0.8, and the acceptable minimum specificity p is 0 The Ratio of α to β was 0.7, and considering the 10% loss to follow-up rate, the final calculation required at least 207 cases to be included as sample objects.
[0078] In one example of this embodiment, Figure 3 As shown in the figure, the pathological analysis data were carried out in Beijing Ditan Hospital and Beijing Youan Hospital between May 2018 and January 2024, and a total of 379 patients (i.e., subjects) signed informed consent and completed liver puncture biopsy. Figure 3After preliminary screening, 91.3% (346) of the patients initially met the diagnostic criteria for metabolic-related fatty liver disease. We further excluded 101 patients diagnosed with chronic hepatitis B, 3 with drug-induced liver injury, 1 with autoimmune hepatitis, and 7 patients with excessive daily alcohol consumption. In addition, 6 patients who could not undergo Golgi protein 73 testing due to insufficient serum samples were also excluded. Finally, 228 patients with metabolic-related fatty liver disease were selected as research subjects. Among these 228 patients, 57.5% (131) were male, such as Figure 4 As shown, the most common metabolic cardiovascular risk factor was excessive body mass index or waist circumference. Fifty-four patients (23.7%) had at least one metabolic cardiovascular risk factor, and 103 patients (45.7%) had 3-5 metabolic cardiovascular risk factors. Figure 4 In the above data, BMI stands for body mass index; HT stands for hypertension; DM stands for diabetes mellitus (eg, type 2 diabetes); TG stands for triglycerides; and HDL stands for high-density lipoprotein cholesterol.
[0079] In one example of this embodiment, Figures 5 to 7 As shown, the correlation analysis specifically includes: correlation analysis between fibrosis degree and clinical characteristics, serum Golgi protein 73 level; and analysis of liver histological pathological characteristics.
[0080] Figure 5 This is a comparison chart of GP73 levels in subjects with different fibrosis grades in one embodiment; according to the degree of liver tissue fibrosis, the subjects were divided into two groups: significant fibrosis (F≥2) and non-significant fibrosis (F0-1) for comparative analysis. The results in Table 1 show that the average age of patients with significant fibrosis is higher (45 vs. 38 years old, p = 0.025), and the proportion of women is also higher (49.4% vs. 38.5%, p = 0.017). In addition, Figure 5 As shown in the results, serum Golgi protein 73 levels in patients with significant fibrosis were significantly higher than those in patients with non-significant fibrosis (128 ng / mL vs. 46 ng / mL, p<0.001). However, the differences between the two groups in transaminase and lipid parameters did not reach statistical significance. The correlation coefficient between serum Golgi protein 73 levels and NAS was 0.269 (p<0.001).
[0081] Table 1 Comparison of clinical characteristics of patients with metabolic-related fatty liver disease at different fibrosis grades
[0082]
[0083] Here, Total stands for total cholesterol (total cholesterol).
[0084] As mentioned above, there were 143 cases with liver fibrosis grade F0-1, accounting for 62.7%; 49 cases with F2 grade, accounting for 21.5%; 36 patients with advanced liver fibrosis (F3-4 grade), accounting for 15.8%; and the proportion of patients diagnosed with significant liver fibrosis (F2-4 grade) was 37.3%. A total of 80 cases were diagnosed with metabolic-related fatty hepatitis, accounting for 35.1%. The study found that there was a statistical correlation between the diagnosis of metabolic-related fatty hepatitis and significant liver fibrosis, with a Pearson correlation coefficient of 0.402 (p<0.001). The area under the receiver operating characteristic curve for metabolic-related fatty hepatitis for significant liver fibrosis was 0.699 (95CI: 0.626-0.772).
[0085] Figure 6 The relationship between the number of combined different metabolic cardiovascular risk factors and the prevalence of significant fibrosis in one embodiment; Figure 7 The figure shows the relationship between the number of combined different metabolic cardiovascular risk factors and the prevalence of metabolic-associated steatohepatitis in an embodiment; in the figure, CMRF: metabolic cardiovascular risk factor; MASH: metabolic-associated steatohepatitis. Patients can be grouped according to the number of metabolic cardiovascular risk factors: single metabolic cardiovascular risk factor, dual metabolic cardiovascular risk factors, and 3 or more metabolic cardiovascular risk factors (metabolic cardiovascular risk factors ≥ 3) to explore the relationship between the number of risk factors and the prevalence of significant liver fibrosis; the results are shown in FIG. Figure 6 As shown in Figure 2, with the increase of metabolic cardiovascular risk factors, the prevalence of significant liver fibrosis was 35.2%, 36.6%, and 38.8%, respectively, but this difference was not statistically significant (p = 0.895); on the contrary, Figure 7 As shown in the data, the prevalence of metabolic-related steatohepatitis increased significantly with the increase in the number of risk factors, which were 20.4%, 23.9%, and 50.5%, respectively, and the difference was statistically significant (p<0.001).
[0086] In an example of an embodiment, 136 patients who underwent liver puncture biopsy were randomly selected as the modeling group, and significant liver fibrosis was used as the positive result of the study. The related factors of significant liver fibrosis were screened out through univariate and multivariate logistic regression analysis. As shown in Table 2, the patient's age, liver elasticity measurement, and serum Golgi protein 73 level are all independent predictors of significant liver fibrosis. Based on the results of multivariate analysis, a non-invasive diagnostic algorithm, namely the GFA algorithm, was constructed, which combines the information of Golgi protein 73, liver elasticity measurement, and age.
[0087] Table 2 shows the results of univariate and multivariate logistic regression analysis of significant liver fibrosis
[0088]
[0089] In Table 2, p-value represents the assumed value or the hypothesized probability.
[0090] In one example of this embodiment, Figure 8 is a receiver operating characteristic curve diagram of different non-invasive diagnostic algorithms for diagnosing significant liver fibrosis in the whole sample (n=228) in one embodiment; Fig. 9 The figure is a receiver operating characteristic curve diagram of different non-invasive diagnostic algorithms for diagnosing significant liver fibrosis in a validation group (n=92) in one embodiment. The reference non-invasive diagnostic algorithm at least includes: FIB algorithm and NFS algorithm.
[0091] In this example, the GFA algorithm based on liver elasticity, liver elasticity measurement values, and the serology-based FI B-4 algorithm and NFS algorithm levels were compared in the entire sample, and the receiver operating characteristic curve was plotted for analysis ( Figure 8 , Table 3). The difference between the diagnostic performance of the GFA algorithm (area under the receiver operating characteristic curve: 0.860, 95% confidence interval: 0.811-0.909) and the area under the receiver operating characteristic curve of liver elasticity measurement (increment 0.070) did not reach statistical significance (p = 0.301). However, the diagnostic performance of the GFA algorithm was significantly better than the area under the receiver operating characteristic curve of the FI B-4 algorithm and the NFS algorithm (p < 0.001). The comparison in the validation group (92 / 228) reached the same conclusion ( Fig. 9 , Table 4), GFA was superior to liver elasticity measurement, FI B-4 algorithm, and NFS algorithm.
[0092] Table 3 shows the performance comparison of different algorithms in diagnosing significant liver fibrosis in the whole sample
[0093]
[0094] Table 4 shows the performance comparison of different algorithms in diagnosing significant liver fibrosis in the validation group
[0095]
[0096] In summary, it was found that elevated serum Golgi protein 73 levels were well correlated with metabolic-related fatty liver hepatitis, and the diagnosis of metabolic-related fatty liver hepatitis was well consistent with significant liver fibrosis (69.9%). Serum Golgi protein 73 was elevated in patients with severe liver inflammation.
[0097] Existing serological markers for liver fibrosis are divided into two types, indirect and direct. Indirect serum markers include a combination of routine laboratory parameters and demographic data such as age. Among them, the FIB-4 has been widely validated and used, with an area under the receiver operating characteristic curve of 0.7-0.8. Recently, some scholars have established a method based on Agile 3+ and 4 scores, which combine liver elasticity determination with inherent demographic data (age, gender and the presence of type 2 diabetes) and serum biomarkers (aspartate aminotransferase, alanine aminotransferase and platelets) to better diagnose advanced fibrosis and cirrhosis in NAFLD, with a diagnostic efficacy of 0.85 under the receiver operating characteristic curve, which is similar to the diagnostic efficacy of liver elasticity (area under the receiver operating characteristic curve 0.83, p = 0.142). This algorithm is similar to the GFA algorithm explored in the present invention, and both are significantly better than FI B-4, but its calculation method is more complicated and includes more reference indicators, which limits its application. The GFA algorithm can be performed with a single serological and liver elasticity determination, and is easy to operate. Serum Golgi protein 73 can be measured using a commercial ELISA kit, which can be established in hospitals or clinics with biochemical laboratories. The price of a single test is only about RMB 100, and it can be performed simultaneously with other liver function tests. The GFA algorithm only requires the patient's age, liver elasticity measurement, and Golgi protein 73 level. It is low-cost and has a high accuracy in predicting significant liver fibrosis. Dynamic monitoring may also be able to predict the occurrence of liver cancer. Further cost-effectiveness and prognosis prediction effects can be carried out through quantitative follow-up studies.
[0098] In conclusion, serum Golgi protein 73 is an independent risk factor for significant liver fibrosis in patients with metabolic-related fatty liver disease. The GFA algorithm based on Golgi protein 73 combined with liver elasticity measurement has a good diagnostic efficacy for significant liver fibrosis and can effectively reduce the need for liver puncture pathological biopsy in patients with metabolic-related fatty liver disease.
[0099] In another embodiment, Figure 2As shown, a system for evaluating liver fibrosis in metabolic-related fatty liver disease based on serum Golgi protein 73 is used in the method described, and the system comprises:
[0100] An evaluation mechanism setting module 100 is used to set a liver tissue pathology evaluation mechanism, wherein the scoring criteria of the evaluation mechanism at least include a fatty degeneration score, a lobular inflammation score, a ballooning degeneration score, and a fibrosis stage;
[0101] An evaluation data acquisition module 200 is used to establish detection conditions for GP73 levels and obtain the detected GP73 results as first detection data, wherein GP73 represents serum Golgi protein 73 of the subject;
[0102] A model building module 300, for building a non-invasive diagnostic model of the subject based on the evaluation mechanism;
[0103] The information acquisition module 400 is used to acquire basic information of the subject and obtain the test result of liver elasticity measurement of the subject as the second test data;
[0104] The result output module 500 is used to input the basic information, the first detection data and the second detection data into the non-invasive diagnosis model to evaluate the liver fibrosis level of the subject.
[0105] The non-invasive diagnosis model is built based on the GFA algorithm, which satisfies:
[0106] GFA=-8.757+1.048×Age(y)+1.028×GP73(ng / mL)+1.588×LSM(kpa);
[0107] Among them, Age (y) is the age of the subject, which can be obtained through the basic information, GP73 (ng / mL) is the GP73 result, and LSM (kpa) is the test result of liver elasticity measurement.
[0108] In an example of this embodiment, the system further includes:
[0109] A comparative data acquisition module, used for selecting a reference non-invasive diagnostic algorithm to evaluate the comparative detection data of liver fibrosis in a subject;
[0110] A comparative data processing module, used for drawing a receiver operating characteristic curve according to the comparative detection data and the output of the non-invasive diagnosis model;
[0111] The algorithm performance evaluation module is used to compare the difference between the diagnostic performance of the GFA algorithm and the reference non-invasive diagnostic algorithm in the area under the receiver operating characteristic curve, so as to compare the performance between different algorithms.
[0112] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A method for evaluating liver fibrosis in metabolic-related fatty liver disease based on serum Golgi protein 73. It is characterized in that The method comprises: Establish a liver tissue pathological evaluation mechanism, the scoring criteria of which at least include fatty degeneration score, lobular inflammation score, ballooning degeneration score, and fibrosis stage; Establishing detection conditions for GP73 levels, and obtaining the detected GP73 results as first detection data, wherein GP73 represents serum Golgi protein 73 of the subject; Based on the evaluation mechanism, construct a non-invasive diagnostic model for the subject; Obtaining basic information of the subject, and obtaining the test result of liver elasticity measurement of the subject as the second test data; The basic information, the first detection data and the second detection data are input into the non-invasive diagnosis model to evaluate the liver fibrosis level of the subject.
2. The method according to claim 1, It is characterized in that The non-invasive diagnosis model is built based on the GFA algorithm, which satisfies: GFA=-8.757+1.048×Age(y)+1.028×GP73(ng / mL)+1.588×LSM(kpa); Among them, Age (y) is the age of the subject, which can be obtained through the basic information, GP73 (ng / mL) is the GP73 result, and LSM (kpa) is the test result of liver elasticity measurement.
3. The method according to claim 2, It is characterized in that The method further comprises: Comparative test data of liver fibrosis evaluation of subjects using a selected reference non-invasive diagnostic algorithm; Drawing a receiver operating characteristic curve based on the comparative detection data and the output of the non-invasive diagnostic model; The difference between the diagnostic performance of the GFA algorithm and the reference non-invasive diagnostic algorithm in the area under the receiver operating characteristic curve was compared to perform a performance comparison between different algorithms.
4. The method according to claim 3, It is characterized in that After the step of drawing the receiver operating characteristic curve, data processing and analysis are performed, specifically including: continuous variable comparison; categorical variable comparison; correlation analysis; multivariate analysis.
5. The method according to claim 3, It is characterized in that The reference non-invasive diagnostic algorithms include at least: FIB algorithm and NFS algorithm.
6. A system for evaluating liver fibrosis in metabolic-related fatty liver disease based on serum Golgi protein 73, used in the method according to any one of claims 1 to 5, It is characterized in that The system comprises: An evaluation mechanism setting module is used to set a liver tissue pathological evaluation mechanism. The scoring criteria of the evaluation mechanism at least include fatty degeneration score, lobular inflammation score, ballooning degeneration score, and fibrosis stage; An evaluation data acquisition module, used to establish detection conditions for GP73 levels and obtain the detected GP73 results as first detection data, wherein GP73 represents serum Golgi protein 73 of the subject; A model building module, used to build a non-invasive diagnostic model for the subject based on the evaluation mechanism; An information acquisition module, used to acquire basic information of the subject and obtain the test result of liver elasticity measurement of the subject as the second test data; The result output module is used to input the basic information, the first detection data and the second detection data into the non-invasive diagnosis model to evaluate the liver fibrosis level of the subject.
7. The system according to claim 6, It is characterized in that in, The non-invasive diagnosis model is built based on the GFA algorithm, which satisfies: GFA=-8.757+1.048×Age(y)+1.028×GP73(ng / mL)+1.588×LSM(kpa); Among them, Age (y) is the age of the subject, which can be obtained through the basic information, GP73 (ng / mL) is the GP73 result, and LSM (kpa) is the test result of liver elasticity measurement.
8. The system according to claim 6, It is characterized in that The system further comprises: A comparative data acquisition module, used for selecting a reference non-invasive diagnostic algorithm to evaluate the comparative detection data of liver fibrosis in a subject; A comparative data processing module, used for drawing a receiver operating characteristic curve according to the comparative detection data and the output of the non-invasive diagnosis model; The algorithm performance evaluation module is used to compare the difference between the diagnostic performance of the GFA algorithm and the reference non-invasive diagnostic algorithm in the area under the receiver operating characteristic curve, so as to compare the performance between different algorithms.
