A marker group for gastric cancer screening and diagnosis and use thereof

Abstract

The application discloses a marker group for screening and diagnosing gastric cancer and application thereof, and belongs to the field of molecular biological technology.The marker group comprises at least four polypeptides shown in sequences of SEQ ID NO.1-36.The marker combination constructed by the application has more excellent sensitivity and specificity when used for the auxiliary diagnosis and early screening of gastric cancer.The selected discriminant peak set (marker combination) of the application is superior to the current routine MRI detection (sensitivity and specificity are 77.42% and 94.12% respectively) of gastric cancer and the detection kit based on miRNA (sensitivity and specificity are 69.4% and 84.6% respectively).The verification result shows that the selected discriminant peak set (marker combination) of the application is superior to the common marker combinations CA72-4, CEA, CA125 and CA19-9 (sensitivity and specificity are 66.3% and 80.2% respectively) for screening gastric cancer.

Description

Technical Field

[0001] This invention belongs to the field of molecular biology technology, specifically relating to a biomarker group for gastric cancer screening and diagnosis and its uses. Background Technology

[0002] Early symptoms of gastric cancer are often subtle, and most patients are diagnosed at middle or late stages. Therefore, reliable early screening plays a crucial role in patient survival. Gastroscopy is the most common method for gastric cancer screening; however, it relies on advanced equipment and skilled endoscopists, is relatively expensive, and can be painful, leading to low patient acceptance. Therefore, it is not suitable for large-scale gastric cancer screening. Conventional endoscopy is suitable for detecting advanced gastric cancer, but its detection rate for early-stage gastric cancer is low.

[0003] Other methods for clinical diagnosis of gastric cancer include serological marker testing and X-ray examination using low-tension double-contrast gastric imaging. However, gastric cancer patients often experience belching and upper abdominal discomfort, which can easily lead to misdiagnosis. Known serum laboratory markers for gastric cancer include carcinoembryonic antigen (CEA), carbohydrate antigen 199 (CA199), and carbohydrate antigen 724 (CA724), with receiver operating characteristic (ROC) areas under the curve (AUC) of 0.564, 0.590, and 0.683, respectively, indicating low diagnostic ability for gastric cancer. The revised NCCN guidelines for gastric cancer diagnosis and treatment have changed the wording of next-generation sequencing (NGS) from "may be considered" to "should be considered" in the clinical pathway. The guidelines state that if available tissue is limited, or patients cannot undergo traditional tissue biopsies, continuous testing of single biomarkers or a limited suite of molecular diagnostic tests will exhaust the sample. The biomarkers to be tested are listed, primarily including: human epidermal growth factor receptor 2 (HER2) overexpression / amplification, immunohistochemical programmed death-ligand 1 (PD-L1) expression, microsatellite instability (MSI), tumor mutational burden (TMB), NTRK gene fusion, RET gene fusion, and BRAFV600E mutation. Additionally, a footnote for diagnostic tests has been added stating that "tumor Epstein-Barr virus (EBV) status is increasingly becoming a potential biomarker in personalized treatment strategies for gastric cancer, but NGS is not currently recommended for clinical diagnosis and treatment."

[0004] Unlike traditional tumor markers, which are protein molecules, emerging tumor markers include CTCs, ctDNA, DNA methylation, and ncRNA. A study by CHENG et al. found that CTCs were detectable in 47.89% of patients with early gastric cancer / precancerous lesions and 4.76% of patients with fundic gland polyps, and that CTC detection could differentiate between early gastric cancer / precancerous lesions and fundic gland polyps. However, CTCs have a short half-life (approximately 1–2.4 h), and their levels in the blood are low, making high-specificity and high-sensitivity detection difficult in gastric cancer. ctDNA can be secreted by primary tumor cells, metastatic tumor cells, and CTCs, and released during apoptosis or necrosis. It contains the genetic characteristics of primary and metastatic tumor cells, such as mutations, amplifications, deletions, and translocations. ctDNA analysis is a more sensitive method than traditional tissue biopsy in identifying tumor-specific genetic variations. For example, due to tumor heterogeneity, the detection rate of FGFR2 amplification by ctDNA analysis is higher than that by tissue biopsy. However, as an effective indicator for gastric cancer screening, diagnosis, and prognostic assessment, ctDNA testing still requires larger-scale clinical cohort studies to validate its application value.

[0005] In recent years, miRNAs have shown promising clinical value as biomarkers in liquid biopsy for gastric cancer diagnosis. In 2019, a study by KONG et al. revealed the relationship between serum miR-25 levels and gastric cancer, finding that serum miR-25 levels were significantly upregulated in gastric cancer patients compared to patients with gastritis and healthy controls. The study demonstrated a sensitivity of 69.4% and a specificity of 84.6% in differentiating between gastric cancer patients and healthy controls, with an area under the receiver operating characteristic (AUC) of 0.768. However, miRNAs still have limitations that require optimization. For example, compared to protein molecules, miRNA molecules are less stable and more prone to degradation. Therefore, further optimization of their detection methods is needed to meet clinical needs. Summary of the Invention

[0006] In view of the above-mentioned shortcomings in the prior art, the present invention provides a set of biomarkers for gastric cancer screening and diagnosis and their uses. When used for auxiliary diagnosis and early screening of gastric cancer, it has excellent sensitivity and specificity and is expected to be applied to the diagnosis and treatment of gastric cancer.

[0007] To achieve the above objectives, the technical solution adopted by the present invention to solve its technical problem is as follows:

[0008] A biomarker set for gastric cancer screening and diagnosis includes at least four of the polypeptides shown in SEQ ID NO. 1-36, with specific sequences shown in Table 1.

[0009] Table 1. Peptide Sequences

[0010]

[0011] Among them, the second amino acid K of peptide 2 is modified with Acetyl, the fourth amino acid G of peptide 6 is modified with Phospho, the first amino acid Q of peptide 8 is modified with Gln->pyro-Glu, the sixth amino acid N is modified with Dehydrated, and the fourth amino acid G of peptide 20 is modified with Dehydrated.

[0012] Furthermore, the biomarker set includes the peptides shown in sequences 1 and 2, as well as the following combinations of peptides:

[0013] The polypeptide combination is one of sequence 3 and sequence 4; sequence 5 and sequence 6; sequence 11 and sequence 12; sequence 30 and sequence 35; sequence 3, sequence 4 and sequence 5; sequence 5, sequence 6 and sequence 7; sequence 11, sequence 12 and sequence 20; sequence 9, sequence 30 and sequence 35.

[0014] Furthermore, the biomarker set includes the peptides shown in sequences 7 and 8, as well as the following combinations of peptides:

[0015] The polypeptide combination is one of the following: sequence 3 and sequence 4; sequence 9 and sequence 10; sequence 3, sequence 4 and sequence 9; or sequence 9, sequence 10 and sequence 11.

[0016] Furthermore, the biomarker group includes peptides shown in sequences 1, 10, and 30, as well as peptides shown in sequences 35, 20, 35, and 9, or sequences 20 and 22.

[0017] Furthermore, the biomarker set includes the peptides shown in sequences 5 and 15, as well as the following combinations of peptides:

[0018] The polypeptide combination is one of the following: sequence 25 and sequence 35; sequence 2 and sequence 16; sequence 9, sequence 25 and sequence 35; sequence 2, sequence 16 and sequence 20.

[0019] Furthermore, the biomarker set includes the polypeptides shown in sequences 4, 8, 19, 20 and / or 22.

[0020] Furthermore, the biomarker set includes the polypeptides shown in sequence 7, sequence 10, sequence 17, sequence 18 and / or sequence 22.

[0021] Furthermore, the biomarker set includes the polypeptides shown in sequences 2, 4, 6, and 9.

[0022] Furthermore, the biomarker group includes polypeptides as shown in SEQ ID NO. 1~36.

[0023] The use of the above biomarkers in the preparation of formulations for gastric cancer screening and diagnosis.

[0024] The above biomarkers may be used in basic medical research for non-diagnostic / therapeutic purposes.

[0025] Further, basic medical research includes Western blotting, immunohistochemistry, or flow cytometry.

[0026] The beneficial effects of this invention are:

[0027] The biomarker combination constructed in this invention has superior sensitivity and specificity when used for auxiliary diagnosis and early screening of gastric cancer, and is expected to be applied to the diagnosis and treatment of gastric cancer. Attached Figure Description

[0028] Figure 1 ROC curve for marker combination 1;

[0029] Figure 2 ROC curve for marker combination 2;

[0030] Figure 3 ROC curve for marker combination 3;

[0031] Figure 4 ROC curve for marker combination 4;

[0032] Figure 5 ROC curve for marker combination 5;

[0033] Figure 6 ROC curve for marker combination 6;

[0034] Figure 7 ROC curve for marker combination 7;

[0035] Figure 8 ROC curve for marker combination 8;

[0036] Figure 9 ROC curve for marker combination 9;

[0037] Figure 10 ROC curve for marker combination 10;

[0038] Figure 11 ROC curve for marker combination 11;

[0039] Figure 12 ROC curve for marker combination 12;

[0040] Figure 13 ROC curve for marker combination 13;

[0041] Figure 14 ROC curve for marker combination 14;

[0042] Figure 15 ROC curve for marker combination 15;

[0043] Figure 16 ROC curve for marker combination 16;

[0044] Figure 17 ROC curve for marker combination 17;

[0045] Figure 18 ROC curve for marker combination 18;

[0046] Figure 19 ROC curve for marker combination 19;

[0047] Figure 20 ROC curve for marker combination 20;

[0048] Figure 21 ROC curve for marker combination 21;

[0049] Figure 22 ROC curve for marker combination 22;

[0050] Figure 23 ROC curve for marker combination 23;

[0051] Figure 24 ROC curve for marker combination 24;

[0052] Figure 25 ROC curve for marker combination 25;

[0053] Figure 26 ROC curve for marker combination 26. Detailed Implementation

[0054] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0055] The patient samples used in this invention are all from Zhongshan Hospital affiliated with Fudan University and have passed ethical review.

[0056] The experimental methods used in this invention are as follows:

[0057] I. Serum Sample Collection

[0058] 1) Sample type: serum.

[0059] 2) Collection requirements: Fasting is required. Use a coagulation tube to draw 5 mL of venous blood, let it stand for 30 min, centrifuge at 3000 rpm for 15 min, and take out about 1 mL of serum and put it into a cryopreservation tube.

[0060] 3) Sample storage:

[0061] Use on the same day; store at 2-8℃.

[0062] If not used on the same day, store at -20℃ for up to 30 days;

[0063] If stored for an extended period (more than one month), it should be kept at -80°C.

[0064] The freeze-thaw cycle should not exceed 3 times.

[0065] II. Extraction of analytes from serum

[0066] 1) After calibrating the mass spectrometer, turn on the Solid Bio Fully Automated Sample Analysis System SPS1000 / SPS4000, and put in the consumables, matching reagent kits and the sample to be tested;

[0067] Select the procedure method "Concentrated loading";

[0068] Run the program:

[0069] a. Opening a hole;

[0070] b. Take at least 10µL of serum sample and activation reagent, mix them in a 1:1 ratio, and place them in the G-row pre-reserved well for later use;

[0071] c. Clean the custom pipette tip in cleaning reagent 1 and cleaning reagent 2 in sequence. Each time, aspirate at least 10µL of liquid and repeat the aspiration and dispensing process at least 3 times.

[0072] d. Process the serum mixture in the G-row wells using the cleaned custom pipette tips. Aspirate at least 10 µL of solution each time, repeating the process at least three times.

[0073] e. Clean the custom pipette tip after adsorbing the serum mixture using cleaning reagent 3. During cleaning, aspirate at least 10µL of liquid each time, repeating the aspiration and dispensing process at least 3 times.

[0074] f. Transfer no less than 10µL of buffer reagent into the H-row pre-reserved hole, and place the customized pipette tip after using cleaning reagent 3 into the liquid to draw no less than 10µL of liquid. Repeat the suction and aspiration at least 3 times.

[0075] g. Transfer at least 10µL of sample matrix solution into the H-row pre-reserved well to complete sample processing;

[0076] h. Spot 2.0 µL of the solution from well H onto the hydrophobic-coated biochip (Wuxi Pimo Technology Co., Ltd.).

[0077] i. Vacuum drying for 240 seconds.

[0078] The main components of each reagent are shown in Table 2.

[0079] Table 2 Reagent Composition

[0080]

[0081] III. Mass Spectrometry Data Acquisition and Upload

[0082] The vacuum-dried and hydrophobic coated biochip (Wuxi Pimo Technology Co., Ltd.) was placed into a mass spectrometer;

[0083] Data acquisition is performed using the pre-defined SP1 voltage (target high voltage), SP2 voltage (pulse high voltage), focusing voltage (lens high voltage), detector voltage (MCP voltage), pulse delay time, acquisition card range, target diameter, laser frequency, calibration method, and laser intensity.

[0084] IV. Quality Control

[0085] 1) After data collection, the data will be uploaded to the "Mass Spectrometry Data Analysis Software";

[0086] 2) The software reads the sample information and signal spectrum, and judges whether the sample and sample pretreatment are qualified according to the quality control model; quality control failure may include a variety of possibilities, including the sample is not a gastric lesion sample, the signal spectrum intensity is not up to standard, etc.

[0087] 3) If the quality control fails, adjust the corresponding parameters according to the quality control results and repeat the serum analyte extraction process;

[0088] 4) If the quality control is qualified, proceed to the next process.

[0089] V. Methods for determining Yin and Yang

[0090] The study of positive cutoff values ​​used gastric cancer samples with clear diagnostic information and normal human samples, covering non-gastric cancer populations such as those with chronic gastritis, gastric ulcers, and gastric polyps. The core algorithm is based on supervised learning using known gastric cancer sample atlases. Through a series of processes including smoothing, noise reduction, and baseline removal, characteristic peaks are selected, a classification model is constructed, and the similarity between the hormone signal atlas and known hormone signal atlases stored in the software is calculated (Cannataro M, Guzzi PH, Mazza T, et al. Preprocessing, Management, and Analysis of Mass Spectrometry Proteomics Data[J]. 2005.). Finally, the similarity score positive cutoff value of the kit is determined by maximizing the Youden index. When the similarity score is less than the positive cutoff value, the sample test result is negative; when the similarity score is greater than or equal to the positive cutoff value, the sample test result is positive. When using the maximum similarity score as the positive cutoff value to assist in the diagnosis of gastric cancer or to perform early screening for gastric cancer, the sensitivity is calculated as: Sensitivity = (Number of true positives / (Number of true positives + Number of false negatives)) × 100%, and Specificity is calculated as: (Number of true negatives / (Number of true negatives + Number of false positives)) × 100%.

[0091] Example 1: Screening and Identification of Biomarkers

[0092] This invention analyzed 350 healthy individuals (180 males and 170 females, aged 20-80 years, with a mean age of 50 years; specific age distribution: 51 cases aged 20-30, 72 cases aged 31-40, 68 cases aged 41-50, 66 cases aged 51-60, 64 cases aged 61-70, and 19 cases aged 71-80) and 350 gastric cancer samples (190 males and 160 females, aged 25-85 years, with a mean age of 55 years; specific age distribution: 44 cases aged 25-35, 36 ...8 cases aged 41-50, 66 cases aged 51-60, 64 cases aged 61-70, and 19 cases aged 71-80). (56 cases aged -45 years, 79 cases aged 46-55 years, 88 cases aged 56-65 years, 63 cases aged 66-75 years, and 20 cases aged 76-85 years.) Time-of-flight mass spectrometry (TOF-MS) was performed. Through first-level mass spectrometry testing, the relative abundance differences of characteristic peak data in normal individuals and gastric cancer patients were comprehensively considered, along with statistical differences (p<0.05, t-test). A machine learning algorithm (random forest) was used to rank the influence factors for feature selection and to assess the matching degree of data in the database. This identified 36 blood peptides with diagnostic capabilities for gastric cancer. The mass-to-charge ratio (m / z), relative abundance, and influence factors of these peptides are shown in Table 1. The relative abundance was normalized to ln(mean signal intensity of cancer patient samples / mean signal intensity of normal samples) based on the normal human sample. The feature peak selection in machine learning was based on feature importance assessment using ensemble learning. Multiple decision trees were constructed to quantify the contribution of each mass-to-charge ratio (m / z) peak in classification / prediction. The feature importance score, or influence factor, is obtained by calculating the mean reduction in impurity caused by the feature when splitting across all tree nodes (Biau, G., Scornet, EA random forest guided tour. TEST 25, 197–227 (2016). https: / / doi.org / 10.1007 / s11749-016-0481-7).

[0093] The specific parameters for first-order mass spectrometry are as follows:

[0094] Ionization method: Matrix-assisted laser desorption / ionization (MALDI), with α-cyano-4-hydroxycinnamic acid (CHCA) as the matrix.

[0095] Quality range: 100-4000 Da.

[0096] Resolution: 20000 (full quality range).

[0097] Laser energy: 30-40% (after optimization).

[0098] Acquisition mode: Positive ion mode.

[0099] Calibration: External quality calibration was performed using the Bruker Peptide Calibration Standard.

[0100] Subsequently, the sequences of these 36 substances in the clinical serum were confirmed using secondary mass spectrometry (MS / MS or TOF / TOF is a peptide identification method recommended by the guidelines of the China Food and Drug Administration and the U.S. Food and Drug Administration). The secondary mass spectrometry data analysis process is as follows:

[0101] Data analysis methods: Mascot software (version 2.8) was used for database searching, with the UniProt Human Proteome Database (released in 2023) as the target database. Search parameters: Enzyme was set to "no digestion", parent ion mass error was allowed ±0.5 Da, fragment ion mass error was allowed ±0.3 Da, fixed modification was cysteine ​​urea methylation, and variable modification was methionine oxidation.

[0102] Sequence confirmation criteria: The confirmation of peptide sequences is based on the matching of fragment ion spectra (b- and y- ions) with theoretical spectra. A Mascot score higher than 30 (p<0.05) is considered significant.

[0103] False positive exclusion: Specificity was verified through reverse database search, and the false positive rate was controlled to below 1%.

[0104] Secondary mass spectrometry parameters:

[0105] Collision-induced dissociation (CID).

[0106] Collision energy: 30 eV.

[0107] Fragment ion mass range: 100-3500 Da.

[0108] Data acquisition: Each sample is scanned at least 1000 times with lasers to improve the signal-to-noise ratio.

[0109] The sequences and specificity of these 36 biomarkers were confirmed by secondary mass spectrometry, ruling out false positives. The specific sequences are shown in Table 1.

[0110] Example 2: Validation of Marker Combinations

[0111] Based on the 36 polypeptide biomarkers identified and confirmed in Example 1 (sequences and mass-to-charge ratios are shown in Table 1), different biomarker combinations were formed as shown in Table 3, and the sensitivity and specificity of the biomarker combinations in the auxiliary diagnosis and early screening of gastric cancer were verified. The analysis process is as follows: For all validation cohort samples, the same MALDI-TOF MS platform and parameters as in Example 1 were used for detection to obtain the mass spectrometry peak intensity data of all biomarkers in Table 1. The entire detection process was blinded, meaning that the experimental operators were unaware of the sample grouping information. The raw mass spectrometry data were processed by baseline correction, smoothing, and normalization (based on the internal standard peak intensity), and then the peak area or intensity value of each biomarker was extracted. Statistical analysis was performed using R software (version 4.0.2). The preprocessed biomarker intensity data were input into a logistic regression model. For auxiliary diagnostic validation, ten-fold cross-validation was performed using all samples from this cohort (Sun T, Liu J, Yuan H, Li X, Yan H. Construction of a risk prediction model for lung infection after chemotherapy in lung cancer patients based on the machine learning algorithm. Front Oncol. 2024 Aug 9;14:1403392. doi: 10.3389 / fonc.2024.1403392. PMID:39184040; PMCID: PMC11341396.). For early screening validation, given the imbalanced sample size, oversampling (SMOTE) was used to process the data before model training and testing (van den Goorbergh R, van Smeden M, Timmerman D, Van Calster B. The harm of class imbalance corrections for riskprediction models: illustration and simulation using logistic regression. JAm Med Inform Assoc. 2022 Aug 16;29(9):1525-1534. doi: 10.1093 / jamia / ocac093.PMID: 35686364; PMCID: PMC9382395.).The data analysis process referenced general guidelines for constructing clinical prediction models (Zweig MH, Campbell G. Receiver-operating characteristic (ROC) plots: a fundamental evaluation tool in clinical medicine. Clin Chem. 1993 Apr;39(4):561-77. Erratum in: Clin Chem 1993 Aug;39(8):1589. PMID: 8472349.). Specific calculation results for performance indicators are shown in Table 3, and the corresponding ROC curves are shown below. Figures 1-26 .

[0112] 1) The auxiliary diagnostic efficacy of various biomarker combinations for gastric cancer was validated using 500 gastric cancer patients (264 males (52.8%), 236 females (47.2%), 48 aged 30-39, 103 aged 40-49, 156 aged 50-59, 125 aged 60-69, and 68 aged 70-80) and 500 healthy individuals (253 males (50.6%), 247 females (49.4%), 55 aged 25-34, 107 aged 35-44, 145 aged 45-54, 123 aged 55-64, and 70 aged 65-75).

[0113] 2) The early screening efficacy of various biomarker combinations for gastric cancer was validated using a sample of 350 gastric cancer patients (178 males (50.9%), 172 females (49.1%), 42 aged 35-44, 85 aged 45-54, 108 aged 55-64, 76 aged 65-74, and 39 aged 75-85) and 3500 healthy individuals (1762 males (50.3%), 1738 females (49.7%), 412 aged 20-29, 698 aged 30-39, 823 aged 40-49, 715 aged 50-59, 523 aged 60-69, and 329 aged 70-80).

[0114] Table 3 Sensitivity and specificity of biomarker combinations for assisted diagnosis and early screening

[0115]

[0116] (Continued from the table above)

[0117]

[0118] (Continued from the table above)

[0119]

[0120] According to Table 3 and Figures 1-26 The detection results show that the discriminant peak set (biomarker combination) selected in this invention is superior to current conventional MRI detection for gastric cancer (sensitivity and specificity of 77.42% and 94.12%, respectively) and miRNA-based detection kits (sensitivity and specificity of 69.4% and 84.6%, respectively). Validation results indicate that the discriminant peak set (biomarker combination) selected in this patent is superior to the commonly used biomarker combination CA72-4, CEA, CA125, and CA19-9 for gastric cancer screening (sensitivity and specificity of 66.3% and 80.2%, respectively).

[0121] Finally, it should be noted that the above specific embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to examples, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications and substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A biomarker set for gastric cancer screening and diagnosis, characterized in that, The biomarker group is selected from one of the following combinations of peptides: A) Sequence 2, Sequence 1, Sequence 3 and Sequence 4; B) Sequences 2, 1, 5, and 6; C) Sequence 2, Sequence 1, Sequence 11 and Sequence 12; D) Sequence 2, Sequence 1, Sequence 30, and Sequence 35; E) Sequence 2, Sequence 1, Sequence 3, Sequence 4 and Sequence 5; F) Sequence 2, Sequence 1, Sequence 5, Sequence 6 and Sequence 7; G) Sequence 2, Sequence 1, Sequence 11, Sequence 12 and Sequence 20; H) Sequence 2, Sequence 1, Sequence 9, Sequence 30 and Sequence 35; I) Sequences 2, 16, 5, and 15; J) Sequence 2, Sequence 16, Sequence 20, Sequence 5 and Sequence 15; K) Sequence 2, Sequence 4, Sequence 6 and Sequence 9; L) Sequences 1 to 36; The amino acid sequences of sequences 1 to 36 are shown in SEQ ID NO. 1 to 36; Specifically, amino acid K at the second position of sequence 2 is modified with Acetyl, amino acid G at the fourth position of sequence 6 is modified with Phospho, amino acid Q at the first position of sequence 8 is modified with Gln->pyro-Glu, amino acid N at the sixth position is modified with Dehydrated, and amino acid G at the fourth position of sequence 20 is modified with Dehydrated.

2. Use of the reagent for detecting the biomarker group of claim 1 in the preparation of preparations for gastric cancer screening and diagnosis.

Images