Pancreatic cancer diagnosis marker based on external vesicle miRNA and application thereof

Abstract

The invention is a divisional application of which the application number is 202411480045.3. The invention provides a pancreatic cancer diagnosis marker based on external vesicle miRNA (micro Ribonucleic Acid) and application of the pancreatic cancer diagnosis marker. The pancreatic cancer diagnosis or prognosis marker based on the extracellular vesicle miRNA relates to a miRNA combination with biological significance in serum or plasma extracellular vesicles, and does not take a conventional reference gene as a reference; the pancreatic cancer diagnosis model constructed by taking the expression quantity ratio of the paired miRNAs in the combination as the input characteristic of the model has high diagnosis sensitivity and specificity on pancreatic cancer, is suitable for different data centers, has good diagnosis stability, has great potential in early pancreatic cancer screening, and has good application prospects. The kit has important practical significance for early diagnosis or prognosis prediction of pancreatic cancer.

Description

[0001] The present application is a divisional application of a Chinese application with the application number 202411480045.3, the title of invention being "Pancreatic cancer diagnostic marker based on exovesicle miRNA and application thereof", and the filing date being October 22, 2024. TECHNICAL FIELD

[0002] The present application belongs to the field of medical detection technology, and specifically relates to a pancreatic cancer diagnostic marker based on exovesicle miRNA and application thereof. BACKGROUND

[0003] Pancreatic cancer (PC) is one of the most deadly malignancies worldwide. Due to rapid disease progression and occult early symptoms, only 15-20% of patients are diagnosed at an early stage. Currently, the incidence and mortality of pancreatic cancer are increasing year by year in the United States, Europe, Japan, and China. It is estimated that by 2050, the global incidence of pancreatic cancer will reach 18.6 per 100,000 people, with an average annual growth rate of 1.1%, which will put a huge pressure on the public health system. The high mortality rate of pancreatic cancer is mainly due to the special anatomical location of the pancreas and non-specific symptoms, resulting in many cases being discovered and diagnosed at an advanced stage when obvious clinical symptoms have already appeared. Globally, the 5-year survival rate of pancreatic cancer is generally less than 10%, with little difference between high-income and low-income countries, and the survival rate is the lowest among all cancers. Despite the progress in diagnosis and treatment in recent years, the high mortality rate of pancreatic cancer is still closely related to late-stage discovery and the limitations of existing treatment methods. Therefore, the development of new screening methods, early diagnosis tools, and treatment strategies is crucial for improving the prognosis of pancreatic cancer patients.

[0004] The diagnosis of pancreatic cancer mainly relies on imaging and serum tumor markers. Imaging methods include ultrasound, computed tomography (CT), and magnetic resonance imaging (MRI), which are commonly used to detect obvious pancreatic masses, but have limitations in early diagnosis, especially for early cases that do not form obvious masses. In addition, the results of imaging diagnosis are easily affected by the experience of the operator and the performance of the equipment. The serum marker CA19-9 protein is the most commonly used indicator in the diagnosis and monitoring of pancreatic cancer, but CA19-9 protein may appear false positive in cases of biliary tract infection, inflammation, or obstruction; and CA19-9 may appear false negative in individuals with Lewis antigen negativity; thus, the early diagnostic effect of CA19-9 protein as a single indicator is not ideal. In contrast, molecular diagnostic techniques provide more accurate means for auxiliary diagnosis by detecting pancreatic cancer-related genes, proteins, and RNA markers. Combining imaging, biomarker detection, and molecular diagnostic techniques can help improve the accuracy of early screening for pancreatic cancer and provide strong support for improving patient survival rates.

[0005] Extracellular vesicles are small vesicles secreted by cells into body fluids or extracellular environment, and play an important role in intercellular communication. Due to the protection of the lipid bilayer, extracellular vesicles can exist stably in various biological fluids and cell culture fluids; this characteristic makes the contents of extracellular vesicles a reliable tumor marker, and miRNA is one of the contents of extracellular vesicles. It has been reported that the expression levels of extracellular vesicle miRNAs are different in different diseases and physiological conditions, and also play an important regulatory role in the progression of malignant tumors. Therefore, extracellular vesicle miRNAs can become a new form of biomarker for early diagnosis and prognosis tracking of cancer.

[0006] However, the biomarkers for early screening of pancreatic cancer have the problems of low sensitivity, low specificity or poor repeatability of diagnostic efficiency. For example, in the prior art, a relatively stable internal reference gene is often selected as a reference to detect the relative expression amount of the marker, but existing literature shows that although the expression of the internal reference gene is relatively stable, the expression amount of the internal reference gene still has some differences under different physiological conditions, resulting in inaccurate relative expression amount of the marker. The marker or marker combination screened in this way will make the repeatability of the diagnostic efficiency poor (i.e. unstable results), and when the sample quantity, sample distribution, etc. change, the accuracy of the diagnostic efficiency will decrease, resulting in a decrease in the reliability of the marker. In addition, different batches or different data centers of multiple sources will also cause the results to be unstable, for example, the marker or marker combination screened in one data set center (results detected by the same center or institution) is applied to other hospitals or data centers, and the results are prone to be unstable (poor repeatability), and even when different batches are used, the results are also unstable. At present, there is still a lack of effective early pancreatic cancer diagnostic kits on the market, so it is urgent to screen pancreatic cancer diagnostic markers or marker combinations with good repeatability, high reliability and high diagnostic efficiency. SUMMARY

[0007] OBJECTIVE

[0008] In view of the problems existing in the prior art method, the present application aims to provide a pancreatic cancer diagnostic or prognostic marker with high sensitivity and specificity, an analysis system, a kit or an application thereof based thereon, including: an analysis system, a kit for predicting pancreatic cancer or evaluating the prognosis of pancreatic cancer, or an application in the preparation of a kit or an analysis system for pancreatic cancer diagnosis, efficacy or prognosis evaluation, or related drug evaluation.

[0009] SOLUTION

[0010] To achieve the above-mentioned object, the present application provides the following technical solutions:

[0011] In a first aspect, the present application provides a diagnostic or prognostic marker panel for pancreatic cancer, said diagnostic or prognostic marker for pancreatic cancer comprises a marker combination consisting of at least two or at least four miRNAs selected from the group consisting of:

[0012] hsa-miR-98, hsa-miR-25, hsa-miR-378i, hsa-miR-150, hsa-miR-501, hsa-miR-155, hsa-miR-193a, hsa-miR-1180.

[0013] Further, said diagnostic or prognostic marker for pancreatic cancer comprises one or several of the following miRNA combinations:

[0014] hsa-miR-98 and hsa-miR-25, optionally hsa-miR-25 is used to calibrate the relative expression of hsa-miR-98 (i.e. the ratio of the expression of hsa-miR-98 and hsa-miR-25);

[0015] hsa-miR-378i and hsa-miR-150, optionally hsa-miR-150 is used to calibrate the relative expression of hsa-miR-378i (i.e. the ratio of the expression of hsa-miR-378i and hsa-miR-150);

[0016] hsa-miR-501 and hsa-miR-150, optionally hsa-miR-150 is used to calibrate the relative expression of hsa-miR-501 (i.e. the ratio of the expression of hsa-miR-501 and hsa-miR-150);

[0017] hsa-miR-378i and hsa-miR-155, optionally hsa-miR-155 is used to calibrate the relative expression of hsa-miR-378i;

[0018] hsa-miR-193a and hsa-miR-155, optionally hsa-miR-155 is used to calibrate the relative expression of hsa-miR-193a (i.e. the ratio of the expression of hsa-miR-193a and hsa-miR-155);

[0019] hsa-miR-150 and hsa-miR-1180, optionally hsa-miR-1180 is used to calibrate the relative expression of hsa-miR-150 (i.e. the ratio of the expression of hsa-miR-150 and hsa-miR-1180).

[0020] Further, the detection of the expression amount of the marker does not rely on a conventional internal reference gene as a relative reference or correction.

[0021] Further, the pancreatic cancer diagnostic or prognostic marker further comprises CA19-9 protein.

[0022] As a preferred embodiment, the pancreatic cancer diagnostic or prognostic marker comprises one or both of the following miRNA combinations 1)~2):

[0023] 1) Combination one: hsa-miR-98, hsa-miR-25, hsa-miR-378i, hsa-miR-150;

[0024] 2) Combination two: hsa-miR-501, hsa-miR-150, hsa-miR-378i, hsa-miR-155, hsa-miR-193a, hsa-miR-1180.

[0025] Further, as a preferred embodiment, the miRNAs in the miRNA pair are serum or plasma miRNAs.

[0026] Further, the miRNAs in the miRNA pair are serum or plasma glycosylated extracellular vesicle miRNAs.

[0027] In a second aspect, an analysis system for predicting pancreatic cancer or evaluating the prognosis of pancreatic cancer is provided, comprising a data analysis module for analyzing the expression amount of a marker combination of a target object to be predicted and calculating the ratio of the expression amount of a miRNA pair or directly taking the ratio of the expression amount of the miRNA pair as an input feature, and further for calculating the prediction value of whether the target object is pancreatic cancer according to the ratio of the expression amount of the miRNA pair; the input feature is selected from at least one of the following ratios of the expression amount of the miRNA pair:

[0028] hsa-miR-98 / hsa-miR-25;

[0029] hsa-miR-378i / hsa-miR-150;

[0030] hsa-miR-501 / hsa-miR-150;

[0031] hsa-miR-378i / hsa-miR-155;

[0032] hsa-miR-193a / hsa-miR-155;

[0033] hsa-miR-150 / hsa-miR-1180.

[0034] Further, the input features are selected from one or more of the following combinations:

[0035] 1) Combination one: hsa-miR-98 / hsa-miR-25, hsa-miR-378i / hsa-miR-150;

[0036] 2) Combination two: hsa-miR-501 / hsa-miR-150; hsa-miR-378i / hsa-miR-155; hsa-miR-193a / hsa-miR-155; hsa-miR-150 / hsa-miR-1180.

[0037] Further, the data analysis module stores a prediction model for outputting the probability value of pancreatic cancer, wherein the construction method of the prediction model comprises: taking the ratio of the expression amounts of the miRNA pairs in the pancreatic cancer, benign, and healthy samples as input variables, and taking whether it is pancreatic cancer as a dependent variable, obtaining the prediction model through multivariate logistic regression calculation, and verifying through a validation set.

[0038] Further, when the detection method is RT-qPCR, the ratio of the expression amounts is the CT difference value of the miRNA pairs (the CT value is a conversion of the expression amount, so the CT difference value represents the ratio of the expression amount).

[0039] Further, the input features further include the expression level of CA19-9 protein of the target object to be predicted, and the analysis system comprises a data analysis module, which is used for analyzing the expression amounts of the marker combinations of the target object to be predicted and calculating the ratio of the expression amounts of the miRNA pairs or directly taking the ratio of the expression amounts of the miRNA pairs as input features, and is further used for calculating the prediction value of whether the target object is pancreatic cancer according to the ratio of the expression amounts of the miRNA pairs and the expression level of CA19-9 protein; the input features are selected from the expression level of CA19-9 protein of the target object to be predicted and the ratio of the expression amounts of the miRNA pairs in the following combinations:

[0040] hsa-miR-98 / hsa-miR-25, hsa-miR-378i / hsa-miR-150.

[0041] Further, the data analysis module stores a prediction model for outputting the probability value of pancreatic cancer, wherein the construction method of the prediction model comprises: taking the ratio of the expression amounts of the miRNA pairs in the pancreatic cancer, benign, and healthy samples as input variables, and taking whether it is pancreatic cancer as a dependent variable, obtaining the prediction model through multivariate logistic regression calculation, and verifying through a validation set.

[0042] In a third aspect, there is provided use of a reagent for detecting the diagnostic or prognostic marker combination of the pancreatic cancer of the first aspect in the preparation of a kit or an analytical system for the diagnosis, efficacy or prognosis evaluation of pancreatic cancer, or the evaluation of related drugs.

[0043] In a fourth aspect, there is provided use of a reagent for detecting the relative expression amount of the diagnostic or prognostic marker combination of the pancreatic cancer of the first aspect in serum or plasma glycosylated extracellular vesicles in the preparation of a kit or an analytical system for the diagnosis, efficacy or prognosis evaluation of pancreatic cancer, or the evaluation of related drugs, optionally, the marker combination is selected from one of the following combinations:

[0044] 1) Combination one: the expression level of hsa-miR-98 relative to hsa-miR-25, the expression level of hsa-miR-378i relative to hsa-miR-150;

[0045] 2) Combination two: the expression level of hsa-miR-501 relative to hsa-miR-150; the expression level of hsa-miR-378i relative to hsa-miR-155; the expression level of hsa-miR-193a relative to hsa-miR-155; the expression level of hsa-miR-150 relative to hsa-miR-1180.

[0046] In a fifth aspect, there is provided a kit for the diagnosis, efficacy or prognosis evaluation of pancreatic cancer, or the evaluation of related drugs, the kit comprising the diagnostic or prognostic marker combination of the pancreatic cancer of the first aspect and / or a detection reagent thereof.

[0047] Further, in pancreatic cancer patients, the expression amount ratio of the miRNA pairs of serum glycosylated extracellular vesicles is significantly different from that of healthy people.

[0048] Further, the detection reagent comprises primers and probes for detecting the corresponding miRNAs, wherein the nucleotide sequences of the primers and probes of each miRNA are as follows:

[0049] The nucleotide sequence of the upstream primer of hsa-miR-150 is shown in SEQ ID NO: 1;

[0050] The nucleotide sequence of the upstream primer of hsa-miR-25 is shown in SEQ ID NO: 2;

[0051] The nucleotide sequence of the upstream primer of hsa-miR-378i is shown in SEQ ID NO: 3;

[0052] The nucleotide sequence of the upstream primer of hsa-miR-98 is shown in SEQ ID NO: 4;

[0053] The nucleotide sequence of the upstream primer of hsa-miR-1180 is shown as SEQ ID NO: 6.

[0054] The nucleotide sequence of the upstream primer of hsa-miR-501 is shown as SEQ ID NO: 7.

[0055] The nucleotide sequence of the upstream primer of hsa-miR-193a is shown as SEQ ID NO: 8.

[0056] The nucleotide sequence of the upstream primer of hsa-miR-155 is shown as SEQ ID NO: 9.

[0057] The nucleotide sequence of the probe is shown as SEQ ID NO: 5.

[0058] The reverse primer is a common downstream primer, and the nucleotide sequence is shown as SEQ ID NO: 10.

[0059] Further, the primers and probes do not contain conventional internal reference genes.

[0060] Beneficial effects

[0061] The pancreatic cancer diagnostic or prognostic marker provided by the present application relates to a pair of miRNAs in serum or plasma extracellular vesicles with biological significance, especially a pair of miRNAs in a specific type of glycosylated extracellular vesicles, and the expression amount ratio of the pair of miRNAs is used as the input feature of a pancreatic cancer diagnostic model, thereby avoiding the problem of unstable diagnostic performance of existing diagnostic markers caused by the instability of internal reference genes, test batches and other differences. Therefore, more accurate, stable and reproducible diagnosis or prognosis evaluation can be achieved; the pancreatic cancer diagnostic model constructed by using the expression amount ratio of the pair of miRNAs as the input feature has high diagnostic sensitivity and specificity, and also has great potential in the screening of early pancreatic cancer. BRIEF DESCRIPTION OF DRAWINGS

[0062] One or more embodiments are illustrated by way of example in the figures that form part of this document, and which should not be construed as limiting the embodiments. Herein, the term "exemplary" is used in the sense of "serving as an example, instance, or illustration". Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments.

[0063] Figure 1 The heat map of the expression amount ratio of the 39 pairs of miRNAs screened in Example 1 on NGS samples.

[0064] Figure 2 The LDA graph of the RT-qPCR detection expression amount ratio of the 39 pairs of miRNAs in Example 2.

[0065] Figure 3 Box plot of CT value difference of RT-qPCR of 2 paired miRNAs in Example 3, in which, advanced stage represents pancreatic cancer advanced stage, early stage represents pancreatic cancer early stage, benign represents benign disease, healthy represents healthy people.

[0066] Figure 4 Pancreatic cancer ROC curve of pancreatic cancer diagnosis model based on 2 paired miRNAs of preferred combination one in Example 3 in different data sets.

[0067] Figure 5 Box plot of model score of model based on 2 paired miRNAs of preferred combination one in Example 3; in which, advanced stage represents pancreatic cancer advanced stage, early stage represents pancreatic cancer early stage, benign represents benign disease, healthy represents healthy people.

[0068] Figure 6 Box plot of CT value difference of RT-qPCR of 4 paired miRNAs in Example 4, in which, advanced stage represents pancreatic cancer advanced stage, early stage represents pancreatic cancer early stage, benign represents benign disease, healthy represents healthy people.

[0069] Figure 7 Pancreatic cancer ROC curve of pancreatic cancer diagnosis model based on 4 paired miRNAs of preferred combination two in Example 4 in different data sets.

[0070] Figure 8 Box plot of model score of model based on 4 paired miRNAs of preferred combination two in Example 4; in which, advanced stage represents pancreatic cancer advanced stage, early stage represents pancreatic cancer early stage, benign represents benign disease, healthy represents healthy people.

[0071] Figure 9 Pancreatic cancer ROC curve of pancreatic cancer diagnosis model based on 2 paired miRNAs of preferred combination one and pancreatic cancer index CA19-9 protein combination in Example 5 in different data sets.

[0072] Figure 10 Pancreatic cancer ROC curve of pancreatic cancer diagnosis model based on 4 paired miRNAs of preferred combination two and pancreatic cancer index CA19-9 protein combination in Example 6 in different data sets.

[0073] Figure 11 Box plot of CT value difference of RT-qPCR of 3 paired miRNAs in Comparative Example 1, in which, advanced stage represents pancreatic cancer advanced stage, early stage represents pancreatic cancer early stage, benign represents benign disease, healthy represents healthy people.

[0074] Figure 12 Pancreatic cancer ROC curve of the pancreatic cancer diagnostic model constructed based on the three pairs of miRNAs of combination three in Comparative Example 1 in different data sets.

[0075] Figure 13 Box plot of the expression levels of hsa-miR-98 and hsa-miR-25 corrected by the internal reference gene U6 (CT value difference of RT-qPCR) in Comparative Example 2, in which, advanced stage represents advanced stage of pancreatic cancer, early stage represents early stage of pancreatic cancer, benign represents benign disease, and healthy represents healthy population. DETAILED DESCRIPTION

[0076] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below. Obviously, the described embodiments are some embodiments but not all embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application. Unless otherwise explicitly indicated, in the entire specification and claims, the term “comprise” or its variants such as “contain” or “include” and the like are understood to include the stated element or component without excluding other elements or components.

[0077] In addition, in order to better illustrate the present application, numerous specific details are given in the following detailed description. Those skilled in the art should understand that the present application can also be implemented without some specific details. In some embodiments, the raw materials, elements, methods, means and the like which are well known to those skilled in the art are not described in detail, so as to highlight the main idea of the present application.

[0078] The reagents, kits, raw materials and equipment used in the following embodiments can be obtained by commercialization, unless otherwise specified. The experimental or detection methods involved in the present application are conventional experimental or detection methods in the art, or are carried out according to the corresponding kit or product instructions, unless otherwise specified.

[0079] Example 1: Screening of characteristic miRNA pairs of pancreatic cancer based on serum extracellular vesicle data

[0080] 1. Collection and preparation of clinical data and samples

[0081] The serum samples of pancreatic cancer patients, benign disease patients and healthy people were collected respectively, and the detailed information is shown in Table 1.

[0082] All pancreatic cancer patients were diagnosed as pancreatic cancer by pathology. Other systemic tumors, tumors metastasized to pancreas and other underlying diseases were excluded.

[0083] Benign disease patients include patients with pancreatitis, benign pancreatic tumors, gallbladder stones, etc.

[0084] Table 1: Statistical table of enrolled cases

[0085]

[0086] In Table 1, the data of the NGS cohort, the data of the RT-qPCR confirmation cohort are data obtained by the applicant from the serum collected from the hospital.

[0087] In Table 1, the data of the RT-qPCR verification cohort 1 and 2 (including miRNA and CA19-9 detection data) are from two different hospitals, and the data are obtained by the detection of the detection reagent provided by Beijing Rejie Biotechnology Co., Ltd. to the hospital.

[0088] That is, the data of the RT-qPCR confirmation cohort, the RT-qPCR verification cohort 1 and the RT-qPCR verification cohort 2 come from different centers.

[0089] In Table 1, RT-qPCR verification cohort 1 & 2 is the data set of RT-qPCR verification cohort 1 and 2 combined.

[0090] 2. Isolation and purification of extracellular vesicles

[0091] Venous blood is collected with a vacuum blood collection tube (coagulant / separation gel), and serum is separated by centrifugation at 2000xg for 10 min at 10-30℃ within 6 hours after blood collection. The serum sample treated as above should be immediately used for detection, or stored at -70℃ or below for no more than 12 months, and repeated freezing and thawing is strictly prohibited. GlyExo-Capture extracellular vesicle extraction reagent is used to isolate extracellular vesicles, and the magnetic beads used are "a lectin-magnetic carrier coupling complex for isolating glycosylated extracellular vesicles in clinical samples" (application number: 202010060055.7).

[0092] 3. Extraction of extracellular vesicle miRNA and NGS sequencing

[0093] Use The total RNA of the extracellular vesicles is extracted by the Mini Kit and evaluated by the high-sensitivity RNA kit of the Qsep100 full-automatic nucleic acid analysis system. Then, the small RNA of the extracellular vesicles is transcribed into a cDNA library by using the Illumina NEBNext small RNA library preparation kit, and the E-Gel Power Snap electrophoresis system and E-Gel SizeSelect II gel are used to select the library of the target size fragments. After checking the quality and concentration of the cDNA library, 75nt, single-end sequencing is performed on the Illumina NextSeq 550 sequencing system, and the sequencing data of a single library is greater than 10M reads.

[0094] 4. Constructing a miRNA interaction network and analyzing NGS sequencing data, obtaining characteristic miRNA pairs of pancreatic cancer by single factor screening and genetic algorithm screening

[0095] (1) Construction of a miRNA interaction network

[0096] 1) Obtain the action targets of miRNAs from the miRTarBase database;

[0097] 2) Based on the action target information of miRNAs obtained in step 1), screen the transcription factors that can be used as the action targets of miRNAs from the human transcription factor database hTFtarget and the AnimalTFDB;

[0098] 3) Based on the transcription factors screened in step 2), obtain the further regulated miRNAs by bioinformatics methods or public databases, thereby constructing the action relationship of miRNA-TF-miRNA, and obtaining the miRNA interaction network.

[0099] (2) Analysis of NGS sequencing data and preparation of miRNA quantitative data

[0100] By analyzing the above NGS sequencing data, the miRNA quantitative data of each sample in the comprehensive sample set containing disease samples and control samples is obtained:

[0101] First, the original sequencing file in fastq format is needed for quality control; for the quality control results, the cutadapt software is used to remove the adapters and low-quality reads; for the qualified data, the exceRpt small RNA analysis process is used for annotation and quantification, thereby obtaining the expression matrix; according to the expression level of the expression matrix, the miRNAs with too low counts are filtered out (which needs to be analyzed according to specific circumstances), and the corrected miRNA quantitative data is obtained.

[0102] (3) Construction of miRNA pair expression ratio feature

[0103] Based on the constructed miRNA interaction network and prepared miRNA quantification data, the expression ratio of the miRNA pair in each sample was calculated.

[0104] In order to avoid the denominator being 0, when calculating the expression ratio of the miRNA pair, the denominator was uniformly processed by adding 1, and the calculation formula was as follows:

[0105] miRNA_a / miRNA_b = counts a / (counts b +1)

[0106] (4) Screening of characteristic miRNA pairs

[0107] Through single factor screening and genetic algorithm screening, the characteristic miRNA pairs of pancreatic cancer were obtained, and the process was basically as follows:

[0108] i) Single factor screening: comparing the expression ratio of each miRNA pair in the disease biological sample group relative to the normal biological sample group, using python's scipy.stats.ttest_ind to calculate p value, and using statsmodels.stats.multitest.fdrcorrection to correct p value, for the corrected p value p-adjusted, based on the threshold 0.05 for screening;

[0109] ii) For the screened miRNA pairs, the log2FoldChange of the change fold of the expression ratio in the disease biological sample group relative to the expression ratio in the normal biological sample group was calculated, and according to the actual situation, the appropriate threshold of log2FoldChange was selected to further screen the appropriate target;

[0110] iii) Using genetic algorithm to further screen 100 times.

[0111] The serum samples of 78 pancreatic cancer patients and 38 benign disease patients were collected as NGS queue (as shown in Table 1), and the miRNA expression in glycosylated extracellular vesicles was measured by NGS to construct paired miRNA. After the above screening program, 39 miRNA pairs were screened. The expression ratio heat map of the 39 paired miRNAs in the NGS queue was drawn, see Figure 1 .

[0112] Figure 1It is shown that the cancer samples and the benign samples have obvious clustering effect on the heat map, that is, the expression ratio of the 39 paired miRNAs performs well on the NGS queue.

[0113] Example 2: Verification of 39 paired miRNAs in RT-qPCR data

[0114] Serum samples of 103 pancreatic cancer patients, 66 benign disease patients and 56 healthy people were collected as the RT-qPCR verification queue (as shown in Table 1). The CT values of the 39 paired miRNAs obtained in Example 1 were measured by the method of RT-qPCR to confirm whether the differences still exist at the level of RT-qPCR.

[0115] The method of RT-qPCR is as follows:

[0116] 1) The total RNA of extracellular vesicles was extracted by using the QIAamp® Circulating Nucleic Acid Mini Kit, and the first strand cDNA synthesis was performed by reverse transcription, and the reaction system and reaction conditions are as follows: Reaction system and volume:

[0117]

[0118] Reaction system Volume (μL) Reverse transcription primer (20 μM) 1 5x Reverse transcription buffer 2 Poly A polymerase (5 U / μL) 0.5 Reverse transcriptase (200 U / μL) 0.5 ATP (10 mM) 0.5 dNTP (10 mM) 0.5 RNA template 5 Total volume 10

[0119] Reaction conditions and time:

[0120] Reaction condition Time (minute) 42℃ 15 85℃ 1 4℃ ∞

[0121] Then, the miRNA RT-qPCR reaction was performed on the ABI 7500 real-time fluorescence quantitative PCR system, and the reaction system and reaction conditions are as follows:

[0122] Reaction system and volume:

[0123] Reaction system Volume (μL) 2x RT-qPCR reaction solution 12.5 Forward primer (10 μM) 2 Reverse primer (10 μM) 2 Probe (10 μM) 1 Rox 0.5 Nuclease-free water 2 cDNA 5 Total volume 25

[0124] Reaction conditions and time:

[0125]

[0126]

[0127] Among them, the primers can be designed according to the conventional method.

[0128] The LDA graph was drawn by the CT value detection result of RT-qPCR, and the LDA graph result of the 39 paired miRNAs in PCR data is as follows: Figure 2 It can be known that the 39 paired miRNAs can obviously distinguish pancreatic cancer, benign disease and healthy samples. Therefore, this.

[0129] ​Example 3: Further screening and validation of the characteristic miRNAs of pancreatic cancer (preferred combination one)

[0130] Based on the RT-qPCR confirmed cohort of pancreatic cancer, benign disease, healthy samples, recursive feature elimination was performed on 39 paired miRNAs using logistic regression model (with diagnostic performance as the screening index, by removing one or more markers, the marker combination with the best diagnostic performance was screened out), and 2 paired miRNAs (preferred combination one) were obtained:

[0131] hsa-miR-98 / hsa-miR-25

[0132] hsa-miR-378i / hsa-miR-150

[0133] Using RT-qPCR quantitative data (RT-qPCR method refers to Example 2), the box plot of preferred combination one (2 paired miRNAs) was displayed.

[0134] Among them, in the RT-qPCR procedure of preferred combination one, the primer sequences used are as follows:

[0135]

[0136] The box plot is shown in Figure 3 : Pancreatic cancer has significant differences (p<0.05) compared with benign disease and healthy samples, especially early pancreatic cancer also has significant differences (p<0.05) compared with benign disease and healthy samples.

[0137] Based on the preferred combination one (hsa-miR-98 / hsa-miR-25, hsa-miR-378i / hsa-miR-150), a pancreatic cancer diagnosis model was constructed, and its diagnostic performance was verified: the RT-qPCR confirmed cohort shown in Table 1 was divided into a training set (24 cases of advanced pancreatic cancer, 38 cases of early pancreatic cancer, 39 cases of benign disease, 34 cases of health) and a test set (16 cases of advanced pancreatic cancer, 25 cases of early pancreatic cancer, 27 cases of benign disease, 22 cases of health) according to 6:4, a model was established using logistic regression on the training set, and was verified on the test set, and was verified on other external verification sets (RT-qPCR verification cohorts 1, 2, 1&2).

[0138] A logistic regression model was established for the 2 paired miRNAs (combination one) of the training set, and the cutoff was determined according to the maximum principle of the training set Youden index. The overall performance of the model was evaluated by the AUC, sensitivity and specificity of the ROC curve.

[0139] The results of the AUC, sensitivity and specificity of the ROC curve are shown in Figure 4 .

[0140] Figure 4 The results show that the preferred combination one (hsa-miR-98 / hsa-miR-25, hsa-miR-378i / hsa-miR-150) can achieve AUC of 0.867 (sensitivity 0.806, specificity 0.808), 0.963 (sensitivity 0.878, specificity 0.918), 0.941 (sensitivity 0.849, specificity 0.898), 0.927 (sensitivity 0.848, specificity 0.845), 0.934 (sensitivity 0.848, specificity 0.873) in the training set, test set, validation set 1 (i.e. validation cohort 1), validation set 2 (i.e. validation cohort 2), validation set 1 & 2 (i.e. validation cohort 1 & 2) respectively in distinguishing pancreatic cancer from other samples (benign disease, healthy), indicating that the preferred combination one (hsa-miR-98 / hsa-miR-25, hsa-miR-378i / hsa-miR-150) not only has excellent diagnostic performance, but also has stable diagnostic performance in different data centers, and there is no situation of unstable results.

[0141] In order to better evaluate the diagnostic performance of the model for early pancreatic cancer, the diagnostic sensitivity and specificity data of the middle and late stages of pancreatic cancer and early pancreatic cancer in different data sets were obtained under the condition of the best cutoff, and the results are shown in Table 2.

[0142] Table 2, sensitivity and specificity of the preferred combination one in different data sets

[0143]

[0144] Table 2 shows that in the evaluation model, the recognition sensitivity of the middle and late stages of pancreatic cancer and early pancreatic cancer in different data sets is relatively high, the recognition sensitivity of the middle and late stages of pancreatic cancer is above 0.750 (i.e. 75.0%), the recognition sensitivity of early pancreatic cancer is above 0.763 (i.e. 76.3%), and it can specifically distinguish other samples (benign disease, healthy), and the correct rate of pancreatic cancer diagnosis is high.

[0145] The sample score distribution box plot of this embodiment is shown in Figure 5 The results show that there is a significant difference between the middle and late stages of pancreatic cancer, early pancreatic cancer and other samples (benign disease, healthy) and the distinguishing effect is good.

[0146] In summary, the preferred combination one (hsa-miR-98 / hsa-miR-25, hsa-miR-378i / hsa-miR-150) of the present application can effectively distinguish pancreatic cancer from other samples (benign disease, healthy), and the recognition sensitivity of early pancreatic cancer is also high, and stable diagnostic results are presented in different source data sets, which can improve the accuracy of early pancreatic cancer, and has important significance for pancreatic cancer diagnosis, efficacy or prognosis evaluation.

[0147] Example 4: Further screening and verification of characteristic miRNAs of pancreatic cancer (preferred combination two)

[0148] Based on the PCR confirmation of the pancreatic cancer, benign and healthy samples in the queue, 39 paired miRNAs were subjected to recursive feature elimination using the SVM model (diagnostic efficiency was used as the screening index, and by removing one or more markers, the marker combination with the best diagnostic performance was screened out), and four paired miRNAs (preferred combination two) were obtained:

[0149] hsa-miR-501 / hsa-miR-150

[0150] hsa-miR-378i / hsa-miR-155

[0151] hsa-miR-193a / hsa-miR-155

[0152] hsa-miR-150 / hsa-miR-1180

[0153] Using RT-qPCR quantitative data (RT-qPCR method refers to Example 2), the box plot of the preferred combination two (4 paired miRNAs) was displayed.

[0154] Among them, in the RT-qPCR procedure of the preferred combination two, the primer sequences used are as follows:

[0155]

[0156] The box plot is shown in Figure 6 Pancreatic cancer has significant differences (p<0.05) compared with benign disease and healthy samples.

[0157] Based on the preferred combination two (hsa-miR-501 / hsa-miR-150, hsa-miR-378i / hsa-miR-155, hsa-miR-193a / hsa-miR-155, hsa-miR-150 / hsa-miR-1180), a pancreatic cancer diagnosis model was constructed, and its diagnostic performance was verified: the RT-qPCR confirmed cohort shown in Table 1 was divided into a training set (24 cases of advanced pancreatic cancer, 38 cases of early pancreatic cancer, 39 cases of benign disease, and 34 cases of health) and a test set (16 cases of advanced pancreatic cancer, 25 cases of early pancreatic cancer, 27 cases of benign disease, and 22 cases of health) according to a ratio of 6:4, a model was established on the training set using logistic regression, and the model was verified on the test set and on other external verification sets (RT-qPCR verification cohorts 1, 2, 1&2).

[0158] A logistic regression model was established for the four paired miRNAs (combination two) of the training set, and the cutoff was determined according to the maximum Youden index of the training set. The overall performance of the model was evaluated by the AUC, sensitivity, and specificity of the ROC curve.

[0159] The AUC, sensitivity, and specificity results of the ROC curve are shown in Table 3. Figure 7

[0160] Figure 7 The results show that the preferred combination two (hsa-miR-501 / hsa-miR-150, hsa-miR-378i / hsa-miR-155, hsa-miR-193a / hsa-miR-155, hsa-miR-150 / hsa-miR-1180) can achieve AUCs of 0.888 (sensitivity 0.806, specificity 0.849), 0.879 (sensitivity 0.780, specificity 0.816), 0.937 (sensitivity 0.935, specificity 0.805), 0.944 (sensitivity 0.834, specificity 0.900), and 0.931 (sensitivity 0.873, specificity 0.851) in the training set, test set, verification set 1 (i.e., verification cohort 1), verification set 2 (i.e., verification cohort 2), and verification set 1&2 (i.e., verification cohort 1&2), respectively, when distinguishing pancreatic cancer from other samples (benign disease, healthy samples), indicating that the preferred combination two (hsa-miR-501 / hsa-miR-150, hsa-miR-378i / hsa-miR-155, hsa-miR-193a / hsa-miR-155, hsa-miR-150 / hsa-miR-1180) not only has excellent diagnostic performance, but also has stable diagnostic performance in different data centers, and there is no situation of unstable results.

[0161] ​In order to better evaluate the model for early diagnosis of pancreatic cancer, the sensitivity and specificity of the model for early and late pancreatic cancer in different data sets were obtained under the optimal cutoff, and the results are shown in Table 3.

[0162] Table 3, sensitivity and specificity of the preferred combination two in different data sets

[0163]

[0164] The results in Table 3 show that the sensitivity of the evaluation model for early and late pancreatic cancer in different data sets is high, the sensitivity for early and late pancreatic cancer is above 0.875 (i.e. 87.5%), and the sensitivity for early and late pancreatic cancer is as high as 0.72 (i.e. 72%), which can specifically distinguish other samples (benign disease, health), and the correct rate of early diagnosis of pancreatic cancer is high.

[0165] The sample score distribution box plot of the embodiment is shown in Figure 8 The results show that the model score has significant difference and good discrimination effect between early and late pancreatic cancer and other samples (benign disease, health).

[0166] In summary, the preferred combination two (hsa-miR-501 / hsa-miR-150, hsa-miR-378i / hsa-miR-155, hsa-miR-193a / hsa-miR-155, hsa-miR-150 / hsa-miR-1180) of the application has high sensitivity for early and late pancreatic cancer, and stable diagnostic results are presented in different data sets, which can improve the accuracy of early pancreatic cancer and has important significance for diagnosis, efficacy or prognosis evaluation of pancreatic cancer.

[0167] Example 5: Construction of pancreatic cancer diagnosis model based on paired miRNA preferred combination one and pancreatic cancer index CA19-9, and verification of its diagnostic performance

[0168] In this embodiment, the preferred combination one (hsa-miR-98 / hsa-miR-25, hsa-miR-378i / hsa-miR-150) and pancreatic cancer index CA19-9 were used for joint modeling (wherein the expression level of CA19-9 was detected by carbohydrate antigen 19-9 detection kit (magnetic microparticle chemiluminescence immunoassay method), which was obtained from Beijing Rejet Biotechnology Co., Ltd. and can be purchased), and the sample information of the training set and the verification set used was consistent with that of Example 3 (the detection of CA19-9 was also detected by the reagent kit provided by the applicant and entrusted to the corresponding hospital). The specific modeling method is as follows:

[0169] The RT-qPCR validation cohort shown in Table 1 was divided into a training set (24 cases of pancreatic cancer in advanced stage, 38 cases of pancreatic cancer in early stage, 40 cases of benign disease, 34 cases of health) and a test set (16 cases of pancreatic cancer in advanced stage, 25 cases of pancreatic cancer in early stage, 26 cases of benign disease, 22 cases of health) in a ratio of 6:4, a model was established using logistic regression on the training set, and was verified on the test set and other external verification sets (RT-qPCR verification cohorts 1, 2, 1 & 2).

[0170] A logistic regression model was established for the two paired miRNAs (combination one) and the pancreatic cancer indicator CA19-9 of the training set, and the cutoff was determined according to the maximum Youden index of the training set. The overall performance of the model was evaluated by the AUC, sensitivity and specificity of the ROC curve.

[0171] The AUC, sensitivity and specificity results of the ROC curve are shown in Table 3. Figure 9

[0172] Figure 9 The results show that the preferred combination one (hsa-miR-98 / hsa-miR-25, hsa-miR-378i / hsa-miR-150) and the pancreatic cancer indicator CA19-9 combined can achieve an AUC of 0.942 (sensitivity 0.839, specificity 0.959) in the training set, 0.975 (sensitivity 0.902, specificity 0.896) in the test set, 0.958 (sensitivity 0.839, specificity 0.924) in verification set 1 (i.e. verification cohort 1), 0.965 (sensitivity 0.795, specificity 0.936) in verification set 2 (i.e. verification cohort 2), and 0.961 (sensitivity 0.811, specificity 0.930) in verification set 1 & 2 (i.e. verification cohort 1 & 2) when distinguishing pancreatic cancer from other samples (benign disease, health), indicating that the preferred combination one (hsa-miR-98 / hsa-miR-25, hsa-miR-378i / hsa-miR-150) and the pancreatic cancer indicator CA19-9 not only further improve the diagnostic performance, but also the diagnostic performance of the results in different data centers is stable, and there is no situation of unstable results.

[0173] In order to better evaluate the diagnostic performance of the model for pancreatic cancer in early stage, the diagnostic sensitivity and specificity data of pancreatic cancer in advanced stage and pancreatic cancer in early stage in different data sets were obtained under the condition of the best cutoff, and the results are shown in Table 4.

[0174] Table 4, sensitivity and specificity of the preferred combination one and the pancreatic cancer indicator CA19-9 combined in different data sets

[0175] ​

[0176] The results in Table 4 show that in the evaluation model of the preferred combination one and the pancreatic cancer index CA19-9 for joint diagnosis, the identification sensitivity of the middle and late stages of pancreatic cancer and the early stage of pancreatic cancer in different data sets is relatively high, the identification sensitivity of the middle and late stages of pancreatic cancer is above 0.811 (i.e. 81.1%), the identification sensitivity of the early stage of pancreatic cancer is above 0.75 (i.e. 75.0%), and other samples (benign diseases, health) can be specifically distinguished, and the pancreatic cancer diagnosis accuracy is high.

[0177] Example 6: Constructing a pancreatic cancer diagnosis model based on the preferred combination two and the pancreatic cancer index CA19-9 and verifying the diagnosis performance thereof

[0178] In this embodiment, the preferred combination two (hsa-miR-501 / hsa-miR-150, hsa-miR-378i / hsa-miR-155, hsa-miR-193a / hsa-miR-155, hsa-miR-150 / hsa-miR-1180) and the pancreatic cancer index CA19-9 are used for joint modeling (wherein the expression level of CA19-9 is detected by a carbohydrate antigen 19-9 detection kit (magnetic microparticle chemiluminescence immunoassay method), which is obtained from Beijing Reelang Biotechnology Co., Ltd. and can be purchased), and the specific modeling method is as follows:

[0179] The RT-qPCR confirmation queue shown in Table 1 is divided into a training set (24 cases of middle and late stages of pancreatic cancer, 38 cases of early stages of pancreatic cancer, 40 cases of benign diseases, and 34 cases of health) and a test set (16 cases of middle and late stages of pancreatic cancer, 25 cases of early stages of pancreatic cancer, 26 cases of benign diseases, and 22 cases of health) according to a ratio of 6:4, a model is established on the training set by using a logistic regression, the model is verified on the test set, and the model is verified on other external verification sets (RT-qPCR verification queues 1, 2, and 1&2).

[0180] A logistic regression model is established for the two paired miRNAs (combination two) and the pancreatic cancer index CA19-9 of the training set, and the cutoff is determined according to the principle of maximizing the training set Youden index. The overall performance of the model is evaluated by the AUC, sensitivity, and specificity of the ROC curve.

[0181] The AUC, sensitivity, and specificity results of the ROC curve are shown in Table 6. Figure 10

[0182] Figure 10 ​The results show that the preferred combination two (hsa-miR-501 / hsa-miR-150, hsa-miR-378i / hsa-miR-155, hsa-miR-193a / hsa-miR-155, hsa-miR-150 / hsa-miR-1180) and the pancreatic cancer indicator CA19-9 in combination can achieve an AUC of 0.963 (sensitivity 0.952, specificity 0.878), 0.970 (sensitivity 1.000, specificity 0.792), 0.955 (sensitivity 0.946, specificity 0.814), 0.978 (sensitivity 0.934, specificity 0.891), and 0.964 (sensitivity 0.939, specificity 0.851) in the training set, the test set, the validation set 1 (i.e., validation cohort 1), the validation set 2 (i.e., validation cohort 2), and the validation set 1 & 2 (i.e., validation cohort 1 & 2) respectively when distinguishing pancreatic cancer from other samples (benign disease, healthy), indicating that the preferred combination two (hsa-miR-501 / hsa-miR-150, hsa-miR-378i / hsa-miR-155, hsa-miR-193a / hsa-miR-155, hsa-miR-150 / hsa-miR-1180) and the pancreatic cancer indicator CA19-9 not only further improve the diagnostic efficiency, but also can stabilize the diagnostic efficiency of the results in different data centers, without the situation of unstable results.

[0183] In order to better evaluate the diagnostic efficiency of the model for early pancreatic cancer, the diagnostic sensitivity and specificity data of the middle and late stages of pancreatic cancer and early pancreatic cancer in different data sets were obtained under the condition of the best cutoff, and the results are shown in Table 5.

[0184] Table 5, sensitivity and specificity of the preferred combination two and the pancreatic cancer indicator CA19-9 in combination in different data sets

[0185]

[0186] Table 5 shows that in the evaluation model of the preferred combination two and the pancreatic cancer indicator CA19-9 in combination, the recognition sensitivity of the middle and late stages of pancreatic cancer and early pancreatic cancer in different data sets is relatively high, the recognition sensitivity of the middle and late stages of pancreatic cancer is above 0.937 (i.e., 93.7%), the recognition sensitivity of early pancreatic cancer is above 0.925 (i.e., 92.5%), and the other samples (benign disease, healthy) can be specifically distinguished, and the pancreatic cancer diagnosis accuracy is high.

[0187] Comparative Example 1:

[0188] Although the advent of big data has improved the convenience of marker screening, there is still a great challenge in screening marker combinations that can be applied to different data centers (different hospitals). Most of the time, the marker combinations with good diagnostic performance obtained by screening are difficult to apply to other data centers (other hospitals or institutions). For example, the following pair of miRNA combinations three:

[0189] hsa-miR-501 / hsa-miR-25

[0190] hsa-miR-20b / hsa-miR-150

[0191] hsa-miR-193a / hsa-miR-150

[0192] Using RT-qPCR quantitative data (RT-qPCR method, refer to Example 2), the box plot of combination three (3 pairs of miRNA) is shown.

[0193] Among them, the RT-qPCR program of preferred combination three uses the following primer sequences:

[0194]

[0195] The box plot is shown in Figure 11 : Pancreatic cancer has significant differences (p < 0.05) compared with benign disease and healthy samples.

[0196] Based on the combination three (hsa-miR-501 / hsa-miR-25, hsa-miR-20b / hsa-miR-150, hsa-miR-193a / hsa-miR-150), a pancreatic cancer diagnosis model is constructed, and its diagnostic performance is verified: The RT-qPCR confirmation queue shown in Table 1 is divided into a training set (24 cases of advanced pancreatic cancer, 38 cases of early pancreatic cancer, 39 cases of benign disease, and 34 cases of health) and a test set (16 cases of advanced pancreatic cancer, 25 cases of early pancreatic cancer, 27 cases of benign disease, and 22 cases of health) according to 6:4. The model is established using logistic regression on the training set, verified on the test set, and verified on other external verification sets (RT-qPCR verification queues 1, 2, 1&2).

[0197] A logistic regression model is established for the 3 pairs of miRNA (combination three) of the training set, and the cutoff is determined according to the maximum principle of the training set. The overall performance of the model is evaluated by the AUC, sensitivity and specificity of the ROC curve.

[0198] The AUC, sensitivity and specificity of the ROC curve are shown in Figure 12 .

[0199] Figure 12 The results show that the combination three (hsa-miR-501 / hsa-miR-25, hsa-miR-20b / hsa-miR-150, hsa-miR-193a / hsa-miR-150) can achieve AUC of 0.886 (sensitivity 0.871, specificity 0.795) and 0.881 (sensitivity 0.829, specificity 0.776) in the training set and the test set respectively in distinguishing pancreatic cancer from other samples (benign disease, healthy), that is, it presents good diagnostic performance in one data center (RT-qPCR confirmation queue), but the AUC is only 0.607 (sensitivity 0.591, specificity 0.669), 0.696 (sensitivity 0.589, specificity 0.800), and 0.717 (sensitivity 0.669, specificity 0.751) in the validation set 1 (i.e. validation queue 1), the validation set 2 (i.e. validation queue 2), and the validation set 1 & 2 (i.e. validation queue 1 & 2) respectively, indicating that although the combination three (hsa-miR-501 / hsa-miR-25, hsa-miR-20b / hsa-miR-150, hsa-miR-193a / hsa-miR-150) presents good diagnostic performance in the randomly divided training set and test set in one data center (RT-qPCR confirmation queue), the diagnostic performance in other data centers (validation set 1 and 2 (i.e. RT-qPCR validation queue 1 and 2) from other hospitals for third-party detection) decreases significantly, that is, the diagnostic result is unstable and difficult to apply to the clinic.

[0200] The diagnostic efficiency of the model for early pancreatic cancer is also evaluated, and the sensitivity and specificity data of the early and advanced pancreatic cancer in the samples in different data sets are obtained under the condition of the best cutoff, and the results are shown in Table 6.

[0201] Table 6, sensitivity and specificity of combination three in different data sets

[0202]

[0203] Table 4 shows that in the evaluation model, the sensitivity and specificity of the early and advanced pancreatic cancer in the randomly divided training set and test set in the same data center are relatively high, especially the recognition sensitivity of the advanced pancreatic cancer in the training set is as high as 0.958 (95.8%), but the sensitivity in other data centers (validation set 1 and 2 from other hospitals for third-party detection) decreases significantly (the recognition sensitivity is as low as less than 0.5), and the diagnostic accuracy also decreases significantly, that is, the diagnostic result is unstable and difficult to apply to the clinic.

[0204] Comparative Example 2:

[0205] As the genes with relatively constant expression in various tissues and cells, the internal reference genes are generally used as the reference for detecting the expression level of the target genes, and play a certain correction role. Common internal reference genes include 18S rRNA, 28S rRNA, U6, etc. Although the expression of the internal reference genes is relatively constant, the expression under different conditions such as physiological or pathological changes can be unstable, which can greatly affect the analysis of the expression level of the target genes, and lead to great difficulty in the selection and application of diagnostic markers. For example, the miRNA pair of hsa-miR-98 / hsa-miR-25 can significantly distinguish pancreatic cancer from other samples (see the box plot of Example 3), but when the conventional internal reference gene U6 is used to correct hsa-miR-98 and hsa-miR-25 respectively, hsa-miR-98 and hsa-miR-25 are used as markers respectively, it is found that hsa-miR-98 can distinguish pancreatic cancer from healthy samples (p<0.05), but cannot distinguish pancreatic cancer from benign pancreatic diseases (see the box plot of Example 4), and hsa-miR-25 can distinguish pancreatic cancer from benign pancreatic diseases (p<0.05), but cannot distinguish pancreatic cancer from healthy samples (see the box plot of Example 5). Figure 3 Figure 13

[0206] The RT-qPCR quantitative data (the RT-qPCR method refers to Example 2) were used to perform the box plot display of hsa-miR-98, hsa-miR-25, and hsa-miR-98 / hsa-miR-25, wherein the box plot of hsa-miR-98 and hsa-miR-25 representing the expression level of hsa-miR-98 and hsa-miR-25 corrected by the internal reference gene U6 is as shown in FIG. 1, and the box plot of hsa-miR-98 / hsa-miR-25 is as shown in FIG. 2. Figure 13 Figure 3

[0207] The primer sequences used for hsa-miR-98, hsa-miR-25, and U6 are as follows:

[0208]

[0209] Figure 13 The results shown in FIGS. 1 to 3 indicate that when the expression levels of hsa-miR-98 and hsa-miR-25 are corrected by U6 as the internal reference gene, the expression level of the pancreatic cancer samples is sometimes not significantly correlated with the expression level of the other samples (for example, hsa-miR-98 cannot significantly distinguish pancreatic cancer from benign patients, and hsa-miR-25 cannot distinguish pancreatic cancer from healthy people), which indicates that when the expression levels of hsa-miR-98 and hsa-miR-25 are corrected by the common internal reference gene (for example, U6), the diagnostic performance is unstable. ​​​​

[0210] In summary, the preferred combination one (hsa-miR-98 / hsa-miR-25, hsa-miR-378i / hsa-miR-150) and the preferred combination two (hsa-miR-501 / hsa-miR-150, hsa-miR-378i / hsa-miR-155, hsa-miR-193a / hsa-miR-155, hsa-miR-150 / hsa-miR-1180) obtained by continuously adjusting the screening method or screening condition are the marker combinations suitable for different data centers, having good diagnostic performance and diagnostic stability, and when combined with the pancreatic cancer index CA19-9, the diagnostic performance can be further improved, and have good clinical application prospects.

[0211] Finally, it should be noted that: the above examples are only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the foregoing examples, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing examples, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A diagnostic or prognostic marker combination for pancreatic cancer, characterized in that: The diagnostic or prognostic marker for pancreatic cancer comprises a marker combination consisting of at least two or at least four of the following miRNAs: hsa-miR-98, hsa-miR-25, hsa-miR-378i and hsa-miR-150.

2. The diagnostic or prognostic marker combination for pancreatic cancer according to claim 1, characterized in that: The diagnostic or prognostic markers for pancreatic cancer include one or more of the following miRNA combinations: hsa-miR-98 and hsa-miR-25, optionally, hsa-miR-25 is used to calibrate the relative expression of hsa-miR-98; hsa-miR-378i and hsa-miR-150, optionally, hsa-miR-150 is used to calibrate the relative expression level of hsa-miR-378i.

3. The diagnostic or prognostic marker combination for pancreatic cancer according to claim 2, characterized in that: The diagnostic or prognostic marker for pancreatic cancer comprises a combination of the following miRNAs: hsa-miR-98, hsa-miR-25, hsa-miR-378i, hsa-miR-150; Optionally, the following miRNA pairs are included: hsa-miR-98 / hsa-miR-25, hsa-miR-378i / hsa-miR-150.

4. The diagnostic or prognostic marker combination for pancreatic cancer according to any one of claims 1 to 3, characterized in that: The miRNA in the miRNA pair is a miRNA in serum or plasma; And / or, the miRNA in the miRNA pair is a miRNA in serum or plasma glycosylated extracellular vesicles.

5. A diagnostic or prognostic marker combination for pancreatic cancer, characterized in that: A combination of a diagnostic or prognostic marker for pancreatic cancer comprising a CA19-9 protein and any one of claims 1 to 4.

6. An analysis system for predicting pancreatic cancer or evaluating the prognosis of pancreatic cancer, characterized in that: The system includes a data analysis module for analyzing the expression levels of a marker combination of a target object to be predicted and calculating the expression level ratio of a miRNA pair or directly using the expression level ratio of the miRNA pair as an input feature, and further for calculating a prediction value of whether the target object has pancreatic cancer based on the expression level ratio of the miRNA pair; the input feature is selected from at least one of the following expression level ratios of miRNA pairs: hsa-miR-98 / hsa-miR-25; hsa-miR-378i / hsa-miR-150.

7. The analysis system according to claim 6, characterized in that The data analysis module stores a prediction model for outputting a probability value of pancreatic cancer, wherein the prediction model is constructed by using the ratio of the expression levels of the miRNA pairs in pancreatic cancer, benign, and healthy samples as input variables and whether the sample is pancreatic cancer as the dependent variable, performing multivariate logistic regression calculation to obtain a prediction model, and testing the model using a validation set; and / or, when the detection method is RT-qPCR, the expression ratio is the CT difference of the miRNA pair; And / or, the input feature further includes the expression level of CA19-9 protein of the target object to be predicted.

8. Use of a reagent for detecting the expression level of the diagnostic or prognostic marker combination for pancreatic cancer according to any one of claims 1 to 5 in the preparation of a kit or analysis system for pancreatic cancer diagnosis, efficacy or prognosis assessment, or related drug evaluation.

9. Use of a reagent for detecting the relative expression level of a combination of diagnostic or prognostic markers for pancreatic cancer described in any one of items 1 to 5 in serum or plasma glycosylated extracellular vesicles in the preparation of a kit or analytical system for pancreatic cancer diagnosis, efficacy or prognosis assessment, or related drug evaluation, wherein the marker combination is optionally selected from the following miRNAs: The expression level of hsa-miR-98 relative to hsa-miR-25, and the expression level of hsa-miR-378i relative to hsa-miR-150.

10. A kit for pancreatic cancer diagnosis, efficacy or prognosis evaluation, or related drug evaluation, characterized in that: The kit comprises the diagnostic or prognostic marker combination for pancreatic cancer according to any one of claims 1 to 5 and / or a detection reagent thereof.

11. The kit according to claim 10, characterized in that In pancreatic cancer patients, the expression ratio of the miRNA pair in their serum glycosylated extracellular vesicles is significantly different from that in healthy subjects.

12. The kit according to claim 10 or 11, characterized in that The detection reagents include primers and probes of the corresponding miRNAs, wherein the sequences of the primers and probes of the miRNAs are as follows: The nucleotide sequence of the upstream primer of hsa-miR-150 is shown in SEQ ID NO: 1; The nucleotide sequence of the upstream primer of hsa-miR-25 is shown in SEQ ID NO: 2; The nucleotide sequence of the upstream primer of hsa-miR-378i is shown in SEQ ID NO: 3; The nucleotide sequence of the upstream primer of hsa-miR-98 is shown in SEQ ID NO:4; The nucleotide sequence of the probe is shown in SEQ ID NO: 5; The reverse primer is a common downstream primer, and its nucleotide sequence is shown in SEQ ID NO:

10.

13. The kit according to claim 12, characterized in that Primers and probes that do not contain conventional internal reference genes.