Protein detection marker for screening early gastric cancer

Abstract

The invention relates to the technical field of early tumor screening and molecular diagnosis, in particular to a protein detection marker for screening early gastric cancer, and particularly relates to a mass spectrum-based multi-protein marker detection method. The protein composition shows excellent diagnostic performance in early gastric cancer detection, achieves an AUC value of 0.95 or above in a clinical queue, and has clinical application potential.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of early screening and molecular diagnosis of tumors, in particular to a protein detection method for screening early gastric cancer, especially a mass spectrometry-based multi-protein marker detection method. BACKGROUND

[0002] Gastric cancer is the fourth most common malignant tumor in the world, and ranks second in cancer-related deaths. Because early gastric cancer often has no obvious symptoms, most patients are in the middle and advanced stages when diagnosed, missing the best opportunity for surgical intervention. However, early diagnosis is crucial for the treatment of gastric cancer, with a five-year survival rate of over 90%. Although gastroscopy is considered the "gold standard" for the early diagnosis of gastric cancer, this method has the characteristics of strong invasiveness, high cost, and dependence on professional operation, limiting its feasibility and acceptance in large-scale screening of the general population.

[0003] Currently, serum tumor markers commonly used in clinical practice, such as carcinoembryonic antigen (CEA), carbohydrate antigen 72-4 (CA72-4), and carbohydrate antigen 19-9 (CA19-9), have generally low sensitivity and specificity in the detection of early gastric cancer, which cannot meet the needs of early screening in the general population or high-risk groups. In addition, these traditional markers are often interfered by age, inflammation, and other non-tumor factors, further reducing their clinical application value. Therefore, developing a non-invasive, convenient, highly sensitive detection method that can be widely used for population screening has become an urgent need in the current field of early screening for gastric cancer.

[0004] The rapid development of proteomics provides a new technical path for the early non-invasive screening of tumors. Blood, as a convenient and abundant source of clinical samples, contains a large number of proteins reflecting the pathological and physiological state of individuals, and is an important carrier for the development of tumor biomarkers. Compared with single markers, multi-marker detection strategies based on protein combinations can more effectively reflect the systemic physiological changes during tumor development, improving detection sensitivity and specificity while enhancing the screening ability for cancer heterogeneity. Mass spectrometry (MS) technology, as a core tool for proteomics research, has high-throughput, high-sensitivity, and high-precision qualitative and quantitative capabilities, and has been widely used in the screening, verification, and clinical translation of disease-related protein markers.

[0005] Currently, there is still a lack of high-performance detection methods based on blood protein combinations for gastric cancer, especially early gastric cancer. Therefore, developing a set of early gastric cancer screening methods based on blood samples, combined with multi-protein marker combinations and high-throughput mass spectrometry detection technology, has important scientific significance and broad clinical application prospects, and is expected to improve the early detection rate of gastric cancer and provide a practical basis for building a molecular screening platform based on mass spectrometry detection. SUMMARY

[0006] Use of a composition comprising at least seven proteins in the preparation of a kit or composition for the in vitro aided diagnosis of cancer, characterized in that the at least seven proteins comprise: CD14, APMAP, IGLL1, IGHV2-5, IGF2, IGLV4-69 and GP1BA.

[0007] The composition further comprises COMP.

[0008] The composition further comprises APOA4.

[0009] The composition further comprises VWF.

[0010] The composition further comprises LDHB.

[0011] The composition further comprises at least one protein selected from FN1, ACTB, ACTG1.

[0012] A kit for the in vitro aided diagnosis of cancer, comprising reagents for detecting at least seven proteins:

[0013] CD14, APMAP, IGLL1, IGHV2-5, IGF2, IGLV4-69 and GP1BA.

[0014] The reagents are also for detecting the protein combination as defined in any one of claims 2-6.

[0015] The beneficial effects of the present invention are:

[0016] 1. High sensitivity and specificity: the protein combination described in the present invention exhibits excellent diagnostic performance in the detection of early gastric cancer, achieving an AUC value of 0.95 or higher in clinical cohorts, and has potential for clinical application.

[0017] 2. Precise detection method based on mass spectrometry: the present invention constructs a standardized protein group detection process, with the advantages of high detection throughput, good repeatability, and high specificity, suitable for clinical detection laboratories and translational medicine platforms.

[0018] 3. Non-invasive, convenient, and easy to promote: the present invention uses plasma samples for detection, which is a non-invasive sampling method, easy for the subjects to accept, and conducive to widespread screening in high-risk populations.

[0019] 4. Suitable for translational application and commercial development: the protein combination and detection process can be further developed into a cancer early screening kit, LDT product or IVD product, suitable for multiple platform detection systems, and has strong prospects for industrial transformation. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 : Volcano Plot of differentially expressed proteins between gastric cancer patients (right) and healthy controls (left). The horizontal axis represents the log2 (fold change) of the protein, and the vertical axis represents the -log 10 (Adjusted P value). Black points represent significantly different proteins (fold change > 2, P < 0.05), and gray points represent non-significant proteins.

[0021] Figure 2 : Receiver operating characteristic curve (ROC) of the model constructed based on LASSO regression, Random Forest (RF), Support Vector Machine (SVM), and Extreme Gradient Boosting (XGBoost) in the validation set.

[0022] Figure 3 : Schematic diagram of the screening process. DETAILED DESCRIPTION

[0023] The screening process of the protein markers of the present patent for early gastric cancer detection is as follows:

[0024] 1. Sample collection:

[0025] Collect plasma or serum samples, and extract the mixture of total proteins by solvent precipitation or magnetic bead enrichment method.

[0026] 2. Protein enzymatic digestion:

[0027] Perform enzymatic digestion on the protein mixture, and the commonly used digestion enzyme is trypsin. This step is used to obtain the peptide segments corresponding to the proteins.

[0028] 3. Peptide separation:

[0029] Use liquid chromatography (LC) to separate the peptides after enzymatic digestion to improve the coverage of mass spectrometry identification.

[0030] 4. Mass spectrometry analysis:

[0031] Introduce the separated peptides into the mass spectrometer for detection and analysis, and obtain the polypeptide mass spectrum of each protein by detecting the mass-to-charge ratio (m / z) of the primary parent ions and secondary fragment ions of the peptides.

[0032] 5. Data processing:

[0033] The characteristic peptide segments of each protein are analyzed by mass spectrometry, and the relative expression amount of each protein is determined according to the primary parent ion spectrum and the m / z value and abundance of the analyzed secondary fragment peaks. The differential proteins between cancer patients and healthy people are analyzed by t-test and other methods, and a model is constructed for screening and discrimination of multiple cancers.

[0034] In the above steps, the separation conditions of liquid chromatography are as follows:

[0035] The chromatographic column specification is a C18 column.

[0036] The mobile phase A is a 0.05%-0.5% formic acid aqueous solution, and the mobile phase B is a 70%-90% acetonitrile solution containing 0.05%-0.5% formic acid.

[0037] The separation gradient time is 30-120 minutes.

[0038] The parameters of mass spectrometry detection are as follows:

[0039] The MS1 scanning range is 375-1550 m / z, the resolution is 60,000-240,000, the AGC is 200%-500%, and the ion injection time is 20-100 ms.

[0040] The MS2 scanning isolation window is 0.5-5 m / z, the resolution is 15,000-60,000, the AGC is 50%-2000%, the ion injection time is 50-300 ms, and the HCD collision energy is 25%-35%.

[0041] In the following examples, the protein marker detection and analysis application are carried out by the following steps:

[0042] 1. Sample collection:

[0043] A total of 167 plasma samples of cancer patients and healthy volunteers (discovery set: 22 gastric cancer patients and 94 healthy people; validation set: 10 gastric cancer patients and 41 healthy people) are collected, and protein mixtures are obtained by direct dilution, solvent precipitation or magnetic bead enrichment. The solvent precipitation method is used in this embodiment, and the specific steps are as follows: a certain volume of plasma is diluted by adding 50-200 mM ammonium bicarbonate solution at a dilution ratio of 1:10, then pre-cooled methanol is added to the plasma dilution solution at a ratio of 2:1, and the mixture is vortexed and mixed, then centrifuged and the supernatant is discarded, and the precipitate is vacuum dried and stored in a -80°C refrigerator for use.

[0044] 2. Protein enzymatic digestion:

[0045] The protein mixture is subjected to enzymatic digestion, and the commonly used digestion enzyme is trypsin. This step is used to obtain the corresponding peptide segments of the protein. The specific steps are as follows: after drying, the precipitated protein is dissolved in 8M urea, 50mM TCEP solution is added as a reducing agent to a final concentration of 5mM, and after heating and denaturation reduction, 100mM IAA solution is added as an alkylating agent to a final concentration of 10mM, and the reaction is carried out at room temperature for 30 minutes. After diluting the urea to a final concentration of less than 1M with 50-200mM ammonium bicarbonate solution, Lys-C enzyme and Trypsin enzyme are added in the order of 1:50-1:500 enzyme / substrate mass ratio, mixed well, and then reacted overnight. 10% TFA is added to a final concentration of 0.1% to terminate the reaction. After desalting by column and vacuum drying, 0.1% FA is used for re-dissolution before use.

[0046] 3. Peptide separation:

[0047] Liquid chromatography (LC) is used to separate the re-dissolved peptide segments to improve the coverage of mass spectrometry identification. The liquid chromatography model is Vanquish Neo UHPLC system and the mass spectrometry model is Orbitrap Exploris 480 (Thermo Fisher Scientific). The chromatographic column specification is 2 microns x 75 microns x 15 centimeters C18 column, the mobile phase A is water containing 0.1% formic acid, the mobile phase B is 80% acetonitrile containing 0.1% formic acid, and the separation gradient time is 102 min.

[0048] 4. Mass spectrometry analysis:

[0049] The separated peptide segments are introduced into the mass spectrometer for detection and analysis. By detecting the mass-to-charge ratio (m / z) of the primary parent ions and secondary fragment ions of the peptide segments, the mass spectrometric spectrum of different polypeptides is obtained. This example uses a non-targeted detection method as an example, but is not limited to non-targeted detection analysis, and is also applicable to targeted detection methods. The mass spectrometry model is Orbitrap Exploris 480 (Thermo Fisher Scientific), the mass spectrometry MS1 scan range is 380-1000 m / z, the resolution is 120,000, the AGC is 300%, and the IT is 50 ms; the MS2 scan isolation window is 4 m / z, the resolution is 30,000, the AGC is 1000%, the ion injection time is 100 ms, and the HCD collision energy is 27%.

[0050] 5. Data processing:

[0051] The mass spectrometry data of the characteristic peptide segments of the target protein were compared with the spectrum generated by database simulation (UniProt Human database, downloaded in Aug 2024). According to the m / z (mass-to-charge ratio) value and relative abundance of the primary parent ion (MS1) spectrum and the dissociated secondary fragment ion (MS / MS), the relative expression level of each protein in the sample was identified and quantified. Subsequently, the t-test statistical method was used to evaluate the protein expression difference between cancer patients and healthy individuals, and the differentially expressed proteins with statistical significance were screened (see Figure 1 The steps are as Figure 3 .

[0052] A total of 447 proteins were identified, of which 258 proteins had an identification rate of >50% in the gastric cancer group and the healthy control group, and were identified in at least 3 samples in each group. Further, 14 proteins were obtained as differential proteins (log2FC>1, P<0.05).

[0053] Based on the above discovered differentially expressed proteins, LASSO regression, random forest (RF), support vector machine (SVM), and gradient boosting decision tree (XGBoost) were used to further construct models for screening analysis of early gastric cancer. The verification results show that the area under the receiver operating characteristic curve (AUC) of the model built by using the protein combination is close to 1 (see Figure 2 and Tables 1 and 2), indicating that it has very high screening and diagnostic performance. After comprehensive comparison, the AUC, sensitivity, specificity, and positive predictive value of the random forest model all reach 1, and the effect of distinguishing gastric cancer and healthy groups is the best.

[0054] It can be seen that the effect of some combinations is not good, for example, the sensitivity of the first 6 groups is only 0.1, which is not suitable for clinical use. At the same time, it can be seen that after adding GP1BA, the sensitivity jumps from 0.1 to 0.5, indicating that the 7-protein combination containing GP1BA has significantly improved detection performance. The AUC of the 8-protein combination formed after the addition of the new COMP protein is higher. The sensitivity (0.7) and Youden index (0.7) of the 9-protein combination after the addition of APOA4 are further improved. The AUC of the 10-protein combination obtained by adding VWF reaches 1. The protein combination obtained by further containing LDHB shows that the AUC / sensitivity / specificity are all 1, and the number of combinations is the smallest. The 11-protein combination containing LDHB is the best combination in the data. In addition, the detection performance remains stable after further adding FN1, ACTB, and ACTG1.

[0055] Table 1. Area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and Youden index of the models built based on the discovered differential protein features using Random Forest (RF) in the validation set.

[0056] Table 2. Area under the receiver operating characteristic curve (AUC), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of various models built based on LASSO regression, Random Forest (RF), Support Vector Machine (SVM), and Extreme Gradient Boosting (XGBoost) in the validation set.

[0057]

Claims

1. Use of a composition comprising at least seven proteins in the preparation of a kit or composition for in vitro auxiliary diagnosis of cancer, characterized in that, The at least seven proteins include: CD14, APMAP, IGLL1, IGHV2-5, IGF2, IGLV4-69, and GP1BA.

2. The use according to claim 1, characterized in that, The composition further comprises COMP.

3. The use according to claim 1, characterized in that, The composition further comprises APOA4.

4. The use according to claim 1, characterized in that, The composition further comprises VWF.

5. The use according to claim 1, characterized in that, The composition further comprises LDHB.

6. The use according to claim 1, characterized in that, The composition further comprises at least one protein selected from FN1, ACTB, and ACTG1.

7. A reagent kit for in vitro auxiliary diagnosis of cancer, characterized in that, It contains reagents for detecting at least seven of the following proteins: CD14, APMAP, IGLL1, IGHV2-5, IGF2, IGLV4-69, and GP1BA.

8. The reagent kit according to claim 7, characterized in that, The reagent is also used to detect at least one protein selected from COMP, APOA4, VWF, LDHB, FN1, ACTB, and ACTG1.

Images