WO2023151651A1 - 非酒精性脂肪性肝炎生物标志物组合物及其应用 - Google Patents

非酒精性脂肪性肝炎生物标志物组合物及其应用 Download PDF

Info

Publication number
WO2023151651A1
WO2023151651A1 PCT/CN2023/075430 CN2023075430W WO2023151651A1 WO 2023151651 A1 WO2023151651 A1 WO 2023151651A1 CN 2023075430 W CN2023075430 W CN 2023075430W WO 2023151651 A1 WO2023151651 A1 WO 2023151651A1
Authority
WO
WIPO (PCT)
Prior art keywords
cut
value
biomarker composition
steatohepatitis
alcoholic
Prior art date
Application number
PCT/CN2023/075430
Other languages
English (en)
French (fr)
Inventor
于君
黄炜燊
张翔
赵贵君
Original Assignee
香港中文大学深圳研究院
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 香港中文大学深圳研究院 filed Critical 香港中文大学深圳研究院
Publication of WO2023151651A1 publication Critical patent/WO2023151651A1/zh

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
    • G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
    • G01N33/68—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving proteins, peptides or amino acids
    • G01N33/6893—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving proteins, peptides or amino acids related to diseases not provided for elsewhere
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
    • G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
    • G01N33/53—Immunoassay; Biospecific binding assay; Materials therefor
    • G01N33/576—Immunoassay; Biospecific binding assay; Materials therefor for hepatitis
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
    • G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
    • G01N33/68—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving proteins, peptides or amino acids
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N2800/00—Detection or diagnosis of diseases
    • G01N2800/08—Hepato-biliairy disorders other than hepatitis
    • G01N2800/085—Liver diseases, e.g. portal hypertension, fibrosis, cirrhosis, bilirubin

Definitions

  • the invention relates to the field of biological detection, in particular to a nonalcoholic steatohepatitis biomarker composition and application thereof.
  • Non-alcoholic fatty liver disease is closely related to obesity, type 2 diabetes and metabolic syndrome. Due to the prevalence of sedentary and less active lifestyles and high-calorie diets, the incidence of NAFLD has been increasing year by year, and it has become the most common liver disease in the world. The average incidence of NAFLD in Asia is 15%-40 %, the incidence rate in my country's cities is as high as 43.3%. The pathological progression of NAFLD is from simple steatosis to non-alcoholic steatohepatitis (NASH), leading to progressive liver fibrosis/cirrhosis, and finally to liver cancer.
  • NASH non-alcoholic steatohepatitis
  • the apoptosis marker cytokeratin 18 (CK-18) fragment is the most commonly validated blood marker, but the sensitivity is only 66% and the specificity is 82% when applied alone; adiponectin, leptin and fiber Protocellular growth factor (FGF21) is only related to abnormal metabolism of the body, and most of them have only been verified in the population undergoing bariatric surgery, and the accuracy of NASH diagnosis is low; lysosomal enzymes (Cathepsin D) etc. The difference is large and cannot be verified. Due to the limited accuracy of single factors in the diagnosis of NASH, it is urgent to establish a marker composition to improve the accuracy of diagnosis.
  • FGF21 Protocellular growth factor
  • hepatic fibrosis is the most important predictor of adverse clinical outcome of the liver, and it is a high-risk NASH.
  • the FDA recommends that the focus of drug development be on NASH with liver fibrosis, which has great health needs and potential benefits.
  • the development of non-invasive diagnostic markers and kits that can diagnose high-risk patients with NASH and liver fibrosis is of great help in finding needed drugs. It is of great significance to monitor the efficacy of drug intervention in patients and clinical medicine.
  • the aim of the present invention is to provide a biomarker composition which shows high sensitivity, high specificity, high positive predictive value and high negative predictive in differentiating non-alcoholic steatohepatitis from simple steatosis value.
  • Non-alcoholic steatohepatitis described in the present invention can adopt three different classification criteria, including:
  • NASH CRN Nonalcoholic Steatohepatitis Research Network
  • NASH CRN Nonalcoholic Steatohepatitis Research Network
  • Another object of the present invention is to provide a diagnostic model of nonalcoholic fatty liver disease and/or a diagnostic model of nonalcoholic steatohepatitis.
  • the non-alcoholic steatohepatitis diagnostic model of the present invention can also compare the risk of non-alcoholic steatohepatitis.
  • Another object of the present invention is to provide an artificial intelligence model capable of diagnosing nonalcoholic fatty liver disease and/or nonalcoholic steatohepatitis according to the level of the biomarker composition in the test sample.
  • Another object of the present invention is to provide a kit for diagnosing nonalcoholic steatohepatitis.
  • a biomarker composition includes related protein markers and/or clinical biochemical markers for the diagnosis of nonalcoholic fatty liver disease and/or nonalcoholic steatohepatitis, the diagnosis
  • the relevant protein markers of nonalcoholic fatty liver disease and/or nonalcoholic steatohepatitis are selected from CXC chemokine ligand 10 (CXCL10), cytokeratin 18 (CK-18), selective autophagy adapter protein ( P62/SQSTM1), carbonic anhydrase III (CA3), squalene epoxidase (SQLE), type III/IV collagen precursor (Pro-C3), or fibroblast growth factor (FGF21) or multiple;
  • the clinical biochemical markers are selected from body mass index (BMI), glycosylated hemoglobin (HbA1c), alanine aminotransferase (ALT), aspartate aminotransferase (AST), low-density lipoprotein cholesterol (LDL-C), total One or more of cholesterol (TC), t
  • the biomarker composition includes at least two protein markers related to the diagnosis of nonalcoholic fatty liver disease and/or nonalcoholic steatohepatitis and/or high-risk nonalcoholic steatohepatitis.
  • said diagnosis of non-alcoholic fatty liver disease and/or non-alcoholic fatty liver disease from the test subject is measured respectively
  • Levels of protein markers associated with hepatitis and/or high-risk non-alcoholic steatohepatitis and the levels of the clinical biochemical markers were then used to establish the optimal marker composition.
  • the biomarker composition is composed of 5 protein markers and 4 clinical biochemical markers (called N9-NASH), including CXCL10, CK-18, P62/SQSTM1, ALT, SQLE, HbA1c , FGF21, PLT and LDL-C.
  • N9-NASH clinical biochemical markers
  • NAFLD nonalcoholic fatty liver disease
  • the biomarker composition consists of 3 protein markers and 2 clinical biochemical markers (called N5-NASH), including CXCL10, CK-18, Pro-C3, AST and BMI.
  • N5-NASH 2 clinical biochemical markers
  • the biomarker composition (referred to as N3-NASH) consists of CXCL10, CK-18 and BMI.
  • the biomarker composition (referred to as N2-Fibrosis) consists of CXCL10 and Pro-C3.
  • N2-Fibrosis can further distinguish high-risk patients with fibrosis in generalized nonalcoholic steatohepatitis.
  • biomarker composition as described above for differentiating non-alcoholic steatohepatitis from simple steatosis.
  • a non-alcoholic fatty liver disease diagnostic model and/or non-alcoholic steatohepatitis diagnostic model its construction method is: collecting non-alcoholic fatty liver disease patients (including non-alcoholic steatohepatitis patients and simple steatosis patients , wherein non-alcoholic steatohepatitis patients include high-risk and low-risk non-alcoholic steatohepatitis patients) blood samples, respectively measure the diagnosis of non-alcoholic fatty liver disease and / or non-alcoholic steatohepatitis and / or The levels of protein markers related to high-risk non-alcoholic steatohepatitis and/or the levels of clinical biochemical markers, the levels of markers in patients with non-alcoholic fatty liver disease and healthy subjects were used support vector machine, logistic regression , Naive Bayesian and 10-fold cross-validation to establish a non-alcoholic fatty liver disease diagnostic model, and/or, the level of markers in patients with non-alcoholic steatohepatitis and patients with simple steatosis using support vector
  • the relevant protein markers for diagnosing nonalcoholic fatty liver disease and/or nonalcoholic steatohepatitis are selected from one of CXCL10, CK-18, P62/SQSTM1, SQLE, CA3, Pro-C3 or FGF21 or more; the clinical biochemical markers are selected from one or more of HbA1c, AST, BMI, ALT, LDL-C, TG, TC, ALP or PLT.
  • the biomarker composition used in the nonalcoholic fatty liver disease diagnostic model is N9-NASH.
  • the biomarker composition used in the nonalcoholic steatohepatitis diagnostic model is N9-NASH, N5-NASH, N3-NASH or N2-Fibrosis.
  • An artificial intelligence model for diagnosing nonalcoholic fatty liver disease and/or an artificial intelligence model for diagnosing nonalcoholic steatohepatitis according to the level of the biomarker composition in the test sample is provided.
  • a kit comprising detection reagents for detecting the level of the biomarker composition.
  • the kit further includes a standard, and the standard includes the biomarker composition.
  • the kit is used to distinguish non-alcoholic fatty liver disease from healthy subjects, and the kit includes the biomarker for detecting the A detection reagent for the level of the marker composition.
  • the kit is used to distinguish non-alcoholic steatohepatitis from simple steatosis, and the kit includes a method for detecting the A detection reagent for the level of the biomarker composition.
  • the kit is used to distinguish high-risk non-alcoholic steatohepatitis from low-risk non-alcoholic steatohepatitis, and the kit includes A detection reagent for detecting the level of the biomarker composition.
  • the kit is useful in distinguishing non-alcoholic fatty liver disease from healthy persons, and/or, distinguishing non-alcoholic steatohepatitis from simple steatosis, and/or, distinguishing high-risk non-alcoholic steatohepatitis from low-risk non-alcoholic steatohepatitis Use in alcoholic steatohepatitis.
  • the method for distinguishing is: providing a sample derived from the subject to be tested, and testing the relevant protein markers for diagnosing nonalcoholic fatty liver disease and/or nonalcoholic steatohepatitis in the samples respectively and the levels of the clinical biochemical markers, and then substituted into support vector machine, logistic regression, naive Bayesian and 10-fold cross-validation to determine the best combination of biomarkers, and ROC curve analysis, in nonalcoholic fatty liver disease
  • the diagnostic model non-alcoholic fatty liver disease and healthy people were distinguished according to the cut-off value; in the non-alcoholic steatohepatitis diagnostic model, the cut-off value was used to distinguish Non-alcoholic steatohepatitis and simple steatosis; in the high-risk non-alcoholic steatohepatitis diagnostic model, according to the Cut-off value to distinguish high-risk and low-risk non-alcoholic steatohepatitis patients.
  • the cut-off critical value is used as a positive judgment value, ⁇ cut-off critical value is judged as positive, and ⁇ cut-off critical value is judged as negative.
  • the cut-off critical value in the non-alcoholic fatty liver disease diagnostic model is preferably 0.58; the preferred cut-off critical value in the non-alcoholic steatohepatitis diagnostic model is 0.47 (The models for diagnosing nonalcoholic fatty liver disease and nonalcoholic steatohepatitis are different machine learning models).
  • the biomarker composition is N9-NASH
  • the Cut-off critical value is 0.58
  • non-alcoholic fatty liver disease and healthy controls can be distinguished The sensitivity and specificity are the highest, if the Cut-off value ⁇ 0.58, it is judged as non-alcoholic fatty liver disease, and the Cut-off value ⁇ 0.58 is judged as healthy;
  • the biomarker composition is N9-NASH
  • the Cut-off critical value is 0.47
  • non-alcoholic steatohepatitis can be distinguished from simple Steatosis is the best. If the Cut-off value ⁇ 0.47, it is diagnosed as non-alcoholic steatohepatitis, and if the Cut-off value is ⁇ 0.47, it is diagnosed as simple steatosis.
  • the cut-off critical values in the nonalcoholic steatohepatitis diagnostic model are preferably 0.27 and 0.61.
  • the biomarker composition is N5-NASH
  • the non-alcoholic steatohepatitis diagnostic model if the Cut-off value is ⁇ 0.61, it is diagnosed as non-alcoholic steatohepatitis, if Cut-off value ⁇ 0.27 is diagnosed as simple steatosis. If 0.27 ⁇ Cut-off value ⁇ 0.61, liver biopsy is required.
  • the Cut-off critical values in the nonalcoholic steatohepatitis diagnostic model are preferably 0.43 and 0.68.
  • the biomarker composition is N3-NASH
  • the non-alcoholic steatohepatitis diagnostic model if the Cut-off value is ⁇ 0.68, it is diagnosed as non-alcoholic steatohepatitis, if Cut-off value ⁇ 0.43, it is diagnosed as simple steatosis. If 0.43 ⁇ Cut-off value ⁇ 0.68, liver biopsy is required.
  • the cut-off critical values in the high-risk nonalcoholic steatohepatitis diagnostic model are preferably 0.37 and 0.30.
  • the biomarker composition is N2-Fibrosis
  • the cut-off value ⁇ 0.37
  • it is diagnosed as high-risk non-alcoholic steatohepatitis , three level management.
  • the cut-off value is ⁇ 0.30
  • it is diagnosed as low-risk non-alcoholic steatohepatitis, and only primary lifestyle intervention is required.
  • 0.30 ⁇ Cut-off value ⁇ 0.37 liver biopsy is required.
  • the sample is blood.
  • the sample is serum.
  • said sample is from a human.
  • Candidate biomarkers of the invention were selected by ranking the importance of protein markers and other clinical variables identified in our previous studies using varImp.
  • the present invention has the following advantages:
  • the present invention identifies a biomarker composition that enables the non-invasive diagnosis of non-alcoholic fatty liver disease and/or non-alcoholic steatohepatitis, the biomarker composition being useful in distinguishing non-alcoholic steatohepatitis from simple steatosis
  • the biomarker composition of the present invention can also distinguish high-risk and low-risk non-alcoholic steatohepatitis. Steatohepatitis.
  • the biomarker composition of the present invention can be detected by widely used methods and can be easily applied in clinical practice. The diagnostic performance of the biomarker compositions of the invention is not affected by age, sex or metabolic state.
  • Figure 1A is a comparison chart of CXCL10, SQLE, CK-18, P62/SQSTM1, FGF21 and CA3 levels in the serum of NAFLD patients and healthy controls;
  • Figure 1B is a comparison chart of CXCL10, SQLE, CK-18, P62/SQSTM1, FGF21 and CA3 levels in the serum of NASH patients and healthy controls;
  • Figure 2 is a chart showing the importance of 17 markers in the diagnosis of NASH
  • Figure 3 is a comparison chart of the performance of N9-NASH and a single biomarker in the diagnosis of healthy controls and NAFLD, as well as in the diagnosis of NASH in NAFLD patients;
  • Figure 4 is a comparison chart of the performance of N9-NASH and a single biomarker in the diagnosis of NASH in the training cohort.
  • Figure 5 is a comparison chart of the performance of N9-NASH and a single biomarker in the diagnosis of NASH in the validation cohort.
  • Figure 6 is a chart showing the importance of markers for the diagnosis of generalized NASH
  • Figure 7 shows the performance of N5-NASH for NASH diagnosis in the discovery and validation cohorts.
  • Figure 8 shows the performance of N3-NASH for NASH diagnosis in the discovery and validation cohorts.
  • Figure 9 shows the two-step method of N3-NASH and N2-Fibrosis to diagnose high-risk NASH.
  • the inventor identified some protein biomarkers related to NASH through preliminary work, combined with clinical biochemical indicators that can identify NAFLD and NASH patients, introduced 24 variables, including 7 serum proteins (CXCL10, SQLE, CK-18, P62/SQSTM1, FGF21, Pro-C3 and CA3) and 17 clinical variables (14 continuous variables: (age, body mass index (BMI), platelets (PLT), serum albumin (Alb), serum alkaline phosphatase level (ALP), serum alanine aminotransferase (ALT), aspartate aminotransferase (AST), bilirubin (bilirubin), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C ), triglycerides (TG), low-density lipoprotein cholesterol (LDL-C), glycosylated hemoglobin (HbA1c)), fasting blood glucose (FBS) and 3 discrete variables (gender, hypertension, diabetes)) to train
  • BMI body mass index
  • PHT serum album
  • liver biopsies are performed in patients with NAFLD primarily because of their abnormal liver function tests and/or abnormal imaging findings. Exclusion criteria include: 1) the daily alcohol consumption of men exceeds 30g, and that of women exceeds 20g; 2) positive hepatitis B surface antigen or anti-HCV antibody, antinuclear antibody titer > 1/160; 3) secondary hepatic steatosis or other Patients with histological features of liver disease. Control subjects were randomly selected from government census databases.
  • Proton magnetic resonance spectroscopy (1H-MRS) was performed to quantify liver triglyceride content in subjects who consented to participate in the study. Control subjects were excluded if: 1) liver triglyceride content exceeded 5%; 2) history of diabetes or hypertension; 3) patients with NAFLD were excluded. Fasting venous blood samples were collected the day before liver biopsy. All patients provided written informed consent to participate in the trial and to collect blood samples specifically for the biomarker study.
  • the independent validation cohort samples were obtained from 217 patients with NAFLD confirmed by liver biopsy.
  • Liver pathology was obtained from ultrasound-guided percutaneous liver biopsy using 16G or 18G needles. After excluding subjects with unexplained histopathology and other causes of liver injury, a total of 201 NAFLD patients were included.
  • NAFLD activity was obtained by calculating the steatosis score + inflammation score + ballooning score in pathology.
  • Liver fibrosis was defined as: grade 0, no fibrosis; grade 1, perisinusoidal or portal fibrosis; grade 2, perisinusoidal/periportal fibrosis; grade 3, bridging fibrosis; and grade 4, cirrhosis.
  • NASH Nonalcoholic Steatohepatitis Research Network
  • Serum was taken out from -80°C for detection of biomarkers. Serum levels of CK-18, FGF21, CXCL10, P62/SQSTM1, SQLE, and CA3 were tested by M30 enzyme-linked immunosorbent assay (ELISA) kit (PEVIVA, Sweden), FGF21ELISA kit (BioVendor, Czech Republic), CXCL10ELISA reagent Kit (R&D Company), P62/SQSTM1ELISA Kit (Cloud ⁇ Clone, USA), SQLE ELISA Kit (Cloud ⁇ Clone, USA), Pro-C3ELISA Kit (Nordic Bioscience, Denmark) and CA3ELISA Kit (Cloud ⁇ Clone , United States) for testing.
  • ELISA enzyme-linked immunosorbent assay
  • Clinical variables included 14 continuous variables (age, body mass index (BMI), platelets (PLT), serum albumin (Alb), serum alkaline phosphatase level (ALP), serum alanine aminotransferase (ALT), aspartate aminotransferase (AST), bilirubin (bilirubin), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), triglycerides (TG), low-density lipoprotein cholesterol (LDL-C), glycated hemoglobin (HbA1c) , fasting blood glucose (FBS)) and 3 discrete variables (sex, hypertension, diabetes).
  • BMI body mass index
  • PHT platelets
  • Album serum albumin
  • ALP serum alkaline phosphatase level
  • ALT serum alanine aminotransferase
  • AST serum alanine aminotransferase
  • AST serum alanine aminotransferase
  • AST serum alanine aminotransfera
  • Correlation analyzes were performed on the covariates with the greatest correlation (absolute value of the Spearman correlation coefficient, abs(rho) ⁇ 0.5). The importance of each variable is ordered by varImp. Key biomarkers and clinical variables were selected by recursive feature elimination. The best diagnostic biomarkers for NASH were identified by using support vector machines, neural networks, naive Bayesian logistic regression and 10-fold cross-validation to build diagnostic models, the selection criteria of markers were checked by Wilcoxon rank sum test.
  • surrogate variable analysis was used in data preprocessing.
  • the values in the experiments are expressed as mean ⁇ standard deviation (SD). Differences in protein levels or clinical variables were determined by the Mann–Whitney U test. A P value of 0.05 was considered statistically significant.
  • the Spearman correlation coefficient was used to assess the association between biomarker levels and related factors, which was used for multiple regression analysis to identify independent factors associated with biomarkers. DeLong test was used to compare the different subgroups AUROC value.
  • NASH-associated proteins CXCL10, CK-18, P62/SQSTM1, SQLE, and FGF21
  • HbA1c, ALT, LDL-C, and PLT blood biochemical variables
  • N9-NASH could distinguish NAFLD patients from healthy controls with an AUROC of 0.999 (95% CI 0.997-1.000) (Fig. 3A), and N9-NASH outperformed individual protein biomarkers (AUROC 0.68-0.87, all P ⁇ 0.01) or clinical variables (AUROC 0.55-0.87, all P ⁇ 0.0001) were superior (Fig. 3B).
  • the sensitivity and specificity of N9-NASH for the diagnosis of NAFLD were 100% and 100%, respectively. 97.2%.
  • N9-NASH in diagnosing NASH in NAFLD patients.
  • NAFLD patients could be identified by N9-NASH with an AUROC of 0.94 (95% CI 0.90-0.99) (Fig. 4A).
  • N9-NASH was significantly more accurate in distinguishing NASH from patients with isolated steatosis: CXCL10 (AUROC 0.66, 95% CI 0.56-0.75, P ⁇ 0.0001), CK-18 (AUROC 0.63 , 95% CI 0.53-0.73, P ⁇ 0.0001), SQLE (AUROC 0.60, 95% CI 0.50-0.70, P ⁇ 0.0001), FGF21 (AUROC 0.48, 95% CI 0.38-0.58, P ⁇ 0.0001), CA3 ( AUROC 0.43, 95%CI 0.32-0.53, P ⁇ 0.0001), or P62/SQSTM1 (AUROC 0.61, 95%CI 0.51-0.71, P ⁇ 0.0001) ( Figure 4A-B); or individual clinical variables (AUROC 0.52-0.63 , all P ⁇ 0.0001).
  • Figure 4C shows the sensitivity, specificity, NPV, and PPV of N9-NASH for NASH diagnosis in the training cohort.
  • N9-NASH had 91.9% sensitivity, 90.5% accuracy, 92.1% NPV and 90.4% PPV (Fig. 4B).
  • N9-NASH performed well in distinguishing NAFLD patients from healthy controls with an AUROC of 0.989 (95% CI 0.976-1.000) (Fig. 3C-D).
  • the specificity was 100%
  • the sensitivity of N9-NASH in the diagnosis of NAFLD was 93.7%
  • the accuracy was 95.1%.
  • N9-NASH significantly outperformed individual protein markers (Fig. 5A-B) or clinical variables.
  • Figure 5C illustrates the sensitivity, specificity, NPV, and PPV of N9-NASH at different cutoffs to distinguish NASH patients from the validation cohort.
  • N9-NASH showed 70.2% sensitivity, 80.2% accuracy, 73.0% NPV, and 90.4% PPV in distinguishing NASH from patients with simple steatosis, significantly higher than single Biomarkers (Figure 5B). Therefore, these results confirmed the diagnostic value of N9-NASH for NASH.
  • the present invention determines the best combination of biomarkers for the diagnosis of NAFLD and NASH, and is called N9-NASH, including 5 protein markers (CXCL10, CK-18, P62/SQSTM1, SQLE and FGF21) and 4 blood biochemical variables (HbA1c, ALT, LDL-cholesterol, and platelets).
  • N9-NASH can distinguish NAFLD patients from healthy controls, and the diagnostic accuracy of AUROC can reach 0.999 in the training cohort and 0.989 in the validation cohort.
  • NASH patients could be identified by N9-NASH with an AUROC of 0.94 in the training cohort and 0.87 in the validation cohort.
  • N9-NASH had 90.6% specificity, 91.9% sensitivity, and 90.4% positive predictive value (PPV) in distinguishing NASH patients from isolated steatosis in the training cohort, while its diagnostic performance was equally good (91.5% in the validation cohort). % specificity performance, 70.2% sensitivity and 90.4% PPV).
  • the Cut-off cut-off value is 0.47
  • the performance of distinguishing non-alcoholic steatohepatitis from simple steatosis is the best. If the Cut-off value ⁇ 0.47, it is diagnosed as non-alcoholic steatohepatitis For steatohepatitis, if the cut-off value is ⁇ 0.47, it is diagnosed as simple steatosis.
  • N5-NASH including CXCL10, CK-18, Pro-C3 , AST and BMI.
  • N5-NASH 2.655*CK-18+3.06*CXCL10-0.066*Pro-C3-0.0802*BMI-0.076*AST-1.62.
  • the biomarker composition is N5-NASH
  • the cut-off cut-off value when the cut-off cut-off value is 0.27 and 0.61, it can have high specificity (>90%) and high sensitivity respectively (>90%) to diagnose non-alcoholic steatohepatitis, if the cut-off value ⁇ 0.61, the diagnosis is non-alcoholic steatohepatitis, if the cut-off value is ⁇ 0.27, the diagnosis is non-alcoholic steatohepatitis Simple steatosis.
  • the cut-off value was 0.61
  • the specificity of N3-NASH in diagnosing generalized NASH was 91.2%
  • the sensitivity was 67.9%
  • the PPV was 91.7%.
  • N3-NASH had a sensitivity of 65.6%, a specificity of 85.4%, an NPV of 38.9%, and a PPV of 94.5% in excluding generalized NASH, with 48 patients (23.8%) in the middle area, further liver biopsy is required.
  • a high cut-off value of 0.68 is used to diagnose generalized NASH, and a low cut-off value of 0.43 is used to exclude simple steatosis, that is, when the biomarker composition is N3-NASH, non-alcoholic In the steatohepatitis diagnostic model, if the Cut-off value ⁇ 0.68, it is diagnosed as non-alcoholic steatohepatitis and requires tertiary care. If the cut-off value is ⁇ 0.43, the diagnosis is simple steatosis, and patients with a cut-off value ⁇ 0.68 only need primary care for lifestyle intervention (Figure 9).
  • patients with N3-NASH cut-off value higher than 0.68 are diagnosed as generalized NASH, and patients with N2-Fibrosis Cut-off value higher than 0.37 are diagnosed as high-risk NASH and need tertiary management.
  • Patients with cut-off values between 0.30 and 0.37 need further liver biopsy for confirmation.
  • Patients with a cut-off cutoff value below 0.30 are diagnosed as low-risk NASH and only require primary management of lifestyle interventions (Figure 9).

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • Data Mining & Analysis (AREA)
  • Pathology (AREA)
  • General Health & Medical Sciences (AREA)
  • Immunology (AREA)
  • Molecular Biology (AREA)
  • Urology & Nephrology (AREA)
  • Chemical & Material Sciences (AREA)
  • Epidemiology (AREA)
  • Databases & Information Systems (AREA)
  • Hematology (AREA)
  • Primary Health Care (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Microbiology (AREA)
  • Biochemistry (AREA)
  • Analytical Chemistry (AREA)
  • Medicinal Chemistry (AREA)
  • Food Science & Technology (AREA)
  • Cell Biology (AREA)
  • Biotechnology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Proteomics, Peptides & Aminoacids (AREA)
  • Communicable Diseases (AREA)
  • Investigating Or Analysing Biological Materials (AREA)

Abstract

提供一种非酒精性脂肪性肝炎生物标志物组合物及其应用，生物标志物组合物包括诊断非酒精性脂肪性肝病和/或非酒精性脂肪性肝炎的相关蛋白标志物和/或临床生化标志物，相关蛋白标志物选自CXCL10、CK-18、P62/SQSTM1、CA3、SQLE、Pro-C3或FGF21中一种或多种；临床生化标志物选自BMI、HDL-C、HbAlc、ALT、AST、LDL-C、TG、TC、ALP或PLT中的一种或多种。通过逻辑回归及人工智能方法建立生物标志物组合物，作为诊断 NAFLD和NASH 的诊断标志物，在NASH诊断方面具有高灵敏度、高特异性、高阳性预测值和高阴性预测值，且不受年龄、性别或代谢状态的影响。

Description

非酒精性脂肪性肝炎生物标志物组合物及其应用 技术领域
本发明涉及生物检测领域，具体涉及一种非酒精性脂肪性肝炎生物标志物组合物及其应用。
背景技术
非酒精性脂肪性肝病(Non-alcoholic fatty liver disease,NAFLD)，与肥胖、2型糖尿病、代谢综合征密切相关。由于久坐少动的生活方式和高热量饮食结构的流行，NAFLD的发病率呈现逐年上升趋势，已成为目前世界上最常见的肝脏疾病，其在亚洲地区的平均人群发病率为15％-40％，在我国城市发病率高达43.3％。NAFLD的病理进展过程为从单纯性脂肪变性到非酒精性脂肪性肝炎(Non-alcoholic steatohepatitis,NASH)，进而导致进展性肝纤维化/肝硬化，最终发展成肝癌。12％-40％的NAFLD患者会进展为NASH，15％-33％的NASH患者可发展为肝硬化，而15％-27％的肝硬化患者会进展成为肝癌。由于肥胖和2型糖尿病患病率的升高，NASH的发病率将在未来十年增加56％。NASH患者进展为肝硬化和肝癌的比例明显高于单纯性脂肪变患者。近期研究表明NASH相关肝癌与其他原因导致的肝癌相比，无有效治疗方法。如何区分单纯性脂肪变性和NASH，对预防疾病的进展至关重要。
单纯性脂肪变性→NASH→肝纤维化/肝硬化→肝癌的疾病进程多呈隐匿性，肝脏活检一直被认为是NASH诊断的金标准，但是由于肝脏活检穿刺是侵入性检查，花费较高，患者的接受度低，限制了其在临床方面的应用。因此迫切需要建立高敏感特异的非侵入性诊断方法以减少肝脏活检的依赖性，并区分需要进一步治疗的患者。然而，NASH诊断标志物的研发严重滞后，至今尚无临床可用的NASH非侵入性诊断以及预后监测的方法。一些诊断标志物被用于非侵入性诊断NASH，然而诊断准确性较低。例如凋亡标志物细胞角蛋白18(CK-18)片段是最常被验证的血液标志物，然而单独应用敏感性仅66％，特异性82％；脂肪细胞因子脂联素、瘦素和纤维原细胞生长因子(FGF21)等只与机体代谢异常相关，并且多数只在做减肥手术的人群中得到验证，对NASH诊断的准确度较低；溶酶体酶(Cathepsin D)等在不同人群中的差异较大，未能得到验证。由于单因子在NASH诊断中的准确度有限，急需建立标志物组合物以提高诊断的准确度。
在NASH中，肝纤维化为肝脏不良临床结局最重要的预测指标，为高危NASH。FDA建议将药物研发的重点放在NASH伴有肝纤维化这一类对健康有着重大需求和潜在效益的领域。开发可以诊断NASH伴有肝纤维化的高危患者的无创诊断标志物和试剂盒对寻找需要药 物干预的患者和监测临床药物疗效意义重大。
发明内容
本发明的目的是提供一种生物标志物组合物，该生物标志组合物在区分非酒精性脂肪性肝炎与单纯性脂肪变性方面显示出高灵敏度、高特异性、高阳性预测值和高阴性预测值。
本发明中所述的非酒精性脂肪性肝炎可采用三种不同的划分标准，包括：
第一种：肝脏脂肪变评分>＝1，炎症评分>＝1，气球样变评分>＝1。
第二种：肝脏脂肪变评分>＝1，炎症评分>＝1，气球样变评分>＝1或根据Nonalcoholic Steatohepatitis Research Network(NASH CRN)定义的NAS评分>＝4。采用第二种划分标准能够确定更多的非酒精性脂肪性肝炎，因此可定义为广义非酒精性脂肪性肝炎。
第三种：广义非酒精性脂肪性肝炎中肝纤维化>＝2，为高危非酒精性脂肪性肝炎，即划分标准为：肝脏脂肪变评分>＝1，炎症评分>＝1，气球样变评分>＝1或根据Nonalcoholic Steatohepatitis Research Network(NASH CRN)定义的NAS评分>＝4，且肝纤维化>＝2。
本发明的另一目的是提供非酒精性脂肪性肝病诊断模型和/或非酒精性脂肪性肝炎诊断模型。本发明的非酒精性脂肪性肝炎诊断模型还能够比较非酒精性脂肪性肝炎风险高低。
本发明的另一目的是提供能够根据生物标志物组合物在待测样本中的水平来诊断非酒精性脂肪性肝病和/或非酒精性脂肪性肝炎的人工智能模型。
本发明的另一目的是提供一种用于诊断非酒精性脂肪性肝炎的试剂盒。
为解决上述技术问题，本发明采用如下技术方案：
一种生物标志物组合物，所述的生物标志物组合物包括诊断非酒精性脂肪性肝病和/或非酒精性脂肪性肝炎的相关蛋白标志物和/或临床生化标志物，所述的诊断非酒精性脂肪性肝病和/或非酒精性脂肪性肝炎的相关蛋白标志物选自CXC趋化因子配体10(CXCL10)、细胞角蛋白18(CK-18)、选择性自噬接头蛋白(P62/SQSTM1)、碳酸酐酶III(CA3)、角鲨烯环氧化酶(SQLE)、III/IV型胶原蛋白前体(Pro-C3)或纤维原细胞生长因子(FGF21)中一种或多种；所述的临床生化标志物选自身高体重指数(BMI)、糖化血红蛋白(HbA1c)、谷丙转氨酶(ALT)、谷草转氨酶(AST)、低密度脂蛋白胆固醇(LDL-C)、总胆固醇(TC)、甘油三酯(TG)、血清碱性磷酸酶(ALP)或血小板(PLT)中的一种或多种。
优选地，所述的生物标志物组合物至少包括两种所述的诊断非酒精性脂肪性肝病和/或非酒精性脂肪性肝炎和/或高危非酒精性脂肪性肝炎的相关蛋白标志物。
优选地，分别测量来自检测对象的所述的诊断非酒精性脂肪性肝病和/或非酒精性脂肪性 肝炎和/或高危非酒精性脂肪性肝炎的相关蛋白标志物的水平和所述的临床生化标志物的水平，然后使用支持向量机、逻辑回归、朴素贝叶斯和10折交叉验证建立最佳标志物组合物。
进一步优选地，所述的生物标志物组合物由5种蛋白标志物和4种临床生化标志物组成(称为N9-NASH)，包含CXCL10、CK-18、P62/SQSTM1、ALT、SQLE、HbA1c、FGF21、PLT和LDL-C。
N9-NASH能够将非酒精性脂肪性肝病(NAFLD)患者与健康者区分开，还能够进一步将非酒精性脂肪性肝病患者中的非酒精性脂肪性肝炎(肝脏脂肪变评分>＝1，炎症评分>＝1，气球样变评分>＝1)和单纯性脂肪变性区分开。
进一步优选地，所述的生物标志物组合物由3种蛋白标志物和2种临床生化标志物组成(称为N5-NASH)，包含CXCL10，CK-18，Pro-C3，AST和BMI。
N5-NASH能够进一步将非酒精性脂肪性肝病患者中的广义非酒精性脂肪性肝炎(肝脏脂肪变评分>＝1，炎症评分>＝1，气球样变评分>＝1或NAS>＝4)和单纯性脂肪变性区分开。
进一步优选地地，所述的生物标志物组合物(称为N3-NASH)由CXCL10，CK-18和BMI组成。
N3-NASH同样能够进一步将非酒精性脂肪性肝病患者中的广义非酒精性脂肪性肝炎(肝脏脂肪变评分>＝1，炎症评分>＝1，气球样变评分>＝1或NAS>＝4)和单纯性脂肪变性区分开。
进一步优选地，所述的生物标志物组合物(称为N2-Fibrosis)由CXCL10和Pro-C3组成。
N2-Fibrosis能够进一步将广义非酒精性脂肪性肝炎中伴有纤维化的高危患者区分开来。
一种如上所述的生物标志物组合物在区分非酒精性脂肪性肝炎和单纯性脂肪变性中的应用。
一种非酒精性脂肪性肝病诊断模型和/或非酒精性脂肪性肝炎诊断模型，其构建方法为：收集非酒精性脂肪性肝病患者(包括非酒精性脂肪性肝炎患者和单纯性脂肪变性患者，其中非酒精性脂肪性肝炎患者包括高危和低危非酒精性脂肪性肝炎患者)的血液样本，分别测量所述的诊断非酒精性脂肪性肝病和/或非酒精性脂肪性肝炎和/或高危非酒精性脂肪性肝炎的相关蛋白标志物的水平和/或所述的临床生化标志物的水平，将非酒精性脂肪性肝病患者和健康者的标志物的水平使用支持向量机、逻辑回归、朴素贝叶斯和10折交叉验证建立非酒精性脂肪性肝病诊断模型，和/或，将非酒精性脂肪性肝炎患者和单纯性脂肪变性患者的标志物的水平使用支持向量机、逻辑回归、朴素贝叶斯和10折交叉验证建立非酒精性脂肪性肝炎诊断模型，和/或，将高危和低危非酒精性脂肪性肝炎患者实验逻辑回归建立诊断高危非酒精性脂 肪性肝炎模型。
其中，所述的诊断非酒精性脂肪性肝病和/或非酒精性脂肪性肝炎的相关蛋白标志物选自CXCL10、CK-18、P62/SQSTM1、SQLE、CA3、Pro-C3或FGF21中一种或多种；所述的临床生化标志物选自HbA1c、AST、BMI、ALT、LDL-C、TG、TC、ALP或PLT中的一种或多种。
进一步优选地，所述的非酒精性脂肪性肝病诊断模型中使用的生物标志组合物为N9-NASH。
进一步优选地，所述的非酒精性脂肪性肝炎诊断模型中使用的生物标志组合物为N9-NASH、N5-NASH、N3-NASH或N2-Fibrosis。
一种根据所述的生物标志物组合物在待测样本中的水平来诊断非酒精性脂肪性肝病人工智能模型和/或诊断非酒精性脂肪性肝炎的人工智能模型。
一种试剂盒，其包括用于检测所述的生物标志物组合物的水平的检测试剂。
优选地，所述的试剂盒还包括标准品，所述的标准品包括所述的生物标志物组合物。
一种如所述的生物标志物组合物在制备试剂盒中的用途，所述的试剂盒用于区分非酒精性脂肪性肝病和健康者，所述的试剂盒包括用于检测所述的生物标志物组合物的水平的检测试剂。
一种如所述的生物标志物组合物在制备试剂盒中的用途，所述的试剂盒用于区分非酒精性脂肪性肝炎和单纯性脂肪变性，所述的试剂盒包括用于检测所述的生物标志物组合物的水平的检测试剂。
一种如所述的生物标志物组合物在制备试剂盒中的用途，所述的试剂盒用于区分高危非酒精性脂肪性肝炎和低危非酒精性脂肪性肝炎，所述的试剂盒包括用于检测所述的生物标志物组合物的水平的检测试剂。
所述的试剂盒在区分非酒精性脂肪性肝病和健康者，和/或，区分非酒精性脂肪性肝炎和单纯性脂肪变性，和/或，区分高危非酒精性脂肪性肝炎和低危非酒精性脂肪性肝炎中的应用。
优选地，所述的区分方法为：提供一种来源于待测对象的样品，分别测试样品中的所述的诊断非酒精性脂肪性肝病和/或非酒精性脂肪性肝炎的相关蛋白标志物和所述的临床生化标志物的水平，然后代入支持向量机、逻辑回归、朴素贝叶斯和10折交叉验证确定最佳生物标志物组合，并进行ROC曲线分析，在非酒精性脂肪性肝病诊断模型中，根据Cut-off值来区分非酒精性脂肪性肝病和健康者；在非酒精性脂肪性肝炎诊断模型中，根据Cut-off值来区 分非酒精性脂肪性肝炎和单纯性脂肪变性；在高危非酒精性脂肪性肝炎诊断模型中，根据Cut-off值来区分高危和低危非酒精性脂肪性肝炎患者。
进一步优选地，所述的Cut-off值由最大化约登指数的ROC分析确定，所述的最大化约登指数＝灵敏度+特异性-1。
Cut-off临界值作为阳性判断值，≥Cut-off临界值即判为阳性，<Cut-off临界值判为阴性。
当所述的生物标志物组合物为N9-NASH时，优选非酒精性脂肪性肝病诊断模型中Cut-off临界值为0.58；优选非酒精性脂肪性肝炎诊断模型中Cut-off临界值为0.47(诊断非酒精性脂肪性肝病和非酒精性脂肪性肝炎的模型为不同的机器学习模型)。
根据一些具有实施方式，当所述的生物标志物组合物为N9-NASH时，非酒精性脂肪性肝病诊断模型中，Cut-off临界值为0.58时，区分非酒精性脂肪性肝病和健康对照敏感性和特异性最高，若Cut-off值≥0.58，则判为非酒精性脂肪性肝病，Cut-off值<0.58判为健康者；
根据一些具有实施方式，当所述的生物标志物组合物为N9-NASH时，非酒精性脂肪性肝炎诊断模型中，Cut-off临界值为0.47时，区分非酒精性脂肪性肝炎和单纯性脂肪变的表现最好，若Cut-off值≥0.47，则诊断为非酒精性脂肪性肝炎，若Cut-off值<0.47，则诊断为单纯性脂肪变性。
当所述的生物标志物组合物为N5-NASH时，优选非酒精性脂肪性肝炎诊断模型中Cut-off临界值为0.27和0.61。
根据一些具有实施方式，当所述的生物标志物组合物为N5-NASH时，非酒精性脂肪性肝炎诊断模型中，若Cut-off值≥0.61，则诊断为非酒精性脂肪性肝炎，若Cut-off值<0.27，则诊断为单纯性脂肪变性。若0.27≤Cut-off值<0.61，则需要进行肝穿刺活检。
当所述的生物标志物组合物为N3-NASH时，优选非酒精性脂肪性肝炎诊断模型中Cut-off临界值为0.43和0.68.
根据一些具有实施方式，当所述的生物标志物组合物为N3-NASH时，非酒精性脂肪性肝炎诊断模型中，若Cut-off值≥0.68，则诊断为非酒精性脂肪性肝炎，若Cut-off值<0.43，则诊断为单纯性脂肪变性。若0.43≤Cut-off值<0.68，则需要进行肝穿刺活检。
当所述的生物标志物组合物为N2-Fibrosis时，优选高危非酒精性脂肪性肝炎诊断模型中Cut-off临界值为0.37和0.30。
根据一些具有实施方式，当所述的生物标志物组合物为N2-Fibrosis时，高危非酒精性脂肪性肝炎诊断模型中，若Cut-off值≥0.37，则诊断为高危非酒精性脂肪性肝炎，需要进行三 级管理。若Cut-off值<0.30，则诊断为低危非酒精性脂肪性肝炎，仅需要进行初级生活方式干预。若0.30≤Cut-off值<0.37，则需要进行肝穿刺活检。
优选地，所述的样品为血液。
进一步优选地，所述的样品为血清。
优选地，所述的样品来自人。
本发明候选生物标志物是通过对我们之前研究中确定的蛋白标志物和其他临床变量的重要性使用varImp进行排序来选择的。
本发明与现有技术相比具有如下优势：
本发明确定了可以非侵入性诊断非酒精性脂肪性肝病和/或非酒精性脂肪性肝炎的生物标志物组合物，该生物标记物组合物在区分非酒精性脂肪性肝炎与单纯性脂肪变性方面显示出高灵敏度、高特异性、高阳性预测值(PPV)和阴性预测值(NPV)，本发明的生物标记物组合物还能够区分非酒精性脂肪性肝炎中的高危和低危非酒精性脂肪性肝炎。本发明的生物标记物组合物可以通过广泛使用的方法进行检测，并且可以很容易地应用于临床实践。本发明的生物标记物组合物的诊断性能不受年龄、性别或代谢状态的影响。
附图说明
图1A为NAFLD患者与健康对照者血清中的CXCL10、SQLE、CK-18、P62/SQSTM1、FGF21和CA3水平对比图；
图1B为NASH患者与健康对照者血清中的CXCL10、SQLE、CK-18、P62/SQSTM1、FGF21和CA3水平对比图；
图2为17个标志物对NASH诊断的重要性排列图；
图3为N9-NASH和单个生物标志物在健康对照者和NAFLD诊断方面以及NAFLD患者中诊断NASH的性能对比图；
图4为N9-NASH和单个生物标志物在训练队列中对NASH诊断的性能对比图。
图5为N9-NASH和单个生物标志物在验证队列中诊断NASH的性能对比图。
图6为标志物对广义NASH诊断的重要性排列图；
图7为N5-NASH在发现和验证队列中对NASH诊断的性能。
图8为N3-NASH在发现和验证队列中对NASH诊断的性能。
图9为N3-NASH和N2-Fibrosis两步法诊断高危NASH的方法。
具体实施方式
下面结合实施例对本发明作进一步描述。但本发明并不限于以下实施例。实施例中采用的实施条件可以根据具体使用的不同要求做进一步调整，未注明的实施条件为本行业中的常规条件。本发明各个实施方式中所涉及到的技术特征只要彼此之间未构成冲突就可以相互组合。
为了提高非侵入性NASH诊断的准确度，发明人经过前期工作鉴定了一些与NASH相关的蛋白质生物标志物，结合能够鉴定NAFLD和NASH患者的临床生化指标，引入24个变量，包括7个血清蛋白(CXCL10，SQLE，CK-18，P62/SQSTM1，FGF21，Pro-C3和CA3)和17个临床变量(14个连续变量：(年龄、体重身高指数(BMI)、血小板(PLT)、血清白蛋白(Alb)、血清碱性磷酸酶水平(ALP)、血清谷丙转氨酶(ALT)、谷草转氨酶(AST)、胆红素(bilirubin)、总胆固醇(TC)、高密度脂蛋白胆固醇(HDL-C)、甘油三酯(TG)、低密度脂蛋白胆固醇(LDL-C)、糖化血红蛋白(HbA1c))、空腹血糖(FBS)和3个离散变量(性别、高血压、糖尿病))以训练标志物模型。本发明确定了NASH的最佳诊断生物标志物组合物。
研究对象：
从香港威尔斯亲王医院招募了374名受试者，在排除36名有其他肝损伤原因的受试者和14名血清样本不足的受试者后，324名受试者包括252名经活检证实为NAFLD的患者和72名健康对照者参加了研究。在NAFLD患者中进行肝活检主要是由于他们的肝功能检查异常和/或影像学结果异常。排除标准包括：1)男性每日饮酒量超过30g，女性超过20g；2)乙肝表面抗原或抗丙肝病毒抗体阳性，抗核抗体滴度>1/160；3)继发性肝脂肪变性或其他肝病组织学特征的患者。对照受试者是从政府人口普查数据库中随机选择的。进行质子磁共振波谱(1H-MRS)以量化同意参与研究的受试者的肝脏甘油三酯含量。如果出现以下情况，则排除对照受试者：1)肝脏甘油三酯含量超过5％；2)糖尿病或高血压病史；3)排除NAFLD患者。在肝活检前一天采集空腹静脉血样。所有患者都提供了参与试验和收集专门用于生物标志物研究的血液样本的书面知情同意书。
独立验证队列标本来自217名肝脏穿刺证实为NAFLD的患者，肝脏病理取自超声引导经皮肝活检，采用16G或者18G针。在排除无法解释的组织病理学以及其他肝损伤原因的受试者后，共201名NAFLD患者纳入。NAFLD活动度通过计算病理中的脂肪变评分+炎症评分+气球样变评分获得。肝纤维化定义为：0级，无纤维化；1级，窦周或门静脉纤维化；2级，肝窦周围/门静脉周围纤维化；3级，桥接纤维化；4级，肝硬化。
诊断标准：
1.NASH:肝脏脂肪变评分>＝1，炎症评分>＝1，气球样变评分>＝1。
2.广义NASH:除1)所示NASH外，还包括根据Nonalcoholic Steatohepatitis Research Network(NASH CRN)定义的NAS评分>＝4。
3.高危NASH:广义NASH中，肝纤维化评分>＝2。
血清蛋白检测：
将血清从-80℃取出，进行生物标志物的检测。血清中CK-18，FGF21，CXCL10，P62/SQSTM1，SQLE，CA3血清水平分别通过M30酶联免疫吸附试验(ELISA)试剂盒(PEVIVA，瑞典)、FGF21ELISA试剂盒(BioVendor，捷克共和国)、CXCL10ELISA试剂盒(R&D公司)、P62/SQSTM1ELISA试剂盒(Cloud‐Clone，美国)、SQLE ELISA试剂盒(Cloud‐Clone，美国)、Pro-C3ELISA试剂盒(Nordic Bioscience,丹麦)和CA3ELISA试剂盒(Cloud‐Clone，美国)进行检测。
算法的训练
我们引入24个变量，包括7个血清蛋白(CXCL10，SQLE，CK-18，P62/SQSTM1，FGF21,Pro-C3和CA3)和17个临床变量以训练标志物模型。临床变量包括14个连续变量(年龄、体重身高指数(BMI)、血小板(PLT)、血清白蛋白(Alb)、血清碱性磷酸酶水平(ALP)、血清谷丙转氨酶(ALT)、谷草转氨酶(AST)、胆红素(bilirubin)、总胆固醇(TC)、高密度脂蛋白胆固醇(HDL-C)、甘油三酯(TG)、低密度脂蛋白胆固醇(LDL-C)、糖化血红蛋白(HbA1c)、空腹血糖(FBS))和3个离散变量(性别、高血压、糖尿病)。对具有最大相关性的协变量进行相关性分析(Spearman相关系数的绝对值，abs(rho)≥0.5)。每个变量的重要性按varImp排序。通过递归特征消除选择关键生物标志物和临床变量。通过使用支持向量机、神经网络、朴素贝叶斯逻辑回归和10折交叉验证，确定NASH的最佳诊断生物标志物，以建立诊断模型，通过Wilcoxon秩和检验检查标记的选择标准。
使用AUROC分析评估生物标志物组合的诊断性能。最佳cut-off值由最大化约登指数(J＝灵敏度+特异性-1)的ROC分析确定。
统计分析
为了消除批次效应，在数据预处理中使用了代理变量分析(sva)。实验中数值用平均值±标准偏差(SD)表示。蛋白质水平或临床变量的差异由Mann-Whitney U检验确定。P值为0.05为具有统计学意义。应用Spearman相关系数评估生物标志物水平与相关因素的关联，用于多元回归分析确定与生物标志物相关的独立因素。DeLong检验用于比较不同亚群的 AUROC值。
实施例一：NASH的生物标志物
为了确定可以诊断NAFLD和NASH的候选生物标志物，我们检测了252名NAFLD患者和72名健康对照者血清中的CXCL10、SQLE、CK-18、P62/SQSTM1、FGF21和CA3水平。在招募的252名NAFLD患者中，129名(51％)患者经病理评估确诊为NASH(肝脏脂肪变评分>＝1，炎症评分>＝1，气球样变评分>＝1)。与健康对照组相比，NAFLD患者中所有6种生物标志物的血清水平均显着升高(P<0.0001，图1A)。在NAFLD患者中，NASH患者(n＝129)的血清CXCL10、SQLE、CK-18和FGF21水平显着高于单纯性脂肪变患者(n＝123)(图1B)。
CXCL10和CK-18与小叶炎症(CXCL10，rho：0.21，P<0.001；CK-18，rho：0.21，P＝0.001)和肝细胞气球样变(CXCL10，rho：0.26，P<0.0001；NAFLD患者的CK-18，rho：0.27，P<0.0001)呈正相关，而SQLE与炎症呈正相关(rho：0.20，P<0.001)。
我们进一步分析了可能作为NAFLD和NASH诊断的候选生物标志物的临床变量。在包括的17个临床变量中，包括年龄、性别、身高、体重、腰围、高血压、糖尿病、血小板、血清谷丙转氨酶、谷草转氨酶、碱性磷酸酶、白蛋白、总胆固醇、总甘油三酯、低密度脂蛋白胆固醇、高密度脂蛋白胆固醇、糖化血红蛋白等。HbA1c(葡萄糖代谢指标)和ALT(肝损伤指标)在对照组、单纯性脂肪变性患者和NASH患者中逐步增加。
我们通过varImp按重要性对所有标志物进行排序，选择了最适合NAFLD和NASH诊断的生物标志物。CXCL10、CK-18、P62/SQSTM1、ALT、SQLE、HbA1c、FGF21、PLT和LDL-C是NASH诊断最重要的前9个变量(图2A)。此外，在所有受试者中，这9个变量之间没有强相关性(|rho|:0.001-0.482)，表明这些标记物彼此不可或缺(图2B)。因此，我们使用这9个独立的生物标志物，包括5个NASH相关蛋白(CXCL10、CK-18、P62/SQSTM1、SQLE和FGF21)和4个血液生化变量(HbA1c、ALT、LDL-C和PLT)来开发NASH诊断模型。采用SVM来训练模型并减少模型过拟合或选择偏差。最终建立了最佳生物标志物组合，并称为N9-NASH，包括上述前9个变量。
我们评估了N9-NASH对健康对照和NAFLD诊断方面的性能。N9-NASH可以将NAFLD患者与健康对照区分开来，AUROC为0.999(95％CI 0.997-1.000)(图3A)，N9-NASH的性能比单个蛋白生物标志物(AUROC 0.68-0.87，所有P<0.01)或临床变量(AUROC 0.55-0.87，所有P<0.0001)优越(图3B)。N9-NASH对NAFLD诊断的敏感性和特异性分别为100％和 97.2％。
然后，我们评估了N9-NASH在NAFLD患者中诊断NASH的性能。在NAFLD患者中，NASH患者可以通过N9-NASH来识别，AUROC为0.94(95％CI 0.90-0.99)(图4A)。与单个蛋白标志物相比，N9-NASH区分NASH与单纯性脂肪变性患者的准确性显着更高：CXCL10(AUROC 0.66,95％CI 0.56-0.75,P<0.0001)，CK-18(AUROC 0.63,95％CI 0.53-0.73,P<0.0001)，SQLE(AUROC 0.60,95％CI 0.50-0.70,P<0.0001)，FGF21(AUROC 0.48,95％CI0.38-0.58,P<0.0001)，CA3(AUROC 0.43,95％CI 0.32-0.53,P<0.0001)，或P62/SQSTM1(AUROC 0.61,95％CI 0.51-0.71,P<0.0001)(图4A-B)；或个体临床变量(AUROC 0.52-0.63,所有P<0.0001)。
图4C显示了N9-NASH在训练队列中对NASH诊断的敏感性、特异性、NPV和PPV。在90.6％的特异性下，N9-NASH具有91.9％敏感性、90.5％准确度、92.1％NPV和90.4％PPV(图4B)。
与训练队列类似，N9-NASH区分NAFLD患者与健康对照方面表现出色，AUROC为0.989(95％CI 0.976-1.000)(图3C-D)。在特异性为100％时，N9-NASH在NAFLD诊断中的敏感性为93.7％，准确率为95.1％。
NASH诊断方面，N9-NASH显著优于单个蛋白标志物(图5A-B)或临床变量。图5C说明了N9-NASH在不同临界值下区分NASH患者与验证队列中的敏感性、特异性、NPV和PPV。在91.5％的特异性下，N9-NASH在区分NASH和单纯性脂肪变患者方面显示出70.2％的敏感性、80.2％的准确度、73.0％的NPV和90.4％的PPV，显着高于单个生物标志物(图5B)。因此，这些结果证实了N9-NASH对NASH的诊断价值。
综上，本发明的确定了NAFLD和NASH诊断的最佳生物标志物组合，并称为N9-NASH，包括5个蛋白标志物(CXCL10、CK-18、P62/SQSTM1、SQLE和FGF21)和4个血液生化变量(HbA1c、ALT、LDL-胆固醇和血小板)。
N9-NASH可以将NAFLD患者与健康对照这区分开来，AUROC的诊断准确性在训练队列中可达0.999，在验证队列中可达0.989。
在NAFLD患者中，可以通过N9-NASH识别NASH患者，训练队列中的AUROC为0.94，验证队列中的AUROC为0.87。
N9-NASH在训练队列中区分NASH患者与单纯性脂肪变性具有90.6％的特异性、91.9％的敏感性和90.4％的阳性预测值(PPV)，而在验证队列中其诊断性能同样出色(91.5％的特异 性，70.2％灵敏度和90.4％PPV)。
非酒精性脂肪性肝病诊断模型中，Cut-off临界值为0.58时，区分非酒精性脂肪性肝病和健康对照敏感性和特异性最高，若Cut-off值≥0.58，则判为非酒精性脂肪性肝病，Cut-off值<0.58判为健康者；
非酒精性脂肪性肝炎诊断模型中，Cut-off临界值为0.47时，区分非酒精性脂肪性肝炎和单纯性脂肪变的表现最好，若Cut-off值≥0.47，则诊断为非酒精性脂肪性肝炎，若Cut-off值<0.47，则诊断为单纯性脂肪变性。
实施例二：广义NASH的生物标志物
NASH的临床试验中通常包含NAS>＝4的患者，进一步地，我们分析了可以区分广义NASH(肝脏脂肪变评分>＝1，炎症评分>＝1，气球样变评分>＝1或NAS>＝4)的标志物。我们从香港队列中纳入145例NAFLD样本，排除7例无法解释的病理诊断，最终138例NAFLD患者纳入发现队列，其中81例根据病理结果诊断为广义NASH，57例为单纯性脂肪变。另一队列中纳入201例样本作为独立验证队列，其中160例为广义NASH，41例为单纯性脂肪变。
通过分析24个变量，包括7个血清蛋白(CXCL10，SQLE，CK-18，P62/SQSTM1，FGF21，Pro-C3和CA3)和17个临床变量(14个连续变量(年龄、BMI、PLT、Alb、ALP、ALT、AST、TC、Bilirubin、HDL-C、TG、LDL-C、HbA1c、FBS)和3个离散变量(性别、高血压、糖尿病))以训练标志物模型。通过随机森林方法筛选重要变量(图6)，朴素贝叶斯和逻辑回归方法建立模型，最终建立了最佳生物标志物组合，并称为N5-NASH，包括CXCL10，CK-18,Pro-C3,AST和BMI。逻辑回归模型中，N5-NASH＝2.655*CK-18+3.06*CXCL10-0.066*Pro-C3-0.0802*BMI-0.076*AST-1.62。
在发现队列中，N5-NASH区分广义NASH中的AUROC为0.881(图7A)。与单个蛋白标志物相比，N5-NASH区分广义NASH与单纯性脂肪变性患者的准确性显着更高：CXCL10(AUROC 0.725,P<0.001)，CK-18(AUROC 0.719,P<0.001)，或Pro-C3(AUROC 0.683,P<0.0001)(图7A-B)。验证队列中，N5-NASH亦可显著区分广义NASH(AUROC＝0.802，图7B)。
当所述的生物标志物组合物为N5-NASH时，非酒精性脂肪性肝炎诊断模型中，Cut-off临界值为0.27和0.61时，可分别高特异性(>90％)和高敏感性(>90％)地诊断非酒精性脂肪性肝炎，若Cut-off值≥0.61，则诊断为非酒精性脂肪性肝炎，若Cut-off值<0.27，则诊断为 单纯性脂肪变性。Cut-off临界值为0.61时，N3-NASH在诊断广义NASH方面的特异性为91.2％，敏感性为67.9％，PPV为91.7％。低Cut-off临界值值0.27时，N3-NASH在排除广义NASH方面的敏感性为90.1％，特异性为54.4％，NPV为79.5％，38个患者(27.5％)在中间区域，需要进行进一步的肝活检检查。
为了增加临床应用的可行性，我们进一步优化了标志物模型，研究是否应用少量的标志物可以达到相似的诊断效率。我们发现CXCL10，CK-18以及BMI的模型组合(N3-NASH)可以高敏感特异的诊断广义NASH(发现队列：AUROC＝0.848；验证队列：AUROC＝0.810)(图8)。逻辑回归模型中，N3-NASH＝2.75*CK-18+3.42*CXCL10+0.08*BMI-16.47。
Cut-off临界值为0.68时，N3-NASH在诊断广义NASH方面的特异性为91.2％，敏感性为62.9％，PPV为91.0％。低Cut-off临界值值0.43时，N3-NASH在排除广义NASH方面的敏感性为90.1％，特异性为49.1％，NPV为77.8％。45个患者(32.6％)在中间区域，需要进行进一步的肝活检检查。在验证队列中，Cut-off临界值为0.68时，N3-NASH在诊断广义NASH方面的特异性为92.7％，敏感性为38.1％，PPV为95.3％。低Cut-off临界值为0.43时，N3-NASH在排除广义NASH方面的敏感性为65.6％，特异性为85.4％，NPV为38.9％，PPV为94.5％，48个患者(23.8％)在中间区域，需要进行进一步的肝活检检查。因此，在实际应用中，选用高Cut-off临界值0.68诊断广义NASH，选用低Cut-off 0.43排除单纯性脂肪变，即当所述的生物标志物组合物为N3-NASH时，非酒精性脂肪性肝炎诊断模型中，若Cut-off值≥0.68，则诊断为非酒精性脂肪性肝炎，需要进行三级护理。若Cut-off值<0.43，则诊断为单纯性脂肪变性，Cut-off值<0.68的患者，仅需要进行生活方式干预的初级护理(图9)。
实施例三：高危非酒精性脂肪性肝炎的生物标志物
为了进一步区分非酒精性脂肪性肝炎中伴有纤维化的高危患者，我们将诊断为广义非酒精性脂肪性肝炎的标本进行分析。119例诊断为广义非酒精性脂肪性肝炎的患者纳入研究，其中发现队列56例，验证队列63例。
通过分析24个变量，包括7个血清蛋白(CXCL10，SQLE，CK-18，P62/SQSTM1，FGF21，Pro-C3和CA3)和17个临床变量(14个连续变量(年龄、BMI、PLT、Alb、ALP、ALT、AST、TC、Bilirubin、HDL-C、TG、LDL-C、HbA1c、FBS)和3个离散变量(性别、高血压、糖尿病))以训练标志物模型。通过逻辑回归方法，最终建立了最佳生物标志物组合，并称为N2-Fibrosis，包括CXCL10和Pro-C3。该逻辑回归模型中 N2-Fibrosis＝0.83*CXCL10+0.03*Pro-C3-3.02。
在发现队列中，N2-Fibrosis区分伴有高危NASH中的AUROC为0.784，验证队列中为0.856。
在实际应用中，N3-NASH Cut-off值高于0.68诊断为广义NASH的患者中，N2-Fibrosis Cut-off临界值高于0.37的患者，诊断为高危NASH，需要进行三级管理。Cut-off值在0.30～0.37之间的患者，需要进一步肝穿刺活检确认。Cut-off临界值低于0.30的患者，诊断为低危NASH，仅需要进行生活方式干预的初级管理(图9)。
以上对本发明做了详尽的描述，其目的在于让熟悉此领域技术的人士能够了解本发明的内容并加以实施，并不能以此限制本发明的保护范围，凡根据本发明的精神实质所作的等效变化或修饰，都应涵盖在本发明的保护范围内。

Claims (19)

  1. 一种生物标志物组合物，其特征在于，所述的生物标志物组合物包括诊断非酒精性脂肪性肝病和/或非酒精性脂肪性肝炎的相关蛋白标志物和/或临床生化标志物，所述的诊断非酒精性脂肪性肝病和/或非酒精性脂肪性肝炎的相关蛋白标志物选自CXCL10、CK-18、P62/SQSTM1、SQLE、CA3、Pro-C3或FGF21中一种或多种；所述的临床生化标志物选自BMI、HDL-C、HbA1c、ALT、AST、LDL-C、TG、TC、ALP或PLT中的一种或多种。
  2. 根据权利要求1所述的生物标志物组合物，其特征在于，所述的生物标志物组合物包括CXCL10、CK-18、AST、P62/SQSTM1、SQLE、Pro-C3、FGF21、BMI、HbA1c、ALT、LDL-C和PLT。
  3. 根据权利要求1所述的生物标志物组合物，其特征在于，所述的生物标志物组合物由CXCL10、CK-18、P62/SQSTM1、ALT、SQLE、HbA1c、FGF21、PLT和LDL-C组成。
  4. 根据权利要求1所述的生物标志物组合物，其特征在于，所述的生物标志物组合物由CXCL10，CK-18，Pro-C3，AST和BMI组成。
  5. 根据权利要求1所述的生物标志物组合物，其特征在于，所述的生物标志物组合物由CXCL10，CK-18和BMI组成。
  6. 根据权利要求1所述的生物标志物组合物，其特征在于，所述的生物标志物组合物由CXCL10和Pro-C3组成。
  7. 一种非酒精性脂肪性肝病和/或非酒精性脂肪性肝炎诊断模型，其特征在于，收集非酒精性脂肪性肝病和/或非酒精性脂肪性肝炎患者的血液样本，分别测量所述的非酒精性脂肪性肝炎相关蛋白标志物的水平和/或所述的临床生化标志物的水平，使用神经网络(Neural Network)、朴素贝叶斯(Bayesian)、逻辑回归(Logistic Regression)和10折交叉验证建立所述的非酒精性脂肪性肝病和/或非酒精性脂肪性肝炎诊断模型，所述的非酒精性脂肪性肝炎相关蛋白标志物选自CXCL10、CK-18、P62/SQSTM1、SQLE、CA3、Pro-C3或FGF21中一种或多种；所述的临床生化标志物选自BMI、HDL-C、HbA1c、ALT、AST、LDL-C、TG、TC、ALP或PLT中的一种或多种。
  8. 根据权利要求7所述的非酒精性脂肪性肝病和/或非酒精性脂肪性肝炎诊断模型，其特征在于，其构建方法具体为：收集非酒精性脂肪性肝病患者(包括非酒精性脂肪性肝炎患者和单纯性脂肪变性患者)的血液样本，分别测量所述的诊断非酒精性脂肪性肝病和/或非酒精性脂肪性肝炎的相关蛋白标志物的水平和/或所述的临床生化标志物的水平，将非酒精性脂肪性肝病患者和健康者的标志物的水平使用支持向量机、逻辑回归、朴素贝叶斯和10折交叉 验证建立非酒精性脂肪性肝病诊断模型，和/或，将非酒精性脂肪性肝炎患者和单纯性脂肪变性患者的标志物的水平使用支持向量机、逻辑回归、朴素贝叶斯和10折交叉验证建立非酒精性脂肪性肝炎诊断模型。
  9. 根据权利要求7所述的非酒精性脂肪性肝病和/或非酒精性脂肪性肝炎诊断模型，其特征在于，所述的非酒精性脂肪性肝病诊断模型中使用的生物标志组合物为权利要求3所述的生物标志物组合物；
    所述的非酒精性脂肪性肝炎诊断模型中使用的生物标志组合物为权利要求3至5中任一项所述的生物标志物组合物。
  10. 根据权利要求7所述的非酒精性脂肪性肝病和/或非酒精性脂肪性肝炎诊断模型，其特征在于，所述非酒精性脂肪性肝炎患者包括高危和低危非酒精性脂肪性肝炎患者，将所述高危和低危非酒精性脂肪性肝炎患者实验逻辑回归建立高危非酒精性脂肪性肝炎诊断模型，所述的非酒精性脂肪性肝病诊断模型中使用的生物标志组合物为权利要求6所述的生物标志物组合物。
  11. 一种根据权利要求1～6中任一项所述的生物标志物组合物在待测样本中的水平，通过建立起来的逻辑回归或人工智能模型来诊断非酒精性脂肪性肝病和/或非酒精性脂肪性肝炎。
  12. 一种试剂盒，其特征在于，所述的试剂盒包括用于检测权利要求1至6中任一项所述的生物标志物组合物的水平的检测试剂。
  13. 根据权利要求12所述的试剂盒，其特征在于，所述的试剂盒还包括标准品，所述的标准品包括权利要求1至6中任一项所述的生物标志物组合物。
  14. 一种如权利要求1至6中任一项所述的生物标志物组合物在制备试剂盒中的用途，其特征在于，所述的试剂盒包括用于检测所述的生物标志物组合物的水平的检测试剂，所述的试剂盒用于区分非酒精性脂肪性肝病和健康者，和/或，所述的试剂盒用于区分非酒精性脂肪性肝炎和单纯性脂肪变性，和/或，所述的试剂盒用于区分高危非酒精性脂肪性肝炎和低危非酒精性脂肪性肝炎。
  15. 根据权利要求14所述的用途，其特征在于，所述的区分方法为：提供一种来源于待测对象的样品，分别测试样品中的所述的诊断非酒精性脂肪性肝病和/或非酒精性脂肪性肝炎的相关蛋白标志物和所述的临床生化标志物的水平，然后代入神经网络、朴素贝叶斯、逻辑回归和10折交叉验证确定最佳生物标志物组合，根据人工智能模型的Cut-off值来区分非酒 精性脂肪性肝病和/或非酒精性脂肪性肝炎和/或单纯性脂肪变性和/或健康者，所述的Cut-off值由最大化约登指数的ROC分析确定，所述的最大化约登指数＝灵敏度+特异性-1。
  16. 根据权利要求14所述的用途，其特征在于，所述的样品为血液。
  17. 根据权利要求14所述的应用，其特征在于，所述的样品来自人。
  18. 一种非酒精性脂肪性肝病和/或非酒精性脂肪性肝炎的诊断方法，其特征在于，测试患者血液样品中权利要求3至5中任一项所述的生物标志物组合物的水平，代入支持向量机、逻辑回归、朴素贝叶斯和10折交叉验证确定最佳生物标志物组合，并进行ROC曲线分析，
    在非酒精性脂肪性肝病诊断模型中，根据Cut-off值来区分非酒精性脂肪性肝病和健康者，若Cut-off值≥Cut-off临界值，则诊断为非酒精性脂肪性肝病，若Cut-off值<Cut-off临界值，则诊断为健康者；
    在非酒精性脂肪性肝炎诊断模型中，根据Cut-off值来区分非酒精性脂肪性肝炎和单纯性脂肪变性，所述的Cut-off值包括低Cut-off临界值和高Cut-off临界值，若患者Cut-off值≥高Cut-off临界值，则诊断为非酒精性脂肪性肝炎，若Cut-off值<低Cut-off临界值，则诊断为单纯性脂肪变性，若低Cut-off临界值≤Cut-off值<高Cut-off临界值，则需要进行肝穿刺活检；
    采用权利要求6所述的生物标志物组合物在高危非酒精性脂肪性肝炎诊断模型中，根据Cut-off值来区分高危非酒精性脂肪性肝炎和低危非酒精性脂肪性肝炎，所述的Cut-off值包括低Cut-off临界值和高Cut-off临界值，若患者Cut-off值≥高Cut-off临界值，则诊断为高危非酒精性脂肪性肝炎，若Cut-off值<低Cut-off临界值，则诊断为低危非酒精性脂肪性肝炎，若低Cut-off临界值≤Cut-off值<高Cut-off临界值，则需要进行肝穿刺活检。
  19. 根据权利要求18所述的诊断方法，其特征在于，当所述的生物标志物组合物为权利要求3所述的生物标志物组合物时，非酒精性脂肪性肝病诊断模型中，Cut-off临界值为0.58时，若Cut-off值≥0.58，则判为非酒精性脂肪性肝病，Cut-off值<0.58判为健康者；非酒精性脂肪性肝炎诊断模型中，Cut-off临界值为0.47，若Cut-off值≥0.47，则诊断为非酒精性脂肪性肝炎，若Cut-off值<0.47，则诊断为单纯性脂肪变性；
    当所述的生物标志物组合物为权利要求4所述的生物标志物组合物时，非酒精性脂肪性肝炎诊断模型中，低Cut-off临界值为0.27、高Cut-off临界值为0.61，若Cut-off值≥0.61，则诊断为非酒精性脂肪性肝炎，若Cut-off值<0.27，则诊断为单纯性脂肪变性，若0.27≤Cut-off值<0.61，则需要进行肝穿刺活检；
    当所述的生物标志物组合物为权利要求5所述的生物标志物组合物时，非酒精性脂肪性 肝炎诊断模型中，低Cut-off临界值为0.43、高Cut-off临界值为0.68，若Cut-off值≥0.68，则诊断为非酒精性脂肪性肝炎，若Cut-off值<0.43，则诊断为单纯性脂肪变性，若0.43≤Cut-off值<0.68，则需要进行肝穿刺活检。
    当所述的生物标志物组合物为权利要求6所述的生物标志物组合物时，高危非酒精性脂肪性肝炎诊断模型中，低Cut-off临界值为0.30、高Cut-off临界值为0.37，若Cut-off值≥0.37，则诊断为高危非酒精性脂肪性肝炎，需要进行三级管理。若Cut-off值<0.30，则诊断为低危非酒精性脂肪性肝炎，仅需要进行初级管理，包括生活方式干预和年度评估，若0.30≤Cut-off值<0.37，则需要进行肝穿刺活检。
PCT/CN2023/075430 2022-02-10 2023-02-10 非酒精性脂肪性肝炎生物标志物组合物及其应用 WO2023151651A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210123907.1 2022-02-10
CN202210123907 2022-02-10

Publications (1)

Publication Number Publication Date
WO2023151651A1 true WO2023151651A1 (zh) 2023-08-17

Family

ID=87544155

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/075430 WO2023151651A1 (zh) 2022-02-10 2023-02-10 非酒精性脂肪性肝炎生物标志物组合物及其应用

Country Status (2)

Country Link
CN (1) CN116577509A (zh)
WO (1) WO2023151651A1 (zh)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2019122342A1 (en) * 2017-12-22 2019-06-27 Metanomics Health Gmbh Method for the assessment of nafld
CN110218788A (zh) * 2018-03-02 2019-09-10 香港中文大学 角鲨烯环氧酶在非酒精性脂肪性肝病的诊断和治疗中的应用
US20200248261A1 (en) * 2017-08-25 2020-08-06 Genfit Non-invasive diagnostic of non-alcoholic fatty liver diseases, non-alcoholic steatohepatitis and/or liver fibrosis
WO2021157631A1 (ja) * 2020-02-04 2021-08-12 デンカ株式会社 非アルコール性脂肪肝炎の検出を補助する方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200248261A1 (en) * 2017-08-25 2020-08-06 Genfit Non-invasive diagnostic of non-alcoholic fatty liver diseases, non-alcoholic steatohepatitis and/or liver fibrosis
WO2019122342A1 (en) * 2017-12-22 2019-06-27 Metanomics Health Gmbh Method for the assessment of nafld
CN110218788A (zh) * 2018-03-02 2019-09-10 香港中文大学 角鲨烯环氧酶在非酒精性脂肪性肝病的诊断和治疗中的应用
WO2021157631A1 (ja) * 2020-02-04 2021-08-12 デンカ株式会社 非アルコール性脂肪肝炎の検出を補助する方法

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
LIU DABIN; WONG CHI CHUN; ZHOU YUNFEI; LI CHUANGEN; CHEN HUARONG; JI FENFEN; GO MINNIE Y.Y.; WANG FEIXUE; SU HAO; WEI HONG; CAI ZO: "Squalene Epoxidase Induces Nonalcoholic Steatohepatitis Via Binding to Carbonic Anhydrase III and is a Therapeutic Target", GASTROENTEROLOGY, ELSEVIER INC., US, vol. 160, no. 7, 27 February 2021 (2021-02-27), US , pages 2467, XP086580504, ISSN: 0016-5085, DOI: 10.1053/j.gastro.2021.02.051 *
WANG XIAOJUAN, ZHANG XIANG, CHU EAGLE S. H., CHEN XIAOTING, KANG WEI, WU FENG, TO KA‐FAI, WONG VINCENT W. S., CHAN HENRY L. Y., CH: "Defective lysosomal clearance of autophagosomes and its clinical implications in nonalcoholic steatohepatitis", THE FASEB JOURNAL, FEDERATION OF AMERICAN SOCIETIES FOR EXPERIMENTAL BIOLOGY, US, vol. 32, no. 1, 1 January 2018 (2018-01-01), US, pages 37 - 51, XP093083668, ISSN: 0892-6638, DOI: 10.1096/fj.201601393R *

Also Published As

Publication number Publication date
CN116577509A (zh) 2023-08-11

Similar Documents

Publication Publication Date Title
Vatsalya et al. Keratin 18 is a diagnostic and prognostic factor for acute alcoholic hepatitis
Tovo et al. Transient hepatic elastography has the best performance to evaluate liver fibrosis in non-alcoholic fatty liver disease (NAFLD)
Liu et al. [Retracted] SOFA Score in relation to Sepsis: Clinical Implications in Diagnosis, Treatment, and Prognostic Assessment
Loomba et al. Validation of serum test for advanced liver fibrosis in patients with nonalcoholic steatohepatitis
WO2022025069A1 (ja) 疾患リスク評価方法、疾患リスク評価装置、及び疾患リスク評価プログラム
Kusvuran Ozkan et al. Can high-sensitivity C-reactive protein and ferritin predict functional outcome in acute ischemic stroke? A prospective study
CN107255711B (zh) 骨桥蛋白用于制备或筛选慢加急性肝衰竭诊断试剂的用途
Cho et al. Point shear wave elastography predicts fibrosis severity and steatohepatitis in alcohol-related liver disease
Ge et al. Monitoring of intestinal inflammation and prediction of recurrence in ulcerative colitis
CN117288963A (zh) 一种hfrep1蛋白用于制备诊断急性心肌梗死、脑梗死的试剂盒的用途
WO2023151651A1 (zh) 非酒精性脂肪性肝炎生物标志物组合物及其应用
CN102175839B (zh) IgG核心岩藻糖基化水平检测方法及其应用
Kamal et al. Serum beta-2 microglobulin is a reliable biomarker to predict diabetic nephropathy
CN116047082B (zh) 一种fgl1蛋白用于制备诊断慢性肾脏病的试剂盒的用途
TWI661198B (zh) 診斷或預斷人類口腔癌的方法
Ekpe et al. The relevance of ascitic lactate dehydrogenase (LDH) and serum ascites albumin gradient (SAAG) in the differential diagnosis of ascites among patients in a Nigerian hospital.
Zheng et al. Machine learning-based mortality prediction models for non-alcoholic fatty liver disease in the general United States population
CN111613327B (zh) 利用基于logistic回归的多发性骨髓瘤诊断模型开发的系统及其应用
Hu et al. Predictive models for Talaromyces marneffei infection in HIV-infected patients using routinely collected data
CN116679042A (zh) 一种fgl1蛋白用于制备诊断急性胰腺炎的试剂盒的用途
CN116840481A (zh) Serpinb5在胃癌前病变标志物中的应用
El-Kafrawy et al. Study and Clinical Significance of Urinary Levels of Glypican 5 in Diabetic Patients
El-Toukhy et al. Assessment of Spleen Stiffness for Prediction of Varices in Patients with Hepatitis C Related Cirrhosis Using Transient Elastography
TWI666449B (zh) 診斷或預斷人類口腔癌的方法
CN117007822A (zh) 一种用于筛查患精神分裂症风险的标志物及其用途

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23752430

Country of ref document: EP

Kind code of ref document: A1