A biomarker for helicobacter pylori infection in active and stable chronic gastritis and application thereof in constructing a high-precision diagnostic model

Abstract

The application belongs to the field of biotechnology and detection, and particularly relates to a biomarker for Helicobacter pylori infection chronic gastritis active stage and stable stage and application thereof in constructing a high-precision diagnosis model. The application is based on unsupervised clustering grouping of clinical indexes of patients with Helicobacter pylori infection chronic gastritis, selects a subgroup for serum proteomics and non-targeted metabolomics analysis, combines clinical indexes for biomarker screening and diagnosis efficiency evaluation. A high-precision diagnosis model (AUC=0.9978) of HGB-CAZPB-Acetic Acid is constructed, and three biological mechanisms of blood function, energy metabolism, cytoskeleton reorganization and cell migration in the active stage and stable stage of Helicobacter pylori infection chronic gastritis are found. The multi-omics model provided by the application can realize systematic interpretation from an upstream regulation mechanism to a downstream functional phenotype, thereby developing a high-precision non-invasive diagnosis tool, which has important prospects for realizing early warning, risk stratification and individualized intervention of gastritis.

Description

Technical Field

[0001] This invention belongs to the field of biotechnology and detection, specifically relating to a biomarker for the active and stable phases of Helicobacter pylori-infected chronic gastritis and its application in constructing a high-precision diagnostic model of HGB-CAZPB-Acetic Acid. Background Technology

[0002] Helicobacter pylori infection is one of the most common types of infection worldwide and the leading cause of chronic gastritis. Approximately 90% of patients with chronic gastritis are infected with Helicobacter pylori, affecting about half of the global population. The clinical manifestations of Helicobacter pylori-induced chronic gastritis are diverse, ranging from asymptomatic to a range of indigestion symptoms such as upper abdominal pain, bloating, loss of appetite, nausea, and belching. While most gastritis patients respond well to intervention after clinical manifestations, recovery from chronic gastritis is a lengthy process. The asymptomatic period before and after a gastritis flare-up is easily overlooked, leading to prolonged chronic gastritis and potentially progressing to gastric ulcers, gastric bleeding, and even gastric cancer.

[0003] Accurate subtyping of Helicobacter pylori-infected chronic gastritis is crucial for developing appropriate treatment plans, assessing disease progression risk, and predicting prognosis. Currently, the diagnosis and subtyping of gastritis primarily rely on clinical symptom assessment, endoscopy, and histopathological examination. Gastroscopy is the gold standard for diagnosing gastritis, allowing direct observation of the gastric mucosa's morphology, color, presence of erosion, bleeding, atrophy, and other changes. Histopathology, through endoscopic mucosal biopsy, clarifies the degree and activity of inflammation, the presence of atrophy, intestinal metaplasia, and dysplasia, providing the most reliable basis for subtyping. However, current subtyping standards also have significant limitations. First, gastroscopy and pathological examination are invasive procedures, unsuitable for large-scale screening or frequent follow-up. Second, they lack dynamic, real-time assessment capabilities; gastroscopy and pathological examination reflect a static state at a specific point in time, making it difficult to frequently and dynamically monitor disease activity and response to treatment.

[0004] Meanwhile, Helicobacter pylori-infected chronic gastritis, a common digestive system disease, often presents with subtle and deceptive clinical manifestations during its stable phase, making it highly prone to relapse or causing secondary damage to the stomach. However, for a long time, there has been a lack of objective and sensitive blood molecular markers to accurately distinguish between the active and stable phases. This situation not only increases the difficulty of clinical diagnosis but also significantly hinders the effective implementation of precision medicine strategies. Therefore, it is of great significance to discover blood biomarkers and mechanistic differences that can differentiate between the active and stable phases of Helicobacter pylori-infected chronic gastritis through clinical diagnostic and testing indicators, as well as multi-level omics technologies. Summary of the Invention

[0005] To address the limitations of current diagnostic methods and the lack of objective, convenient, and quantifiable biomarkers, this invention provides a biomarker for the active and stable phases of Helicobacter pylori infection in chronic gastritis.

[0006] This invention further constructs a high-precision diagnostic model for HGB-CAZPB-Acetic Acid based on the above-mentioned biomarkers.

[0007] The technical solution adopted by the present invention to achieve the above objectives is as follows: This invention provides a biomarker for the active and stable phases of Helicobacter pylori infection in chronic gastritis, wherein the biomarker includes blood markers, protein markers, and metabolite markers; The blood biomarkers are HGB, UA, CK, RDW-CV, CHE, and GGT; The protein biomarkers are TFRC, CAPZB, ACTR2, ACTR3, CFL1, and ENO3; The metabolite markers are: MADN0339 (2-isopropylmalic acid), MADN2955 (20-carboxyleukotriene B4), MEDN4058 (sodium creatine phosphate), MEDP0200 (diethyl glutarate), MEDP0407 (retinol), MEDP1171 (butenoyl-PAF), and MEDP3439. 097 (dehydrocholesterol), MEDP3557 (PC(22:6 / 18:0)), MEDP4231 (1-octadecyl lysophosphatidylcholine), MEDP4242 (acetic acid), MW0057371HN (PC(20:4(5Z,8Z,11Z,14Z) / 24:0)), MEDP3440 097 (cholest-4-en-3-one) and MEDP3442 097 (yeast sterol).

[0008] Preferably, joint determination is based on biomarkers, and the specific determination principle is as follows: Based on hematological function: assessment was made using HGB, RDW-CV, TFRC, and butenoyl-PAF; Based on energy metabolism: the determination was made by CK, ENO3, sodium creatine phosphate, diethyl glutarate, and acetic acid; Based on cytoskeleton reorganization and cell migration: the determination was made by ARCT2, ARCT3, CFL1, CAZPB, PC (22:6 / 18:0), 1-octadecyl lysophosphatidic acid and PC (20:4(5Z,8Z,11Z,14Z) / 24:0).

[0009] Preferably, HGB, RDW-CV, TFRC, and butenoyl-PAF are a combination of biomarkers used to determine the active phase of Helicobacter pylori-infected chronic gastritis based on abnormal blood function.

[0010] Preferably, CK, ENO3, sodium creatine phosphate, diethyl glutarate, and acetic acid are a combination of biomarkers used for screening disease activity in patients susceptible to anemia based on abnormal energy metabolism.

[0011] Preferably, ARCT2, ARCT3, CFL1, CAZPB, PC (22:6 / 18:0), 1-octadecyl lysophosphatidylcholine and PC (20:4(5Z,8Z,11Z,14Z) / 24:0) are used as a combination of biomarkers to determine the dynamic changes in the mucosal layer during the active phase of Helicobacter pylori-infected chronic gastritis and to identify patients in the stable phase.

[0012] The present invention also provides a high-precision diagnostic model of HGB-CAZPB-Acetic Acid constructed based on the above-mentioned biomarkers.

[0013] Preferably, the high-precision diagnostic model utilizes HGB, acetic acid, and CAZPB to form a combined diagnostic biomarker.

[0014] The present invention further provides the application of the above-mentioned biomarkers or high-precision diagnostic models in the preparation of kits for rapid detection and high-precision non-invasive diagnosis of the active and stable phases of Helicobacter pylori-infected chronic gastritis.

[0015] This invention utilizes unsupervised clustering to group patients with Helicobacter pylori-infected chronic gastritis, selecting subgroups for serum proteomics and non-targeted metabolomics analysis. Combined with clinical indicators, biomarker screening and diagnostic efficacy evaluation are conducted. A high-precision diagnostic model (AUC=0.9978) for HGB-CAZPB-Acetic Acid was constructed, and three major biological mechanisms of Helicobacter pylori-infected chronic gastritis in active and stable phases—blood function, energy metabolism, cytoskeleton remodeling, and cell migration—were identified. The multi-omics model provided by this invention enables a systematic interpretation from upstream regulatory mechanisms to downstream functional phenotypes, thereby developing high-precision non-invasive diagnostic tools with significant potential for early warning, risk stratification, and personalized intervention for gastritis.

[0016] The beneficial effects of this invention are as follows: (1) This invention addresses the problem of the lack of objective blood indicators for the classification of Helicobacter pylori-infected chronic gastritis. It utilizes clinical sample data to mine reliable blood biochemical indicators for classification. Through serum proteomics and metabolomics, it explains the significance of clinical indicators as classification indicators for Helicobacter pylori-infected gastritis in the active and stable phases from a mechanistic perspective. It also screens highly specific blood markers for auxiliary classification, helping to accurately reflect the active and stable phases of Helicobacter pylori-infected gastritis.

[0017] (2) This invention further screens and discovers potential therapeutic targets, promoting the advancement of gastritis towards precision medicine. It overcomes the limitations of current reliance on endoscopy and unclear identification during the stable phase of chronic gastritis, providing a non-invasive, dynamic monitoring tool for the subtyping of Helicobacter pylori-infected chronic gastritis. By revealing the interaction network of clinical indicators, proteins, and metabolites, it not only assists in diagnosis, but also, through proteomics and metabolomics analysis of the functional regulation of proteins and the dynamic changes of metabolites in the blood, reveals the occurrence and development of digestive system diseases at the molecular mechanism level, providing core technical support for the discovery of clinical blood diagnostic biomarkers.

[0018] (3) The ability of this invention to accurately distinguish between the active and stable phases of gastritis is a key link in achieving precision medicine and has decisive significance for treatment decisions and prognostic assessment. This study successfully constructed a high-precision diagnostic model HGB-CAZPB-acetic acid (AUC=0.9978) through integrated analysis of clinical indicators and multi-omics data, demonstrating significant clinical translational potential. It also deeply elucidated the intrinsic mechanism of the active and stable phases of Helicobacter pylori-infected chronic gastritis, laying a research foundation for precise intervention in the active and stable phases of gastritis. Attached Figure Description

[0019] Figure 1 The relative expression levels of the first six factors in the active and stable phases of Helicobacter pylori-infected chronic gastritis. p<0.001 Figure 2 Differential protein analysis of Helicobacter pylori-infected chronic gastritis during active and stable phases and a healthy control group; including: (A) Clustering results of the three groups based on partial least squares discriminant analysis (PLS-DA) (A: active phase, B: stable phase, C: healthy control group); (B) Comparison of Venn diagrams of differential protein expression levels among the three groups; (C) Comparison of volcano diagrams between the active and stable phases; (D) Comparison of volcano diagrams between the healthy control group and the active phase; (E) Comparison of volcano diagrams between the healthy control group and the stable phase; (F) Comparison of heatmaps between the active and stable phases; (G) Comparison of heatmaps between the healthy control group and the active phase; (H) Comparison of heatmaps between the healthy control group and the stable phase; Figure 3TFRC, CAPZB, ACTR2, ACTR3, CFL1, and ENO3 are significantly differentially expressed proteins that can distinguish between the active and stable phases of chronic gastritis caused by Helicobacter pylori infection. Figure 4 Serum metabolomics analysis of H. pylori-infected chronic gastritis in active and stable phases and healthy controls; (A) Serum metabolite classification analysis; (B) Cluster analysis of H. pylori-infected chronic gastritis in active and stable phases and healthy controls based on partial least squares discriminant analysis (PLS-DA) (A: active phase, B: stable phase, C: healthy control); (C) Volcano plot of H. pylori-infected chronic gastritis in active and stable phases; (D) Volcano plot of H. pylori-infected chronic gastritis in healthy controls and active phases; (E) Volcano plot of H. pylori-infected chronic gastritis in healthy controls and stable phases; (F) Bubble plot of KEGG analysis of H. pylori-infected chronic gastritis in active and stable phases; (G) Bubble plot of KEGG analysis of H. pylori-infected chronic gastritis in healthy controls and active phases; (H) Bubble plot of KEGG analysis of H. pylori-infected chronic gastritis in healthy controls and stable phases. Figure 5 MADN0339 (2-Isopropylmalic acid), MADN2955 (20-CarboxyLTB4), MEDN4058 (Sodium creatine phosphate), MEDP0200 (Diethyl glutarate), MEDP0407 (Retinol), MEDP1171 (Butyryl-PAF), MEDP3439 097 (destosterol), MEDP3557 (PC (22:6 / 18:0)), MEDP4231 (1-octadecyl lysophosphatidic acid), MEDP4242 (acetic acid), MW0057371HN (PC (204 (5Z, 8Z, 11Z, 14Z) / 24:0)), MEDP3440 097 (cholesterol-4-en-3-one) and MEDP3442 097 (yeast sterol) is a metabolite that can distinguish between the active and stable phases of chronic gastritis caused by Helicobacter pylori infection; Figure 6A classification model of Helicobacter pylori-infected chronic gastritis active and stable phases based on clinical indicators, differential proteins, and metabolites; (A) ROC curve of the Helicobacter pylori-infected chronic gastritis staging model based on blood function (HGB AUC=0.8428, RDW-CV AUC=0.5683, TFRC AUC=0.6833, butyryl-PAF AUC=0.7722, combined diagnostic efficacy AUC=0.9200); (B) ROC curve of the Helicobacter pylori-infected chronic gastritis staging model based on energy metabolism (CKAUC=0.7183, ENO3). (AUC=0.7533, sodium creatine phosphate AUC=0.8133, diethyl glutarate AUC=0.7300, acetic acid AUC=0.9456, combined diagnostic efficacy AUC=0.9767); (C) ROC curve of the Helicobacter pylori infection chronic gastritis staging classification model based on cytoskeleton reorganization and cell migration (ARCT2 AUC=0.7700, ARCT3 AUC=0.6800, CFL1 AUC=0.6789, CAZPB AUC=0.8456, PC (22:6 / 18:0) AUC=0.8411, 1-octadecyl lysophosphatidylcholine AUC=0.0700, PC (20:4 (5Z, 8Z, 11Z, 14Z) / 24:0) AUC=0.7389, combined diagnostic efficacy AUC=0.9967). (D) ROC curve of the diagnostic combination of HGB, acetic acid and CAZPB, AUC=0.9978. Detailed Implementation

[0020] The technical solution of the present invention will be further explained and described below through specific embodiments.

[0021] Example 1 (a) Research Subjects This invention included patients with Helicobacter pylori (HP)-infected gastritis who visited the Guang'anmen Hospital of the China Academy of Chinese Medical Sciences between January 2, 2018 and March 16, 2023. The scope of diagnosis and treatment covered examination, surgery, hospitalization, and follow-up. Clinical and follow-up data were collected using an electronic data acquisition system and case report forms. A total of 1409 patients with HP-infected gastritis were included, with a wide age range (15-87 years). All enrolled patients had no history of upper gastrointestinal diseases, hypertension, or cardiovascular diseases, and were in varying degrees of chronic Helicobacter pylori-infected gastritis.

[0022] (II) Unsupervised clustering analysis and grouping Serum samples and demographic data from all patients were collected and tested according to standardized procedures, and a total of 69 clinical biochemical parameters were analyzed. To subtype the disease, K-means unsupervised clustering (K=2) was applied to the standardized 69 data points from 1409 patients, resulting in two main subgroups. The clustering quality was assessed using silhouette coefficients, and the subgroups were defined as active and stable phases of Helicobacter pylori-associated chronic gastritis, based on clinical diagnostic analysis. The biomarker optimization process included: ranking the top ten indicators by contribution, visualizing effect sizes using forest plots, and iteratively reselecting clinical indicators until clear intergroup separation was achieved.

[0023] (III) Proteomics 1. Protein extraction and trypsin digestion Serum samples from 30 patients were randomly selected from each group, and healthy controls (HC) were recruited for comparison. Serum samples were incubated with magnetic beads in binding buffer (50 mM Tris, 10 mM EDTA) at room temperature, followed by magnetic separation, bead washing, and resuspension. Proteins were treated with reducing agent (0.1% DOC, 100 mM TCEP, 400 mM CAA; 95°C, 15 min) and digested with trypsin (enzyme-protein ratio 1:50, 12 h). The digestion reaction was terminated with 0.1% formic acid, and peptides were desalted and dried.

[0024] 2. Liquid chromatography-mass spectrometry (LC–MS / MS) Peptides were analyzed using an Orbitrap Fusion Lumos Tribrid mass spectrometer (Thermo Fisher Scientific) coupled with an EASY-nLC1000 nanoliter liquid chromatography system. A 10 cm C18 column (75 µm inner diameter, 3 µm XB-C18 packing material) was used, with a gradient elution program of 350 nL / min for 75 minutes. Mobile phase B (99.5% acetonitrile + 0.5% formic acid) was added from 3% to 100% in mobile phase A (99.5% water + 0.5% formic acid). The total run time (including loading / washing) was approximately 90 minutes. MS1 full scan (m / z 300-1400) was acquired on the Orbitrap analyzer at 70,000 resolution and 2.0 kV electrospray voltage.

[0025] 3. Proteomics data analysis Differentially expressed genes (DEGs) were identified using limma software (v3.58.1), with selection criteria of corrected p-value <0.05 and |FC|>1.5 (FC>1.5: upregulated; FC<0.67: downregulated). GO annotation was performed using the UniProt-GOA database, and pathway enrichment was analyzed using KEGG. Differentially expressed proteins (DEPs) accession numbers / sequences were retrieved from the STRING database (v10.1) to construct a protein-protein interaction network.

[0026] (iv) Metabolomics 1. Sample preparation and extraction Hydrophilic metabolites: Thaw samples frozen at -80°C on ice, vortex for 10 seconds, and extract with 300 µL of acetonitrile-methanol (1:4) mixture (containing internal standard). After vortexing for 3 minutes and centrifugation (12,000 rpm, 10 minutes, 4°C), take 200 µL of the supernatant and incubate at -20°C for 30 minutes, then centrifuge again (12,000 rpm, 3 minutes, 4°C) for LC-MS / MS analysis.

[0027] Hydrophobic metabolites: Thawed samples on ice were extracted with 1 mL of a methyl tert-butyl ether-methanol (3:1) mixture (containing internal standard), vortexed for 15 min, mixed with 200 µL of water, centrifuged (12,000 rpm, 10 min), and 200 µL of the upper organic phase was collected. After vacuum drying, the mixture was reconstituted with 200 µL of acetonitrile-isopropanol (1:1) and analyzed by LC-MS / MS.

[0028] 2. UPLC conditions Hydrophilic metabolites: Analyzed under three different conditions using an LC-ESI-MS / MS system. In positive ion mode, a T3 C18 column was used with a mobile phase of 0.1% formic acid (A) and acetonitrile (B) containing 0.1% formic acid. The gradient program was: phase B increased from 5% to 20% within 2 minutes, to 60% within 3 minutes, to 99% within 1 minute, held for 1.5 minutes, returned to 5% within 0.1 minutes, and equilibrated for 2.4 minutes. The same gradient was used in negative ion mode on the same column. A third analysis was performed in negative ion mode using a HILIC column. The mobile phase was 20 mM ammonium formate dissolved in acetonitrile-water-methanol in different proportions. The gradient program was: phase B increased from 5% to 30% within 3.5 minutes, to 95% within 2 minutes, held for 1 minute, and then rapidly reequilibrated. The column temperature was maintained at 40°C, the flow rate at 0.4 mL / min, and the injection volume at 5 µL.

[0029] Hydrophobic compounds: UPLC column: Thermo Accucore™ C30 (2.6 μm, 2.1 mm × 100 mm); solvent system: mobile phase A was acetonitrile / water (60 / 40, V / V, containing 0.1% formic acid and 10 mmol / L ammonium formate), mobile phase B was acetonitrile / isopropanol (10 / 90 V / V, containing 0.1% formic acid and 10 mmol / L ammonium formate); gradient program: phase B increased from 20% to 30% at 2 min, to 60% at 4 min, to 85% at 9 min, to 90% at 14 min, to 95% at 15.5 min, held for 17.3 min, and decreased to 20% at 20 min; flow rate: 0.35 mL / min; column temperature: 45°C; injection volume: 2 µL. The eluent was alternately introduced into the ESI-triple quadrupole-linear ion trap (QTRAP)-MS system.

[0030] 3. QTOF-MS / MS Data acquisition employed an information-dependent acquisition mode, controlled using Analyst TF 1.7.1 software (Sciex, Concord, Canada). Ion source parameters were set as follows: Ion source gas 1 (GAS1) 50 psi; Ion source gas 2 (GAS2) 50 psi; Curtain gas (CUR) 25 psi; Ion source temperature (TEM) 550°C; Declustering voltage (DP) 60V / -60V for positive / negative ion modes; Ion spray voltage (ISVF) 5000V / -4000V for positive / negative ion modes. Flush ion scanning parameters were set as follows: mass range 25-1000 Da; accumulation time 40 ms; collision energy 30V / -30V for positive / negative ion modes; collision energy diffusion 15; charge state 1 to 1; intensity threshold 100 cps; exclusion of isotope peaks within 4 Da; mass tolerance 50 ppm. 4. ESI-Q TRAP-MS / MS Hydrophilic compounds: Mass spectrometry analysis was performed using a QTRAP LC-MS / MS system (Sciex) equipped with an ESI turbine ion spray interface, controlled by Analyst 1.6.3 software. ESI parameters: ion source temperature 500°C; spray voltage (IS) 5500V / -4500V for positive / negative ion modes; nebulizer gas (GSI) 55psi, auxiliary gas (GSII) 60psi, curtain gas (CUR) 25psi; collision gas (CAD) high mode. Tuning / calibration was performed using 10 / 100 μmol / L polypropylene glycol solutions in QQQ / LIT modes. Metabolite-specific MRM ion pairs were monitored based on elution time.

[0031] Hydrophobic compounds: ESI parameters: turbine spray; ion source temperature 500°C; spray voltage (IS) 5500V / -4500V for positive / negative ion modes; nebulizer gas (GSI) 45psi, auxiliary gas (GSII) 55psi, curtain gas (CUR) 35psi; collision gas (CAD) in medium mode. Tuning / calibration were the same as above. MRM experiments used nitrogen as the collision gas (5psi), and the declustering voltage and collision energy for each ion pair were optimized individually. Ion pair combinations were monitored according to the metabolite elution time.

[0032] 5. Qualitative and quantitative analysis of metabolites All sample extracts were mixed in equal volumes to form a quality control sample, which was then subjected to non-targeted analysis on an LC-QTOF-MS / MS platform. Multiple ion pairs and retention times of metabolites were extracted and identified using the MWDB standard database, the Maiwei integrated database-all AI prediction database, and MetDNA software, enabling precise metabolite identification. By integrating the above data with Maiwei's proprietary targeted database, a new project-specific spectral library was constructed. Finally, based on the Q-Trap instrument platform, the MRM mode was used for precise quantitative analysis of all sample metabolites. Metabolite quantification employed triple quadrupole mass spectrometry in multiple reaction monitoring mode. During MRM operation, the first quadrupole first screens for precursor ions of the target compound while excluding interference from other substances; these precursor ions undergo collision-induced dissociation in the second quadrupole to generate a series of compound-specific characteristic fragment ions; the third quadrupole filters and selects representative fragment ions, effectively eliminating interference from non-target ions, thereby improving quantitative accuracy and repeatability. After acquiring LC-MS data from multiple samples, the extracted ion chromatographic peaks of all metabolites were integrated and calibrated for the chromatographic peaks of the same metabolite in different samples to ensure data consistency.

[0033] 6. Metabolomics data analysis In the two-group comparative analysis, differentially expressed metabolites were screened using variable importance projection values ​​(VIP>1) and p-values ​​(P<0.05, Student's t-test). In the multi-group comparative analysis, differentially expressed metabolites were determined based on VIP values ​​(VIP>1) and p-values ​​(P<0.05, ANOVA). Identified metabolites were annotated using the KEGG Compound database (http: / / www.kegg.jp / kegg / compound / ), and then the annotated metabolites were mapped to the KEGG Pathway database (http: / / www.kegg.jp / kegg / pathway.html).

[0034] 7. Statistical Analysis Data analysis and visualization were performed using GraphPad Prism 9.4.0 and R software version 4.3.3. All results are expressed as mean ± standard deviation (M ± SD). One-way ANOVA was used for comparisons among multiple groups, t-tests were used for comparisons between two groups, and non-parametric tests were used where applicable.

[0035] Example 2: Unsupervised clustering of patients based on clinical biochemical indicators This study included 1,409 patients with Helicobacter pylori-infected chronic gastritis (590 males and 819 females; male-to-female ratio 1:1.39), aged 15-87 years. Among them, 665 patients (63.4%) were aged 20-50 years (male-to-female ratio 1:1.20). Another 479 patients (45.7%) were over 60 years old (male-to-female ratio 1:1.55).

[0036] To identify key variables for subsequent cluster analysis and modeling, this invention conducted a systematic feature analysis. First, based on the effect size forest plot assessment results, the top ten contributing indicators were selected. The results showed significant differences in the distinguishing importance of different blood biochemical indicators. Among them, six indicators—hemoglobin (HGB, effect size = 1.731), uric acid (UA, effect size = 1.320), creatine kinase (CK, effect size = 0.568), red blood cell distribution width coefficient of variation (RDW-CV, effect size = 0.288), cholinesterase (CHE, effect size = 0.272), and gamma-glutamyl transferase (GGT, effect size = 0.265)—had significantly higher effect sizes than other indicators (p < 0.001). This indicates that these six indicators are core features for distinguishing patient categories and should be prioritized as key variables.

[0037] Unsupervised cluster analysis was used to validate the selected indicators to differentiate the severity of Helicobacter pylori-infected chronic gastritis. To optimize the final model, six of the most significant indicators (HGB, UA, CK, RDW-CV, CHE, GGT, p<0.001) were selected from the top ten indicators. Clustering based on these six factors yielded high cumulative explained variance (50.6%; 36%+14.6%), clear cluster separation, and very few outliers. The silhouette coefficients were concentrated above 0.25, confirming high intra-cluster homogeneity and significant inter-cluster heterogeneity. Thirty patients were randomly selected from each cluster for clinical diagnosis. Patients in the active phase were associated with a higher incidence of adverse clinical reactions and more severe gastric mucosal damage. These six indicators constitute a core biomarker combination for differentiating the clinical status of active and stable phases of Helicobacter pylori-infected chronic gastritis, and this combination shows significant application potential as a reliable and practical basis for clinical stratification.

[0038] A comparative analysis of six key blood indicators in patients with active and stable Helicobacter pylori-infected gastritis was conducted using box plots and radar charts. Figure 1 The results showed that all indicators differed significantly between groups (p<0.001), reflecting different pathophysiological characteristics. Hemoglobin (HGB), a key indicator for assessing oxygen-carrying capacity and anemia, was significantly reduced in the active phase, suggesting anemia and supporting its diagnostic value. Red blood cell distribution width variation coefficient (RDW-CV) is an indicator of red blood cell size variability. Its elevation in the active phase suggests that mucosal inflammation / erosion and related microbleeds exacerbate iron loss or hinder nutrient absorption, leading to red blood cell size variability and elevated RDW-CV. Uric acid (UA), as the end product of purine metabolism, showed elevated levels in the active phase, possibly due to accelerated immune cell turnover and increased purine breakdown during inflammation. Creatine kinase (CK) was significantly elevated in the active phase, which may be related to disease-related energy metabolism changes rather than direct muscle damage. Gamma-glutamyl transferase (GGT) was elevated in the active phase. Conversely, cholinesterase (CHE) was decreased in patients in the active phase, possibly due to the inflammatory metabolic burden inhibiting hepatic synthesis. In summary, these six indicators clearly outline the distinct clinicopathological features of active and inactive phases of Helicobacter pylori-infected chronic gastritis. The interrelationships among these indicators constitute a potential biomarker set reflecting disease activity, providing crucial laboratory evidence for a deeper understanding of the systemic effects of this disease and contributing to improvements in clinical staging and management strategies.

[0039] Example 3: Mechanism study of active and stable phases of Helicobacter pylori-infected chronic gastritis From the patient cohort classified according to clinical indicators, 30 patients were selected as group A (active phase: AP) and group B (stable phase: SP), and another 20 healthy individuals were selected as group C (healthy controls: HC) for subsequent multi-omics analysis. Partial least squares discriminant analysis (PLS-DA) showed that the three groups exhibited a significant separation trend in the principal component space. Figure 2 A). Venn diagram analysis revealed 55, 37, and 23 differentially abundant molecules in the C vs A, C vs B, and A vs B comparisons, respectively, including 94 unique differentially abundant molecules ( ). Figure 2 .B). Volcano map ( Figure 2 The .CE diagram details the number of molecules that were significantly up / downregulated in each comparison group and their statistical significance, while the heatmap ( Figure 2 The .FH (color gradient) visually presents the expression patterns of these differentially expressed molecules across the three groups, with the color gradient clearly reflecting the continuous changes from low to high abundance. These results collectively confirm the existence of unique molecular profiles at different stages of gastritis, providing a multi-dimensional and objective basis for disease subtyping.

[0040] To identify key differentially expressed proteins between the active and stable phases of Helicobacter pylori-infected chronic gastritis, this study constructed protein-protein interaction (PPI) networks for groups A (active phase), B (stable phase), and C (healthy group) to characterize phase-specific molecular regulatory features. In a direct comparison between the active and stable phases, several core actin cytoskeleton-related proteins showed significant differences. The actin filament stabilizer CAPZA was downregulated in the active phase, indicating a weakened braking mechanism for filament extension, leading to a more dynamic cytoskeleton formation, thereby promoting immune cell migration and phagocytosis. The core component of the Arp2 / 3 complex, ACTR3, was upregulated, driving branched actin nucleation and phagocytic cup formation; while ACTR2 downregulation suggests inflammation-mediated reprogramming of the Arp2 / 3 complex for precise cytoskeleton remodeling. The actin depolymerization factor CFL1 was upregulated, accelerating cytoskeleton turnover and providing structural flexibility for efficient phagocytosis. The glycolytic enzyme ENO3 was simultaneously upregulated, reflecting enhanced glycolytic activity to meet increased energy demands, consistent with its correlation with the clinical blood marker CK. Transferrin receptor TFRC is specifically downregulated during active phase, impairing iron uptake, thereby reducing hemoglobin synthesis and increasing RDW-CV, consistent with the common anemia manifestations in active gastritis. Relative expression analysis among the three groups ( Figure 3 The differential expression patterns were shown: TFRC and ACTR2 were specifically altered during the active phase, suggesting their potential as biomarkers for the active phase; CAPZB and ACTR2 were uniquely elevated during the stable phase, indicating that they may be biomarkers for the stable phase of the disease; CFL1 and ENO3 were continuously upregulated throughout the course of the disease, making them potential biomarkers for distinguishing patients from healthy individuals.

[0041] Example 4: Metabolic reprogramming mechanism in active and stable phases of Helicobacter pylori-infected chronic gastritis The identified metabolites were classified by structure and then visualized using a ring diagram. Figure 4 A), exhibiting a highly complex composition dominated by lipids and lipid molecules. The main categories include glycerophosphates (19.8%), amino acids / metabolites (12.15%), glycerides (9.76%), fatty acids (7.44%), and sterol esters (6.26%), highlighting the central role of energy and protein-related metabolites in the gastric mucosal metabolome.

[0042] Further, through the orthogonal partial least squares discriminant analysis (OPLS-DA) model ( Figure 4 (.B) shows that the samples from each group exhibit clear separation based on this metabolomic profile. Volcano plot analysis identified 44 upregulated and 80 downregulated metabolites during the active phase compared to the stable phase. Figure 4 .CE, Table S6-8). KEGG pathway enrichment analysis revealed metabolic pathways with significant differences between the comparison groups. Pathway analysis comparing the active and stationary phases ( Figure 4 The .F) showed significant enrichment of ABC transporter proteins, starch / sucrose metabolism, and carbohydrate digestion / absorption pathways, reflecting malnutrition and carbohydrate metabolism.

[0043] To accurately identify the key metabolites that contribute most significantly to differentiating different stages of Helicobacter pylori-infectious chronic gastritis, this invention generates variable importance projection (VIP) score plots for each comparison group based on the OPLS-DA model. By screening metabolites with VIP scores >1.0 and high confidence, 13 core metabolites were finally identified as having high discriminative value in all three comparison groups, including: MADN0339 (2-isopropylmalic acid), MADN2955 (20-carboxyleukotriene B4), MEDN4058 (sodium creatine phosphate), MEDP0200 (diethyl glutarate), MEDP0407 (retinol), MEDP1171 (butenoyl-PAF), and MEDP3439. 097 (dehydrocholesterol), MEDP3557 (PC(22:6 / 18:0)), MEDP4231 (1-octadecyl lysophosphatidylcholine), MEDP4242 (acetic acid), MW0057371HN (PC(20:4(5Z,8Z,11Z,14Z) / 24:0)), MEDP3440 097 (cholest-4-en-3-one) and MEDP3442 097 (Yeast Sterol). Specifically, 2-isopropylmalic acid, an intermediate in leucine synthesis, reflects changes in cellular demand for protein synthesis precursors under stress. 20-Carboxyleukotriene B4 plays an important role in innate immunity by inducing phagocyte reaggregation, promoting the release of antimicrobial effector molecules, and enhancing pathogen uptake and clearance. In terms of energy metabolism, sodium creatine phosphate, as an intracellular energy reserve, directly reflects tissue energy homeostasis imbalance through its fluctuations; acetic acid, which also functions as a metabolic hub and histone acetylation substrate, can serve as an indicator of the body's overall energy metabolism status. Retinol, closely related to mucosal barrier integrity and immune regulation, affects epithelial barrier function and immune regulation capacity through changes in its expression level. Butenoyl-PAF, a characteristic metabolite of the platelet-activating factor pathway, directly reflects systemic anemia and inflammatory flare-ups during active gastritis, a finding corroborated by clinical indicators such as HGB, RDW-CV, and the differentially expressed protein TFRC.

[0044] PC(22:6 / 18:0), 1-octadecyl lysophosphatidylcholine, and PC(20:4(5Z,8Z,11Z,14Z) / 24:0) play important roles in cytoskeleton dynamics and signal transduction. The lysophosphatidylcholine derivative 1-octadecyl lysophosphatidylcholine also plays a crucial role in cell signal transduction. Figure 5Metabolic alterations indicate that active Helicobacter pylori-infected chronic gastritis is characterized by synergistic metabolic dysregulation, featuring energy metabolism reprogramming, cholesterol / phospholipid remodeling, and impaired barrier function.

[0045] Example 5: Evaluation of the efficacy of biomarker screening and diagnosis Based on the aforementioned research, we identified significantly different indicators from clinical blood tests, serum proteomics, and metabolomics. To discover biomarkers that can accurately distinguish between the active and stable phases of Helicobacter pylori-infected chronic gastritis, this invention integrates clinical parameters and multi-omics data, summarizing key differential indicators into three types of biological mechanisms affecting the disease, thereby evaluating their diagnostic efficacy.

[0046] The first type of mechanism involves hematological functions, including HGB (AUC=0.8428), RDW-CV (AUC=0.5683), TFRC (AUC=0.6833), and butenoyl-PAF (AUC=0.7722). The combined diagnostic efficacy of these indicators reached 0.9200 (AUC). Figure 6 (A). These biomarkers suggest that during active gastritis, impaired gastric mucosal function leads to impaired iron absorption, resulting in reduced hemoglobin synthesis and anemia. Erythropoiesis is subsequently affected, producing smaller red blood cells, as evidenced by elevated RDW-CV. Furthermore, the continuous release of inflammatory factors during this phase suppresses bone marrow hematopoiesis and promotes downregulation of TFRC. Downregulation of butenoyl-PAF further provides evidence for a complex metabolic disorder in the body.

[0047] The second type of mechanism involves energy metabolism, including CK (AUC=0.7183), ENO3 (AUC=0.7533), sodium creatine phosphate (AUC=0.8133), diethyl glutarate (AUC=0.7300), and acetic acid (AUC=0.9456). The combined diagnostic efficacy of these markers reached 0.9767 (AUC=0.9767). Figure 6 (B). These indicators suggest persistent gastric mucosal inflammation during the active phase: inflammatory cell infiltration consumes large amounts of oxygen, potentially impairing microcirculation and leading to local hypoxia; inflammatory factors simultaneously inhibit mitochondrial function, promoting a shift in metabolism from aerobic oxidation to anaerobic glycolysis, as evidenced by the upregulation of ENO3. The upregulation of diethyl glutarate and acetic acid further indicates a disruption in the traditional energy supply state dominated by the TCA cycle and mitochondrial oxidation.

[0048] The third type of mechanism involves cytoskeleton remodeling and cell migration, including ARCT2 (AUC=0.7700), ARCT3 (AUC=0.6800), CFL1 (AUC=0.6789), CAZPB (AUC=0.8456), PC (22:6 / 18:0) (AUC=0.8411), 1-octadecyl lysophosphatidylcholine (AUC=0.0700), and PC (20:4(5Z,8Z,11Z,14Z) / 24:0) (AUC=0.7389). The combined diagnostic efficacy of these markers reached 0.9967 (AUC=0.9967). Figure 6 (C). This group of biomarkers outlines the cellular activity mechanisms of gastric mucosal damage during active gastritis: upregulation of ARCT3, which promotes cell migration, indicates that cells are undergoing internal cytoskeleton remodeling, while downregulation of ARCT2 may weaken cytoskeleton stability, creating conditions for cell morphological changes and migration. Downregulation of CFL1, a key regulator of actin depolymerization, usually indicates disruption of microfilament homeostasis. The important scaffold protein CAZPB plays a central role in membrane wrinkle formation, cell migration, and invasion; its upregulation is one of the strongest indicators of enhanced cell migration capacity, directly driving pseudopodia formation by promoting subcellular actin polymerization. From a metabolic perspective, phosphatidylcholine (PC(22:6 / 18:0) and PC(20:4(5Z,8Z,11Z,14Z) / 24:0)) and lysophosphatidic acid (1-octadecyl lysophosphatidic acid) constitute a closely related lipid signaling molecule pair.

[0049] Based on an in-depth exploration of the significance of differential indicators between the active and stable phases of Helicobacter pylori-infected chronic gastritis and healthy controls, this invention, to improve the convenience and simplicity of diagnostic criteria, screened HGB, acetic acid, and CAZPB to form a combined diagnostic combination based on their biological origin and function. HGB represents a clinical indicator of hematological function, acetic acid represents serum metabolomics and energy metabolism, and CAZPB represents a serum proteomic marker involved in cytoskeleton remodeling and cell migration. This combination achieved a diagnostic efficacy of 0.9978 (…). Figure 6 The study demonstrated its powerful ability to accurately distinguish between the active and stable phases of Helicobacter pylori-infected chronic gastritis. This finding provides crucial evidence for the development of high-precision, non-invasive diagnostic tools for assessing gastritis activity.

[0050] This invention, by integrating clinical biochemical indicators, serum proteomics, and metabolomics, systematically reveals profound molecular differences between the active and stable phases of Helicobacter pylori-infected chronic gastritis, and successfully constructs a high-precision diagnostic model. A healthy state exhibits a baseline level of dynamic equilibrium in biochemistry, proteomics, and metabolomics, without inflammation or metabolic disturbances. In contrast, even clinically mild stable gastritis reflects a deviation from homeostasis, indicating a subclinical chronic inflammatory state.

[0051] Multi-omics data showed that bacterial infection and phagocytosis pathways were continuously activated in the proteome, and carbohydrate and amino acid metabolism were altered, indicating that sustained low-level immune surveillance and tissue repair activities against Helicobacter pylori maintained a compensatory balance. In stark contrast, the active phase exhibited systemic and multi-level pathophysiological dysregulation. Clinically, decreased HGB and increased RDW-CV suggested anemia and impaired erythropoiesis during the active phase. At the proteomic level, pathways such as FcγR-mediated phagocytosis and actin cytoskeleton regulation were strongly activated, and upregulation of proteins such as ACTR3 and CFL1 drove immune cell migration and infiltration, exacerbating mucosal inflammation. At the metabolomic level, an energy crisis landscape was observed: dramatic fluctuations in key energy metabolites and upregulation of the glycolytic enzyme ENO3 together indicated that, driven by inflammatory hypoxia and mitochondrial dysfunction, the metabolic mode shifted from aerobic oxidation to anaerobic glycolysis. Therefore, the healthy, stable, and active phases of Helicobacter pylori-infectious chronic gastritis represent three different states: homeostasis, immune tolerance-based metabolic compensation, and acute inflammatory metabolic reprogramming.

[0052] The core discovery of this invention lies in the fact that the differences between the active and stable phases are not isolated phenomena, but rather interconnected through three major biological mechanisms: hematological dysfunction, energy metabolism reprogramming, and cytoskeleton remodeling / cell migration, forming a complete pathological network. Hematological dysfunction is the most direct manifestation: decreased HGB and increased RDW-CV during the active phase are not only clinical findings but also reflect underlying molecular alterations. Proteomics shows that TFRC-specific downregulation impairs iron utilization, mechanistically explaining insufficient HGB synthesis; simultaneously, metabolomics reveals downregulation of the platelet-activating factor pathway metabolite butenoyl-PAF, which is associated with potential microcirculatory disturbances and micromucosal hemorrhage. These factors collectively contribute to anemia. The resulting anemia and decreased oxygen-carrying capacity then trigger a second mechanism: energy metabolism reprogramming. Insufficient oxygen supply to tissues forces cells to use alternative energy sources: clinically, elevated CK levels indicate mobilization of the phosphocreatine buffer system; at the proteomic level, upregulation of the glycolytic enzyme ENO3 reflects full activation of the glycolytic pathway; and in the metabolome, disordered TCA cycle intermediates indicate impaired mitochondrial oxidative phosphorylation, reflecting that cells are shifting to a faster but less efficient energy metabolism mode to meet the inflammatory demands of Helicobacter pylori-induced gastritis.

[0053] The third major mechanism, cytoskeleton remodeling and cell migration, acts as the executors of inflammatory responses and tissue repair. When energy metabolism reprogramming provides the energy basis for cellular activities, proteins identified by the proteome, such as ACTR3, CFL1, and CAPZB, directly regulate actin cytoskeleton remodeling, driving immune cell migration to the site of infection and promoting phagocytosis. Metabolomics further elucidates changes in the microenvironment: alterations in membrane lipids affect membrane fluidity and signal transduction, jointly promoting cell migration and inflammatory responses. These three mechanisms do not operate in parallel but form a vicious cycle: Helicobacter pylori infection first triggers mucosal damage and bleeding, manifesting as hematological dysfunction; the resulting anemia and local hypoxia initiate anaerobic glycolysis—a marker of energy metabolism reprogramming; the altered metabolism then provides the energy and signals needed for cytoskeleton remodeling and immune cell migration; these cellular responses further exacerbate local inflammation and tissue damage, ultimately worsening anemia and metabolic disorders. This self-reinforcing pathological network provides a systemic explanation for understanding the multi-system effects of active Helicobacter pylori-infected chronic gastritis.

[0054] The diagnostic indicators screened based on the above mechanisms not only exhibited high specificity and synergy, but also demonstrated significant clinical translational potential. This research process ultimately refined an integrated diagnostic model consisting of HGB (clinical indicator), CAZPB (protein biomarker), and acetic acid (metabolite), achieving an AUC of 0.9978 and demonstrating excellent diagnostic performance. HGB, as a core representative of hematological function, CAZPB, a key protein in cytoskeleton remodeling, and acetic acid, acting as a pivotal molecule in energy metabolism, correspond to the aforementioned three core mechanisms, enabling the model to capture disease characteristics from multiple dimensions.

[0055] Furthermore, diagnostic marker combinations constructed based on different mechanisms each possess unique characteristics and clinical significance: the hematology combination (HGB, RDW-CV, TFRC, butenoyl-PAF) is suitable for screening disease activity in patients susceptible to anemia; the energy metabolism combination (CK, ENO3, creatine phosphate, diethyl glutarate, acetic acid) reflects metabolic stress and treatment response; and the cytoskeleton combination (ACTR2 / 3, CFL1, CAZPB, PC (22:6 / 18:0), PC (20:4 / 24:0), 1-octadecyl LPA) is used to assess dynamic changes at the mucosal level and identify patients in the stable phase. In clinical practice, these biomarkers support the development of non-invasive blood diagnostic kits, which are expected to reduce reliance on endoscopy and enable convenient disease staging and monitoring, particularly important in primary healthcare or large-scale screening. Simultaneously, they represent potential therapeutic targets, such as regulating cell migration through the CAZPB pathway or improving local energy supply by adjusting acetic acid metabolism, opening new avenues for precision treatment of Helicobacter pylori-infectious chronic gastritis. In summary, this invention not only constructs a high-performance diagnostic model, but also provides a system-level biological interpretation through multi-omics integration, laying a translational foundation for precision medicine of Helicobacter pylori-infected chronic gastritis.

[0056] Effect verification To systematically verify the effectiveness and reliability of the high-precision diagnostic model (hemoglobin (HGB), acetic acid, and CAZPB) constructed in this invention for the clinical diagnosis of Helicobacter pylori (HP)-infected gastritis, a targeted clinical validation trial was conducted at the Guang'anmen Hospital of the China Academy of Chinese Medical Sciences. During the study, strict clinical research guidelines were followed, and baseline clinical data and long-term follow-up data of HP-infected gastritis patients were systematically collected. Data collection utilized an electronic data collection system (EDC) and case report forms (CRF).

[0057] This invention included 105 patients with *Helicobacter pylori*-infected gastritis who met the diagnostic and inclusion / exclusion criteria. The participants ranged in age from 26 to 70 years, covering young, middle-aged, and elderly individuals. All participants underwent routine clinical examinations related to *Helicobacter pylori* infection and gastritis to confirm the clinical diagnosis.

[0058] Based on this, the HGB-CAZPB-Acetic Acid high-precision diagnostic model provided by this invention was used to conduct parallel testing and verification on the above-mentioned subjects. The verification results showed that the consistency between the detection results of the diagnostic model of this invention and the clinical routine examination results was as high as 98.10%. This consistency result fully demonstrates that the diagnostic model constructed by this invention has excellent diagnostic accuracy, can effectively identify H. pylori-infected gastritis, and can provide reliable technical support for the rapid and accurate diagnosis of H. pylori-infected gastritis in clinical practice, thus having significant clinical application value.

[0059] Table 1. Validation of the effectiveness of the HGB-CAZPB-Acetic Acid high-precision diagnostic model. .

Claims

1. A biomarker for the active and stable phases of Helicobacter pylori infection in chronic gastritis, characterized in that, The biomarkers include blood markers, protein markers, and metabolite markers; The blood biomarkers are HGB, UA, CK, RDW-CV, CHE, and GGT; The protein biomarkers are TFRC, CAPZB, ACTR2, ACTR3, CFL1, and ENO3; The metabolite markers are: MADN0339 (2-isopropylmalic acid), MADN2955 (20-carboxyleukotriene B4), MEDN4058 (sodium creatine phosphate), MEDP0200 (diethyl glutarate), MEDP0407 (retinol), MEDP1171 (butenoyl-PAF), and MEDP3439. 097 (dehydrocholesterol), MEDP3557 (PC(22:6 / 18:0)), MEDP4231 (1-octadecyl lysophosphatidylcholine), MEDP4242 (acetic acid), MW0057371HN (PC(20:4(5Z,8Z,11Z,14Z) / 24:0)), MEDP3440 097 (cholest-4-en-3-one) and MEDP3442 097 (yeast sterol).

2. The biomarker according to claim 1, characterized in that, The joint determination based on biomarkers follows the following specific principles: Based on hematological function: assessment was made using HGB, RDW-CV, TFRC, and butenoyl-PAF; Based on energy metabolism: the determination was made by CK, ENO3, sodium creatine phosphate, diethyl glutarate, and acetic acid; Based on cytoskeleton reorganization and cell migration: the determination was made by ARCT2, ARCT3, CFL1, CAZPB, PC (22:6 / 18:0), 1-octadecyl lysophosphatidic acid and PC (20:4(5Z,8Z,11Z,14Z) / 24:0).

3. The biomarker according to claim 2, characterized in that, HGB, RDW-CV, TFRC, and butenoyl-PAF are a combination of biomarkers based on abnormal blood function used to determine the active phase of Helicobacter pylori-infected chronic gastritis.

4. The biomarker according to claim 2, characterized in that, CK, ENO3, sodium creatine phosphate, diethyl glutarate, and acetic acid are a combination of biomarkers used to screen for disease activity in patients susceptible to anemia based on abnormal energy metabolism.

5. The high-precision diagnostic model according to claim 2, characterized in that, ARCT2, ARCT3, CFL1, CAZPB, PC (22:6 / 18:0), 1-octadecyl lysophosphatidylcholine and PC (20:4(5Z,8Z,11Z,14Z) / 24:0) are a combination of biomarkers used to determine the dynamic changes in the mucosal layer during the active phase of Helicobacter pylori-infected chronic gastritis and to identify patients in the stable phase.

6. A high-precision diagnostic model of HGB-CAZPB-Acetic Acid constructed based on the biomarkers described in any one of claims 1-5.

7. The high-precision diagnostic model according to claim 6, characterized in that, The high-precision diagnostic model utilizes HGB, acetic acid, and CAZPB to form a combined diagnostic biomarker.

8. The application of a biomarker as described in any one of claims 1-5 or a high-precision diagnostic model as described in claim 6 or 7 in the preparation of a kit for rapid detection and high-precision non-invasive diagnosis of the active and stable phases of chronic gastritis caused by Helicobacter pylori infection.

Images