A gastric cancer diagnostic marker based on extracellular vesicle miRNA and its application

Abstract

The present invention provides a diagnostic or prognostic marker for gastric cancer, a gastric cancer diagnostic kit based thereon, and its use. The gastric cancer diagnostic or prognostic marker of the present invention involves biologically significant miRNA pairs found in extracellular vesicles of serum or plasma. A gastric cancer diagnostic model constructed using the expression ratio of the miRNA pairs as a model feature has high diagnostic sensitivity and specificity for gastric cancer and has significant potential for screening for early gastric cancer, thus having important practical significance for the early diagnosis or prognosis of gastric cancer.

Description

Technical Field

[0001] The present invention belongs to the field of medical detection technology, and specifically relates to a gastric cancer diagnostic marker based on extracellular vesicle miRNA and its application. Background Art

[0002] Gastric cancer is an epithelial malignancy and one of the most common tumors of the digestive system. Gastric cancer has long ranked among the top three cancers in my country, and it is a cancer with a high incidence and poor prognosis. According to the International Agency for Research on Cancer's Global Cancer Statistics Report, nearly half of all new cases and deaths from gastric cancer worldwide in 2020 occurred in China. Despite significant advances in surgical techniques and treatment options for gastric cancer, the five-year overall survival rate remains only 35.1%. This is because most patients with early-stage gastric cancer do not exhibit specific symptoms, making accurate diagnosis and timely treatment difficult. Compared to the cumbersome and challenging treatment of advanced gastric cancer, most early-stage gastric cancers can be treated endoscopically, avoiding laparotomy and gastrectomy, thereby better preserving the functional integrity of the stomach. Advancing the treatment window for gastric cancer patients can improve prognosis and survival. Therefore, it is imperative to improve the difficulty in detecting early-stage gastric cancer, increase its diagnosis rate, and explore more objective, accurate, and convenient diagnostic indicators and methods.

[0003] Traditional gastric cancer diagnostic techniques include gastroscopy, CT, and serum tumor marker detection. Generally speaking, gastroscopy is currently the most accurate method for detecting gastric cancer lesions in the early stages. However, biopsy under gastroscopy requires obtaining the corresponding tissue from the patient's body, which causes certain trauma and can easily cause the patient to feel pain during the examination. In addition, gastroscopy is difficult and costly to operate, and is easily limited by factors such as the patient's physical condition, making it inefficient. CT can generally only detect tumors after they have grown to a certain extent, and therefore cannot be used for early screening of gastric cancer. Currently, common serum tumor markers for gastric cancer in clinical practice include pepsinogen and carcinoembryonic antigen. However, these markers have poor diagnostic specificity and sensitivity for gastric cancer, cannot accurately identify non-gastric cancer samples, and have limited value in screening for early gastric cancer. Therefore, there is an urgent need in this field to develop an early diagnostic marker for gastric cancer with excellent diagnostic performance.

[0004] Extracellular vesicles (EVs) are small sacs secreted by cells and released into body fluids or the extracellular environment. Measuring only 30-100 nanometers in diameter, they play a crucial role in intercellular communication. Protected by a lipid bilayer, EVs can exist stably in various biological fluids and cell culture media; this property makes their contents reliable tumor markers, and miRNAs are one type of EV content. EV miRNAs are reported to be expressed at varying levels in different diseases and physiological conditions and to play a crucial regulatory role in the progression of malignant tumors. Therefore, EV miRNAs may become a novel biomarker for early cancer diagnosis and prognosis tracking. Summary of the Invention

[0005] Purpose of the Invention

[0006] In response to the problems existing in the above-mentioned prior art methods, the present invention aims to provide a gastric cancer diagnostic or prognostic marker that can identify early gastric cancer and has high sensitivity and specificity for gastric cancer diagnosis, a kit based on the marker, or its application, including: application in constructing a model for gastric cancer diagnosis, efficacy or prognosis evaluation, or application in the preparation of a drug or kit for gastric cancer diagnosis, efficacy or prognosis evaluation.

[0007] Solution

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] In a first aspect, the present invention provides a diagnostic or prognostic marker for gastric cancer, wherein the diagnostic or prognostic marker for gastric cancer comprises a miRNA pair consisting of any two of the following miRNAs, or any combination of the miRNA pairs:

[0010] hsa-miR-16, hsa-miR-10a, hsa-miR-128, hsa-miR-146a, hsa-miR-191, hsa-miR-215, hsa-miR-21, hsa-miR-221, hsa-miR-222, hsa-miR-26a, h sa-miR-26b, hsa-miR-19b, hsa-miR-363, hsa-miR-106b, hsa-miR-186, hsa-miR-629, hsa-miR-339, hsa-miR-103a, hsa-miR-140, hsa-miR-181.

[0011] In a preferred embodiment, the diagnostic or prognostic marker for gastric cancer comprises a miRNA pair selected from the following:

[0012] hsa-miR-103a / hsa-miR-140, hsa-miR-16 / hsa-miR-10a, hsa-miR-16 / hsa-miR-128, hsa-miR- 16 / hsa-miR-146a, hsa-miR-16 / hsa-miR-191, hsa-miR-16 / hsa-miR-215, hsa-miR-16 / hsa-mi R-21, hsa-miR-16 / hsa-miR-221, hsa-miR-16 / hsa-miR-222, hsa-miR-16 / hsa-miR-26a, hsa-m iR-16 / hsa-miR-26b, hsa-miR-19b / hsa-miR-128, hsa-miR-26b / hsa-miR-10a, hsa-miR-26b / h sa-miR-21, hsa-miR-363 / hsa-miR-106b, hsa-miR-363 / hsa-miR-128, hsa-miR-363 / hsa-miR- 222, hsa-miR-215 / hsa-miR-26b, hsa-miR-221 / hsa-miR-363, hsa-miR-186 / hsa-miR-221, hsa -miR-629 / hsa-miR-363, hsa-miR-629 / hsa-miR-339, hsa-miR-191 / hsa-miR-21, hsa-miR-191 / hsa-miR-222, hsa-miR-21 / hsa-miR-181a, hsa-miR-26b / hsa-miR-140, and combinations of any two, three or more of the foregoing.

[0013] In a further preferred embodiment, the diagnostic or prognostic marker for gastric cancer comprises or consists of a combination of any three selected from the following miRNA pairs:

[0014] hsa-miR-16 / hsa-miR-10a, hsa-miR-16 / hsa-miR-128, hsa-miR-16 / hsa-miR-146a, hsa-miR-16 / hsa-miR-191, hsa-miR-16 / h sa-miR-215, hsa-miR-16 / hsa-miR-21, hsa-miR-16 / hsa-miR-221, hsa-miR-16 / hsa-miR-222, hsa-miR-16 / hsa-miR-26a, hsa- miR-16 / hsa-miR-26b, hsa-miR-19b / hsa-miR-128, hsa-miR-363 / hsa-miR-106b, hsa-miR-363 / hsa-miR-128, hsa-miR-363 / hs a-miR-222, hsa-miR-221 / hsa-miR-363, hsa-miR-186 / hsa-miR-221, hsa-miR-629 / hsa-miR-363, hsa-miR-629 / hsa-miR-339.

[0015] In some preferred embodiments, the diagnostic or prognostic marker for gastric cancer comprises or consists of the following miRNA pairs:

[0016] hsa-miR-16 / hsa-miR-221, hsa-miR-186 / hsa-miR-221, hsa-miR-629 / hsa-miR-363, hsa-miR-221 / hsa-miR-363, hsa-miR-629 / hsa-miR-339.

[0017] In other preferred embodiments, the diagnostic or prognostic marker for gastric cancer comprises or consists of the following miRNA pairs:

[0018] hsa-miR-215 / hsa-miR-26b, hsa-miR-26b / hsa-miR-10a, hsa-miR-26b / hsa-miR-21.

[0019] In other preferred embodiments, the diagnostic or prognostic marker for gastric cancer comprises or consists of the following miRNA pairs:

[0020] hsa-miR-103a / hsa-miR-140, hsa-miR-26b / hsa-miR-10a, hsa-miR-26b / hsa-miR-21.

[0021] In other preferred embodiments, the diagnostic or prognostic marker for gastric cancer comprises or consists of the following miRNA pairs:

[0022] hsa-miR-191 / hsa-miR-21, hsa-miR-191 / hsa-miR-222, hsa-miR-21 / hsa-miR-181a, hsa-miR-26b / hsa-miR-140.

[0023] Preferably, the miRNA in the miRNA pair is a miRNA in serum or plasma, preferably a miRNA in serum extracellular vesicles.

[0024] In a second aspect, the present invention provides a model for gastric cancer diagnosis, therapeutic efficacy or prognosis assessment, wherein the model is constructed by a construction method comprising the following steps:

[0025] The expression ratio of the miRNA pair in the diagnostic or prognostic marker for gastric cancer as described in the first aspect above is used as the model feature, and a machine learning method is used to perform modeling.

[0026] The machine learning method can be a machine learning method commonly used in the field of modeling, such as logistic regression, support vector machine, tree model, neural network, etc.

[0027] In a third aspect, the present invention provides use of the diagnostic or prognostic marker for gastric cancer or a detection reagent thereof as described in the first aspect above in the preparation of a drug or a kit for gastric cancer diagnosis, efficacy or prognosis evaluation.

[0028] In a fourth aspect, the present invention provides a drug or kit for gastric cancer diagnosis, efficacy or prognosis evaluation, wherein the drug or kit comprises the gastric cancer diagnostic or prognostic marker or its detection reagent as described in the first aspect above.

[0029] In gastric cancer patients, the expression ratio of the above miRNA pairs in their serum extracellular vesicles is significantly different from that in patients with benign gastric diseases and healthy people.

[0030] Beneficial effects

[0031] The diagnostic or prognostic markers for gastric cancer provided by the present invention involve biologically significant miRNA pairs in serum or plasma extracellular vesicles. The expression ratio of the miRNA pair is used as the characteristic of the gastric cancer diagnostic model, avoiding the instability when the expression level of a single miRNA is used as a characteristic, thereby enabling more accurate diagnosis or prognostic evaluation; the gastric cancer diagnostic model constructed using the expression ratio of the miRNA pair as the model characteristic has high diagnostic sensitivity and specificity, and also has significant potential in screening for early gastric cancer. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] One or more embodiments are exemplarily illustrated by the accompanying figures, and these exemplary illustrations do not limit the embodiments. The term "exemplary" is used herein to mean "serving as an example, example, or illustration." Any embodiment described herein as "exemplary" is not necessarily to be construed as superior or preferred over other embodiments.

[0033] Figure 1 The clustering effect of the expression ratios of the 26 paired miRNAs screened in Example 1 on the NGS samples is shown; (A) is an LDA graph, and (B) is a heat map, with different grayscales representing healthy, benign, and gastric cancer samples, respectively.

[0034] Figure 2 This is the clustering effect of the expression ratios of the 26 paired miRNAs screened in Example 1 on the qPCR confirmation cohort; it is an LDA graph, and different grayscales represent healthy, benign, and gastric cancer samples, respectively.

[0035] Figure 3 ROC curves of the models constructed for 45 three-miRNA pair combinations.

[0036] Figure 4 The figure shows the diagnostic effect of gastric cancer on the gastric cancer diagnostic model constructed based on the preferred combination of paired miRNAs. (A) shows the ROC curve, and (B) shows the confusion matrix.

[0037] Figure 5 The diagram shows the predictive effect of the gastric cancer diagnostic model constructed based on the preferred combination of paired miRNAs on different disease types, wherein (A) shows the ROC curve and (B) shows the sample score box plot.

[0038] Figure 6 The figure shows the diagnostic performance of the gastric cancer diagnostic model constructed based on the preferred combination 2 of paired miRNAs in the validation set; wherein, (A) shows the ROC curve, and (B) shows the sample score box plot.

[0039] Figure 7 The figure shows the diagnostic performance of the gastric cancer diagnostic model constructed based on the preferred combination of paired miRNAs in the validation set; wherein, (A) shows the ROC curve, and (B) shows the sample score box plot.

[0040] Figure 8 The figure shows the diagnostic performance of the gastric cancer diagnostic model constructed based on the preferred combination of four paired miRNAs in the validation set; wherein, (A) shows the ROC curve, and (B) shows the sample score box plot. DETAILED DESCRIPTION

[0041] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. Unless otherwise expressly stated, throughout the specification and claims, the term "including" or its variations such as "comprising" or "including" will be understood to include the stated elements or components, without excluding other elements or other components.

[0042] In addition, in order to better illustrate the present invention, numerous specific details are provided in the following detailed description. It should be understood by those skilled in the art that the present invention can be practiced without certain specific details. In some embodiments, raw materials, components, methods, means, etc. that are well known to those skilled in the art are not described in detail in order to highlight the main purpose of the present invention.

[0043] Unless otherwise specified, the reagents, kits, raw materials, and equipment used in the following examples can all be purchased from commercial sources. The experiments or detection methods involved in the present invention, unless otherwise specified, are all conventional in the art, or are performed with reference to the corresponding kits or product instructions.

[0044] Example 1: Screening of characteristic miRNA pairs for gastric cancer based on serum extracellular vesicle data

[0045] 1. Collection and preparation of clinical data and samples

[0046] A retrospective analysis was performed and serum samples were collected from patients with gastric cancer, patients with benign gastric diseases, and healthy subjects. The sample information is shown in Table 1.

[0047] Table 1. Statistics of selected cases

[0048]

[0049] All gastric cancer patients were diagnosed with gastric cancer through electronic gastroscopy biopsy. Other systemic tumors, tumors metastasized to the stomach, and other underlying diseases were excluded. In addition, the patients did not receive chemotherapy or radiotherapy before surgery. Benign diseases included gastric polyps, gastritis, and gastric ulcers.

[0050] As shown in Table 1, the dataset was divided into four cohorts. The NGS cohort was used to discover characteristic miRNA pairs for gastric cancer. The qPCR confirmation cohort (with the same samples as the aforementioned NGS cohort) and the qPCR validation cohort were used for modeling and validation of the preferred miRNA pair combinations, respectively. The multicenter validation cohort was used for expanded validation of the miRNA pair combinations.

[0051] 2. Isolation and purification of extracellular vesicles

[0052] Collect blood in a 5 mL vacuum blood collection tube (coagulant + separation gel) and centrifuge at 2000 × g for 10 minutes at room temperature within 2 hours after sampling to separate the serum; transfer the resulting serum to a 1.5 mL centrifuge tube within 4 hours and centrifuge at 3000 × g for 10 minutes at 4°C, discarding the precipitate to remove any cell debris and apoptotic bodies; then, draw out the upper layer of serum, aliquot it, and store it at -80°C. Repeated freezing and thawing is strictly prohibited.

[0053] Extracellular vesicles from serum samples were isolated using the GlyExo-Capture kit, using the company's patented "lectin-magnetic carrier coupled complex for isolating glycosylated exosomes from clinical samples" magnetic beads.

[0054] If the extracellular vesicles of the serum sample cannot be isolated in time, the sample can be stored at -20°C or below, and freezing and thawing are strictly prohibited.

[0055] 3. Extraction and NGS sequencing of extracellular vesicle miRNA

[0056] use Total RNA from extracellular vesicles was extracted using the Illumina NEBNext Small RNA Library Preparation Kit according to the kit's instructions and evaluated using the high-sensitivity RNA kit of the Qsep100 fully automated nucleic acid analysis system. The extracellular vesicle small RNAs were then compiled into a cDNA library using the Illumina NEBNext Small RNA Library Preparation Kit. The library was purified using the E-Gel Power Snap electrophoresis system and E-Gel SizeSelect II gels to select and purify fragments of the desired size. After checking the quality and concentration of the cDNA library, 75nt single-end sequencing was performed on the Illumina NextSeq 550 Sequencing System, generating over 10 million reads per library.

[0057] 4. Construct miRNA interaction network and analyze NGS sequencing data to obtain characteristic miRNA pairs for gastric cancer through single factor screening and genetic algorithm screening.

[0058] (1) Construction of miRNA interaction network

[0059] 1) Obtain miRNA targets from the miRTarBase database;

[0060] 2) Based on the miRNA target information obtained in step 1), screening transcription factors that can serve as miRNA targets from the human transcription factor database hTFtarget and AnimalTFDB;

[0061] 3) Based on the transcription factors screened in step 2), the miRNAs that are further regulated are obtained through bioinformatics methods or public databases, thereby constructing the miRNA-TF-miRNA interaction relationship and obtaining the miRNA interaction network.

[0062] (2) Analyze NGS sequencing data and prepare miRNA quantification data

[0063] By analyzing the above NGS sequencing data, miRNA quantitative data of each sample in the comprehensive sample set including disease samples and control samples were obtained:

[0064] First, the original sequencing file in fastq format is required for quality control. Based on the quality control results, cutadapt software is used to remove adapters and low-quality reads. For the data that pass the quality control, the exceRpt small RNA analysis process is used for annotation and quantification to obtain the expression matrix. Based on the expression level of the expression matrix, miRNAs with too low counts are filtered out (specific analysis is required for specific cases) to obtain corrected miRNA quantitative data.

[0065] (3) Construction of miRNA expression ratio features

[0066] Based on the miRNA interaction network constructed above and the prepared miRNA quantitative data, the expression ratios of the miRNA pairs with interacting relationships in each sample were calculated.

[0067] In order to avoid the phenomenon of the denominator being 0, when calculating the miRNA expression ratio, the denominator is uniformly added with "1". The calculation formula is as follows:

[0068] miRNA_a / miRNA_b=counts a / (count b +1)

[0069] (4) Screening of characteristic miRNA pairs

[0070] Through single-factor screening and genetic algorithm screening, characteristic miRNA pairs for gastric cancer were obtained. The basic process is as follows:

[0071] i) Univariate screening: Compare the expression ratios of each miRNA pair in the diseased biological sample group relative to the normal biological sample group, calculate the p-value using Python's scipy.stats.ttest_ind, and correct the p-value using statsmodels.stats.multitest.fdrcorrection. Screen based on the corrected p-value p-adjusted, based on a threshold of 0.05;

[0072] ii) For the screened miRNA pairs, calculate the logarithm of the fold change of their expression ratio in the disease biological sample group relative to the expression ratio in the normal biological sample group, i.e., log2FoldChange, and select an appropriate threshold for log2FoldChange based on the actual situation to further screen suitable targets;

[0073] The genetic algorithm screening was performed more than 100 times.

[0074] After the above screening procedure, 26 paired miRNAs were obtained, and their detailed information is shown in Table 2 below.

[0075] Table 2. Details of 26 paired miRNAs

[0076]

[0077] In addition, the clustering effect diagram of the expression ratios of these 26 paired miRNAs on the NGS cohort was drawn, see Figure 1 ; Figure 1 The expression ratios of these 26 paired miRNAs can clearly distinguish healthy, benign gastric disease and gastric cancer samples on the LDA graph ( Figure 1 -A), and there is also an obvious clustering effect on the heat map ( Figure 1 -B).

[0078] These 26 paired miRNAs performed well in grouping and clustering cancer and non-cancer samples in the NGS cohort.

[0079] Example 2: qPCR validation of 26 paired miRNAs

[0080] In this example, 26 paired miRNAs screened in Example 1 were validated by qPCR, and the samples were the same as those used for NGS.

[0081] The specific verification method is as follows:

[0082] First, perform first-strand cDNA synthesis. The reaction system and reaction conditions are as follows:

[0083]

[0084] Reaction conditions Time (minutes) 42℃ 60 85℃ 5 4℃ ∞

[0085] Then, the cDNA from the previous step was diluted 5-fold with nuclease-free water and the miRNA qPCR reaction was performed on the ABI 7500 Real-Time Fluorescence Quantitative PCR System. The reaction system and reaction conditions were as follows:

[0086]

[0087]

[0088] In the above RT-qPCR procedure, the primer sequences used are as follows:

[0089]

[0090] The classification effect diagram of the samples was made according to the qPCR results. Figure 2 ,Depend on Figure 2 It can be seen that these 26 paired miRNAs can clearly distinguish healthy, benign gastric disease and gastric cancer samples on the LDA graph ( Figure 2 ). Therefore, these 26 paired miRNAs also have the potential to become clinical markers for gastric cancer based on qPCR data.

[0091] In addition, in addition to these 26 paired miRNAs, we anticipate and demonstrate through experiments that other pairs or combinations of miRNAs included in these 26 paired miRNAs will also have the potential to become clinical markers for gastric cancer.

[0092] Subsequently, 45 three-miRNA pair combinations were obtained by randomly selecting any three miRNA pairs from the 26 paired miRNAs, as detailed in Example 3. Then, the five paired miRNAs for the preferred combination one were obtained using a random forest model, as detailed in Example 4; the three paired miRNAs for the preferred combination two were obtained using a logistic regression model, as detailed in Example 6; the three paired miRNAs for the preferred combination three were obtained using a support vector machine model, as detailed in Example 7; and the four paired miRNAs for the preferred combination four were obtained using a Gaussian naive Bayes model, as detailed in Example 8.

[0093] Example 3: Predictive effects of 45 three-miRNA combinations selected from 26 paired miRNAs

[0094] In this example, three miRNA pairs were randomly selected from the 26 paired miRNAs to form a combination, obtaining a total of 45 three-miRNA pair combinations. Logistic regression modeling was performed on each of the 45 three-miRNA pair combinations to evaluate their predictive performance for gastric cancer.

[0095] The ROC curves of each model are as follows Figure 3 As shown, Figure 3 The results showed that the AUCs of these 45 three-miRNA pair combinations were all above 0.89, and the union of the combinations covered 26 paired miRNAs.

[0096] It can be seen that the combination of any three of the 26 paired miRNAs can achieve good gastric cancer prediction results, which further supports the reliability of the 26 paired miRNAs as clinical markers for gastric cancer.

[0097] Example 4: Construction of a gastric cancer diagnostic model based on paired miRNA preferred combination 1 and verification of its diagnostic performance

[0098] In this embodiment, 5 paired miRNAs (hsa-miR-16 / hsa-miR-221, hsa-miR-186 / hsa-miR-221, hsa-miR-629 / hsa-miR-363, hsa-miR-221 / hsa-miR-363, hsa-miR-629 / hsa-miR-339) selected from the above 26 paired miRNAs are selected as preferred combination one. A gastric cancer diagnostic model is constructed based on this preferred combination one, and its diagnostic performance is verified.

[0099] The samples of the 101 qPCR confirmation cohorts shown in Table 1 were divided into training and test sets in a ratio of 7:3, and a random forest model was established on the training set. In addition, serum glycosylated extracellular vesicles from 54 gastric cancer cases, 35 benign gastric diseases, and 50 healthy samples (i.e., the "qPCR validation cohort" in Table 1) were collected, and miRNA was sequenced by qPCR as an external validation set.

[0100] The ROC curve of the model is as follows Figure 4 As shown; Figure 4 It shows that the AUC of the training set, test set, and validation set are 1.000, 0.920, and 0.930 respectively ( Figure 4 -A); In addition, the confusion matrix of each data set was checked under the optimal cutoff of the training set, which showed that the sensitivity and specificity of the test set and validation set were 82.35%, 81.48% and 85.71%, 83.53% ( Figure 4 -B), are at a relatively high level.

[0101] The above results show that the preferred combination consisting of the above five paired miRNAs has relatively stable predictive performance in various data sets, and has the characteristics of high sensitivity and strong specificity. Therefore, it can better diagnose and screen gastric cancer and has great clinical application value.

[0102] Example 5: Analysis of the predictive effect of the gastric cancer diagnostic model based on the preferred combination 1 on early gastric cancer samples

[0103] In order to verify the early screening effect of the gastric cancer diagnostic model based on the preferred combination 1 for gastric cancer, this example uses the model constructed in Example 4 to analyze the prediction effects of gastric cancer vs. benign gastric disease, gastric cancer vs. health, early gastric cancer vs. benign gastric disease, early gastric cancer vs. health, late gastric cancer vs. benign gastric disease, and late gastric cancer vs. health.

[0104] ROC curve and prediction score box plot are shown in Figure 5 ,Depend on Figure 5 As can be seen, the model has good predictive effects on both early gastric cancer and benign gastric disease samples, with AUCs above 0.9. The boxplots of the prediction scores also show significant differences. This indicates that the combination of these five paired miRNAs has a good discriminatory effect on both early gastric cancer and benign gastric disease samples, meeting the criteria for early cancer screening.

[0105] Example 6: Construction of a gastric cancer diagnostic model based on the second paired miRNA preferred combination and verification of its diagnostic performance

[0106] In this embodiment, three paired miRNAs (hsa-miR-215 / hsa-miR-26b, hsa-miR-26b / hsa-miR-10a, hsa-miR-26b / hsa-miR-21) selected from the above 26 paired miRNAs are selected as the preferred combination 2. A gastric cancer diagnostic model is constructed based on the preferred combination 2, and its diagnostic performance is verified.

[0107] The samples of the 101 qPCR confirmation cohorts shown in Table 1 were divided into training and test sets in a ratio of 7:3, and a model was established using logistic regression on the training set. In addition, serum glycosylated extracellular vesicles from 54 gastric cancer patients, 35 benign gastric diseases, and 50 healthy samples (i.e., the "qPCR validation cohort" in Table 1) were collected, and miRNA was sequenced by qPCR as an external validation set.

[0108] The ROC curve and sample score distribution box plot of the external validation set are as follows Figure 6 As shown; Figure 6 The results showed that the combination of these three paired miRNAs achieved an AUC of 0.869 on the validation set, with a sensitivity and specificity of 78.79% and 89.40%, respectively. Furthermore, the validation set score boxplot showed significant differences in scores for gastric cancer, benign gastric disease, and healthy samples, demonstrating good discrimination.

[0109] Example 7: Construction of a gastric cancer diagnostic model based on paired miRNA preferred combination three and verification of its diagnostic performance

[0110] In this embodiment, three paired miRNAs (hsa-miR-103a / hsa-miR-140, hsa-miR-26b / hsa-miR-10a, hsa-miR-26b / hsa-miR-21) selected from the above 26 paired miRNAs are selected as the preferred combination three. A gastric cancer diagnostic model is constructed based on the preferred combination three, and its diagnostic performance is verified.

[0111] The samples of the 101 qPCR confirmation cohorts shown in Table 1 were divided into training and test sets in a ratio of 7:3, and a linear SVM model was established on the training set. In addition, serum glycosylated extracellular vesicles from 592 gastric cancer patients, 452 benign gastric diseases, and 118 healthy samples (i.e., the "multicenter validation cohort" in Table 1) were collected, and miRNA was sequenced by qPCR as an external validation set.

[0112] The ROC curve and sample score distribution box plot of the multicenter validation cohort are shown in the figure. Figure 7 As shown; Figure 7 The results show that the combination of these six paired miRNAs achieved an AUC of 0.967 on the validation set, with a sensitivity and specificity of 90.74% and 98.82%, respectively. Furthermore, the validation set score boxplot shows significant differences in scores for gastric cancer, benign gastric disease, and healthy samples, with good discrimination.

[0113] Example 8: Construction of a gastric cancer diagnostic model based on paired miRNA preferred combination 4 and verification of its diagnostic performance

[0114] In this embodiment, four paired miRNAs (hsa-miR-191 / hsa-miR-21, hsa-miR-191 / hsa-miR-222, hsa-miR-21 / hsa-miR-181a, hsa-miR-26b / hsa-miR-140) selected from the above 26 paired miRNAs are selected as the preferred combination four. A gastric cancer diagnostic model is constructed based on the preferred combination four, and its diagnostic performance is verified.

[0115] The samples of the 101 qPCR confirmation cohorts shown in Table 1 were divided into training and test sets in a ratio of 7:3, and the model was established using Gaussian Bayesian. In addition, serum glycosylated extracellular vesicles from 592 gastric cancer patients, 452 benign gastric diseases, and 118 healthy samples (i.e., the "multicenter validation cohort" in Table 1) were collected, and miRNA was sequenced by qPCR as an external validation set.

[0116] The ROC curve and sample score distribution box plot of the multicenter validation cohort are shown in the figure. Figure 8 As shown; Figure 8The results show that the combination of these six paired miRNAs achieved an AUC of 0.888 on the validation set, with a sensitivity and specificity of 84.48% and 81.76%, respectively. Furthermore, the validation set score boxplot shows significant differences in scores for gastric cancer, benign gastric disease, and healthy samples, demonstrating good discrimination.

[0117] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A diagnostic marker for gastric cancer, characterized in that The diagnostic markers for gastric cancer consist of the following miRNAs: hsa-miR-16, hsa-miR-221, hsa-miR-186, hsa-miR-629, hsa-miR-363, and hsa-miR-339; Among them, when diagnosing gastric cancer, the expression ratios of hsa-miR-16 and hsa-miR-221, the expression ratios of hsa-miR-186 and hsa-miR-221, the expression ratios of hsa-miR-629 and hsa-miR-363, the expression ratios of hsa-miR-221 and hsa-miR-363, and the expression ratios of hsa-miR-629 and hsa-miR-339 were used as analysis features.

2. A diagnostic marker for gastric cancer, characterized in that: The diagnostic markers for gastric cancer are composed of the following miRNAs: hsa-miR-215, hsa-miR-26b, hsa-miR-10a, hsa-miR-21; Among them, when diagnosing gastric cancer, the expression ratios of hsa-miR-215 to hsa-miR-26b, the expression ratios of hsa-miR-26b to hsa-miR-10a, and the expression ratios of hsa-miR-26b to hsa-miR-21 were used as analysis features.

3. A diagnostic marker for gastric cancer, characterized in that The diagnostic markers for gastric cancer are composed of the following miRNAs: hsa-miR-103a, hsa-miR-140, hsa-miR-26b, hsa-miR-10a, hsa-miR-21; Among them, when diagnosing gastric cancer, the expression ratios of hsa-miR-103a to hsa-miR-140, the expression ratios of hsa-miR-26b to hsa-miR-10a, and the expression ratios of hsa-miR-26b to hsa-miR-21 were used as analysis features.

4. A diagnostic marker for gastric cancer, characterized in that The diagnostic markers for gastric cancer are composed of the following miRNAs: hsa-miR-191, hsa-miR-21, hsa-miR-222, hsa-miR-181a, hsa-miR-26b, and hsa-miR-140; Among them, when diagnosing gastric cancer, the expression ratios of hsa-miR-191 to hsa-miR-21, the expression ratios of hsa-miR-191 to hsa-miR-222, the expression ratios of hsa-miR-21 to hsa-miR-181a, and the expression ratios of hsa-miR-26b to hsa-miR-140 were used as analysis features.

5. The diagnostic marker for gastric cancer according to any one of claims 1 to 4, characterized in that The miRNA is miRNA in serum or plasma.

6. The diagnostic marker for gastric cancer according to claim 5, characterized in that The miRNA is the miRNA in serum extracellular vesicles.

7. Use of the gastric cancer diagnostic marker detection reagent according to any one of claims 1 to 6 in the preparation of a kit for gastric cancer diagnosis.

8. A kit for diagnosing gastric cancer, characterized in that: The kit comprises a detection reagent for detecting the expression level of each miRNA in the diagnostic marker for gastric cancer according to claim 1, wherein The forward primer used to detect the expression level of hsa-miR-16 is shown in SEQ ID NO: 5, and the reverse primer is shown in SEQ ID NO: 22; The forward primer used to detect the expression level of hsa-miR-221 is shown in SEQ ID NO: 11, and the reverse primer is shown in SEQ ID NO: 22; The forward primer used to detect the expression level of hsa-miR-186 is shown in SEQ ID NO: 6, and the reverse primer is shown in SEQ ID NO: 22; The forward primer used to detect the expression level of hsa-miR-629 is shown in SEQ ID NO: 17, and the reverse primer is shown in SEQ ID NO: 22; The forward primer used to detect the expression level of hsa-miR-363 is shown in SEQ ID NO: 16, and the reverse primer is shown in SEQ ID NO: 22; The forward primer used to detect the expression level of hsa-miR-339 is shown in SEQ ID NO: 15, and the reverse primer is shown in SEQ ID NO:

22.

9. A kit for diagnosing gastric cancer, characterized in that: The kit comprises a detection reagent for detecting the expression level of each miRNA in the diagnostic marker for gastric cancer according to claim 2, wherein: The forward primer used to detect the expression level of hsa-miR-215 is shown in SEQ ID NO: 9, and the reverse primer is shown in SEQ ID NO: 22; The forward primer used to detect the expression level of hsa-miR-26b is shown in SEQ ID NO: 14, and the reverse primer is shown in SEQ ID NO: 22; The forward primer used to detect the expression level of hsa-miR-10a is shown in SEQ ID NO: 2, and the reverse primer is shown in SEQ ID NO: 22; The forward primer used to detect the expression level of hsa-miR-21 is shown in SEQ ID NO: 10, and the reverse primer is shown in SEQ ID NO:

22.

10. A kit for diagnosing gastric cancer, characterized in that: The kit comprises a detection reagent for detecting the expression level of each miRNA in the diagnostic marker for gastric cancer according to claim 3, wherein The forward primer used to detect the expression level of hsa-miR-103a is shown in SEQ ID NO: 18, and the reverse primer is shown in SEQ ID NO: 22; The forward primer used to detect the expression level of hsa-miR-140 is shown in SEQ ID NO: 19, and the reverse primer is shown in SEQ ID NO: 22; The forward primer used to detect the expression level of hsa-miR-26b is shown in SEQ ID NO: 14, and the reverse primer is shown in SEQ ID NO: 22; The forward primer used to detect the expression level of hsa-miR-10a is shown in SEQ ID NO: 2, and the reverse primer is shown in SEQ ID NO: 22; The forward primer used to detect the expression level of hsa-miR-21 is shown in SEQ ID NO: 10, and the reverse primer is shown in SEQ ID NO:

22.

11. A kit for diagnosing gastric cancer, characterized in that: The kit comprises a detection reagent for detecting the expression level of each miRNA in the diagnostic marker for gastric cancer according to claim 4, wherein The forward primer used to detect the expression level of hsa-miR-191 is shown in SEQ ID NO: 7, and the reverse primer is shown in SEQ ID NO: 22; The forward primer used to detect the expression level of hsa-miR-21 is shown in SEQ ID NO: 10, and the reverse primer is shown in SEQ ID NO: 22; The forward primer used to detect the expression level of hsa-miR-222 is shown in SEQ ID NO: 12, and the reverse primer is shown in SEQ ID NO: 22; The forward primer used to detect the expression level of hsa-miR-181a is shown in SEQ ID NO: 20, and the reverse primer is shown in SEQ ID NO: 22; The forward primer used to detect the expression level of hsa-miR-26b is shown in SEQ ID NO: 14, and the reverse primer is shown in SEQ ID NO: 22; The forward primer used to detect the expression level of hsa-miR-140 is shown in SEQ ID NO: 19, and the reverse primer is shown in SEQ ID NO:

22.

12. The kit according to any one of claims 8 to 11, characterized in that In gastric cancer patients, the expression ratio of the miRNA pair in their serum extracellular vesicles is significantly different from that in patients with benign gastric diseases or healthy people.

Images