A set of gene mutations and their use in diagnosing pancreaticobiliary cancer

Abstract

The application belongs to the field of biological medicine, and particularly relates to a group of gene mutations and application thereof in diagnosis of pancreaticobiliary duct cancer. Specifically, the application provides a group of gene mutations for detecting pancreaticobiliary duct cancer, wherein the gene mutations include one or more of AKT1, KRAS, APC, NRAS, ARID1A, PIK3CA, AXIN1, PPP2R1A, BAP1, PTEN, BRAF, SMAD4, CDKN2A, TERT, TP53, EGFR, FBXW7, FGFR2, HRAS, IDH1 and IDH2.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of biological medicine, and particularly relates to a group of gene mutations and application thereof in diagnosis of pancreatobiliary tract cancer. BACKGROUND

[0002] Pancreatobiliary tract cancer (Pancreatobiliary tract cance) includes bile tract cancer (BTC) and pancreatic cancer.

[0003] Bile duct cancer originates from cholangiocytes in different anatomical locations, such as intrahepatic, extrahepatic and gallbladder, or can directly originate from hepatocytes. Bile duct cancer includes cholangiocarcinoma (CCA), gallbladder cancer (GBC) and ampullary cancer. Cholangiocarcinoma is the second most common primary liver cancer, accounting for about 3% of all gastrointestinal tumors. Cholangiocarcinoma is further divided into intrahepatic (iCCA), perihilar (pCCA) or distal (dCCA) cholangiocarcinoma. Although bile duct cancer is relatively uncommon worldwide, its global incidence has increased rapidly in recent years, and the incidence is highest in East and South Asia (such as Thailand and China) and parts of South America. Risk factors for bile duct cancer include primary sclerosing cholangitis (PSC), liver flukes, fibropolycystic liver disease (such as bile duct adenomas and bile duct papillomatosis), biliary and gallbladder stones, viral hepatitis and chemical carcinogen exposure. Pancreatic cancer is one of the highest mortality rates of cancer worldwide, and is the seventh leading cause of cancer death worldwide due to its poor prognosis, and specifically includes pancreatic head cancer, pancreatic tail cancer, diffuse cancer, etc.

[0004] Diagnosis of pancreaticobiliary cancer is challenging. As early symptoms are nonspecific or even absent, most patients are diagnosed at an advanced stage. Late diagnosis contributes to the poor prognosis of BTC patients, with a 5-year overall survival of less than 20%. Patients with biliary strictures and jaundice can have cholangiocarcinoma, gallbladder cancer or pancreatic cancer, and it is difficult to distinguish between malignant and benign strictures (iatrogenic bile duct injury, primary sclerosing cholangitis (PSC) and choledocholithiasis). Conventionally, pancreaticobiliary cancer is diagnosed by a combination of various methods, including clinical examination, imaging, endoscopic procedures, pathological evaluation and biochemical tests (e.g. CA19-9). However, these methods have some limitations, for example, CA19-9 is not suitable for patients who are Lewis antigen negative (7% of the general population), and the sensitivity and specificity of the above methods are not satisfactory. It is reported that about 15-24% of patients who underwent surgery for malignant biliary strictures were eventually diagnosed as benign.

[0005] Therefore, there is an urgent need to develop a better detection method for diagnosing pancreaticobiliary cancer with high sensitivity, specificity and safety. SUMMARY

[0006] Patients with pancreaticobiliary cancer usually have poor clinical prognosis, with a 5-year overall survival rate of less than 20%. This is mainly related to late diagnosis. In addition, accurate preoperative differentiation between malignant and benign diseases can avoid unnecessary trauma. Therefore, there is an urgent need to develop a detection method for diagnosing malignant pancreaticobiliary cancer with high sensitivity, specificity and safety.

[0007] To achieve the above technical purposes, the present application provides the following technical solutions:

[0008] In a first aspect, the present application provides a set of gene mutations for detecting pancreaticobiliary cancer, wherein the gene mutations comprise one or more of AKT1, KRAS, APC, NRAS, ARID1A, PIK3CA, AXIN1, PPP2R1A, BAP1, PTEN, BRAF, SMAD4, CDKN2A, TERT, TP53, EGFR, FBXW7, FGFR2, HRAS, IDH1, and IDH2.

[0009] More specifically, as verified in the embodiments of the present application, the diagnosis of pancreaticobiliary tract cancer can be made more sensitively by using the presence of mutation in at least one of the following genes as a criterion: AKT1, KRAS, APC, NRAS, ARID1A, PIK3CA, AXIN1, PPP2R1A, BAP1, PTEN, BRAF, SMAD4, CDKN2A, TERT, TP53, EGFR, FBXW7, FGFR2, HRAS, IDH1, IDH2.

[0010] The term "pancreaticobiliary tract cancer" as used herein can also be referred to as "pancreatobiliary tract cancer" which includes bile tract cancer (BTC) and pancreatic cancer; the bile tract cancer includes cholangiocarcinoma (CCA), gallbladder cancer (GBC) and ampullary cancer; the pancreatic cancer includes head of pancreas cancer, tail of pancreas cancer and diffuse cancer.

[0011] Preferably, the non-cancer patient in the present application is a subject who has not been diagnosed with cancer for at least 12 months, and optionally, the non-cancer patient can have the following symptoms of non-malignant tumors: gallstones, biliary obstruction, biliary stenosis, pancreatic space-occupying lesions, pancreatic cysts, pancreatitis.

[0012] In another aspect, the present application provides the use of a reagent for detecting the mutation of the aforementioned genes in the preparation of a product for diagnosing pancreaticobiliary tract cancer.

[0013] More specifically, the diagnosis of pancreaticobiliary tract cancer refers to the differentiation between patients with malignant tumors (pancreaticobiliary tract cancer) and patients without malignant tumors, and the patients without malignant tumors can have benign diseases such as gallstones, biliary obstruction, biliary stenosis, pancreatic space-occupying lesions, pancreatic cysts, pancreatitis, etc.

[0014] Preferably, the product includes a kit, a chip, a diagnostic system, etc.

[0015] In an alternative embodiment, the reagent for detecting the genetic mutation comprises any of the following methods: TaqMan probe method, sequencing method, chip method, MALDI-TOF MS detection, PCR-RFLP, PCR-SSCP, AS-PCR, SNaPshot method, SNPlex typing system, SNPStream analysis system, Sequenom typing system, DHPLC, DGGE.

[0016] Preferably, the genetic mutation of the present application is detected from a sample of a subject.

[0017] Preferably, the sample is taken from the biliary tract.

[0018] Specifically, the sample comprises bile, exfoliated cells in the biliary tract, tissue sample.

[0019] More specifically, the exfoliated cells in the biliary tract, tissue sample can be a sample obtained by ERCP biopsy / brushing.

[0020] The term "biliary tract" is the general term for the ducts that carry bile from the liver to the duodenum. It is divided into intrahepatic and extrahepatic parts.

[0021] The term "ERCP" refers to the technique of inserting a duodenoscope to the descending part of the duodenum, finding the duodenal papilla, inserting a contrast catheter into the papilla opening through the biopsy channel, and taking x-ray photographs after injecting contrast agent to show the pancreatic and biliary ducts.

[0022] Preferably, the method of collecting bile includes but is not limited to duodenal drainage, gallbladder puncture, and direct surgical collection.

[0023] More preferably, the sample needs to be processed, and the processing includes the step of DNA extraction.

[0024] Preferably, the processing can also include steps of purification, quality control, etc.

[0025] Preferably, the subject includes a suspected pancreaticobiliary cancer patient.

[0026] Preferably, the genetic mutation of the present application can also be used simultaneously with other types of markers, including expression markers, methylation markers.

[0027] Preferably, the methylation markers comprise any one or more of: 3-OST-2, EBF3, RASSFl, APC, EYA4, RUNX3, BNIP3, FHIT, SALL3, CCND2, FOXEl, SEPT9, CD1D, GSTPl, SFRPl, CDH1, hMLH1, SLIT2, CDH13, KCNK12, SLIT3, CDKN2A, MGMT, SOX17, CDKN2B, NDRG4, TERT, CDOl, NPTX2, TFPI2, CLEC11, NXPH1, TIMP3, CNRIP1, PENK, TMEFF2 (HPP1), DAPK1, PRKCB VIM, DCLK1, PTCHD2, ZSCAN18, DLC1, RARbeta2 (RARB).

[0028] Preferably, the methylation markers comprise one or more of SOX17, 3-OST-2, NXPH1, SEPT9 and TERT.

[0029] Preferably, the methylation markers can be detected by methods known in the art, specifically for example: pyrosequencing, bisulfite sequencing, methylation microarray, qPCR, digital PCR, next generation sequencing, whole genome bisulfite sequencing, DNA enrichment, reduced representation bisulfite sequencing, HPLC, MassArray, methylation specific PCR, or a combination thereof.

[0030] In another aspect, the present application also provides a diagnostic system for diagnosing pancreaticobiliary cancer, which reports a diagnosis conclusion according to whether the subject has at least one of the mutations provided by the present application.

[0031] More specifically, the diagnosis criterion is that having at least one mutation provided by the present application is diagnosed as a patient with pancreaticobiliary cancer. More specifically, not having any of the mutations of the genes described in the present application is a non-cancer patient (may have other non-malignant tumor diseases).

[0032] Preferably, the system comprises:

[0033] (1) a sample collection and processing device for completing the following steps: collecting a sample from a subject, processing the sample;

[0034] (2) a nucleic acid sequence determination device;

[0035] (3) a calculation device for obtaining a diagnosis conclusion according to whether the subject has at least one of the mutations provided by the present application.

[0036] Preferably, the sample comprises bile, exfoliated cells in the biliary tract, a tissue sample.

[0037] Preferably, the processing comprises steps of purification, quality control, DNA extraction, etc.

[0038] Preferably, the nucleic acid sequence determining device can detect whether the subject has any gene mutation provided by the present application by implementing any one of the following methods: TaqMan probe method, sequencing method, chip method, MALDI-TOF MS detection, PCR-RFLP, PCR-SSCP, AS-PCR, SNaPshot method, SNPlex typing system, SNPStream analysis system, Sequenom typing system, DHPLC, DGGE.

[0039] Preferably, the system can further comprise a methylation detecting device for detecting methylation markers.

[0040] In another aspect, the present application also provides a method for diagnosing pancreaticobiliary cancer, which determines whether the patient has pancreaticobiliary cancer according to whether the subject has at least one of the gene mutations provided by the present application.

[0041] Preferably, the method provided by the present application can also be used in combination with other diagnostic methods, such as abdominal ultrasound, clinical examination, endoscopic surgery, biochemical test, radiological imaging (CT, MRI, MRCP), etc.

[0042] The present application has the following beneficial effects:

[0043] The technical solution provided by the present application can obtain bile as a sample for detection in a non-invasive sampling manner, and has the characteristics of high specificity and high sensitivity; the disease can be accurately diagnosed at an earlier stage, the patient can be treated early, the cure possibility can be improved, and the survival period can be prolonged; at the same time, for non-cancer patients, unnecessary surgical trauma is avoided. BRIEF DESCRIPTION OF DRAWINGS

[0044] Figure 1 is the subject selection standard and research involved flowchart of the present application.

[0045] Figure 2 is the basic information statistical chart of the subject involved in the present application.

[0046] Figure 3 is the verification of the diagnostic performance of each diagnostic model in different data sets.

[0047] Figure 4 is the statistical result of the detection of subjects in the training set and the validation set.

[0048] Figure 5 is the verification of the diagnostic performance of the methylation markers.

[0049] Figure 6 is the comparison result of the diagnostic performance of each diagnostic model in different data sets and the diagnostic performance of CA19-9.

[0050] Figure 7 is the result statistics of the consistency of gene mutation detection results in brush test samples and biopsy samples. DETAILED DESCRIPTION

[0051] The application will be further described below in conjunction with the embodiments. The following description is only the preferred embodiments of the application and does not limit the application in other forms. Any skilled person in the art can modify the disclosed technical content to obtain equivalent embodiments. Any simple modification or equivalent change made in accordance with the technical essence of the application to the following embodiments falls within the protection scope of the application.

[0052] Example 1, Screening, identification and validation of mutant and methylated genes

[0053] 1. Study population and experimental design

[0054] The study population was selected from 338 patients with pancreatic and biliary diseases admitted to the following five hospitals from November 2018 to October 2020. The five hospitals are: Chinese Academy of Medical Sciences Cancer Hospital, Sichuan Dazhou Central Hospital, Capital Medical University Affiliated Beijing Chaoyang Hospital, Beijing University of Chinese Medicine Dongfang Hospital, Hebei Province Yangxian County People's Hospital. 79 patients were excluded due to various reasons such as non-biliary cancer or insufficient bile DNA, and finally 259 patients were selected for molecular detection.

[0055] Among the 259 patients, 209 patients diagnosed as malignant or benign were selected as the training group (n=104) and the validation group (n=105), of which 116 cases were malignant tumors and 93 cases were benign diseases. Malignant tumors were confirmed as biliary malignant tumors (pancreaticobiliary cancer) after ERCP (retrograde cholangiopancreatography) biopsy / brushing (ERCP-obtained biopsies / brushings) diagnosis. Benign disease patients include 78 patients with cholelithiasis, and no malignant lesions were found in the surgical specimens of the 78 patients, but chronic cholangitis was found. Among them, 15 cases were followed up for more than 12 months, and no malignant lesions were found after ERCP to remove gallstones.

[0056] In addition, considering the low sensitivity of ERCP pathology in some patients with bile duct cancer (BTC), 50 patients diagnosed as negative or suspicious by ERCP biopsy / brushing were classified as an independent test cohort. In the test cohort, 40 patients with malignant tumors were confirmed by pathological evaluation (surgical resection specimens, biopsy / brushing obtained by percutaneous needle or ERCP, n = 21), radiological imaging (n = 2) or clinical criteria (n = 17) during follow-up, while 10 patients with benign diseases were confirmed by surgical pathology (n = 4), radiological imaging (n = 1) or no malignant tumor was found after at least 12 months of follow-up (n = 5).

[0057] Patient enrollment and study design are shown in Figure 1 The study was approved by the ethics committees of the above five hospitals (ID: NCC2018JJJ-001).

[0058] 2. Sample preparation

[0059] All patients obtained bile samples, of which 181 patients obtained bile samples by ERCP before treatment, and 78 patients with cholelithiasis obtained bile samples during cholecystectomy.

[0060] The obtained bile samples were centrifuged at 12000 rpm per minute for 10 minutes to separate the supernatant and particles. DNA was extracted from the bile samples using the TIANAMP Genomic DNA Kit (Tiangen Biotech, Beijing, China). RQ-PCR was performed using Taqman probes for human GAPDH gene to determine the quality of DNA.

[0061] In addition to the bile samples, paired biopsy or brush biopsy tissues obtained during ERCP were subjected to NGS (Next-generation sequencing technology) detection sequencing in 34 and 9 patients, and genomic DNA was extracted using QIAamp DNA Mini Kit (Qiagen, USA).

[0062] Further, CA19-9 detection was performed on the serum of most patients by electrochemiluminescence (ECL) technology using Roche E601 system (Roche Diagnostics, Switzerland).

[0063] 3. Analysis of gene mutations and gene methylation using BileScreen

[0064] Table 1, mutant genes and genes with methylation modification found by sequencing of the present application

[0065]

[0066]

[0067] 400 ng of DNA was fragmented by sonication, followed by end-repair. Then, the DNA fragments were digested by methylation-sensitive restriction enzyme Hha I (R0139S, New England Biolabs, MA, USA) and targeted gene sequencing based on amplicon- targeted capture technology by Mutation Capsule technology. The detailed protocol can be found in Qu, C. et al. Detection of early-stage hepatocellular carcinoma in asymptomatic HBsAg-seropositive individuals by liquid biopsy. Proceedings of the National Academy of Sciences of the United States of America. 116, 6308-6312 (2019).

[0068] Briefly, the fragments of restriction enzyme Hha I digested DNA were processed using KAPA Hyper Prep kit (Roche, Switzerland) to obtain sequencing library by a series of steps, including end-repair, adenosine A tailing, custom adapter ligation and three rounds of PCR amplification (the first round using common sequence primers, the last two rounds using primers containing target-specific and common sequences). 23 mutant genes and 44 genes with methylation modification were sequenced (Table 1).

[0069] 4. Data processing and mutation / methylation detection

[0070] Gene sequencing was performed using the Digital (UID) high-throughput sequencing platform. Briefly, a unique UID tag was added to each fragment in the sample before PCR amplification, and then the library was amplified. After sequencing, the sequences of the fragments were aligned, and the repeated fragments with the same UID marker were combined while retaining the naturally repeated fragments with different UID markers. The end coordinates of the reads were calculated according to the start coordinates and cigar information. According to the start coordinates and end coordinates of the reads, the reference sequence corresponding to the reads was cut from the reference genome. The reads were re-aligned with the hg19 genome to obtain the start and end positions of the mutations. The definition of an effective UID (EUID, Effective UID) family is a UID (Unique Identifier, UID) family containing at least two reads and at least 80% of the same read types. The frequency of each mutation is calculated by dividing the number of alternative EUID families by the sum of alternative and reference families. We further manually checked the mutations (in IGV) and annotated the candidate variant genes using VEP (Ensembl Variant Effect Predictor).

[0071] There were at least four EUID families detected. For hotspot mutations of oncogenes including KRAS (G12, G13, Q61 and A146), the limitation of detection (LOD, limitation of detection) was set to 0.5%. For other mutant genes, including common non-hotspot tumor suppressor genes such as TP53, Smad4 and rare mutations, in order to reduce the occurrence of false positives, the LOD was set to 1%. Since there is no matching white blood cell to exclude germ line mutations, the detected mutations in the sample are screened by germ line and somatic mutation databases to determine the highest likelihood of germ line mutations. Mutations with a frequency of ≥0.1% in the germ line mutation database (1000AF, ESP6500 AA / EA, Exac AF) are first excluded. Further screening of the passed mutations, for those with a higher frequency (≥40%), if they appear in the COSMIC database with less than 10 samples, they are likely to be germ line mutations, and are therefore further excluded.

[0072] In terms of methylation analysis, clusters with HHA I restriction sites at the ends are unmethylated sequences, and molecules with at least one HHA I restriction site and the ends are not restriction sites are methylated sequences. The methylation ratio of each base is the number of methylated molecules divided by the sum of methylated and non-methylated molecules.

[0073] 5. Construction of BileScreen diagnostic model

[0074] AKT1, KRAS, APC, NRAS, ARID1A, PIK3CA, AXIN1, PPP2R1A, BAP1, PTEN, BRAF, SMAD4, CDKN2A, TERT, TP53, EGFR, FBXW7, FGFR2, HRAS, IDH1, IDH2 mutations all appear malignant tumors (as shown in Table 2 below).

[0075] Table 2, detection results of mutations

[0076]

[0077]

[0078]

[0079]

[0080]

[0081]

[0082]

[0083]

[0084]

[0085]

[0086]

[0087]

[0088]

[0089] In the 44 methylation modification genes, using the training set, 5 methylation modification genes SOX17, 3-OST-2, NXPH1, SEPT9 and TERT were screened out by using stepwise penalty Logistic regression method for constructing diagnostic model. The above-mentioned 5 methylation modification gene markers were used for penalty Logistic regression of the training set, and the leave-one-out cross-validation was used. The performance of the model was evaluated by the area under the receiver operating characteristic curve (ROC curve), sensitivity and specificity indexes. The cutoff value of methylation was determined according to the Youden index of ROC analysis.

[0090] In the BileScreen model, a mutation and methylation integration was positive if one of them was positive. The performance of the BileScreen model was further evaluated in the separate validation and test sets.

[0091] In addition, since the selected benign cases in the training and validation sets were mostly young women with gallstones, there was a certain tendency in age and gender between the malignant and benign patients. In order to exclude the influence of such artificial sample selection on the diagnostic prediction results, 52 age- and gender-matched malignant patients and 52 benign patients were assigned to the training set. Therefore, the age and gender distribution in the validation set was not uniform Figure 2 ). However, all mutations and methylation were not significantly correlated with age and gender (correlation coefficient <0.5, or Wilcoxon test P>0.05).

[0092] 6. Statistical analysis

[0093] The role of individual mutation genes or methylation-modified genes in predicting disease status was evaluated using ROC analysis (pROC package) and Wilcoxon test. The gene markers of the diagnostic model were screened using the penalized logistic regression method (glmnet package). In the training set, the ROC curve used the original score as input, and the Youden index was used alone to determine the optimal cutoff point for methylation. In addition, ROC analysis was used to compare the performance of different methods, with the "0 or 1" value determined by the corresponding cutoff point as input. Sensitivity and specificity were calculated using standard 2x2 contingency tables. All R package-related analyses were based on R software (V.3.6.3).

[0094] 7. Results

[0095] 1) BileScreen model establishment based on gene mutation and methylation modification

[0096] Sequencing and data analysis were performed on 104 patients in the training set, including 52 patients with malignant tumors diagnosed by ERCP and 52 patients with benign diseases, of which a part was confirmed by surgical pathology that the tissue site did not occur cancer, and the other part was patients with bile duct stones, and at least 12 months of follow-up were determined as benign diseases, the clinical characteristics are summarized as Figure 2 .

[0097] DNA analysis of bile samples was performed using mutation capsule technology, detecting 23 mutated genes and 44 methylation modification genes. The most common mutated genes in cancer were TP53 (50%) and KRAS (46%). CTNNB1 and GNAS mutations were detected simultaneously in both cancer and benign disease patients; therefore, these genes were not significantly associated with malignancy, and CTNNB1 and GNAS mutations were excluded from the BileScreen model.

[0098] Table 3. Detection accuracy of each dataset

[0099]

[0100]

[0101] Using gene mutations to distinguish between patients with pancreatic and cholangiocarcinoma and subjects with non-cancer (benign) diseases:

[0102] Mutations in AKT1, KRAS, APC, NRAS, ARID1A, PIK3CA, AXIN1, PPP2R1A, BAP1, PTEN, BRAF, SMAD4, CDKN2A, TERT, TP53, EGFR, FBXW7, FGFR2, HRAS, IDH1, and IDH2 were considered positive if at least one mutation was detected. The sensitivity of using mutations alone to distinguish between patients with pancreatocholangiocarcinoma and non-cancer subjects was 81%, the specificity was 100%, and the AUC was 0.90 (Table 3). Figure 3 A).

[0103] Using methylation markers to differentiate between patients with pancreatic and cholangiocarcinoma and subjects with non-cancer (benign) diseases:

[0104] For methylation biomarkers, a diagnostic model was constructed using a stepwise penalized logistic regression method, selecting five markers: SOX17, 3-OST-2, NXPH1, SEPT9, and TERT. Figure 3 (Table 4). Using the leave-one-out method, methylation markers alone can effectively identify pancreaticocholangiocarcinoma patients from non-cancerous cases, with a sensitivity of 88%, specificity of 98%, and AUC of 0.93 (Table 3). Figure 3 The cutoff value for the methylation score was 0.422, which produced the largest Youden index. Figure 5 ).

[0105] Table 4. Diagnostic model constructed based on 5 methylation biomarkers

[0106]

[0107] Finally, when gene mutations and methylation markers were combined, i.e. BileScreen:

[0108] A positive was defined as either positive in both, which further improved the performance with a sensitivity of 94%, a specificity of 98% and an AUC of 0.96 (Table 3, Figure 3 ). BileScreen was validated in an additional 105 cases (validation cohort), of which 64 were malignant and 41 were benign Figure 4 BileScreen accurately predicted the disease status in 59 malignant and 40 non-cancer cases. BileScreen showed a sensitivity of 92% and a specificity of 98% in the validation cohort with an AUC of 0.95 (Table 3, Figure 3 ). If only gene mutations were used, the sensitivity and specificity were 78% and 100%, respectively. The sensitivity and specificity of methylation marker analysis alone were 81% and 98%, respectively (Table 3).

[0109] 2. BileScreen model for detection of suspicious malignant tumors

[0110] We further validated BileScreen in 50 patients with inconclusive ERCP results (test cohort), as these 50 patients had ERCP diagnoses of "suspected malignancy" or "cannot exclude cancer". These cases were followed up for at least 12 months, of which 40 out of 50 were found to be malignant. The remaining 10 cases did not develop cancer during the follow-up period and were diagnosed as benign. Out of the 40 malignant lesions, 36 were positive, and out of the 10 benign lesions, 2 were positive, with a sensitivity of 90% and a specificity of 80% Figure 3 For those patients who could not be definitively diagnosed by ERCP, the BileScreen results were significantly correlated with the clinical outcome (P < 0.001, continuity-corrected Chi-square test).

[0111] Mutations or methylation alone could distinguish cancer from benign patients with a sensitivity of 75% and 80%, respectively, a specificity of 90% and 80%, respectively, and an AUC of 0.83 and 0.8, respectively (Table 3, Figure 3 ).

[0112] 3. Comparison of CA19-9 and BileScreen model detection results in each group

[0113] In the training and validation cohorts, we screened 85 and 74 patients, respectively, for which serum CA19-9 data were available Figure 2, serum CA19-9 and BileScreen were directly compared in these patients. Serum CA19-9 had an AUC of 0.78 and 0.81 in discriminating between benign and malignant in the two cohorts, respectively Figure 6 ), with a cutoff of >27 U / mL, serum CA19-9 had a sensitivity of 88% and 91% and a specificity of 67% and 70%, respectively (Table 5).

[0114] Table 5. Validation results of each model and CA19-9 in different datasets

[0115]

[0116] In contrast, BileScreen had a sensitivity of 93% and 94% and a specificity of 98% and 96%, respectively. Across the training and validation cohorts, CA19-9 had a sensitivity and specificity of 90% and 68%, respectively, both lower than BileScreen’s 93% and 97%. Thus, BileScreen outperformed serum CA19-9 in detecting pancreaticobiliary cancers, particularly in terms of specificity.

[0117] In addition, serum CA19-9 results in the 38 patients in the test cohort were as follows Figure 6 As shown in Figure A, serum CA19-9 had a very low specificity (14%) and an AUC of 0.51, with a sensitivity of 84% (Table 4, Figure 6 ) compared to the training and validation cohorts. In contrast, BileScreen had a sensitivity and specificity of 87% and 86%, respectively, in this group.

[0118] 4. Comparison of gene mutation results in bile and ERCP biopsy / brushing samples

[0119] In the test cohort, 34 patients had biopsy samples obtained by ERCP and 9 patients had brush samples. We analyzed both the genetic and tissue samples in a head-to-head comparison study. Of the 43 cases, 70 mutations were present in both sample types, 5 mutations were detected only in bile (brush samples), and 9 were detected only in tissue (biopsy samples) Figure 7 Thus, 93% (70 / 75) of mutations detected in bile were also detected in tissue, while additional mutations were only found in bile, and 89% (70 / 79) of tissue-derived mutations were detected in bile. In addition, of the 43 samples, 36 had mutations detected in at least one type of sample, of which 34 (94%) had at least one common mutation detected between the two sample types. Thus, the concordance rate of mutation status detected by bile versus tissue was 95% (41 / 43).

Claims

1. A set of gene mutations for detecting pancreaticobiliary cancer, the gene mutations consisting of AKT1, KRAS, APC, NRAS, ARID1A, PIK3CA, AXIN1, PPP2R1A, BAP1, PTEN, BRAF, SMAD4, CDKN2A, TERT, TP53, EGFR, FBXW7, FGFR2, HRAS, IDH1, IDH2.

2. Use of reagents for detecting the gene mutations of claim 1 in the preparation of a product for diagnosing pancreaticobiliary cancer.

3. The use of claim 2, wherein the reagents for detecting the gene mutations comprise reagents used in any one or more of the following methods: TaqMan probe method, sequencing method, chip method, flight mass spectrometer detection, restriction fragment length polymorphism method, single strand conformation polymorphism method, allele-specific PCR, SNaPshot method, SNPlex typing system, SNPStream analysis system, Sequenom typing system, denaturing high performance liquid chromatography method, denaturing gradient gel electrophoresis method.

4. The use of claim 2, wherein the gene mutations are detected from a sample from a subject.

5. The use of claim 4, wherein the sample is taken from the biliary tract; the sample comprises bile, exfoliated cells in the biliary tract, a tissue sample.

6. The use of claim 5, wherein the exfoliated cells in the biliary tract, the tissue sample is a sample obtained by ERCP biopsy / brushing.

7. The use of claim 2, wherein the gene mutations of claim 1 are further used in combination with methylation markers, the methylation markers comprising one or more of SOX17, 3-OST-2, NXPH1, SEPT9 and TERT.

8. The use of claim 7, wherein the methylation markers further comprise any one or more of EBF3, RASSF1, APC, EYA4, RUNX3, BNIP3, FHIT, SALL3, CCND2, FOXE1, CD1D, GSTP1, SFRP1, CDH1, hMLH1, SLIT2, CDH13, KCNK12, SLIT3, CDKN2A, MGMT, CDKN2B, NDRG4, CDO1, NPTX2, TFPI2, CLEC11, TIMP3, CNRIP1, PENK, TMEFF2, DAPK1, PRKCB, VIM, DCLK1, PTCHD2, ZSCAN18, DLC1, RARβ2.

9. A diagnostic system for diagnosing pancreaticobiliary cancer, the system comprising a computing device for diagnosing whether a subject has at least one of the gene mutations of claim 1.

10. The diagnostic system of claim 9, the system comprising: (1) a sample collection and processing device for performing the following steps: collecting a sample from a subject, processing the sample; (2) a nucleic acid sequence determination device. ​ ​ (3) a computing device for diagnosing whether the subject has at least one of the mutations in the genes according to claim 1.

11. The system of claim 10, wherein the sample is selected from the group consisting of bile, cells taken from the biliary tract, tissue, blood, urine, saliva, and tears.

12. The system of claim 11, wherein the cells taken from the biliary tract, tissue sample is a sample taken from an ERCP biopsy / brushing.

13. The system of claim 10, wherein the nucleic acid sequence determining device is capable of detecting whether the subject has at least one of the mutations in the genes according to claim 1 by any one of the following methods: TaqMan probe method, sequencing method, chip method, mass spectrometry, restriction fragment length polymorphism method, single strand conformation polymorphism method, allele-specific PCR, SNaPshot method, SNPlex typing system, SNPStream analysis system, Sequenom typing system, denaturing high performance liquid chromatography method, denaturing gradient gel electrophoresis method.

14. The diagnostic system according to any one of claims 9-13, further comprising a methylation detecting device for detecting methylation markers.

15. The system of claim 14, wherein the methylation markers further comprise any one or more of EBF3, RASSFl, APC, EYA4, RUNX3, BNIP3, FHIT, SALL3, CCND2, FOXEl, CD1D, GSTPl, SFRPl, CDHl, hMLHl, SLIT2, CDH13, KCNK12, SLIT3, CDKN2A, MGMT, CDKN2B, NDRG4, CDOl, NPTX2, TFPI2, CLEC11, TIMP3, CNRIP1, PENK, TMEFF2, DAPKl, PRKCB, VIM, DCLKl, PTCHD2, ZSCAN18, DLC1, RARβ2. ​

Images