A gastric cancer biomarker composition, screening method and application thereof

Abstract

The application discloses a gastric cancer biomarker composition and a screening method and application thereof, and belongs to the technical field of bioinformatics; the screening method of the gastric cancer biomarker composition comprises the following steps: collecting blood samples of gastric cancer patients and healthy controls, performing high-throughput plasma proteomics detection, and screening out differentially expressed proteins; the candidate protein markers are verified by using a prospective cohort in the UK BioSample Bank, 17 kinds of core proteins are finally determined, and the gastric cancer biomarker composition is obtained; and a gastric cancer risk prediction model is constructed by combining clinical risk factors and the protein expression levels of the gastric cancer biomarkers. The gastric cancer risk prediction model constructed by the application can realize precise stratification of gastric cancer risks, and decision curve analysis shows that the net benefit of the gastric cancer risk prediction model is significantly better than that of a traditional screening method, so that the gastric cancer risk prediction model provides an efficient tool for early warning and personalized intervention of gastric cancer.

Description

Technical Field

[0001] This invention relates to the field of bioinformatics technology, specifically to a gastric cancer biomarker composition, its screening method, and its application. Background Technology

[0002] Gastric cancer (GC) is the fifth most common malignant tumor worldwide. It has a significant insidious onset, with more than 80% of patients being diagnosed at an advanced stage. The five-year survival rate is only about 35.1%, while the five-year survival rate for patients with early-stage gastric cancer can reach more than 90%. This huge difference highlights the importance of early screening and diagnosis.

[0003] Currently, gastroscopy screening is considered the most effective means of reducing gastric cancer mortality. However, this screening method faces many challenges in its practical promotion: First, gastroscopy is invasive, and patients generally experience discomfort and fear, leading to poor patient compliance; second, the examination is relatively expensive, and the cost of a single gastroscopy examination may place a certain economic burden on ordinary families; third, gastroscopy is highly dependent on experienced endoscopists and sophisticated examination equipment, making it difficult to popularize on a large scale in areas with relatively scarce medical resources.

[0004] In clinical practice, primary screening methods based on epidemiological risk factors have been widely used, including Helicobacter pylori (H. pylori) infection detection, pepsinogen (PG) detection, and gastrin-17 (G-17) detection. However, these traditional screening methods have significant limitations: the positive rate of H. pylori infection in the general population is as high as 50%-70%, but the proportion that eventually develops into gastric cancer is less than 3%, making its specificity insufficient when used alone as a screening indicator; the sensitivity of pepsinogen detection for early gastric cancer is only 60%-70%, and it is easily interfered with by other gastric diseases; and the risk assessment models built based on these traditional biomarkers have a gastric cancer detection rate of only about 1% in practical applications, with poor concentration effects in high-risk groups.

[0005] With the development of omics technologies, blood proteomics has provided a new research direction for early screening of gastric cancer. Circulating proteins, as direct reflectors of disease phenotypes, have unique advantages: on the one hand, as end products of gene expression, proteins can more directly reflect the pathophysiological state of the body; on the other hand, compared with metabolomics, proteomics has higher disease specificity. In recent years, advancements in liquid chromatography-tandem mass spectrometry (LC-MS / MS) technology have made large-scale plasma proteomics analysis possible, with detection sensitivity reaching the femtomolar level, providing technical support for the discovery of gastric cancer biomarkers.

[0006] However, several key issues remain to be addressed in current proteomics research: First, most studies employ small-sample case-control designs, resulting in a large number of differentially expressed proteins but insufficient specificity; second, biomarker validation is often limited to single centers, lacking independent validation through large-scale prospective cohorts; and third, existing predictive models often rely solely on proteomics data, failing to effectively integrate clinical risk factors and genetic information, thus limiting the models' generalization ability. These limitations severely hinder the translation of proteomics biomarkers into clinical applications.

[0007] The establishment of large biobanks such as the UK Biobank offers a significant opportunity to overcome the aforementioned bottlenecks. These repositories not only contain long-term follow-up data from hundreds of thousands of samples but also integrate multi-omics information, including genomics and proteomics, providing an ideal platform for biomarker discovery and validation. In conclusion, by fully utilizing these resources, it is hoped that tools for predicting gastric cancer risk can be developed. Summary of the Invention

[0008] The purpose of this invention is to provide a gastric cancer biomarker composition, its screening method, and its application, which improves the enrichment effect of magnetic particles on proteins, thereby improving the screening efficiency of gastric cancer-related biomarkers and has important clinical application value.

[0009] The technical solution adopted by the present invention to achieve the above objectives is as follows:

[0010] A screening kit for a gastric cancer biomarker composition includes EDTA anticoagulant tubes, a protein enrichment reagent, and magnetic particles; the protein enrichment reagent includes an enrichment buffer, a digestion buffer, and a termination buffer; the digestion buffer contains trypsin, and the amount of trypsin used is 0.01-0.05 wt% of the total digestion buffer; the magnetic particles include enrichment magnetic beads, which are oleic acid-modified magnetite nanoparticles; the gastric cancer biomarker composition includes CTSD, KRT19, GGH, ITGA11, FABP5, CTRC, MMP7, SERPINA12, GGT5, NCF2, SCGB3A2, GYS1, SFRP1, ATP6AP2, LSP1, LGALS1, and ALDH3A1. The gastric cancer biomarker composition proposed in this invention includes 17 core proteins. By combining clinical risk factors and the protein expression levels of the gastric cancer biomarker composition, a gastric cancer risk prediction model can be constructed, which can achieve precise stratification of gastric cancer risk. The net benefit of the gastric cancer risk prediction model is significantly better than that of traditional screening methods, providing an efficient tool for early warning and personalized intervention of gastric cancer.

[0011] Preferably, the method for screening the gastric cancer biomarker composition includes collecting blood from gastric cancer patients and healthy controls and preparing plasma; enriching plasma proteins with magnetic particles; screening differentially expressed proteins based on liquid chromatography-tandem mass spectrometry (LC-MS / MS); and then independently validating the composition through a prospective cohort study to obtain the gastric cancer biomarker composition. The magnetic particles include enriched magnetic beads, which are oleic acid-modified magnetite nanoparticles. This invention utilizes enriched magnetic beads to efficiently capture low-abundance proteins in plasma, combines this with liquid chromatography-tandem mass spectrometry (LC-MS / MS) technology to accurately screen differentially expressed proteins between gastric cancer patients and healthy controls, and validates the composition through a prospective cohort study to ultimately obtain a gastric cancer biomarker composition with high specificity and sensitivity. This method not only significantly improves the detection sensitivity and screening accuracy of trace proteins but also has good clinical applicability. Its non-invasive blood testing method and standardized procedure provide a reliable tool for early diagnosis and treatment monitoring of gastric cancer.

[0012] Preferably, the magnetic particles also include modified enriched magnetic beads.

[0013] More preferably, in the preparation of the modified enrichment magnetic beads, the enrichment magnetic beads are ultrasonically dispersed in xylene, and then, under the action of azobisisobutyronitrile (AIBN), vinyl decanoate and 4-hydroxybutyl acrylate are added for polymerization to obtain the modified enrichment magnetic beads; the enrichment magnetic beads are oleic acid-modified iron oxide magnetic nanoparticles. This invention enhances the affinity of the enrichment magnetic beads for target proteins and reduces non-specific adsorption by functionalizing the surface coating layer of the enrichment magnetic beads, significantly improving the enrichment fold of the prepared modified enrichment magnetic beads; at the same time, it enhances the repeatability and stability of the modified enrichment magnetic beads, providing excellent technical support for high-throughput proteomics detection.

[0014] More preferably, the ratio of enrichment magnetic beads to xylene is 1g:50-200mL.

[0015] More preferably, the mass ratio of enriched magnetic beads to azobisisobutyronitrile is 1:0.01-0.05.

[0016] More preferably, the mass ratio of enriched magnetic beads to vinyl decanoate is 1:0.4-1.

[0017] More preferably, the mass ratio of enriched magnetic beads to 4-hydroxybutyl acrylate is 1:0.1-1.

[0018] More preferably, the preparation of modified enriched magnetic beads specifically involves,

[0019] Under a nitrogen atmosphere, enriched magnetic beads were ultrasonically dispersed in xylene, and an alkenyl unsaturated compound was added and mixed thoroughly. Azobisisobutyronitrile (AIBN) was then added, and the mixture was heated at 60-80°C for 4-12 hours. After the reaction, the beads were washed 2-5 times with petroleum ether and vacuum dried to obtain the modified enriched magnetic beads. The enriched magnetic beads are oleic acid-modified iron(III) oxide magnetic nanoparticles. The alkenyl unsaturated compound includes at least one of vinyl decanoate, 4-hydroxybutyl acrylate, and 2-nonenoic acid. This invention further introduces 2-nonenoic acid repeating units onto the surface of the enriched magnetic beads, which helps to further enhance intermolecular forces, synergistically enhance the affinity with target proteins, improve the anti-interference performance of the prepared modified enriched magnetic beads, and thus further improve the enrichment factor and repeatability stability of the prepared modified enriched magnetic beads.

[0020] More preferably, the enriched magnetic beads are oleic acid-modified iron oxide magnetic nanoparticles.

[0021] More preferably, the ratio of enrichment magnetic beads to xylene is 1g:50-200mL.

[0022] More preferably, the mass ratio of enriched magnetic beads to vinyl decanoate is 1:0.4-1.

[0023] More preferably, the mass ratio of enriched magnetic beads to 4-hydroxybutyl acrylate is 1:0.1-1.

[0024] More preferably, the mass ratio of enriched magnetic beads to 2-nonenoic acid is 1:0.1-0.8.

[0025] More preferably, the mass ratio of enriched magnetic beads to azobisisobutyronitrile is 1:0.01-0.05.

[0026] More preferably, the volume ratio of xylene to petroleum ether is 1:1-4.

[0027] Preferably, a method for screening a gastric cancer biomarker composition specifically includes,

[0028] S1. Plasma preparation: Fasting blood samples were collected from participants. After collection, the blood samples were added to EDTA anticoagulant tubes at 0-10℃ for 0.1-1h and centrifuged at 1500-2000g for 5-15min to obtain plasma. When the participant was a pathologically diagnosed individual who had not received anti-tumor treatment, plasma from gastric cancer patients was obtained. When the participant was an individual with no malignant lesions on gastroscopy, plasma from healthy controls was obtained.

[0029] S2. Plasma pretreatment: Mix plasma and enrichment buffer, add magnetic particles and incubate for 20-40 min. Wash with phosphate buffer solution, add digestion buffer, and incubate at 30-40℃ for 12-48 h. Add stop buffer, centrifuge at 15000-17000g for 10-30 min, collect the supernatant and use reverse-phase resin to remove salt and buffer to complete desalting. When the plasma is from a gastric cancer patient, obtain the supernatant sample of the gastric cancer patient; when the plasma is from a healthy control, obtain the supernatant sample of the healthy control.

[0030] S3. Liquid Chromatography-Tandem Mass Spectrometry (LS-MS / MS) Analysis: The supernatant was analyzed using LS-MS / MS technology to obtain LS-MS / MS data. Chromatographic separation was performed using a C18 reversed-phase column with a particle size of 1-2 μm, an inner diameter of 100-200 μm, and a length of 20-40 cm. Mobile phase A was a 0.1-0.3% (w / w) aqueous formic acid solution, and mobile phase B was an acetonitrile solution, a mixture of acetonitrile, formic acid, and deionized water, with acetonitrile comprising 70-90 wt% of the deionized water volume and formic acid comprising 0.1-0.3 wt% of the deionized water volume. The gradient elution program was as follows: 40-80 minutes, phase B linearly increased from 5-10% to 40-60%. The mass spectrometer uses a data-independent acquisition (DIA) mode with a first-stage scan resolution of 50,000-70,000 and a second-stage scan resolution of 20,000-40,000, and a collision energy of 20-40%.

[0031] S4. Proteomics Analysis: Differentially expressed proteins were screened based on liquid chromatography-tandem mass spectrometry (LC-MS / MS) data. These proteins were then validated in a validation set using Cox proportional hazards regression analysis and LASSO-Cox regression analysis to identify a gastric cancer biomarker composition and determine its protein expression levels. The validation set consisted of prospective cohort data from the UK Biobank.

[0032] More preferably, in step S2, the magnetic particles are enriched magnetic beads or modified enriched magnetic beads, and the enriched magnetic beads are oleic acid-modified iron oxide magnetic nanoparticles.

[0033] More preferably, in step S2, the volume ratio of plasma to enrichment buffer is 1:2-10.

[0034] More preferably, in step S2, the ratio of plasma to enriched magnetic beads is 1g:2-10mL.

[0035] More preferably, the digestion buffer in step S2 contains trypsin, and the amount of trypsin used accounts for 0.01-0.05 wt% of the total amount of digestion buffer.

[0036] More preferably, the volume ratio of plasma to digestion buffer in step S2 is 1:2-10.

[0037] More preferably, the volume ratio of plasma to termination buffer in step S2 is 1:2-10.

[0038] This invention also discloses a method for constructing a gastric cancer risk prediction model, including: constructing a gastric cancer risk prediction model based on risk assessment data, calculating a weighted risk score, and classifying gastric cancer risk levels; the risk assessment data includes protein characteristic data of a gastric cancer biomarker composition and / or clinical risk factors. This invention, through a two-stage research design, systematically screens gastric cancer-related plasma protein biomarkers and develops a gastric cancer risk prediction model based on proteomics characteristics, which has significant clinical value for achieving precise screening and personalized prevention of gastric cancer; simultaneously, through multi-dimensional data integration, it provides optimized new ideas for the prevention and screening of gastric cancer.

[0039] Preferably, clinical risk factors include age, sex, smoking status, physical activity level, and family history of cancer.

[0040] Preferably, the formula for calculating the weighted risk score is: Weighted Risk Score = exp(β1×X1+β2×X2+...+βn×Xn), where Xn represents the level of each variable and βn is the corresponding coefficient obtained from the Cox regression model.

[0041] Preferably, a method for constructing a gastric cancer risk prediction model specifically includes:

[0042] S1. Obtain protein characteristic data of the gastric cancer biomarker composition, where the protein characteristic data is the protein expression level.

[0043] S2. Clinical risk factor analysis:

[0044] Prospective cohort data were obtained from the UK Biobank and baseline characteristics were statistically analyzed. A univariate Cox proportional hazards regression model was used to screen for potential clinical variables associated with the risk of gastric cancer.

[0045] Clinical risk factors were obtained by using a backward multivariate Cox regression analysis model based on potential clinical variables associated with gastric cancer risk.

[0046] S3. Construct a gastric cancer risk prediction model:

[0047] Prospective cohort data were obtained from the UK Biobank and randomly divided into training and test sets with a data volume distribution ratio of 6-9:1-4.

[0048] Based on the training set data, a gastric cancer risk prediction model was constructed using risk assessment data; the risk assessment data included protein characterization data of the gastric cancer biomarker composition and / or clinical risk factors.

[0049] The gastric cancer risk prediction model was validated based on the test set data.

[0050] S5. Gastric Cancer Risk Stratification:

[0051] The weighted risk score is calculated based on the gastric cancer risk prediction model. The formula for calculating the weighted risk score is: Weighted risk score = exp(β1×X1+β2×X2+...+βn×Xn), where Xn represents the level of each variable and βn is the corresponding coefficient obtained from the Cox regression model.

[0052] The risk cutoff value is determined using the "surv_cutpoint" function of the "survminer" class in R, and the class is divided into low-risk, medium-risk, and high-risk groups.

[0053] This invention, based on plasma proteomics, develops a novel gastric cancer risk prediction model. It features minimally invasive sampling and simple operation, making it suitable for large-scale population screening and significantly improving screening adherence. In this invention, circulating proteins, as reflectors of disease phenotypes, possess high sensitivity and specificity, enabling earlier and more accurate identification of individuals at high risk for gastric cancer. Bidirectional validation using a discovery set and a large prospective validation set ensures the reliability of the biomarkers and the generalization ability of the model, providing solid evidence for clinical translation. Furthermore, this invention innovatively combines proteomics data with clinical factors and genetic information to construct a multi-dimensional prediction system, achieving precise assessment of gastric cancer risk.

[0054] This invention utilizes vinyl decanoate, 4-hydroxybutyl acrylate, and 2-nonenoic acid to functionalize the surface coating of enrichment magnetic beads, resulting in the following beneficial effects: the modified enrichment magnetic beads prepared by this invention exhibit high enrichment folds and excellent repeatability, with enrichment folds ranging from 48 to 113 times and relative standard deviations from 0.82 to 4.71%. Therefore, this invention discloses a modified enrichment magnetic bead with high enrichment folds and excellent repeatability. Its application in biomarker screening helps improve biomarker screening efficiency and provides reliable technical support for high-throughput proteomics detection. Attached Figure Description

[0055] Figure 1 This is a schematic diagram of blood sample collection and information statistics.

[0056] Figure 2 This is a schematic diagram of proteomic analysis for gastric cancer biomarkers.

[0057] Figure 3The results of the screening for gastric cancer biomarkers.

[0058] Figure 4 This is a schematic diagram illustrating the construction of a gastric cancer risk prediction model.

[0059] Figure 5 The AUC results are for a gastric cancer risk prediction model built based on the training set.

[0060] Figure 6 For calibration curves.

[0061] Figure 7 The ROC results for the gastric cancer risk prediction model constructed based on the training set in Example 1 are shown.

[0062] Figure 8 The ROC results for Example 2 are the results of constructing a gastric cancer risk prediction model based on the training set.

[0063] Figure 9 To validate the AUC results of the gastric cancer risk prediction model based on the test set.

[0064] Figure 10 For calibration curves.

[0065] Figure 11 The ROC results of the gastric cancer risk prediction model are used to verify the test set in Example 1.

[0066] Figure 12 Example 2 uses the test set to validate the ROC results of the gastric cancer risk prediction model.

[0067] Figure 13 Stratification of gastric cancer risk.

[0068] Figure 14 This represents the decision curve result.

[0069] Figure 15 SEM image of the modified and enriched magnetic beads. Detailed Implementation

[0070] The present invention will now be described in further detail with reference to specific embodiments. The given embodiments are merely illustrative of the invention and not intended to limit its scope. The embodiments provided below can serve as a guide for further improvements by those skilled in the art and do not constitute a limitation on the invention in any way.

[0071] Unless otherwise specified, the experimental methods used in the following examples are conventional methods. Unless otherwise specified, the materials and reagents used in the following examples are commercially available.

[0072] Example 1:

[0073] A gastric cancer biomarker composition comprising CTSD, KRT19, GGH, ITGA11, FABP5, CTRC, MMP7, SERPINA12, GGT5, NCF2, SCGB3A2, GYS1, SFRP1, ATP6AP2, LSP1, LGALS1, and ALDH3A1.

[0074] A method for screening gastric cancer biomarker compositions, comprising,

[0075] S1. Plasma Preparation: Fasting blood samples were collected from participants. After collection, the blood samples were incubated at 4°C for 1 hour, then added to EDTA anticoagulant tubes and centrifuged at 4°C and 1800g for 10 minutes to obtain plasma. Plasma from gastric cancer patients was obtained when the participant was pathologically diagnosed and had not received antitumor treatment; plasma from healthy controls was obtained when the participant had no malignant lesions on gastroscopy. EDTA anticoagulant tubes were purchased from BD Biosciences, USA.

[0076] S2. Plasma Pretreatment: Plasma and enrichment buffer were mixed, enrichment magnetic beads were added and incubated for 30 min. After washing with phosphate buffer solution, digestion buffer was added and incubated at 37℃ for 24 h. Stop buffer was added, and the mixture was centrifuged at 16000g for 20 min. The supernatant was collected and desalted by removing salt and buffer solution using a reverse-phase resin. When the plasma was from a gastric cancer patient, a supernatant sample from the gastric cancer patient was obtained; when the plasma was from a healthy control, a supernatant sample from the healthy control was obtained. Enrichment buffer, digestion buffer, and stop buffer were all purchased from Huzhou Shenke Biotechnology Co., Ltd. The enrichment magnetic beads were oleic acid-modified iron oxide magnetic nanoparticles. Trypsin was added to the digestion buffer and mixed for plasma pretreatment. The amount of trypsin used was 0.02 wt% of the total digestion buffer. The volume ratio of plasma to enrichment buffer was 1:4, the volume ratio of enrichment magnetic beads to plasma was 1 g: 5 mL, the volume ratio of plasma to digestion buffer was 1:5, and the volume ratio of plasma to stop buffer was 1:5.

[0077] S3. Liquid Chromatography-Tandem Mass Spectrometry (LS-MS / MS): A Thermo Scientific Orbitrap Eclipse mass spectrometer coupled with a UltiMate 3000 UHPLC system was used to analyze the supernatant samples from gastric cancer patients and healthy controls, obtaining LS-MS / MS data. Chromatographic separation was performed using a C18 reversed-phase column with a particle size of 1.9 μm, an inner diameter of 150 μm, and a length of 30 cm. Mobile phase A was a 0.1% formic acid aqueous solution, and mobile phase B was an acetonitrile solution, a mixture of acetonitrile, formic acid, and deionized water, with acetonitrile comprising 80 wt% of the deionized water volume and formic acid comprising 0.1 wt% of the deionized water volume. The gradient elution program was: 60 minutes, with phase B linearly increasing from 8% to 50%. The mass spectrometer used a data-independent acquisition (DIA) mode with a first-stage scan resolution of 60,000 and a second-stage scan resolution of 30,000, and a collision energy of 32%.

[0078] S4. Proteomics Analysis: Based on liquid chromatography-tandem mass spectrometry (LC-MS / MS) data, differentially expressed proteins were screened. These proteins were then validated in a validation set using Cox proportional hazards regression analysis and LASSO-Cox regression analysis to identify a gastric cancer biomarker composition and determine its protein expression levels. The validation set consisted of prospective cohort data from the UK Biobank, comprising 52,552 cases, including data from 92 gastric cancer patients and 52,460 healthy controls.

[0079] Methods for constructing gastric cancer risk prediction models include,

[0080] S1. Obtain protein characteristic data of the gastric cancer biomarker combination group. The protein characteristic data is the protein expression level.

[0081] S2. Clinical risk factor analysis:

[0082] Prospective cohort data were obtained from the UK Biobank and baseline characteristics were statistically analyzed. A univariate Cox proportional hazards regression model was used to screen for potential clinical variables associated with the risk of gastric cancer.

[0083] Clinical risk factors were obtained by using a backward multivariate Cox regression analysis model based on potential clinical variables associated with gastric cancer risk.

[0084] S3. Construct a gastric cancer risk prediction model:

[0085] Prospective cohort data were obtained from the UK Biobank and randomly divided into training and test sets with a data volume distribution ratio of 7:3.

[0086] Based on the training set data, a gastric cancer risk prediction model was constructed using risk assessment data, which consisted of clinical risk factors.

[0087] The gastric cancer risk prediction model was validated based on the test set data.

[0088] S4. Gastric Cancer Risk Stratification:

[0089] Based on prospective cohort data, the weighted risk score was calculated using the best-performing gastric cancer risk prediction model. The calculation formula is: Weighted risk score = exp(β1×X1+β2×X2+...+βn×Xn), where Xn represents the level of each variable and βn is the corresponding coefficient obtained from the Cox regression model.

[0090] The risk cutoff value is determined using the "surv_cutpoint" function of the "survminer" class in R, and the class is divided into low-risk, medium-risk, and high-risk groups.

[0091] Example 2:

[0092] A gastric cancer biomarker composition, as in Example 1.

[0093] A method for screening a gastric cancer biomarker composition is the same as in Example 1.

[0094] The method for constructing a gastric cancer risk prediction model is the same as in Example 1, except that the risk assessment data in step S4 consists of clinical risk factors and protein characteristic data of a gastric cancer biomarker composition.

[0095] Example 3:

[0096] A gastric cancer biomarker composition, as in Example 1.

[0097] The preparation of modified enriched magnetic beads includes,

[0098] Under a nitrogen atmosphere, enriched magnetic beads were ultrasonically dispersed in xylene, followed by the addition of vinyl decanoate and 4-hydroxybutyl acrylate, and then mixed thoroughly. Azobisisobutyronitrile (AIBN) was added, and the mixture was heated at 70°C for 6 hours. After the reaction, the mixture was washed three times with petroleum ether and vacuum dried to obtain the modified enriched magnetic beads. The enriched magnetic beads were oleic acid-modified magnetite nanoparticles. The mass ratio of enriched magnetic beads to xylene was 1 g:100 mL; the mass ratio of enriched magnetic beads to vinyl decanoate was 1:1; the mass ratio of enriched magnetic beads to 4-hydroxybutyl acrylate was 1:1; the mass ratio of enriched magnetic beads to AIBN was 1:0.03; and the volume ratio of xylene to petroleum ether was 1:2.

[0099] A method for screening a gastric cancer biomarker composition, compared with Example 1, except that in step S2 the enrichment magnetic beads are replaced with the modified enrichment magnetic beads prepared in this example, and the other conditions are the same as in Example 1.

[0100] The method for constructing a gastric cancer risk prediction model is the same as in Example 1, except that the risk assessment data in step S4 consists of clinical risk factors and protein characteristic data of a gastric cancer biomarker composition.

[0101] Example 4:

[0102] A gastric cancer biomarker composition, as in Example 1.

[0103] The preparation of modified enriched magnetic beads was the same as in Example 3, except that the mass ratio of enriched magnetic beads to vinyl decanoate was changed to 1:0.4.

[0104] A method for screening a gastric cancer biomarker composition, compared with Example 1, except that in step S2 the enrichment magnetic beads are replaced with the modified enrichment magnetic beads prepared in this example, and the other conditions are the same as in Example 1.

[0105] The method for constructing a gastric cancer risk prediction model is the same as in Example 1, except that the risk assessment data in step S4 consists of clinical risk factors and protein characteristic data of a gastric cancer biomarker composition.

[0106] Example 5:

[0107] A gastric cancer biomarker composition, as in Example 1.

[0108] The preparation of the modified enrichment magnetic beads was the same as in Example 3, except that the mass ratio of enrichment magnetic beads to 4-hydroxybutyl acrylate was changed to 1:0.1.

[0109] A method for screening a gastric cancer biomarker composition, compared with Example 1, except that in step S2 the enrichment magnetic beads are replaced with the modified enrichment magnetic beads prepared in this example, and the other conditions are the same as in Example 1.

[0110] The method for constructing a gastric cancer risk prediction model is the same as in Example 1, except that the risk assessment data in step S4 consists of clinical risk factors and protein characteristic data of a gastric cancer biomarker composition.

[0111] Example 6:

[0112] A gastric cancer biomarker composition, as in Example 1.

[0113] The preparation of modified enriched magnetic beads includes,

[0114] Under a nitrogen atmosphere, enriched magnetic beads were ultrasonically dispersed in xylene, followed by the addition of vinyl decanoate, 4-hydroxybutyl acrylate, and 2-nonenoic acid, and then mixed thoroughly. Azobisisobutyronitrile (AIBN) was added, and the mixture was heated at 70°C for 6 hours. After the reaction, the mixture was washed three times with petroleum ether and vacuum dried to obtain the modified enriched magnetic beads. The enriched magnetic beads were oleic acid-modified magnetite nanoparticles. The mass ratio of enriched magnetic beads to xylene was 1 g:100 mL; the mass ratio of enriched magnetic beads to vinyl decanoate was 1:1; the mass ratio of enriched magnetic beads to 4-hydroxybutyl acrylate was 1:1; the mass ratio of enriched magnetic beads to 2-nonenoic acid was 1:0.8; the mass ratio of enriched magnetic beads to AIBN was 1:0.03; and the volume ratio of xylene to petroleum ether was 1:2.

[0115] A method for screening a gastric cancer biomarker composition, compared with Example 1, except that in step S2 the enrichment magnetic beads are replaced with the modified enrichment magnetic beads prepared in this example, and the other conditions are the same as in Example 1.

[0116] The method for constructing a gastric cancer risk prediction model is the same as in Example 1, except that the risk assessment data in step S4 consists of clinical risk factors and protein characteristic data of a gastric cancer biomarker composition.

[0117] Example 7:

[0118] A gastric cancer biomarker composition, as in Example 1.

[0119] The preparation of modified enriched magnetic beads was the same as in Example 6, except that the mass ratio of enriched magnetic beads to 2-nonenoic acid was changed to 1:0.1.

[0120] A method for screening a gastric cancer biomarker composition, compared with Example 1, except that in step S2 the enrichment magnetic beads are replaced with the modified enrichment magnetic beads prepared in this example, and the other conditions are the same as in Example 1.

[0121] The method for constructing a gastric cancer risk prediction model is the same as in Example 1, except that the risk assessment data in step S4 consists of clinical risk factors and protein characteristic data of a gastric cancer biomarker composition.

[0122] Comparative Example 1:

[0123] A gastric cancer biomarker composition, as in Example 1.

[0124] The preparation of the modified enriched magnetic beads was the same as in Example 3, except that 4-hydroxybutyl acrylate was not added.

[0125] A method for screening a gastric cancer biomarker composition, compared with Example 1, except that in step S2 the enrichment magnetic beads are replaced with the modified enrichment magnetic beads prepared in this example, and the other conditions are the same as in Example 1.

[0126] The method for constructing a gastric cancer risk prediction model is the same as in Example 1, except that the risk assessment data in step S4 consists of clinical risk factors and protein characteristic data of a gastric cancer biomarker composition.

[0127] Comparative Example 2:

[0128] A gastric cancer biomarker composition, as in Example 1.

[0129] The preparation of the modified enriched magnetic beads was the same as in Example 3, except that vinyl decanoate was not added.

[0130] A method for screening a gastric cancer biomarker composition, compared with Example 1, except that in step S2 the enrichment magnetic beads are replaced with the modified enrichment magnetic beads prepared in this example, and the other conditions are the same as in Example 1.

[0131] The method for constructing a gastric cancer risk prediction model is the same as in Example 1, except that the risk assessment data in step S4 consists of clinical risk factors and protein characteristic data of a gastric cancer biomarker composition.

[0132] Comparative Example 3:

[0133] A gastric cancer biomarker composition, as in Example 1.

[0134] The preparation of the modified enriched magnetic beads was the same as in Example 6, except that vinyl decanoate and 4-hydroxybutyl acrylate were not added.

[0135] A method for screening a gastric cancer biomarker composition, compared with Example 1, except that in step S2 the enrichment magnetic beads are replaced with the modified enrichment magnetic beads prepared in this example, and the other conditions are the same as in Example 1.

[0136] The method for constructing a gastric cancer risk prediction model is the same as in Example 1, except that the risk assessment data in step S4 consists of clinical risk factors and protein characteristic data of a gastric cancer biomarker composition.

[0137] Experimental example:

[0138] 1. Screening of gastric cancer biomarker compositions

[0139] The gastric cancer biomarker composition was screened according to the screening method of a gastric cancer biomarker composition in Example 1. Figure 1 This diagram illustrates the blood sample collection and statistical analysis. A gastric cancer group and a non-gastric cancer group were established. The gastric cancer group included 100 fasting blood samples from gastric cancer patients collected at Zhejiang Cancer Hospital, while the non-gastric cancer group included 94 fasting blood samples from healthy controls collected at Zhejiang Cancer Hospital. Blood samples from both groups were used to screen for gastric cancer biomarker combinations, and basic information on clinical risk factors in the blood samples was statistically analyzed.

[0140] Figure 2 This is a schematic diagram of proteomic analysis for gastric cancer biomarkers. Figure 2As shown, in proteomics analysis, based on liquid chromatography-tandem mass spectrometry (LC-MS / MS) data, searches were performed using Spectronaut 17.4 software and the UniProt Human Protein Database (2022 version). Trypsin digestion was allowed, with a maximum of two missed cleavage sites. The fixed modification was cysteine ​​carboxymethylation, with variable modifications including methionine oxidation and N-terminal acetylation. The false positive rate (FDR) threshold was <1%. Antibody-based protein analysis was performed using the Olink Explore 3072 platform, which includes protein assays, data processing, and quality control procedures across the entire study scope, yielding proteomics characteristics. Based on these characteristics, differentially expressed proteins (DEPs) were defined as follows: upregulated protein expression when the Log2 fold change (Log2FC) was greater than 0, and downregulated protein expression when Log2FC was less than 0. Multiple validation was performed with a false positive rate (FDR) <0.05 as the significance threshold, resulting in the screening of 2306 differentially expressed proteins. Subsequently, Cox proportional hazards regression analysis was used to validate the differentially expressed proteins on a validation set, which consisted of prospective cohort data from the UK Biobank. Proteins were considered successfully validated if their directional effects were consistent and P < 0.05. A total of 25 proteins were validated. Table 1 shows the validation results for these 25 proteins, where FC represents fold change, FDR represents false discovery rate, HR represents hazard ratio, and CI represents confidence interval. As shown in Table 1, among the 25 retested validated proteins, 20 were positively correlated with the risk of gastric cancer (hazard ratio HR > 1, P < 0.05), and 5 were negatively correlated with the risk of gastric cancer (hazard ratio HR < 1, P < 0.05).

[0141] Table 1. Validation results of 25 proteins

[0142]

[0143] Further optimization of characteristics was achieved through LASSO-Cox regression analysis to obtain a gastric cancer biomarker composition. Figure 3 For the screening results of gastric cancer biomarkers, FC represents fold change, FDR represents false discovery rate, and HR represents hazard ratio. Therefore, the gastric cancer biomarker composition includes CTSD, KRT19, GGH, ITGA11, FABP5, CTRC, MMP7, SERPINA12, GGT5, NCF2, SCGB3A2, GYS1, SFRP1, ATP6AP2, LSP1, LGALS1, and ALDH3A1.

[0144] 2. Construction of a gastric cancer risk prediction model

[0145] Based on baseline characteristics from prospective cohort data, a univariate Cox proportional hazards regression model was used to screen for potential clinical factors associated with gastric cancer risk. Subsequently, a backward-stepping multivariate Cox regression model was used to determine the final clinical risk factors included in the predictive model, and the hazard ratios (HRs) and 95% confidence intervals (CIs) of each clinical risk factor were calculated. Clinical risk factors included age, sex, education level, smoking status, physical activity level, and family history of cancer. Table 2 shows the hazard ratios and 95% confidence intervals for each clinical risk factor; HR represents the hazard ratio, and CI represents the confidence interval.

[0146] Table 2. Hazard ratios and 95% confidence intervals for clinical risk factors

[0147]

[0148] According to the method of constructing the gastric cancer risk prediction model in Example 1, a gastric cancer risk prediction model based on clinical risk factors is constructed, and a weighted risk score is output to achieve gastric cancer risk stratification; according to the method of constructing the gastric cancer risk prediction model in Example 2, a gastric cancer risk prediction model based on a combination of clinical risk factors and gastric cancer biomarkers is constructed, and a weighted risk score is output to achieve gastric cancer risk stratification. Figure 4 This is a schematic diagram illustrating the construction of a gastric cancer risk prediction model.

[0149] The performance of the gastric cancer risk prediction model based on clinical risk factors in Example 1 and the gastric cancer risk prediction model based on a combination of clinical risk factors and gastric cancer biomarkers in Example 2 were evaluated using the area under the curve (AUC). Time-dependent receiver operating characteristic (ROC) curves were plotted. A bootstrap method was used to compare the performance differences of each model with 500 stratified samples, and calibration curves were plotted to assess the consistency between the predicted probability of no gastric cancer and the actual observed value.

[0150] Figure 5 The AUC results for the gastric cancer risk prediction models built based on the training set are shown. The blue solid line represents the AUC results of the gastric cancer risk prediction model based on clinical risk factors obtained in Example 1, and the purple solid line represents the AUC results of the gastric cancer risk prediction model built in Example 2 based on a combination of clinical risk factors and gastric cancer biomarkers. Figure 5 As shown, the area under the curve (AUC) of the gastric cancer risk prediction model constructed in Example 1 was 0.767, and the 95% confidence interval was 0.711–0.823; the AUC of the gastric cancer risk prediction model constructed in Example 2 was 0.823, and the 95% confidence interval was 0.780–0.866. The gastric cancer risk prediction model constructed based on a combination of clinical risk factors and gastric cancer biomarkers showed a significant performance improvement compared to other models (P < 0.05). Figure 6For calibration curves, the solid blue line represents the calibration curve of the gastric cancer risk prediction model based on clinical risk factors constructed in Example 1, and the dashed purple line represents the calibration curve of the gastric cancer risk prediction model constructed in Example 2 based on a combination of clinical risk factors and gastric cancer biomarkers. Figure 6 It can be seen that the gastric cancer risk prediction model constructed based on the combination of clinical risk factors and gastric cancer biomarkers has good consistency with actual observers in predicting the probability of gastric cancer occurrence.

[0151] Figure 7 The image shows the ROC results of the gastric cancer risk prediction model built based on the training set in Example 1. The red solid line represents the predictive efficacy for 5-year gastric cancer risk, the blue solid line represents the predictive efficacy for 10-year gastric cancer risk, and the yellow solid line represents the predictive efficacy for 15-year gastric cancer risk. Figure 7 It can be seen that the predictive power of the model for the risk of gastric cancer in 5 years is 0.833, the predictive power for the risk of gastric cancer in 10 years is 0.801, and the predictive power for the risk of gastric cancer in 15 years is 0.767. Figure 8 The image shows the ROC results of the gastric cancer risk prediction model built based on the training set in Example 2. The red solid line represents the predictive power for 5-year gastric cancer risk, the blue solid line represents the predictive power for 10-year gastric cancer risk, and the yellow solid line represents the predictive power for 15-year gastric cancer risk. The model has a predictive power of 0.877 for 5-year gastric cancer risk, 0.848 for 10-year gastric cancer risk, and 0.813 for 15-year gastric cancer risk.

[0152] Figure 9 To validate the AUC results of the gastric cancer risk prediction model based on the test set, the blue solid line represents the AUC results of Example 1 validating the gastric cancer risk prediction model constructed based on clinical risk factors, and the purple solid line represents the AUC results of Example 2 validating the gastric cancer risk prediction model constructed based on a combination of clinical risk factors and gastric cancer biomarkers. Figure 9 As can be seen, the area under the curve (AUC) of the gastric cancer risk prediction model validated in Example 1 was 0.750; the AUC of the gastric cancer risk prediction model validated in Example 2 was 0.835. The gastric cancer risk prediction model constructed based on the combination of clinical risk factors and gastric cancer biomarkers showed a significant performance improvement compared to other models (P < 0.05). Figure 10 For calibration curves, the solid blue line represents the calibration curve of the gastric cancer risk prediction model constructed based on clinical risk factors in Example 1, and the dashed purple line represents the calibration curve of the gastric cancer risk prediction model constructed based on a combination of clinical risk factors and gastric cancer biomarkers in Example 2. Figure 10 It can be seen that the gastric cancer risk prediction model constructed based on the combination of clinical risk factors and gastric cancer biomarkers has good consistency with actual observers in predicting the probability of gastric cancer occurrence.

[0153] Figure 11 The ROC results for Example 1, based on the test set, validate the gastric cancer risk prediction model. The red solid line represents the predictive efficacy for 5-year gastric cancer risk, the blue solid line represents the predictive efficacy for 10-year gastric cancer risk, and the yellow solid line represents the predictive efficacy for 15-year gastric cancer risk. Figure 11 It can be seen that the predictive power of the model for the risk of gastric cancer in 5 years is 0.733, the predictive power for the risk of gastric cancer in 10 years is 0.721, and the predictive power for the risk of gastric cancer in 15 years is 0.756. Figure 12 The ROC results for Example 2, based on the test set, validate the gastric cancer risk prediction model. The red solid line represents the predictive efficacy for 5-year gastric cancer risk, the blue solid line represents the predictive efficacy for 10-year gastric cancer risk, and the yellow solid line represents the predictive efficacy for 15-year gastric cancer risk. Figure 12 It can be seen that the predictive power of the model for the risk of gastric cancer in 5 years is 0.830, the predictive power for the risk of gastric cancer in 10 years is 0.797, and the predictive power for the risk of gastric cancer in 15 years is 0.834.

[0154] Figure 13 To stratify the risk of gastric cancer, participants were divided into low, intermediate, and high risk groups based on risk scores. Significant gradient differences in gastric cancer incidence rates were observed among the different risk groups (P<0.001). Table 3 shows the gastric cancer risk ratios for different risk stratification groups. As shown in Table 3, compared to the low-risk group, the relative risk of gastric cancer in the intermediate-risk group was 7.47 times higher (95% confidence interval: 2.27–24.57); the relative risk in the high-risk group was 45.30 times higher (95% confidence interval: 14.21–144.38).

[0155] Table 3 Gastric cancer risk ratios in different risk stratification groups

[0156]

[0157] Figure 14 This represents the decision curve result. (From...) Figure 14 It can be seen that the gastric cancer risk prediction model constructed based on a combination of clinical risk factors and gastric cancer biomarkers shows a higher net benefit compared to other models.

[0158] 3. SEM analysis of modified enriched magnetic beads

[0159] The modified enriched magnetic beads prepared in Example 5 were ultrasonically dispersed in anhydrous ethanol, and then the mixture was dropped onto a clean copper grid. After drying, the mixture was observed using a transmission electron microscope.

[0160] Figure 15 SEM images of modified and enriched magnetic beads. Figure 15 It can be seen that the modified enriched magnetic beads are regular spherical in shape, have a coating layer on the surface, and have a large particle size distribution and uneven size.

[0161] 4. Enrichment factor of modified enriched magnetic beads

[0162] The enrichment magnetic beads used in Example 1, as well as the modified enrichment magnetic beads prepared in Examples 3-7 and Comparative Examples 1-3, were collected. Using a cytochrome C standard peptide as a model, the cytochrome C standard peptide was diluted to a concentration of 50 nmol / L to obtain a diluted cytochrome C solution. The signal-to-noise ratio (S / N) of the cytochrome C mass spectrum before enrichment was detected by liquid chromatography-tandem mass spectrometry (LC-MS / MS). Then, the enrichment magnetic beads used in Example 1, as well as the modified enrichment magnetic beads prepared in Examples 3-7 and Comparative Examples 1-3, were applied to the diluted cytochrome C solution for enrichment. The S / N of the cytochrome C mass spectrum after enrichment was detected by LC-MS / MS. The enrichment fold of the modified enrichment magnetic beads = S / N of the mass spectrum after enrichment / S / N of the mass spectrum before enrichment. Table 4 shows the enrichment folds.

[0163] Table 4 Enrichment Factors

[0164]

[0165] As shown in Table 4, the enrichment factor of the modified enriching magnetic beads prepared in Examples 3-5 of the present invention is higher than that in Example 1. This is because, in the preparation of the modified enriching magnetic beads, Examples 3-5 introduced a coating layer composed of vinyl decanoate and 4-hydroxybutyl acrylate units on the surface of the enriching magnetic beads. The enrichment factor of the modified enriching magnetic beads prepared in Example 3 is higher than that in Examples 4 and 5. This is because, in the preparation of the modified enriching magnetic beads, the amounts of vinyl decanoate and 4-hydroxybutyl acrylate used are different. The enrichment factor of the modified enriching magnetic beads prepared in Example 3 is higher than that in Comparative Examples 1 and 2. This is because, in the preparation of the modified enriching magnetic beads, Example 3 uses vinyl decanoate and 4-hydroxybutyl acrylate synergistically for modification, while Comparative Example 1 uses only vinyl decanoate and Comparative Example 2 uses only 4-hydroxybutyl acrylate. This indicates that, compared to introducing a coating layer composed of vinyl decanoate units or a coating layer composed of 4-hydroxybutyl acrylate units alone onto the surface of enriched magnetic beads, synergistically introducing a coating layer composed of vinyl decanoate and 4-hydroxybutyl acrylate units onto the surface of enriched magnetic beads can effectively improve the enrichment factor of the prepared modified enriched magnetic beads.

[0166] The enrichment factor of the modified enriching magnetic beads prepared in Examples 6-7 of this invention is higher than that in Example 5 because, in the preparation of the modified enriching magnetic beads, Examples 6-7 introduced a coating layer composed of vinyl decanoate, 4-hydroxybutyl acrylate, and 2-nonenoic acid units on the surface of the enriching magnetic beads. The enrichment factor of the modified enriching magnetic beads prepared in Example 6 is higher than that in Example 7 because the amount of 2-nonenoic acid used in the preparation of the modified enriching magnetic beads is different. The enrichment factor of the modified enriching magnetic beads prepared in Example 6 is higher than that in Comparative Example 3 because, in the preparation of the modified enriching magnetic beads, Comparative Example 3 only introduced a coating layer composed of 2-nonenoic acid units on the surface of the enriching magnetic beads. This indicates that adding 2-nonenoic acid units to the surface of the enriching magnetic beads helps to further improve the enrichment factor of the prepared modified enriching magnetic beads.

[0167] 5. Repeatability of modified enriched magnetic beads

[0168] The enrichment magnetic beads used in Example 1, as well as the modified enrichment magnetic beads prepared in Examples 3-7 and Comparative Examples 1-3, were collected. Using cytochrome C standard peptide as a model, the cytochrome C standard peptide was diluted to a concentration of 50 nmol / L to obtain a cytochrome C dilution. The enrichment magnetic beads used in Example 1, as well as the modified enrichment magnetic beads prepared in Examples 3-7 and Comparative Examples 1-3, were then applied to the cytochrome C dilution for enrichment. Analysis was performed by liquid chromatography-tandem mass spectrometry (LC-MS / MS) to detect the signal-to-noise ratio (S / N) of the mass spectrometric peaks of enriched cytochrome C. Each experiment was repeated four times, and the relative standard deviation (%) was calculated. Table 5 shows the relative standard deviation (%).

[0169] Table 5. Relative Standard Deviation (%)

[0170]

[0171] As shown in Table 5, the relative standard deviation of the modified enriched magnetic beads prepared in Examples 3-5 of this invention is lower than that in Example 1. This is because, in the preparation of the modified enriched magnetic beads, Examples 3-5 introduced a coating layer composed of vinyl decanoate and 4-hydroxybutyl acrylate units on the surface of the enriched magnetic beads. The relative standard deviation of the modified enriched magnetic beads prepared in Example 3 is lower than that in Examples 4 and 5. This is because, in the preparation of the modified enriched magnetic beads, the amounts of vinyl decanoate and 4-hydroxybutyl acrylate used are different. The relative standard deviation of the modified enriched magnetic beads prepared in Example 3 is lower than that in Comparative Examples 1 and 2. This is because, in the preparation of the modified enriched magnetic beads, Example 3 uses vinyl decanoate and 4-hydroxybutyl acrylate synergistically for modification, while Comparative Example 1 uses only vinyl decanoate and Comparative Example 2 uses only 4-hydroxybutyl acrylate. This indicates that, compared to introducing a coating layer composed of vinyl decanoate units or a coating layer composed of 4-hydroxybutyl acrylate units alone onto the surface of enriched magnetic beads, synergistically introducing a coating layer composed of vinyl decanoate and 4-hydroxybutyl acrylate units onto the surface of enriched magnetic beads can effectively improve the repeatability stability of the prepared modified enriched magnetic beads.

[0172] The relative standard deviation of the modified enriched magnetic beads prepared in Examples 6-7 of this invention is lower than that in Example 3 because, in the preparation of the modified enriched magnetic beads, Examples 6-7 introduced a coating layer composed of vinyl decanoate, 4-hydroxybutyl acrylate, and 2-nonenoic acid units on the surface of the enriched magnetic beads. The relative standard deviation of the modified enriched magnetic beads prepared in Example 6 is lower than that in Example 7 because the amount of 2-nonenoic acid used in the preparation of the modified enriched magnetic beads is different. The relative standard deviation of the modified enriched magnetic beads prepared in Example 6 is lower than that in Comparative Example 3 because, in the preparation of the modified enriched magnetic beads, Comparative Example 3 only introduced a coating layer composed of 2-nonenoic acid units on the surface of the enriched magnetic beads. This indicates that adding 2-nonenoic acid units to the surface of the enriched magnetic beads helps to further improve the repeatability and stability of the prepared modified enriched magnetic beads.

[0173] The conventional operations in the operation steps of this invention are well known to those skilled in the art and will not be described in detail here.

[0174] The embodiments described above provide a detailed explanation of the technical solutions of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the present invention. Any changes and modifications made within the scope of the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A screening kit for a gastric cancer biomarker composition, comprising: ethylenediaminetetraacetic acid anticoagulant tubes, a protein enrichment reagent, and magnetic particles; wherein the protein enrichment reagent comprises an enrichment buffer, a digestion buffer, and a termination buffer; wherein the digestion buffer contains trypsin, and the amount of trypsin used accounts for 0.01-0.05 wt% of the total digestion buffer; wherein the magnetic particles comprise modified enrichment magnetic beads, and the enrichment magnetic beads are oleic acid-modified magnetite nanoparticles; wherein the gastric cancer biomarker composition comprises CTSD, KRT19, GGH, ITGA11, FABP5, CTRC, MMP7, SERPINA12, GGT5, NCF2, SCGB3A2, GYS1, SFRP1, ATP6AP2, LSP1, LGALS1, and ALDH3A1; wherein, in the preparation of the modified enrichment magnetic beads, the enrichment magnetic beads are ultrasonically dispersed in xylene, and then, under the action of azobisisobutyronitrile, vinyl decanoate and 4-hydroxybutyl acrylate are added for polymerization reaction to obtain the modified enrichment magnetic beads.

2. The screening kit for a gastric cancer biomarker composition according to claim 1, characterized in that, The method for screening the gastric cancer biomarker composition includes collecting blood from gastric cancer patients and healthy controls and preparing plasma, enriching plasma proteins with magnetic particles, screening differentially expressed proteins based on liquid chromatography-tandem mass spectrometry analysis, and then independently validating the composition through a prospective cohort to obtain the gastric cancer biomarker composition; the magnetic particles include modified enriched magnetic beads, which are oleic acid-modified iron oxide magnetic nanoparticles.

3. The screening kit for a gastric cancer biomarker composition according to claim 1, characterized in that, The ratio of enriched magnetic beads to xylene is 1g:50-200mL.

4. The screening kit for a gastric cancer biomarker composition according to claim 1, characterized in that, The mass ratio of the enriched magnetic beads to azobisisobutyronitrile is 1:0.01-0.

05.

5. The screening kit for a gastric cancer biomarker composition according to claim 1, characterized in that, The mass ratio of the enriched magnetic beads to vinyl decanoate is 1:0.4-1.

6. The screening kit for a gastric cancer biomarker composition according to claim 1, characterized in that, The mass ratio of the enriched magnetic beads to 4-hydroxybutyl acrylate is 1:0.1-1.

Images