Extracellular vesicle protein marker combination for colorectal cancer diagnosis and application of extracellular vesicle protein marker combination

Abstract

The invention discloses an extracellular vesicle protein marker combination for colorectal cancer diagnosis and application of the extracellular vesicle protein marker combination, and belongs to the technical field of medical molecular diagnosis. The marker for colorectal cancer diagnosis is a combination of CTTN, HNRNPK and PSMC6 protein, is derived from tissue extracellular vesicles or plasma extracellular vesicles, and has relatively high correlation with colorectal cancer diagnosis. The biomarker provided by the invention can assist the existing colorectal cancer diagnosis means, and can accurately identify colorectal cancer patients with normal CEA and CA19-9 levels so as to improve the detection rate of the colorectal cancer patients.

Description

Technical Field

[0001] The present invention belongs to the technical field of medical molecular diagnosis, and specifically relates to an extracellular vesicle protein marker combination for colorectal cancer diagnosis and application thereof. Background Art

[0002] Colorectal cancer (CRC) is the third most common malignant tumor in the world and one of the main causes of cancer-related death. In my country, the incidence and mortality of CRC are rising year by year, with 550,000 new cases each year, accounting for 12.2% of all malignant tumor cases, and 280,000 deaths, accounting for 9.5% of all deaths from malignant tumors. In recent years, many advances have been made in the treatment of CRC, but due to the lack of typical symptoms and early diagnosis methods in the early stage, most patients are diagnosed in the middle and late stages with poor prognosis; the 5-year survival rate of CRC patients with distant metastasis is less than 20%. Therefore, promoting disease prevention screening and early diagnosis is crucial for the prevention and treatment of CRC. Colonoscopy is the "gold standard" for detecting CRC, but its high invasiveness and potential surgical risks lead to low patient compliance; tumor markers in traditional physical examinations, such as carcinoembryonic antigen (CEA) and carbohydrate antigen 19-9 (CA19-9), have poor sensitivity and specificity and limited clinical value. Therefore, finding and identifying non-invasive and specific diagnostic markers, developing new therapeutic targets, and revealing the molecular regulatory mechanisms of CRC occurrence and development are scientific issues that need to be urgently addressed in this research field.

[0003] Liquid biopsy is a new type of non-invasive detection technology that collects human body fluids as test samples to reflect the molecular characteristics and dynamic changes of tumors. The results of liquid biopsy are more consistent with the characteristics of tumor heterogeneity, making up for the shortcomings of imaging examinations and pathological biopsies in obtaining comprehensive biological information of tumors. At present, liquid biopsies mainly collect and detect tumor-related information such as circulating tumor cells (CTCs), circulating tumor DNA (ctDNA), and extracellular vesicles (EVs) released by tumor cells in body fluids, which can reflect the dynamic progression of tumors. Among them, extracellular vesicles have the advantages of large amount of information, easy enrichment, strong stability, and high detection rate, making them a "dark horse" in tumor liquid biopsy. Extracellular vesicles are lipid bilayer membrane vesicles with a diameter of about 30-150nm secreted by different types of cells, and are widely present in various body fluids. There is increasing evidence that cancer-derived extracellular vesicles play a key role in tumor cell-to-cell communication by transferring and exchanging oncogenic molecules (including proteins, noncoding RNA, and lipids), thereby participating in promoting tumorigenesis, tumor proliferation, metastasis, angiogenesis, and immune escape. Many studies have confirmed the functions of cancer-derived extracellular vesicle proteins, which transmit cell signals to the tumor microenvironment to synchronize phenotypic patterns between different cells. When the body is diseased, the content of proteins contained in extracellular vesicles will change accordingly as the disease progresses. Extracellular vesicles in body fluids carrying cancer-related proteins are expected to become new markers for early diagnosis of cancer. Summary of the invention

[0004] Based on the above background, the present invention provides a novel extracellular vesicle marker that can be used for CRC screening and diagnosis. The marker is obtained by proteomics screening. After verification by a clinical sample set, the combination of CTTN, HNRNPK and PSMC6 proteins is identified as a CRC diagnostic biomarker and is used to construct a prediction model.

[0005] The proteomics screening process of the present invention comprises the following steps:

[0006] (1) The inventors analyzed tissue extracellular vesicles and plasma extracellular vesicles samples from 40 CRC patients and 20 healthy controls using data-independent acquisition mass spectrometry (DIA-MS) technology and discovered candidate marker protein combinations;

[0007] (2) The candidate biomarkers selected in (1) were validated in 40 independent plasma extracellular vesicle samples (20 CRC patients and 20 healthy controls) using parallel reaction monitoring (PRM) technology, and finally 6 proteins were screened for validation (CTTN, HNRNPK, PSMC6, NAP1L1, EIF3B, MYH9).

[0008] (3) In an independent cohort of 490 cases (251 CRC patients, 130 benign intestinal disease controls, and 109 healthy controls), the candidate markers in (2) were validated by ELISA, and three proteins (CTTN, HNRNPK, PSMC6, hereinafter referred to as ColonTrack) were finally determined as CRC diagnostic biomarkers and their diagnostic efficacy. Among them, 344 samples were used as the modeling cohort (176 CRC patients, 91 benign intestinal disease controls, and 77 healthy controls), and 146 samples were used as the Zhongshan Hospital sample validation set (Internal validation) (75 CRC patients, 39 benign intestinal disease controls, and 32 healthy controls).

[0009] (4) ColonTrack was validated in an external validation cohort of 274 samples (142 CRC patients, 66 benign intestinal disease controls, and 66 healthy controls) to determine the efficacy of ColonTrack as a diagnostic biomarker for CRC.

[0010] Based on this, the present invention provides the following technical solutions:

[0011] In a first aspect, the present invention provides a use of a biomarker or a reagent for detecting a biomarker in at least one of the following:

[0012] a1) Use in the preparation and / or screening of products for screening colorectal cancer;

[0013] a2) Use in the preparation and / or screening of products for diagnosis or auxiliary diagnosis of colorectal cancer;

[0014] a3) Application in the preparation and / or screening of products for distinguishing or assisting in distinguishing colorectal cancer from benign intestinal diseases.

[0015] Preferably, the colorectal cancer is early colorectal cancer, and the early colorectal cancer refers to colorectal cancer in stage I according to the AJCC stage.

[0016] Preferably, the benign intestinal diseases refer to all benign non-malignant intestinal diseases, including but not limited to Crohn's disease, ulcerative colitis, colorectal adenoma or polyp.

[0017] The biomarkers are a combination of proteins or genes encoding them as shown in b1)-b3):

[0018] b1) CTTN protein or its encoding gene;

[0019] b2) HNRNPK protein or its encoding gene;

[0020] b3) PSMC6 protein or its encoding gene.

[0021] The biomarkers are derived from extracellular vesicles of tumor tissue and / or extracellular vesicles of plasma.

[0022] The products include but are not limited to reagents, test kits, chips, test strips, membrane strips or detection platforms.

[0023] The reagents for detecting biomarkers include any reagents required for detecting the expression levels of the proteins shown in b1) to b3) or their encoding genes by RT-PCR, RT-qPCR, biochip detection, Southern blotting, in situ hybridization, flow cytometry, enzyme-linked immunosorbent assay, immunoblotting, immunohistochemistry, immunofluorescence or fluorescent dye method.

[0024] In some embodiments of the present invention, the reagent for detecting a biomarker is selected from at least one of the following:

[0025] c1) Antibodies specific for proteins shown in b1)-b3);

[0026] c2) primers for specifically amplifying genes encoding proteins shown in b1) to b3);

[0027] c3) A probe that specifically recognizes a gene encoding a protein shown in b1) to b3) or its transcript.

[0028] Preferably, the reagents for detecting biomarkers include any reagents required for detecting the expression levels of the proteins shown in b1) to b3) by flow cytometry, enzyme-linked immunosorbent assay, immunoblotting or immunohistochemistry.

[0029] In some embodiments of the present invention, the reagent for detecting biomarkers is selected from antibodies specific to proteins shown in b1)-b3), and the antibodies can be coupled to one of acridinium ester, HRP, ALP or FITC.

[0030] Most preferably, the reagents for detecting biomarkers include any reagents required for detecting the expression levels of the proteins shown in b1) to b3) by enzyme-linked immunosorbent assay.

[0031] In a second aspect, the present invention provides a kit, comprising the reagent for detecting a biomarker according to the first aspect of the present invention.

[0032] Furthermore, the kit also includes other reagents involved in RT-PCR, RT-qPCR, biochip detection, Southern blotting, in situ hybridization, enzyme-linked immunosorbent assay, immunoblotting, immunohistochemistry, immunofluorescence or fluorescent dye detection systems.

[0033] Furthermore, the kit also includes reagents for extracting extracellular vesicles from tissues and / or plasma.

[0034] In a specific embodiment of the present invention, PEG chemical precipitation method is preferably used to extract tissue and / or plasma extracellular vesicles; the morphology of the extracellular vesicles is then observed by transmission electron microscopy; the size and concentration of plasma extracellular vesicles are analyzed by nanoparticle tracking analyzer; and the expression of plasma extracellular vesicle surface markers (TSG101, CD81, CD9, Calnexin) is analyzed by protein blotting.

[0035] In a third aspect, the present invention provides a model for colorectal cancer screening and / or diagnosis and / or differentiation of colorectal cancer and benign intestinal diseases, wherein the model comprises an information acquisition module and a diagnosis module.

[0036] The information acquisition module is used to execute the step of acquiring the biomarker expression level information of the subject, wherein the biomarker is selected from a combination of CTTN protein, HNRNPK protein and PSMC6 protein, and the biomarker expression level information is derived from the extracellular vesicles of the tumor tissue and / or the plasma extracellular vesicles of the subject.

[0037] The diagnostic module is used to execute the step of judging the health status of the subject according to the biomarker expression level information.

[0038] Preferably, the colorectal cancer is early colorectal cancer, and the early colorectal cancer refers to colorectal cancer in stage I according to the AJCC stage.

[0039] Preferably, the benign intestinal diseases refer to all benign non-malignant intestinal diseases, including but not limited to Crohn's disease, ulcerative colitis, colorectal adenoma or polyp.

[0040] Specifically, the model diagnoses whether a subject has colorectal cancer by detecting the expression level of biomarker proteins, and the diagnosis includes that compared with healthy people, the expression level of biomarker proteins in the tissues and / or plasma extracellular vesicles of colorectal cancer patients is significantly changed; the significant change means that compared with healthy people, the expression level of biomarker proteins in the tissues and / or plasma extracellular vesicles of colorectal cancer patients has a statistically significant difference, that is, the P value is <0.05.

[0041] Those skilled in the art can obtain the above-mentioned biomarker protein expression level information in a large number of samples, and perform modeling according to conventional machine learning methods to obtain the model described in the present invention, wherein the model includes but is not limited to random forest, kNN or SVM.

[0042] In a fourth aspect, the present invention provides an application of the model described in the third aspect of the present invention in preparing a product for colorectal cancer screening and / or diagnosis and / or differentiation of colorectal cancer and benign intestinal diseases, the application comprising the following steps:

[0043] S1: quantitatively detecting the expression level of a biomarker of a batch of samples, wherein the biomarker is selected from a combination of CTTN protein, HNRNPK protein and PSMC6 protein;

[0044] S2: Perform data analysis and machine learning algorithms on biomarker expression level data;

[0045] S3: Quantitatively detect the expression level of the biomarker in the subject's sample, and determine whether the subject has colorectal cancer based on the probability score in the machine learning algorithm.

[0046] Preferably, the colorectal cancer is early colorectal cancer, and the early colorectal cancer refers to colorectal cancer in stage I according to the AJCC stage.

[0047] Preferably, the benign intestinal diseases refer to all benign non-malignant intestinal diseases, including but not limited to Crohn's disease, ulcerative colitis, colorectal adenoma or polyp.

[0048] The judgment method is: if the probability score of the sample to be tested is between 0.5 and 1, it is judged that the patient has colorectal cancer; if the probability score of the sample to be tested is between 0 and 0.5, it is judged that the patient does not have colorectal cancer.

[0049] The batch samples described in step S1 include a training set and a test set.

[0050] In a specific embodiment of the present invention, the data analysis described in step S2 includes using R (version 4.2.2) to compare the significant differences in protein content between different groups: using SPSS25.0 and GraphPad Prism 10 to perform significant difference analysis on the relative content of all proteins in different groups of diseases; screening the characteristic variables through the mrmr-selection package (version 0.2.6), and finally obtaining the biomarker protein combination.

[0051] The machine learning algorithm described in the present invention is a conventional technical means in the field, including: inputting training set sample data, training model, evaluating model, inputting test set sample data for blind testing, and evaluating test set blind testing results.

[0052] The technical solution provided by the present invention has the following technical advantages:

[0053] 1) The biomarker protein provided by the present invention is derived from the extracellular vesicles in the plasma of the subject. The extracellular vesicle protein is wrapped by a lipid bilayer and has better stability than free protein. The detection can be completed by collecting blood samples, and the patient compliance is good.

[0054] 2) The biomarkers provided by the present invention are significantly better than the digestive tract tumor markers CEA and CA19-9 in the diagnosis of colorectal cancer, especially the diagnosis of early colorectal cancer, and the differentiation of colorectal cancer and benign intestinal diseases.

[0055] 3) The kit for screening and / or diagnosing colorectal cancer prepared according to the biomarkers provided by the present invention is suitable for all people to be examined, and the detection is convenient and the operation is simple; the detection indicators have a high correlation with the early diagnosis of colorectal cancer, and the detection results are accurate; most of the reagents in the kit are relatively common, and the detection cost is low. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 Flowchart for proteomics-based screening and confirmation of biomarker proteins.

[0057] Figure 2 Flowchart for proteomics-based validation of protein biomarkers for diagnostic efficacy.

[0058] Figure 3 The results of extracellular vesicle identification are shown in Figure 1. A, transmission electron microscopy results; B, particle size distribution; C, protein blotting results.

[0059] Figure 4 This is a graph showing the results of screening CRC diagnostic markers based on proteomics technology provided by the present invention.

[0060] Figure 5This is a graph showing the expression levels of the diagnostic markers provided by the present invention in an independent large cohort of samples and the results of constructing a diagnostic model.

[0061] Figure 6 This is a result diagram of the diagnostic model provided by the present invention for identification and early diagnosis efficacy of CRC. DETAILED DESCRIPTION

[0062] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only some embodiments of the present invention, not all. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0063] The experimental methods involved in the examples of the present invention are conventional methods unless otherwise specified, and are performed according to the techniques or conditions described in the literature in the field or according to the product instructions. The materials and reagents used in the examples of the present invention can be purchased from commercial channels unless otherwise specified.

[0064] Data analysis methods

[0065] Each experiment involved in the embodiments of the present invention was repeated at least 3 times. The data are presented in the form of mean ± the standard error (SE). When comparing two groups, if the sample satisfies the normal distribution and the homogeneity of variance, Ttest is used; if the sample satisfies the normal distribution but does not satisfy the homogeneity of variance, Welch t'test is used; if the sample does not satisfy the normal distribution, the non-parametric test Wilcoxon rank sum test is used. When multiple groups are compared, if the sample satisfies the normal distribution and the homogeneity of variance, One-way ANOVA is used; if the sample satisfies the normal distribution but does not satisfy the homogeneity of variance, the Welch one-way ANOVA test is used; if the sample does not satisfy the normal distribution, the non-parametric test Kruskal-Wallis test is used. When p < 0.05, the difference is statistically significant.

[0066] Study subjects

[0067] The subjects included in the clinical study included patients with CRC, patients with benign intestinal diseases and healthy controls. The clinical study was approved by the hospital ethics committee and agreed to by all subjects, who signed a written informed consent.

[0068] Extraction and identification of EVs

[0069] Extraction of extracellular vesicles from tissues:

[0070] To isolate EVs from CRC and tumor-adjacent tissues, the collected tissues were first cut into 300 μm thick sections using a Leica CM1860 cryostat and then incubated at 37°C for 15 min in 1640 medium containing a papain dissociation system (Worthington Biochemical, USA). The tissue digest was filtered through a 70 μm filter and transferred to a new centrifuge tube containing protease and phosphatase inhibitors. The samples were centrifuged at 300 × g (10 min) and 2,000 × g (10 min) at 4°C to remove cells and tissue debris. The supernatant was then centrifuged at 12,000 × g for 30 min at 4°C and filtered using a 0.22 μm filter (Millipore, USA). The supernatant was diluted and transferred to an ultracentrifuge tube (Thermo Fisher Scientific, USA) and centrifuged at 118,000 × g for 120 min at 4°C using an F37L rotor. The pellet was resuspended in 1 mL PBS and then purified using an Exosupur column (Echobiotech, China) and ultrafiltration. The EVs samples were finally stored at -80°C for future use to minimize freeze-thaw cycles.

[0071] Extraction of plasma extracellular vesicles:

[0072] 3-5 mL of peripheral blood was collected from CRC patients and healthy controls in the hospital, and the supernatant (plasma) was collected by centrifugation at 2,500 prm for 10 minutes at room temperature. The clear supernatant collected after centrifugation at 10,000 × g for 1 hour at 4°C was the plasma with large vesicles removed. The above plasma was incubated with PEG6000 at 4°C for 3 hours, and then centrifuged at 10,000 × g for 1 hour at 4°C to obtain extracellular vesicle precipitation, which was resuspended in 100 μL phosphate buffer to obtain an extracellular vesicle suspension.

[0073] Transmission electron microscopy and immunoblotting were used to characterize the extracellular vesicles. Figure 3 As shown in Figure 2, the collected extracellular vesicles are clearly visible, with a concave surface and a disc-shaped or saucer-like vesicle structure ( Figure 3 A), with diameters ranging from 100 to 200 nm ( Figure 3 B), extracellular vesicle marker proteins TSG101, CD81, CD9, and Calnexin were identified, which met the extracellular vesicle identification criteria ( Figure 3 C). The above results show that we have successfully isolated extracellular vesicles (Tissue in the figure represents tissue, T_EV represents extracellular vesicles from tumor tissue, and P_EV represents extracellular vesicles from normal adjacent tissue).

[0074] Screening of biomarker proteins

[0075] The inventors analyzed tissue extracellular vesicles and plasma extracellular vesicle samples from 40 CRC patients and 20 healthy controls by data independent acquisition mass spectrometry (DIA-MS) technology; the selected candidate biomarkers were verified in 40 independent plasma extracellular vesicle samples (20 CRC patients and 20 healthy controls) by parallel reaction monitoring (PRM) technology, and finally 6 proteins were screened for verification (P6 Panel: MYH9, CTTN, EIF3B, PSMC6, NAP1L1, HNRNPK), see for details. Figure 4 As shown (in the figure, CRC represents the plasma of CRC patients, HC represents the plasma of healthy controls, PT represents the primary tumor tissue of CRC, and NM represents the adjacent tumor tissue of CRC).

[0076] Then, if Figure 5 As shown in the figure (CRC represents plasma of CRC patients, BD represents plasma of patients with benign intestinal diseases, and HC represents plasma of healthy controls), the inventors verified the candidate markers by enzyme-linked immunosorbent assay in 344 samples (176 CRC patients, 91 benign intestinal disease controls, and 77 healthy controls): 96-well plates were coated with anti-CTTN, HNRNPK, PSMC6, NAP1L1, EIF3B, and MYH9 antibodies at 37°C for 4 hours, standard wells, test sample wells, and blank wells were set respectively, 7 standard wells were set, 100 μL of standards of different concentrations were added in sequence, 100 μL of standard diluent was added to the blank wells, and 100 μL of test samples were added to the remaining wells. The ELISA plate is coated and incubated at 37°C for 1 hour; the liquid is discarded, and the plate is spun dry without washing; then the HRP-labeled monoclonal antibody against the above protein is incubated with the extracellular vesicle suspension at room temperature and incubated at 37°C for 1 hour; the liquid is discarded, and then the plate is washed with a freshly prepared PBS eluent containing 5% Tween-20, 350 μL per well each time, and soaked for 1-2 minutes; 90 μL of TMB substrate is added to each well, the ELISA plate is coated, and the plate is color-developed at 37°C in the dark, and the reaction time is controlled to be 10-20 minutes, not exceeding 30 minutes; when there is an obvious gradient blue in the first 3 to 4 wells of the standard wells and the gradient is not obvious in the last 3 to 4 wells, 50 mL of 2M sulfuric acid stop solution can be added to each well to stop the color development reaction, and then the optical density value (OD value) of each well is measured at a wavelength of 450 nm using an ELISA reader.

[0077] Results: The combination of three proteins (CTTN, HNRNPK, and PSMC6) in extracellular vesicles was identified as a precise diagnostic model for CRC, and this model was named ColonTrack.

[0078] Efficacy analysis of validation set samples in CRC diagnosis

[0079] The inventors validated the ColonTrack diagnostic model using internal and external independent cohort samples and determined that the model has high efficacy in the accurate diagnosis of CRC. Figure 6 The results of model validation are shown (Internal validation represents the Zhongshan sample validation set, External validation represents the external independent cohort sample validation set, Specificity represents specificity, Sensitivity represents sensitivity, AUC represents the area under the subject curve, CI represents confidence interval, and CRA represents intestinal adenoma). The specific results are as follows:

[0080] (1) In the Zhongshan Hospital sample validation set (Internal validation): ColonTrack had an area under the ROC curve (AUC) of 0.986 (95% CI = 0.973-0.999) for differentiating CRC patients from non-CRC patients (including patients with benign intestinal diseases and healthy controls) in the cohort. The sensitivity and specificity for CRC diagnosis were 96.0% and 91.5%, respectively, which was superior to the auxiliary diagnostic efficacy of gastrointestinal tumor markers CEA [0.622 (95% CI = 0.531-0.712)] and CA19-9 [0.695 (95% CI = 0.610-0.780)] for CRC. The model accuracy was 93.8% ( Figure 6 A left). When distinguishing benign intestinal diseases (BD) from CRC patients, the area under the ROC curve (AUC) of ColonTrack for detecting CRC was 0.988 (95% CI = 0.974-1.000), and the sensitivity and specificity for CRC diagnosis were 96.0% and 92.3%, respectively, which was superior to the auxiliary diagnosis of CRC by digestive tract tumor markers CEA [0.576 (95% CI = 0.464-0.688)] and CA19-9 [0.686 (95% CI = 0.585-0.788)], and the model accuracy reached 94.7% ( Figure 6 A right).

[0081] (2) In the external validation set, ColonTrack had an area under the ROC curve (AUC) of 0.974 (95% CI = 0.953-0.995) for differentiating CRC patients from non-CRC patients (including patients with benign intestinal diseases and healthy controls) in the cohort. The sensitivity and specificity for CRC diagnosis were 92.3% and 94.7%, respectively, which was superior to the auxiliary diagnostic efficacy of gastrointestinal tumor markers CEA [0.555 (95% CI = 0.474-0.637)] and CA19-9 [0.545 (95% CI = 0.465-0.626)] for CRC. The model accuracy was 93.4% ( Figure 6 B left). When distinguishing between patients with benign intestinal diseases and CRC, the area under the ROC curve (AUC) of ColonTrack for detecting CRC was 0.977 (95% CI = 0.957-0.998), and the sensitivity and specificity for CRC diagnosis were 92.3% and 97.0%, respectively, which was superior to the auxiliary diagnosis of CRC by digestive tract tumor markers CEA [0.512 (95% CI = 0.421-0.603)] and CA19-9 [0.516 (95% CI = 0.422-0.609)], and the model accuracy reached 93.8% ( Figure 6 B right).

[0082] (3) In the Zhongshan Hospital sample validation set (Internal validation): ColonTrack had an area under the ROC curve (AUC) of 1.000 (95% CI = 1.000-1.000) for differentiating CRA from early CRC patients (stage I) for detecting CRC. The sensitivity and specificity for CRC diagnosis reached 100% and 88.9%, respectively, which was superior to the auxiliary diagnostic efficacy of gastrointestinal tumor markers CEA [0.635 (95% CI = 0.449-0.821)] and CA19-9 [0.606 (95% CI = 0.356-0.855)] for CRC. The model accuracy reached 91.2% ( Figure 6 C).

[0083] (4) In the external validation set, ColonTrack had an area under the ROC curve (AUC) of 0.969 (95% CI = 0.913-1.000) for differentiating CRA from early CRC patients (stage I). The sensitivity and specificity for CRC diagnosis were 93.1% and 100%, respectively, which was superior to the auxiliary diagnostic efficacy of gastrointestinal tumor markers CEA [0.588 (95% CI = 0.419-0.756)] and CA19-9 [0.547 (95% CI = 0.392-0.702)] for CRC. The model accuracy was 97.3% ( Figure 6 D).

[0084] The present invention firstly uses data-independent acquisition mass spectrometry (DIA-MS) proteomics quantitative technology to screen candidate biomarkers of extracellular vesicles of colorectal cancer cells; then the peptide samples obtained by enzymatic hydrolysis are subjected to parallel reaction monitoring (PRM) targeted quantitative analysis to determine 6 colorectal cancer-related proteins; the related proteins are quantitatively analyzed by ELISA to verify that 3 colorectal cancer characteristic proteins are obtained as biomarkers. The 3 colorectal cancer characteristic proteins provided by the present invention can be used for the screening and diagnosis of colorectal cancer, as well as the differentiation of colorectal cancer from benign intestinal diseases.

[0085] The three colorectal cancer characteristic proteins provided by the present invention show good diagnostic efficacy when used as biomarkers for the diagnosis of colorectal cancer, and their diagnostic efficacy is significantly better than that of digestive tract tumor markers. In addition, the biomarkers described in the present invention can be used to distinguish colorectal cancer from early colorectal cancer, which is beneficial to the early screening and diagnosis of colorectal cancer.

[0086] A person skilled in the art may also use any conventional protein detection method to detect the expression levels of the three proteins in the extracellular vesicles of the sample, and perform machine learning modeling according to conventional statistical methods, and use the constructed machine learning model to judge the health status of the subject.

[0087] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments may still be modified, or some or all of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. Use of a biomarker or a reagent for detecting a biomarker in at least one of the following: a1) Use in the preparation and / or screening of products for screening colorectal cancer; a2) Use in the preparation and / or screening of products for diagnosis or auxiliary diagnosis of colorectal cancer; a3) Use in the preparation and / or screening of products for distinguishing or assisting in distinguishing colorectal cancer from benign intestinal diseases; The biomarkers are a combination of proteins or genes encoding them as shown in b1)-b3): b1) CTTN protein or its encoding gene; b2) HNRNPK protein or its encoding gene; b3) PSMC6 protein or its encoding gene.

2. The use according to claim 1, characterized in that: The colorectal cancer is early colorectal cancer, which refers to colorectal cancer in AJCC stage I; the benign intestinal diseases are benign intestinal diseases other than malignant tumors, including Crohn's disease, ulcerative colitis, colorectal adenoma or polyp.

3. The use according to claim 1, characterized in that: The biomarkers are derived from extracellular vesicles of tumor tissue and / or extracellular vesicles of plasma.

4. The use according to claim 1, characterized in that: The reagents for detecting biomarkers include any reagents required for detecting the expression levels of the proteins shown in b1) to b3) or their encoding genes by RT-PCR, RT-qPCR, biochip detection, Southern blotting, in situ hybridization, flow cytometry, enzyme-linked immunosorbent assay, immunoblotting, immunohistochemistry, immunofluorescence or fluorescent dye method.

5. The use according to claim 4, characterized in that: The reagent for detecting biomarkers is selected from at least one of the following: c1) Antibodies specific for proteins shown in b1)-b3); c2) primers for specifically amplifying genes encoding proteins shown in b1) to b3); c3) A probe that specifically recognizes a gene encoding a protein shown in b1) to b3) or its transcript.

6. A kit comprising the reagent for detecting a biomarker according to any one of claims 1 to 5.

7. The kit according to claim 6, characterized in that The kit also includes other reagents involved in RT-PCR, RT-qPCR, biochip detection, Southern blotting, in situ hybridization, enzyme-linked immunosorbent assay, immunoblotting, immunohistochemistry, immunofluorescence or fluorescent dye detection systems.

8. The kit according to claim 7, characterized in that The kit also includes reagents for extracting extracellular vesicles from tissue and / or plasma.

9. A model for colorectal cancer screening and / or diagnosis and / or differentiation of colorectal cancer from benign intestinal diseases, the model comprising an information acquisition module and a diagnosis module; The information acquisition module is used to execute the step of acquiring biomarker expression level information of the subject, wherein the biomarker is selected from a combination of CTTN protein, HNRNPK protein and PSMC6 protein, and the biomarker expression level information is derived from extracellular vesicles of tumor tissue and / or plasma extracellular vesicles of the subject; The diagnostic module is used to execute the step of judging the health status of the subject according to the biomarker expression level information.

10. Use of the model according to claim 9 in the preparation of a product for colorectal cancer screening and / or diagnosis and / or differentiation of colorectal cancer and benign intestinal diseases, the use comprising the following steps: S1: quantitatively detecting the expression level of a biomarker of a batch of samples, wherein the biomarker is selected from a combination of CTTN protein, HNRNPK protein and PSMC6 protein; S2: Perform data analysis and machine learning algorithms on biomarker expression level data; S3: Quantitatively detect the expression level of the biomarker in the subject's sample, and determine whether the subject has colorectal cancer based on the probability score in the machine learning algorithm.

Images