Serum protein combination marker for monitoring liver fibrosis reversion of chronic hepatitis B patients after long-term treatment and application of serum protein combination marker

Abstract

The invention discloses a serum protein combination marker for monitoring liver fibrosis reversion of chronic hepatitis B patients after long-term treatment and application thereof. The invention first discloses a serum protein combination marker for monitoring liver fibrosis reversion of chronic hepatitis B patients after long-term antiviral treatment, and the serum protein combination marker comprises SERPINA7, CD163 and CFHR4. The invention further discloses application of the marker in preparation of products for monitoring liver fibrosis reversion of chronic hepatitis B patients after long-term treatment. By detecting the content ratio of the serum protein combination marker before and after long-term antiviral treatment of chronic hepatitis B patients, the liver fibrosis reversion state can be accurately monitored, and the performance is obviously superior to that of instantaneous elastography and conventional noninvasive indexes. In addition, the serum protein combined marker is based on blood detection, is noninvasive and convenient, and can provide a reliable basis for treatment effect evaluation and clinical decision.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of biological medicine. More particularly, it relates to serum protein biomarker for monitoring liver fibrosis reversal in chronic hepatitis B patients after long-term treatment and its application. BACKGROUND

[0002] Long-term antiviral therapy (AVT) for chronic hepatitis B (CHB) is a key strategy to control disease progression and reduce the risk of end-stage liver disease. For CHB patients receiving long-term AVT, accurate monitoring of the reversal status of liver fibrosis is an important link to evaluate treatment effect, predict clinical prognosis and guide individualized management. However, the existing liver fibrosis monitoring methods have significant limitations in the specific treatment background of long-term AVT, and it is difficult to meet the needs of accurate and dynamic assessment.

[0003] Liver biopsy (LBx): As the "gold standard" for liver fibrosis assessment, its inherent invasiveness, operational risk, sampling error and poor patient compliance make it impossible to be used as a dynamic monitoring tool for repeated assessment during long-term treatment. Transient elastography (TE): TE has important value in the assessment of liver fibrosis in naive patients, but its results are significantly affected by the activity of liver inflammation. In the context of long-term AVT, with the effective control and resolution of intrahepatic inflammation, TE values often decrease, but this decrease is difficult to reliably distinguish between the resolution of inflammation and the true degradation and reversal of fibrosis matrix, resulting in inaccurate reflection of the dynamic changes of fibrosis remodeling process. Serum markers and emerging non-invasive diagnostic methods (NITs): including traditional serum models such as APRI, FIB-4 and some emerging non-invasive methods, which have certain value in the baseline fibrosis assessment of naive patients. However, during long-term AVT, the patient's internal environment undergoes complex and continuous dynamic changes (such as sustained viral suppression, changes in host metabolic state, potential viral variation / drug resistance, and the interweaving of inflammation resolution and tissue repair processes). These changes often lead to a significant decrease in the sensitivity and specificity of the above serum markers and NITs, and their ability to distinguish between fibrosis reversal and non-reversal states (diagnostic performance) is significantly insufficient, which cannot meet the clinical needs for accurate monitoring.

[0004] Although long-term AVT can effectively suppress viral replication and control inflammation, there is high heterogeneity in the proportion of patients achieving significant reversal of liver fibrosis, and the risk of long-term liver-related adverse events (such as liver decompensation and hepatocellular carcinoma) in patients who do not achieve fibrosis reversal is still significantly increased. Unlike short-term treatment, the reversal of liver fibrosis after long-term AVT involves a more complex and long-term pathophysiological process, including but not limited to: dynamic degradation and remodeling balance of extracellular matrix (ECM) components; resolution and immune regulation of persistent (even low-level) inflammation; reversal and apoptosis of activated hepatic stellate cells (HSCs); long-term effects of liver tissue repair and regeneration mechanisms. This unique, long-term evolving pathological mechanism makes it difficult for existing monitoring tools that rely on a single mechanism or static indicators (whether a transient "snapshot" of LBx, physical property measurements of TE, or models based on conventional biochemical indicators) to comprehensively, dynamically, and accurately capture and reflect the true progress of fibrosis reversal.

[0005] Notably, preliminary studies have shown that after long-term AVT, the serum proteomic characteristics of patients who achieve fibrosis reversal and those who do not may undergo significant remodeling, suggesting the presence of potential specific biomarkers that can reflect this complex pathological process. However, existing studies have obvious limitations: lack of systematic, large cohort studies on longitudinal, dynamic serum proteomic changes in patients undergoing long-term AVT; inability to deeply analyze the clear correlation between these changes and the pathophysiological mechanisms of fibrosis reversal / progression (such as specific ECM metabolic pathways, inflammation / repair signaling pathways); and failure to discover and verify specific serum protein combination markers with high diagnostic performance suitable for long-term dynamic monitoring based on these findings.

[0006] Therefore, in view of the unique and complex pathophysiological mechanisms of liver fibrosis reversal in the specific clinical scenario of long-term antiviral therapy, and the widespread shortcomings of existing monitoring tools, there is an urgent need to develop new, non-invasive, high-sensitivity, and high-specificity monitoring tools. Among them, based on serum proteomic analysis that better reflects long-term dynamic pathological processes, discovering and verifying specific combination biomarkers that can accurately distinguish fibrosis reversal status has become a key scientific problem and important clinical need that needs to be addressed. SUMMARY

[0007] One object of the present application is to provide a serum protein combination marker for monitoring liver fibrosis reversal in chronic hepatitis B patients after long-term treatment, so as to achieve monitoring of the liver fibrosis reversal status in chronic hepatitis B patients after 260 weeks of antiviral therapy.

[0008] Another object of the present application is to improve the application of the above-mentioned serum protein combination marker in the preparation of products for monitoring the reversal status of liver fibrosis in chronic hepatitis B patients after long-term antiviral therapy.

[0009] To achieve the above object, the present application adopts the following technical solutions:

[0010] The present application first provides a serum protein combination marker for monitoring the reversal of liver fibrosis in chronic hepatitis B patients after long-term antiviral treatment, which comprises serpin family A member 7 (SERPINA7), CD163 molecule (CD163) and complement factor H-related protein 4 (CFHR4); wherein the UniprotKB IDs of the SERPINA7, CD163 and CFHR4 are P05543, Q86VB7 and Q92496, respectively.

[0011] In a specific embodiment of the present application, the serum protein combination marker is derived from tissues or blood, etc.; in a preferred embodiment of the present application, the protein marker is derived from blood.

[0012] The present application further provides the use of the above-mentioned serum protein combination marker or its detection reagent in the preparation of a product for monitoring the reversal status of liver fibrosis in chronic hepatitis B patients after long-term antiviral treatment.

[0013] In a specific embodiment of the present application, the detection reagent detects the content ratio of the serum protein combination marker after long-term antiviral treatment to that before antiviral treatment in chronic hepatitis B patients.

[0014] In a specific embodiment of the present application, the detection reagent comprises any one or more of the reagents used in ELISA, suspension chip, immunoblotting, MSD electrochemiluminescence technology and mass spectrometry.

[0015] In a preferred embodiment of the present application, the detection reagent is the reagent used in mass spectrometry; the content ratio of the above-mentioned serum protein combination marker after long-term antiviral treatment to that before antiviral treatment is detected by mass spectrometry, i.e. the content ratio; more preferably, the mass spectrometry comprises 4D-DIA-MS and / or PRM-MS.

[0016] In a specific embodiment of the present application, the product can be a kit.

[0017] The present application further provides a kit for monitoring the reversal status of liver fibrosis in chronic hepatitis B patients after long-term antiviral treatment, which comprises a reagent having the ability to detect the content ratio of the above-mentioned serum protein combination marker after long-term antiviral treatment to that before antiviral treatment in chronic hepatitis B patients.

[0018] The application also provides a model for monitoring the liver fibrosis reversal state of a chronic hepatitis B patient after long-term antiviral treatment, which is based on the content ratio of the above-mentioned serum protein combination marker before and after long-term antiviral treatment of the chronic hepatitis B patient, and the model formula is as follows: P = 1 / (1+exp(-(2.2135-0.2958xSERPINA7'-1.6642xCD163'+0.293xCFHR4')), if P>0.68 (critical value) is determined to be reversed, otherwise not reversed; in the formula, SERPINA7', CD163' and CFHR4' all represent the content ratio of SERPINA7, CD163 and CFHR4 before and after long-term antiviral treatment of the chronic hepatitis B fibrosis patient; wherein the UniprotKB IDs of the SERPINA7, CD163 and CFHR4 are P05543, Q86VB7 and Q92496, respectively.

[0019] The application further provides a system for monitoring the liver fibrosis reversal state of a chronic hepatitis B patient after long-term antiviral treatment, comprising:

[0020] a detection device for obtaining the content ratio of the above-mentioned serum protein combination marker before and after long-term antiviral treatment of the chronic hepatitis B fibrosis patient;

[0021] a data processing device for calculating P by the above-mentioned model;

[0022] a result determination device for monitoring the liver fibrosis reversal state of the chronic hepatitis B patient after long-term antiviral treatment according to P: if P>critical value, it is determined to be reversed, otherwise not reversed; the critical value is 0.68.

[0023] In the application, the period of long-term antiviral treatment is ≥156 weeks; in a specific embodiment of the application, the period of long-term antiviral treatment is 260 weeks.

[0024] The beneficial effects of the application are as follows:

[0025] The serum protein combination marker (SERPINA7, CD163 and CFHR4) of the application can accurately monitor the liver fibrosis reversal state by detecting the content ratio before and after long-term antiviral treatment (260 weeks) of the chronic hepatitis B patient, and the performance is significantly better than that of transient elastography and conventional non-invasive indicators. In addition, the serum protein combination marker is based on blood detection, non-invasive and convenient, and can provide reliable basis for treatment effect evaluation and clinical decision-making. BRIEF DESCRIPTION OF DRAWINGS

[0026] The specific embodiments of the application will be further described in detail below with reference to the accompanying drawings.

[0027] Figure 1 To find the differential abundance of proteins (DAPs) between patients with CHB in the discovery cohort training set after long-term AVT and those without reverse fibrosis, and the diagnostic efficiency (AUC value) of these proteins for liver fibrosis reversal: A shows the Venn diagram of specific or common DAPs between patients with and without reverse fibrosis after long-term AVT, wherein, ↑: up-regulation; ↓: down-regulation; B shows the diagnostic efficiency (AUC value) of the expression abundance of DAPs in patients with and without reverse fibrosis after long-term AVT for liver fibrosis reversal.

[0028] Figure 2 To validate the internal validation results based on the discovery cohort training set: A shows the comparison of the performance (AUC value) of the serum protein combined marker (compared with conventional NITs: LSM, APRI, FIB4) in monitoring the fibrosis reversal status after 260 weeks of long-term AVT; B shows the calibration curve of the monitoring probability of the serum protein combined marker and conventional NITs.

[0029] Figure 3 The serum protein combined marker model based on the PRM-MS validation group is used to monitor the fibrosis reversal status after 260 weeks of long-term antiviral therapy, and the receiver operating characteristic (ROC) curve is shown. DETAILED DESCRIPTION

[0030] In order to more clearly illustrate the present application, the present application will be further described below in conjunction with preferred embodiments and drawings. Similar components are denoted by the same reference numerals in the drawings. Those skilled in the art should understand that the specific description below is illustrative rather than limiting, and should not limit the scope of protection of the present application.

[0031] The population of the present application was screened from the nationwide multicenter randomized controlled clinical trial and its extension study (ClinicalTrials.gov Identifier: NCT01938781, NCT01938820, NCT03777969, NCT02849132) to select chronic hepatitis B (CHB) patients undergoing antiviral therapy (AVT);

[0032] Among them, the inclusion criteria are as follows:

[0033] 1) Age ≥ 18 years old;

[0034] 2) Paired liver biopsy (LBx) before treatment (i.e. baseline or initial treatment) and after treatment (AVT treatment for 260 weeks) is available, or clinical diagnosis of compensated cirrhosis (without LBx) before treatment and LBx after treatment is available;

[0035] 3) Paired blood samples available at pre-treatment (i.e. baseline or naive) and post-treatment LBx time points;

[0036] 4) HBsAg positive for at least 6 months prior to AVT;

[0037] 5) HBV DNA level >20,000 IU / ml for HBeAg positive patients, or HBV DNA level >2,000 IU / ml for HBeAg negative patients;

[0038] 6) LBx core length ≥0.5 cm and contains at least 5 portal tracts.

[0039] Exclusion criteria are as follows:

[0040] 1) Current or past presence of liver decompensation (e.g. ascites, variceal bleeding or hepatic encephalopathy);

[0041] 2) Serum alpha-fetoprotein level >100 ng / mL or creatinine level >1.5 times the upper limit of normal;

[0042] 3) Co-infection with hepatitis C virus or human immunodeficiency virus;

[0043] 4) Other serious liver disease or major organ dysfunction;

[0044] 5) Pregnancy or lactation.

[0045] Patients who met the inclusion and exclusion criteria were divided into discovery group and validation group for subsequent serum biomarker discovery.

[0046] To ensure consistency within all participating clinical research centers, a standardized protocol was implemented, including:

[0047] 1) All clinical research centers uniformly adopted the same patient enrollment eligibility criteria;

[0048] 2) All patients received a standardized AVT regimen mainly with entecavir;

[0049] 3) According to the established follow-up protocol, systematic clinical evaluation was performed every six months;

[0050] 4) The main outcome focused on histological fibrosis reversal and the incidence of predetermined clinical endpoints.

[0051] All enrolled patients signed a written informed consent. The present invention has been approved by the Ethics Committee of Beijing Friendship Hospital, Capital Medical University (2016-P2-021-04) and follows the relevant principles of the "Helsinki Declaration".

[0052] The methods used in the following examples are as follows:

[0053] 1. Histological assessment

[0054] Formalin-fixed paraffin-embedded liver biopsy (LBx) tissue blocks were serially sectioned (5 μm thickness), stained with reticulin, and independently assessed by two pathologists blinded to the experimental conditions, LBx time point of sampling, and other clinical information. Cases with discordant assessments were reviewed by a third senior pathologist. Histological changes were assessed using the Ishak score and progression-uncertain-reversal (PIR) score, and classified as fibrosis reversers or fibrosis non-reversers:

[0055] Fibrosis reversers were defined as follows:

[0056] For patients with paired LBx: a decrease in Ishak score of >1 point from baseline (ΔIshak >1) after treatment, or stable Ishak score after treatment but with PIR score showing predominantly reversal changes.

[0057] For patients with clinical diagnosis of compensated cirrhosis before treatment: Ishak stage ≤4 (ΔIshak >1) after treatment, or Ishak stage 5-6 after treatment but with histology showing predominantly reversal changes.

[0058] Fibrosis non-reversers were defined as follows: patients not meeting the above criteria for fibrosis reversers.

[0059] Baseline absence of steatosis was defined as follows:

[0060] For patients with paired LBx: histological assessment before treatment showing <5% steatosis area.

[0061] For patients with clinical diagnosis of compensated cirrhosis before treatment: absence of steatosis was determined by the absence of evidence of fatty liver on ultrasound examination.

[0062] 2. Clinical diagnosis of compensated cirrhosis

[0063] Diagnosis of compensated cirrhosis required meeting any of the following criteria:

[0064] 1) Endoscopy: if the patient had undergone endoscopy, the presence of esophageal and / or gastric varices had to be confirmed and the possibility of non-cirrhotic portal hypertension excluded.

[0065] 2) No endoscopy: if endoscopy was not performed, at least two of the following four criteria had to be met:

[0066] Imaging examinations: Imaging examinations such as ultrasound, CT, or MRI show typical manifestations of cirrhosis, including irregular liver surface, granular or nodular liver parenchyma, with or without splenomegaly (defined as spleen thickness >4.0 cm or greater than 5 intercostal spaces).

[0067] Hematological indicators: Routine blood tests showed platelet count <100×10^9 / L with no other explainable reasons.

[0068] Abnormal serum and coagulation function: serum albumin level <35 g / L, or prothrombin time prolonged by more than 3 seconds (after discontinuation of thrombolytic or anticoagulant drugs for more than 7 days), or decreased cholinesterase level (excluding the influence of anticholinesterase drugs).

[0069] Liver stiffness measurement: Transient elastography (LSM) showed that the liver stiffness was >12.4 kPa.

[0070] 3. Processing of serum samples

[0071] 1) Collection and Pretreatment: Blood samples were collected from patients with chronic hepatitis B fibrosis before (baseline) and after antiviral treatment using standard venipuncture into vacutainer tubes. Serum was separated by centrifugation at 2000 g for 10 minutes at 4°C. Serum was rapidly aliquoted and stored in a -80°C freezer.

[0072] 2) Removal of high-abundance proteins: Take 40 μL of frozen serum and apply High-abundance proteins (e.g., albumin, which accounts for 70–90% of total serum protein) were removed using an Albumin / IgG Depletion Kit (Millipore, USA). The depleted samples were lyophilized and concentrated, reconstituted with SDS lysis buffer (Beyotime, China), and then centrifuged at 12,000 g for 10 min at room temperature to collect the clear supernatant.

[0073] 3) Protein quantification: The protein concentration in the supernatant was determined using the bicinchoninic acid (BCA) method.

[0074] 4) Electrophoresis quality control: Take a sample containing 10 μg of protein and perform 12% SDS-PAGE electrophoresis separation. TM Coomassie blue staining was performed using an LG device (GenScript, China), and images were recorded and analyzed using an automated digital gel imaging system (Tanon, China).

[0075] 5) Proteolytic pre-treatment: Based on the quantification results, equal amount of protein (50 pg) was taken and the volume was adjusted to a uniform concentration. Dithiothreitol (DTT, Thermo Fisher, USA) was added to a final concentration of 5 mM and incubated at 55 °C for 30 min for reduction reaction. Subsequently, iodoacetamide (Thermo Fisher, USA) was added to a final concentration of 10 mM and alkylated at room temperature for 15 min in the dark. Six-fold volume of pre-cooled acetone was added and the proteins were precipitated at -20 °C overnight. The precipitate was collected by centrifugation at 8,000 g for 10 min at 4 °C and dissolved in 100 pL of 50 mM ammonium bicarbonate solution (Thermo Fisher, USA).

[0076] 6) Trypsin digestion: Trypsin (1 pg, Likuso, China) was added and the digestion was performed at 37 °C overnight. The reaction was stopped by adding phosphoric acid (Sigma, USA) (pH ~ 3).

[0077] 7) Peptide desalting: The peptide mixture generated from the digestion was desalted and purified using SOLA™ SPE columns (Thermo Fisher, USA). The desalted peptides were concentrated by freeze-drying.

[0078] 8) iRT peptide addition: The freeze-dried peptides were reconstituted using a solution containing iRT standard peptides (1 : 10 ratio, Biognosys, Switzerland) to prepare the final peptide sample (i.e., the processed serum sample) for the following mass spectrometry analysis.

[0079] 4、4D-DIA-MS technology

[0080] 1) Chromatographic separation: The processed serum samples were analyzed using the Evosep One system (Evosep Biosystems, Odense, Denmark) coupled with a timsTOF Pro 2 mass spectrometer (Bruker Daltonics, MA, USA). The processed serum samples were first loaded on an Evotip C18 trap column and separated on a 15 cm long analytical column using the standardized 30 samples per day (30SPD) method with an effective gradient length of 15 min at a flow rate of 1 pL / min. The mobile phase consisted of solvent A (0.1% (v / v) formic acid in water) and solvent B (0.1% (v / v) formic acid in acetonitrile). Mass spectrometry analysis was performed in PASEF mode with key parameters including: capillary voltage 1.5 kV, dry gas flow rate 3.0 L / min, temperature 180 °C, mass-to-charge (m / z) scan range 100-1700, and gradient mode collision energy (20-59 eV).

[0081] 2) Mass spectrometry data acquisition (TimsTOF Pro MS):

[0082] DDA mode spectral library construction: ion mobility (1 / K0) range of precursor ion fragmentation: 0.85-1.3Vs / cm 2 Intensity-triggered tandem mass spectrometry (MS / MS) analysis was performed (isolation width: ±1 m / z; TopN=12, i.e., the first 12 ions with the highest intensity were selected for fragmentation).

[0083] DIA mode analysis of samples using an extended ion mobility range of 0.6-1.6 Vs / cm 2 ; Use a variable isolation window covering the full m / z range; synchronize cycle time with chromatographic peak width.

[0084] 3) Data processing and quantification (Spectronaut Pulsar TM v15.3):

[0085] Database search: Based on the UniProt human proteome database (filter conditions: reviewed_yes + taxonomy_9606), trypsin digestion was set (allowing up to 2 missed cleavage sites). The fixed modification was cysteine ​​alkylation, and the variable modification was methionine oxidation.

[0086] Spectral library construction and DIA analysis: A spectral library was constructed using DDA data, and DIA data were analyzed based on it. A 1% FDR threshold was applied for quality control at the PSM, peptide, and protein levels.

[0087] Quantification strategy: Protein quantification is based on MS2 fragment ion intensities. Peptide abundance is calculated as the average of the intensities of the three most promiscuous precursor ions, while protein abundance is calculated as the average of the abundances of the three most abundant unique peptides for that protein.

[0088] Data filtering and preprocessing: Immunoglobulins and proteins with >50% missing values ​​were removed. Remaining missing values ​​were imputed using the missForest algorithm. All sample data were globally normalized using the vsn algorithm, and the final output was an abundance matrix representing protein content (based on MS signal intensity).

[0089] 5. Serum proteome expression analysis

[0090] Sample division: Using the caret software package, stratified sampling was performed on the discovery group samples according to fibrosis reversal status, and the samples were divided into a model training set and an internal test set in a ratio of 60%:40%.

[0091] Differential Abundance Proteins (DAPs) identification: Based on the protein abundance matrix of the training set, the ratio of MS signal intensity of each patient's post-antiviral therapy (Post) to pre-antiviral therapy (Pre) (referred to as Post / Pre) was calculated. Using paired t-test or Wilcoxon signed rank test (according to the data distribution), combined with the fold change (FC) threshold (FC>1.2), DAPs with statistical significance (p<0.05) after long-term AVT were screened.

[0092] DAPs feature analysis: Using the VENNY 2.1 online tool, through the Venn diagram analysis, the specific and common DAPs between the fibrosis reversal patients and non-reversal patients in the training set were identified and displayed.

[0093] 6. Machine learning (ML) model construction

[0094] Feature selection: Based on the DAPs identified in the training set, the predictive ability of the Post / Pre ratio of each protein on the fibrosis reversal state was evaluated using univariate logistic regression (stats package implementation), and the significantly related features were screened.

[0095] Model training: Using the selected protein features, an L2 regularized logistic regression model (ridge regression) was constructed. During the model training process, the regularization penalty coefficient λ was optimized by 10-fold cross-validation. This step was completed using the glmnet software package.

[0096] 7. PRM-MS technology verification

[0097] Serum protein preparation and spectrum library construction based on data-dependent acquisition (DDA) strictly followed the same protocol as the aforementioned 4D-data-independent acquisition mass spectrometry (4D-DIA-MS) experiment to ensure methodological homogeneity. Targeted PRM-MS analysis was performed on a timsTOF Pro 2 mass spectrometer (Bruker Daltonics) coupled with a nanoElute liquid chromatography system (Bruker Daltonics): Trypsin-digested peptides were separated on a reversed-phase C18 chromatographic column (75 μm x 25 cm, 1.6 μm, 120 A, Dr. Maisch GmbH) using a 60 min gradient of 0.1% formic acid in water (solvent A) and 0.1% formic acid in acetonitrile (solvent B) at a flow rate of 300 nL / min. Separation was performed on an Ion Opticks (Melbourne, Australia) at a flow rate of 300 nL / min with mobile phases A (0.1% formic acid in water) and B (0.1% formic acid in acetonitrile) using a 60-min nonlinear gradient program (0–45 min: 2% → 22% B; 45–50 min: 22% → 37% B; 50–55 min: 37% → 80% B; 55–60 min: 80% B). The mass spectrometer was operated in PRM-PASEF mode in positive ion mode. The precursor ion list was configured using the timsControl PRM module. Key parameters included a capillary voltage of 1.4 kV, a scan mass-to-charge ratio range of 100–1700, and an ion mobility (1 / K0) window of 0.6–1.6 Vs / cm. 2 , ion accumulation / release time of 100 ms, and a preset retention time window of 10 minutes for scheduled acquisition.

[0098] Data processing (SpectroDive v11.8 software): First, the extracted ion chromatogram (XIC) peaks of the fragment ions were automatically extracted and manually reviewed to ensure accurate alignment of retention times across samples. Peptide and protein quantification was then performed by summing the peak areas of all diagnostic ions of the target peptides, correcting them using spike-in standards to obtain relative peptide abundances, and then integrating the abundances of all target peptides for the same protein to obtain the relative content of the protein. Finally, data preprocessing was performed, and the missForest algorithm was applied to fill missing values ​​in the sparse data. The vsn algorithm was used to globally normalize the protein abundance matrix. Finally, protein abundance data represented by MS signal intensity were output for analysis.

[0099] Example 1 Screening of serum protein combination markers and model construction for monitoring liver fibrosis reversal in chronic hepatitis B patients after long-term antiviral treatment

[0100] 36 patients with chronic hepatitis B (CHB) before and after long-term antiviral therapy (AVT, 260 weeks) as the discovery group, 54 patients with chronic hepatitis B (CHB) before and after long-term antiviral therapy (AVT, 260 weeks) as the validation group, as shown in Table 1. The patient's demographic characteristics and related laboratory monitoring indicators were recorded in detail when the sample was collected, including alanine aminotransferase, aspartate aminotransferase, total bilirubin, albumin, platelets, international normalized ratio INR, HBsAg, HBeAg, HBV DNA, liver elasticity value LSM, APRI and FIB4; In addition, 15 patients in the discovery group had paired liver biopsy (LBx) before and after 260 weeks of antiviral therapy (Ishak score≥2 at baseline), and the remaining 21 patients were diagnosed as compensated cirrhosis at baseline, and only had LBx after 260 weeks of antiviral therapy. The validation group had liver biopsy (LBx) before and after 260 weeks of antiviral therapy, and according to the histological evaluation method, 69.4% (25 / 36) of patients in the discovery group achieved fibrosis reversal after long-term antiviral therapy, and 77.8% (42 / 54) of patients in the validation group achieved liver fibrosis reversal after long-term antiviral therapy.

[0101] Table 1 Characteristics of CHB patients receiving long-term AVT in the discovery group and the validation group

[0102]

[0103]

[0104]

[0105] II. Mass spectrometry detection

[0106] The paired serum samples of the discovery group were analyzed by 4D-DIA-MS technology for longitudinal serum proteomics analysis, and the abundance matrix of serum proteins was obtained. According to the reversal state, the discovery group was stratified sampling, randomly divided into training set (training) and testing set (testing), containing 60% and 40% of patients respectively, the training set was used for machine learning (ML) model development, and the testing set was reserved for subsequent validation. The differential abundance proteins (DAPs) of patients in the training set of the discovery group who achieved fibrosis reversal and did not reverse after long-term AVT were analyzed, and the results are shown in Table 2. Figure 1As shown in Table 2, 90 unique down-regulated proteins (indicated as "reversal-specific down" in the figure) and 33 up-regulated proteins (indicated as "reversal-specific up" in the figure) were identified in the reversal group, while 13 unique down-regulated proteins (indicated as "non-reversal-specific down" in the figure) and 14 up-regulated proteins (indicated as "non-reversal-specific up" in the figure) were identified in the non-reversal group. There were 22 proteins commonly down-regulated and 10 proteins commonly up-regulated in both groups.

[0107] It is worth noting that the ratio of MS signal intensity after anti-viral therapy (AVT) based on DAPs (referred to as Post / Pre) to that before anti-viral therapy was used to evaluate the status of liver fibrosis reversal after 260 weeks of anti-viral therapy by drawing a receiver operating characteristic (ROC) curve, and the results are shown in Table 2. Figure 1 As shown in Table 2, the post / pre ratio of 33 DAPs in the fibrosis reversal group after long-term AVT reached a clinically relevant discrimination ability (AUC>0.70), highlighting their potential as biomarkers for fibrosis reversal.

[0108] Table 2 DAPs of CHB patients in the training set in the fibrosis reversal group and the non-reversal group after long-term AVT

[0109]

[0110]

[0111]

[0112]

[0113]

[0114] Note: AUC, area under the curve; R-specific, DAPs specific to fibrosis reversal; NR-specific, DAPs specific to non-fibrosis reversal; common, DAPs common to both fibrosis reversal and non-reversal; FC, fold change.

[0115] III. Model development

[0116] In machine learning (ML) model development based on the discovery set training set of fibrosis reversers and non-reversers in DAPs of long-term AVT, univariate logistic regression was used for feature vector screening, and then regularized logistic regression with Ridge penalization was used for modeling. The final selected protein combination included three proteins: serine protease inhibitor family A member 7 (SERPINA7), CD163 molecule (CD163), and complement factor H-related protein 4 (CFHR4). The formula of its model is as follows: P = 1 / (1+exp(-(2.2135-0.2958xSERPINA7'-1.6642xCD163'+0.293xCFHR4')); if P>0.68 (cut-off value) is determined as reversed, otherwise as non-reversed; in the formula, SERPINA7', CD163' and CFHR4' are all expressed as the ratio of the MS signal intensity of SERPINA7, CD163 and CFHR4 of the patient with fibrosis of chronic hepatitis B after 260 weeks of antiviral treatment to that before antiviral treatment (Post / Pre, i.e. content ratio); wherein the UniprotKB ID of the SERPINA7, CD163 and CFHR4 is P05543, Q86VB7 and Q92496 respectively, see Table 2.

[0117] The model based on the above serum protein combination marker was used to monitor the liver fibrosis reversal status of the training set and test set of patients with fibrosis of chronic hepatitis B after 260 weeks of antiviral treatment, and the performance of the model was analyzed by drawing the receiver operating characteristic (ROC) curve, as shown in Table 3, and the AUC of the training set and test set reached 0.87 (95% confidence interval [CI]: 0.71-1.00) and 0.68 (95% CI: 0.28-1.00) respectively. Table 3 Area under the curve (AUC) of serum protein combination marker model in the training set and test set of 4D-DIA-MS discovery set for monitoring fibrosis reversal status after long-term ATV

[0118]

[0119] IV. Validation

[0120] 1. Internal validation

[0121] To avoid model overfitting, Bootstrap resampling method was used for internal validation in the training set. Resampling with replacement was performed 1,000 times from the training set, and the boot and calibrate software package was used to generate a calibration curve. The AUC value performance curve of the serum protein combination marker model for monitoring fibrosis reversal status after 260 weeks of AVT is shown in Figure 2.Figure 2 As shown in FIG. 2A, the AUC value of the serum protein combination marker model (indicated as "serum protein combination" in the figure) was stable at 0.87 after 1,000 times of Bootstrap self-sampling correction; the calibration curve of monitoring probability was as shown in FIG. 2B, which showed that the monitoring result was highly consistent with the actual observation (Brier score = 0.138), indicating that the model monitoring was highly reliable. Figure 2 As shown in FIG. 2A and FIG. 2B, the serum protein combination marker model showed a significantly better performance than the conventional non-invasive indicators (NITs) in monitoring the state of fibrosis reversal, such as LSM Post / Pre (AUC = 0.59, Brier = 0.214), APRI Post / Pre (AUC = 0.72, Brier = 0.184) and FIB-4 Post / Pre (AUC = 0.77, Brier = 0.195). Figure 2

[0122] 2. External validation

[0123] The PRM-MS technology was used for external validation in the validation group. The serum protein combination marker was qualitatively and quantitatively analyzed by the PRM-MS technology in the validation group, and the model based on the serum protein combination marker was used to monitor the state of liver fibrosis reversal after 260 weeks of antiviral therapy, and the performance of the model was analyzed by drawing the ROC curve, as shown in FIG. 3, the AUC value for distinguishing the state of fibrosis reversal was 0.75 (95% CI: 0.63-0.88), which confirmed the universality of the model. Figure 3

[0124] 3. Subgroup analysis

[0125] The stability of the serum protein combination marker model in specific clinical relevant populations (such as patients with baseline Ishak ≥ 3, baseline non-steatosis patients and patients with ΔIshak ≥ 1 after AVT) was verified, and the performance of the model was analyzed by drawing the ROC curve for monitoring the state of fibrosis reversal after 260 weeks of antiviral therapy based on the serum protein combination marker model, and the AUC value was further obtained, as shown in Table 4, which showed that the model showed consistent performance in the 4D-DIA-MS discovery group (training set, test set and their combination) and the PRM-MS validation group in the clinical relevant populations.

[0126] Table 4 AUC values of serum protein combination marker for monitoring the state of fibrosis reversal after long-term antiviral therapy in each clinical relevant subgroup in the 4D-DIA-MS discovery group and the PRM-MS validation group ​​

[0127]

[0128] In summary, the serum protein panel model described above exhibits robust performance in multiple validation frameworks and clinical subgroups, establishing its potential as a non-invasive biomarker for monitoring the state of fibrosis reversal in CHB patients receiving long-term AVT.

[0129] Application of serum protein panel model in Example 2

[0130] Serum samples from CHB patients before and after 260 weeks of antiviral treatment were collected, and the following serum protein panel markers were analyzed qualitatively and quantitatively (the method can be ELISA, suspension chip, MSD electrochemical luminescence technology, mass spectrometry such as 4D-DIA-MS and PRM-MS, or immunoblotting): serine protease inhibitor family A member 7 (SERPINA7), CD163 molecule (CD163), and complement factor H-related protein 4 (CFHR4). Then the ratio of the content of the above serum protein panel markers after 260 weeks of antiviral treatment to that before antiviral treatment (Post / Pre) in the serum samples was calculated, and the Post / Pre of the above serum protein panel markers was substituted into the formula P = 1 / (1+exp(-(2.2135-0.2958xSERPINA7'-1.6642xCD163'+0.293xCFHR4'))); the state of liver fibrosis reversal after long-term antiviral treatment of CHB patients was monitored according to P: if P>0.68 (critical value) it was determined to be reversed, otherwise it was not reversed. Two patients were monitored using the above method as follows:

[0131] Patient 1, using 4D-DIA-MS technology to detect SERPINA7, CD163 and CFHR4 in serum after antiviral treatment (260 weeks) and before antiviral treatment (0 weeks), the Post / Pre were 0.80, 0.76 and 1.21 (i.e. SERPINA7' = 0.80, CD163' = 0.76 and CFHR4' = 1.21) respectively, and after calculation by substituting into the formula, the P value was 0.74395>critical value 0.68, so it was determined to be reversed; the liver biopsy fibrosis score of this patient before treatment was 2, and the liver biopsy fibrosis score after 260 weeks of antiviral treatment was 1, and the pathological diagnosis result was reversed. The model is consistent with the pathological monitoring results.

[0132] Patient 2, using 4D-DIA-MS technology to detect the serum of SERPINA7, CD163 and CFHR4 after antiviral treatment (260 weeks) and before antiviral treatment (0 weeks) Post / Pre are: 1.02, 1.00 and 1.10 (i.e. SERPINA7'=1.02, CD163'=1.00 and CFHR4'=1.10), after calculation P value is 0.63874<critical value 0.68, so it is determined as not reversed; the patient was clinically diagnosed as compensated cirrhosis before treatment, and the liver biopsy pathological fibrosis score was 5 after 260 weeks of antiviral treatment, the PIR score was uncertain, and the pathological diagnosis result was not reversed. The model is consistent with the pathological monitoring result.

[0133] Obviously, the above embodiments of the present application are only examples for clearly illustrating the present application, and are not limitations on the embodiments of the present application. For ordinary skilled in the art, other different forms of changes or variations can be made on the basis of the above description, and all the embodiments cannot be exhausted here. Any obvious changes or variations derived from the technical solutions of the present application are still within the protection scope of the present application.

Claims

1. A serum protein combination marker for monitoring the reversal of liver fibrosis in patients with chronic hepatitis B after long-term antiviral treatment, characterized in that: The serum protein combination marker includes: SERPINA7, CD163 and CFHR4; wherein the UniprotKB IDs of SERPINA7, CD163 and CFHR4 are P05543, Q86VB7 and Q92496 respectively.

2. The serum protein combination marker according to claim 1, characterized in that The source of the serum protein combination marker is tissue or blood; preferably, the source of the serum protein combination marker is blood.

3. Use of the serum protein combination marker or its detection reagent according to claim 1 or 2 in the preparation of a product for monitoring the reversal status of liver fibrosis in patients with chronic hepatitis B after long-term antiviral treatment.

4. The use according to claim 3, characterized in that The detection reagent is used to detect the content ratio of the serum protein combination marker in chronic hepatitis B patients after long-term antiviral treatment and before antiviral treatment.

5. The use according to claim 4, characterized in that The detection reagents include any one or more reagents used in ELISA, suspension chip, immunoblotting, MSD electrochemiluminescence technology and mass spectrometry.

6. The use according to claim 5, characterized in that The detection reagent is a reagent used in mass spectrometry; preferably, the mass spectrometry includes 4D-DIA-MS and / or PRM-MS.

7. A kit for monitoring the reversal status of liver fibrosis in patients with chronic hepatitis B after long-term antiviral treatment, characterized in that: The kit comprises a detection reagent for detecting the ratio of the content of the serum protein combination marker in chronic hepatitis B patients after long-term antiviral treatment to that before treatment according to claim 1.

8. A model for monitoring the reversal of liver fibrosis in patients with chronic hepatitis B after long-term antiviral treatment, characterized in that: The model is based on the ratio of the content of the serum protein combination marker described in claim 1 in patients with chronic hepatitis B after long-term antiviral treatment to that before antiviral treatment, and the model formula is as follows: P = 1 / (1+exp(-(2.2135-0.2958×SERPINA7'-1.6642×CD163'+0.293×CFHR4')), if P>0.68, it is judged to be reversal, otherwise it is not reversed; where SERPINA7', CD163' and CFHR4' all represent the ratio of the content of SERPINA7, CD163 and CFHR4 in patients with chronic hepatitis B fibrosis after long-term antiviral treatment to that before antiviral treatment.

9. A system for monitoring the reversal status of liver fibrosis in patients with chronic hepatitis B after long-term antiviral treatment, characterized in that: include: A detection device for obtaining the ratio of the serum protein combination marker content of claim 1 in a chronic hepatitis B patient after long-term antiviral treatment and before antiviral treatment; A data processing device for calculating P using the model according to claim 8; The result determination device is used to determine the reversal status of liver fibrosis in chronic hepatitis B patients after long-term antiviral treatment based on P: if P> critical value, it is determined to be reversed, otherwise it is not reversed; the critical value is 0.

68.

10. The serum protein combination marker according to claim 1 or 2, or the use according to any one of claims 3 to 6, or the kit according to claim 7, or the model according to claim 8, or the system according to claim 9, characterized in that: The period of the long-term antiviral treatment is ≥156 weeks; preferably, the period of the long-term antiviral treatment is 260 weeks.

Images