A biomarker for extrahepatic cholangiocarcinoma prognosis evaluation, a prognosis risk evaluation system and application thereof

Abstract

The application discloses a biomarker for prognosis evaluation of extrahepatic cholangiocarcinoma, a prognosis risk evaluation system and application, and belongs to the technical field of bioinformatics. The application adopts the combination of miR-34c-5p, miR-100-5p, miR-193b-5p, miR-122-5p, miR-5588-5p and miR-122-3p as a biomarker, can effectively improve the accuracy of differential diagnosis of extrahepatic cholangiocarcinoma, can distinguish extrahepatic cholangiocarcinoma patients from healthy people and benign biliary disease patients with high precision, and overcomes the limitation that traditional tumor markers such as CA19-9 are prone to false positive results in benign biliary diseases. Meanwhile, the expression level change of the biomarker combination can dynamically reflect the postoperative tumor load change, and is helpful for accurately evaluating the surgical effect.

Description

Technical Field

[0001] This invention relates to the field of bioinformatics technology, and in particular to a biomarker, prognostic risk assessment system, and application for the prognostic evaluation of extrahepatic cholangiocarcinoma. Background Technology

[0002] The information disclosed in the background section of this invention is intended only to enhance the understanding of the overall background of the invention and is not necessarily to be construed as an admission or in any way implying that such information constitutes prior art known to those skilled in the art.

[0003] Extrahepatic cholangiocarcinoma (eCCA), a malignant tumor originating from the epithelium of the extrahepatic bile ducts, has an extremely poor clinical prognosis, with a five-year survival rate of less than 15%. Due to the lack of specific symptoms in the early stages of the disease, over 70% of patients have missed the opportunity for surgical treatment by the time of diagnosis. Surgical resection (R0 resection), currently the only potentially curative treatment, still carries a high risk of recurrence. Current imaging and serological markers are insufficient for assessing postoperative recurrence risk and residual lesions, making it difficult to support precise clinical monitoring of disease progression.

[0004] In clinical diagnosis, carbohydrate antigen 19-9 (CA19-9) and carcinoembryonic antigen (CEA) are the mainstream tumor markers, but their application in cholangiocarcinoma faces significant limitations. CA19-9 exhibits unsatisfactory sensitivity and specificity, and it is often abnormally elevated in benign biliary tract diseases such as cholelithiasis (CBDS) and cholangitis, easily leading to false-positive results. Furthermore, its correlation with patient prognosis is weak. CEA's diagnostic efficacy for cholangiocarcinoma is even more limited. This necessitates a clinical need for more sensitive, specific, stable, and non-invasive blood biomarkers to meet the needs of early diagnosis of eCCA, evaluation of surgical efficacy, and postoperative prognosis.

[0005] MicroRNAs (miRNAs), a class of non-coding small RNAs approximately 20-24 nucleotides in length, participate in gene expression regulation through post-transcriptional regulation. They are typically secreted actively by tumor cells, encapsulated in exosomes or protein complexes, and exhibit good stability in plasma. These characteristics make them a research hotspot in the field of liquid biopsy for cancer. While some studies have proposed single miRNA biomarkers for cholangiocarcinoma, their clinical applications still have significant limitations: in disease differentiation, their accuracy in distinguishing extrahepatic cholangiocarcinoma from benign biliary diseases (such as cholelithiasis and cholangitis) is insufficient; in postoperative monitoring, their sensitivity to changes in tumor burden and surgical efficacy (such as assessment of radical resection status) is low; in prognostic assessment, their association with overall survival and recurrence-free survival is weak, making them difficult to use as independent prognostic indicators; furthermore, some studies involve complex and costly detection methods, making them unsuitable for large-scale screening and long-term follow-up monitoring. These shortcomings make it difficult for existing miRNA biomarkers to meet the integrated clinical needs for eCCA diagnosis, efficacy monitoring, and prognostic prediction. Summary of the Invention

[0006] In view of this, the present invention provides a biomarker, a prognostic risk assessment system and application for the prognostic evaluation of extrahepatic cholangiocarcinoma (eCCA). The present invention can achieve high-precision differentiation between eCCA and healthy individuals and benign biliary diseases (such as cholelithiasis and cholangitis). It can also be used for early postoperative efficacy assessment and can predict patients' overall survival (OS) and recurrence-free survival (RFS). When this biomarker is used in combination with CA19-9, it can further improve diagnostic accuracy.

[0007] In a first aspect, the present invention provides the application of a reagent for detecting the expression level of a biomarker in the preparation of a product for the differential diagnosis of extrahepatic cholangiocarcinoma; wherein the biomarker is a combination of miR-34c-5p, miR-100-5p, miR-193b-5p, miR-122-5p, miR-5588-5p and miR-122-3p.

[0008] Secondly, the present invention provides the application of a reagent for detecting the expression level of a biomarker in the preparation of a product for evaluating the postoperative efficacy of extrahepatic cholangiocarcinoma; wherein the biomarker is a combination of miR-34c-5p, miR-100-5p, miR-193b-5p, miR-122-5p, miR-5588-5p and miR-122-3p.

[0009] In this invention, the postoperative efficacy assessment preferably distinguishes between R0 resection and R1 resection surgical status. R0 resection refers to complete resection with no tumor residue under a microscope, while R1 resection refers to incomplete resection with visible tumor residue under a microscope.

[0010] Thirdly, the present invention provides the application of a reagent for detecting the expression level of a biomarker in the preparation of a product for predicting the prognosis of extrahepatic cholangiocarcinoma; wherein the biomarker is a combination of miR-34c-5p, miR-100-5p, miR-193b-5p, miR-122-5p, miR-5588-5p and miR-122-3p.

[0011] In this invention, the prediction of prognosis for extrahepatic cholangiocarcinoma is preferably the prediction of overall postoperative survival and recurrence-free survival.

[0012] Preferably, the detection includes in vitro detection of the concentration of the biomarker in a biological sample, wherein the biological sample is plasma, peripheral blood, serum, bile, exosomes, free RNA, or dried blood spots.

[0013] Preferably, the product includes at least one of reagents, reagent kits, test strips, chips, and systems.

[0014] Fourthly, the present invention provides a system for the diagnosis, postoperative efficacy assessment, or prognostic assessment of extrahepatic cholangiocarcinoma, comprising:

[0015] The data acquisition module is used to acquire the expression levels of biomarkers in the biological samples of the subjects; the biomarkers are a combination of miR-34c-5p, miR-100-5p, miR-193b-5p, miR-122-5p, miR-5588-5p and miR-122-3p;

[0016] The model prediction module is used to construct a prediction model based on the expression levels of biomarkers obtained by the data acquisition module and output a risk score.

[0017] The data output module is used to output prediction results based on the risk score.

[0018] Preferably, the method for obtaining the expression levels of biomarkers in the subject's biological samples includes Northern hybridization, microarray analysis, real-time quantitative PCR (qPCR), and high-throughput sequencing technologies. miRNAs that are stably expressed in plasma, such as miR-21-5p, miR-16-5p, miR-24-3p, and U6 snRNA, can be selected as internal controls; the expression levels of each miRNA can be determined using a 2-1... -ΔΔCt Z-score standardization, min-max normalization, or the use of relative expression proportions are all applicable to different detection platform standards.

[0019] Preferably, the prediction model is constructed based on a machine learning algorithm, which is selected from any one of the following: logistic regression, linear regression, random forest, neural network, support vector machine, Bayesian classification, gradient boosting, K-nearest neighbors, or decision tree.

[0020] Preferably, the formula for calculating the risk score is:

[0021] Risk score = -1.217 + 0.0264 × miR-34c-5p expression level + 0.0412 × miR-100-5p expression level + 0.0110 × miR-193b-5p expression level + 0.0113 × miR-122-5p expression level + 0.0156 × miR-5588-5p expression level + 0.0172 × miR-122-3p expression level;

[0022] A risk score of -0.53 or higher is considered high risk; a risk score of -0.53 or lower is considered low risk.

[0023] Preferably, the model prediction module is used to construct a prediction model based on the expression levels of biomarkers and CA19-9 obtained by the data acquisition module, and output a risk score;

[0024] The formula for calculating the risk score is as follows:

[0025]

[0026] Where, joint factor = α 0+ α 1× Composite Index + α 2× CA19-9 serum concentration;

[0027] Composite index = -1.217 + 0.0264 × miR-34c-5p expression level + 0.0412 × miR-100-5p expression level + 0.0110 × miR-193b-5p expression level + 0.0113 × miR-122-5p expression level + 0.0156 × miR-5588-5p expression level + 0.0172 × miR-122-3p expression level;

[0028] α 0 = -3.08443; α 1 = 0.95618; α 2 = 0.09363; CA19-9 serum concentration is a value in "U / mL";

[0029] A risk score of ≥0.35 is considered high risk; a risk score of <0.35 is considered low risk.

[0030] Fifthly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, can realize the functions of the above-described system.

[0031] Compared with the prior art, the present invention has achieved the following beneficial effects:

[0032] (1) This invention uses a combination of miR-34c-5p, miR-100-5p, miR-193b-5p, miR-122-5p, miR-5588-5p and miR-122-3p as biomarkers, which can effectively improve the accuracy of differential diagnosis of extrahepatic cholangiocarcinoma. It can accurately distinguish patients with extrahepatic cholangiocarcinoma from healthy people and patients with benign biliary tract diseases, overcoming the limitation of traditional tumor markers such as CA19-9 in benign biliary tract diseases that are prone to false positive results. At the same time, the change in the expression level of this combination of biomarkers can dynamically reflect the postoperative tumor burden change, which helps to accurately assess the surgical efficacy.

[0033] (3) The combination of biomarkers of the present invention has high diagnostic sensitivity. In the eCCA and healthy population cohort (cohort 1), its diagnostic accuracy for eCCA reached 75%, with an AUC of 82.06%. In the eCCA and CBDS patient cohort (cohort 2), the diagnostic accuracy reached 83.75%, with an AUC of 94.81%. The combination of this biomarker with the tumor marker CA19-9 can further improve the diagnostic accuracy. In cohort 1, its diagnostic accuracy reached 91.46%, with an AUC of 93.59%, and in cohort 2, its diagnostic accuracy reached 90.00%, with an AUC of 97.94%.

[0034] (4) Based on the excellent discriminative ability of the miRNA index I combined with the CA19-9 model, a decision curve analysis (DCA) was performed on it. The model achieved the highest standardized net benefit within the clinically relevant threshold (0.10~0.80). When the prevalence was assumed to be 0.30, the clinical benefit of the combined model was always better than that of miRNA index I and CA19-9 alone, and better than the "all treatment / no treatment" strategy.

[0035] (5) The risk scoring model constructed based on the combination of the above biomarkers can effectively predict the overall survival and recurrence-free survival of patients, and can provide support for risk stratification and adjuvant treatment decision-making as an independent prognostic factor, thus solving the problem of weak correlation between existing biomarkers and prognosis.

[0036] (6) In clinical applications, the detection of the above biomarkers can be achieved through conventional techniques such as standard qPCR. Related products such as reagents, kits, and detection systems are easy to operate and have low cost, making them suitable for large-scale screening and long-term follow-up monitoring, thereby meeting the integrated clinical needs of early diagnosis, surgical efficacy evaluation, and postoperative prognosis of extrahepatic bile duct cancer. Attached Figure Description

[0037] The accompanying drawings, which form part of this specification, are used to provide a further understanding of the invention. The illustrative embodiments and descriptions of the invention are used to explain the invention and do not constitute an undue limitation thereof. Obviously, those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0038] Figure 1 This is a schematic diagram illustrating the biomarker screening, construction, and model evaluation for the diagnosis and prediction of eCCA according to an embodiment of the present invention.

[0039] Figure 2 This document presents the diagnostic efficacy and calibration curves of the miRNA index I model and the combined factor of miRNA index I and CA19-9 in embodiments of the present invention; wherein, A is the diagnostic efficacy verification of miRNA index I and CA19-9 in cohort 1; B is the calibration curve of the miRNA index I model; C is the diagnostic efficacy verification of miRNA index I and CA19-9 in cohort 2; and D is the calibration curve of the miRNA index I model.

[0040] Figure 3 This is a decision curve analysis of an embodiment of the present invention, wherein A is a decision curve analysis (DCA) of the combined model (miRNA index I + CA19-9) under different hypothetical eCCA prevalence rates (0.15, 0.30, 0.50); ​​B is a comparison of the standardized net benefits of the combined model, miRNA index I and CA19-9 when the hypothetical prevalence rate is 0.30.

[0041] Figure 4 This invention illustrates the changes in miRNA index I and CA19-9 in eCCA patients before and after surgery. A represents the postoperative expression level of miRNA index I; B represents the postoperative expression level of CA19-9; C represents the comparison of the change in miRNA index I on day 7 after surgery between R0 and R1 resection patients; and D represents the comparison of the change in CA19-9 on day 7 after surgery between R0 and R1 resection patients.

[0042] Figure 5This invention relates to the correlation between Kaplan-Meier survival analysis of miRNA index I and overall survival (A) and recurrence-free survival (B) after surgery for extrahepatic bile duct cancer. Detailed Implementation

[0043] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0044] The technical solution of the present invention will be further described below with reference to specific embodiments.

[0045] Example

[0046] 1. Sample collection and processing:

[0047] (1) The study subjects included patients with extrahepatic bile duct cancer, healthy individuals undergoing physical examinations, and patients with common bile duct stones (some of whom also had acute cholangitis). All sample collection procedures were approved by the ethics committee.

[0048] (2) Collect 5 mL of venous blood into a tube containing EDTA anticoagulant. Centrifuge the sample at 1500 g for 10 minutes (4℃) to separate the plasma. Centrifuge the supernatant at 12000 g for 10 minutes to remove cell debris. Store the plasma at -80℃.

[0049] 2. RNA extraction and reverse transcription

[0050] (1) Add 750 μL SparkZol LS reagent to every 250 μL of plasma, mix thoroughly, and lyse at room temperature for 5 minutes.

[0051] (2) Add 150 μL of chloroform, shake vigorously for 15 seconds, incubate at room temperature for 3 minutes, and centrifuge at 12000 g for 15 minutes (4℃). Take the supernatant (about 450~500 μL), add an equal volume of isopropanol, and add 1 μL of GlycoBlue to assist precipitation.

[0052] (3) Incubate at room temperature for 10 minutes, centrifuge at 12000 g for 10 minutes (4℃), wash twice with 75% ethanol, air dry for 5~10 minutes, and then resuspend in 15~20 μL of RNase-free water.

[0053] (4) Reverse transcription was performed using the Vazyme miRNA first-strand cDNA synthesis kit.

[0054] 3. qRT-PCR detection and normalization

[0055] (1) The Bio-Rad CFX Connect real-time fluorescence quantitative PCR instrument was used. The reaction system was 20 μL and ChamQUniversal SYBR qPCR Master Mix was used.

[0056] (2) Thermal cycling conditions: pre-denaturation at 95℃ for 30 s, followed by 40 cycles (95℃ for 10 s, 60℃ for 30 s), and finally melting curve analysis.

[0057] (3) RNA expression level using 2 -ΔΔCt The internal reference was hsa-miR-21-5p (due to its stable expression and high abundance in the samples). Three replicate wells were set for each sample, and repeatability CV < 3% was considered acceptable.

[0058] 4. miRNA Feature Selection and Model Construction

[0059] Figure 1 This diagram illustrates the biomarker screening, construction, and model evaluation for the diagnosis and prediction of eCCA in this embodiment. The specific steps are as follows:

[0060] (1) Screening of candidate miRNAs: Differentially expressed miRNAs were screened by sequencing plasma samples from 20 eCCA patients and 10 healthy individuals. The edgeR analysis method was used to screen for candidate miRNAs upregulated in the plasma of tumor patients with |log2FC|≥2 and FDR<0.05.

[0061] (2) Feature selection: The importance of candidate miRNA features was scored using the random forest algorithm, with mtry set to the default value (square root of the number of features for classification tasks); set.seed=123 ensured reproducibility. A random forest model containing 500 decision trees was constructed with miRNA expression level as the independent variable and sample grouping as the dependent variable. The average Gini impurity reduction (MDG) and average accuracy reduction (MDA) of each miRNA across all trees were calculated as feature importance scores. A cross-filtering strategy was adopted: miRNAs with both MDA and MDG higher than their respective medians were retained. The threshold was confirmed using minimum OOB error / 5-fold cross-validation, and the results were consistent, ultimately yielding 13 miRNAs with the highest diagnostic value:

[0062]

[0063]

[0064] Where T is the number of trees in the forest (T=500); S j,m It is the m-th tree with features Xj The set of all nodes to be split; N t It is the number of samples at node t; N It is the total number of samples; Δ Gini It is the amount of Gini impurity reduction before and after the node splits; OOB m It is the set of out-of-bag samples of the m-th tree; I(·) It is an indicator function; it takes a value of 1 if the prediction is correct, and 0 otherwise. This is the original prediction result; It is a variable X j Scrambled prediction results

[0065] (3) qPCR validation and modeling: Thirteen miRNAs were validated by qPCR in plasma samples from 81 patients with eCCA and 83 healthy individuals. Six miRNAs were screened using bidirectional stepwise logistic regression based on the Akaike Information Criterion (AIC) to construct a diagnostic model.

[0066]

[0067] Among them, miRNA index I is the composite index. β 0 represents the intercept term. β i Let be the logistic regression coefficient of the i-th miRNA. Expression level i The expression level of the i-th miRNA is measured.

[0068] The final miRNA index I was -1.217 + 0.0264×miR-34c-5p expression level + 0.0412×miR-100-5p expression level + 0.0110×miR-193b-5p expression level + 0.0113×miR-122-5p expression level + 0.0156×miR-5588-5p expression level + 0.0172×miR-122-3p expression level.

[0069] (4) Validation and Comparative Analysis: Validation was conducted in multiple independent cohorts, including healthy controls, patients with cholelithiasis (CBDS), and pre- and post-operative paired samples. The optimal diagnostic threshold of the model was determined, and the diagnostic accuracy (ROC, AUC), calibration (calibration curve, Hosmer-Lemeshow [HL] test, MAE), and net clinical benefit (DCA) of the model were evaluated. The diagnostic performance was compared with CA19-9 and CEA, and a joint model was constructed by combining CA19-9 and CEA.

[0070] (i) Determining the optimal threshold for the model: The model threshold is calculated by evaluating sensitivity, specificity, and the Youden index. The Youden index measures the classifier's ability to maximize the true negative rate while maintaining a high true positive rate. The maximum value of the Youden index corresponds to the maximum diagnostic critical point of the method. The model-defined threshold is the maximum value of the Youden index, calculated using the following formula:

[0071]

[0072]

[0073]

[0074] Among them, TP represents true positive, which refers to the number of patients diagnosed by the gold standard who are judged as patients by the diagnostic model; FN represents false negative, which refers to the number of patients diagnosed by the gold standard who are judged as non-patients by the diagnostic model; FP represents false positive, which refers to the number of non-patients diagnosed by the gold standard who are judged as patients by the diagnostic model; and TN represents true negative, which refers to the number of non-patients diagnosed by the gold standard who are judged as non-patients by the diagnostic model.

[0075] In this embodiment, the optimal threshold for miRNA index I is -0.53. The calculated value of miRNA index I is used as the risk score. When the risk score is ≥ -0.53, it is diagnosed as high risk of eCCA; when the risk score is < -0.53, it is diagnosed as low risk of eCCA.

[0076] (ii) Evaluation of model calibration: Calibration refers to the degree of consistency between the predicted risk of the prediction model and the actual risk. It is an important dimension for measuring the accuracy of model prediction. A goodness-of-fit test P>0.05 indicates that the model calibration is good.

[0077] In this embodiment, the goodness-of-fit test for miRNA index I was HL p=0.051, indicating that the model was well calibrated. The goodness-of-fit test in the eCCA and CBDS patient cohorts was HL p=0.997.

[0078] (iii) Evaluation of model accuracy: Accuracy refers to the model's ability to correctly distinguish between individuals at high risk and those at low risk of the outcome, i.e., the model's ability to correctly classify whether the research event has occurred. Calculations showed that the accuracy of miRNA index I in diagnosing eCCA in this embodiment reached 75%, with an AUC of 82.06%. In the eCCA and CBDS patient cohorts, the diagnostic accuracy reached 83.75%, with an AUC of 94.81%.

[0079] (iv) Construction of the prediction model for CA19-9: The above miRNA index I and CA19-9 serum concentration (in U / mL) were included in a binary logistic regression model, and the combination factor was calculated:

[0080] Joint factor = α 0+ α 1× miRNA index I+ α 2× CA19-9 serum concentration;

[0081] in α 0 represents the intercept, which is -3.08443; α 1 and α 2 represents the regression coefficients for miRNA index I and CA19-9, which are 0.95618 and 0.09363, respectively.

[0082] The risk score P corresponding to the combined factor is calculated using the following formula:

[0083]

[0084] The diagnostic threshold was set at 0.35. When P ≥ 0.35, the risk of the subject having eCCA was considered high; otherwise, the risk was considered low.

[0085] When combined with the tumor marker CA19-9, the diagnostic accuracy of the combined factor reached 91.46%, with an AUC of 93.59%, which was statistically significant compared to miRNA index I and CA19-9.

[0086] Figure 2 The diagnostic efficacy and calibration curves of the miRNA index I model and the combination factor of miRNA index I and CA19-9 are presented. Figure 2 In the figure, A represents the diagnostic efficacy validation of miRNA index I and CA19-9 in cohort 1. The AUC of miRNA index I was 82.06%, indicating that the combined factor significantly enhanced diagnostic efficacy. Figure 2 In the figure, B represents the calibration curve for miRNA index I. The figure shows HL p=0.051, indicating that the model calibration is good. Figure 2 In the figure, C represents the diagnostic efficacy validation of miRNA index I and CA19-9 in cohort 2. The AUC of miRNA index I was 94.81%, significantly stronger than that of CA19-9. Combining CA19-9 can significantly improve diagnostic efficacy. Figure 2 D in the figure represents the calibration curve of miRNA index I. The figure shows HL p=0.997, indicating that the model is well calibrated.

[0087] (v) Evaluation of the model’s net clinical benefit: Decision curve analysis (DCA) was used to evaluate the model’s net benefit at different threshold probabilities.

[0088] like Figure 3 As shown in A, the combined model achieved the highest standardized net benefit within the clinically relevant thresholds (0.10–0.80), as... Figure 3 As shown in B, when the prevalence rate is assumed to be 0.30, the clinical benefit of the combination model is consistently superior to miRNAindex I and CA19-9 alone, and also superior to the "all treatment / no treatment" strategy.

[0089] 5. Other applications of diagnostic models

[0090] (1) Postoperative monitoring (postoperative efficacy assessment)

[0091] Figure 4 The changes in miRNA index I and CA19-9 before and after surgery in patients with eCCA. Figure 4 In the figure, A and B represent the expression levels of miRNA index I and CA19-9 on postoperative day 7, and it can be seen that both showed a significant decrease. Figure 4 C in the figure represents the change in miRNA index I on day 7 post-surgery (ΔmiRNA index I = preoperative miRNA index I - postoperative miRNA index I on day 7). The comparison between R0 resection and R1 resection patients showed that the decrease in miRNA index I was smaller in patients undergoing R1 resection (p = 0.043). Figure 4 The figure D shows the change in CA19-9 on postoperative day 7 (ΔCA19-9 = preoperative CA19-9 - postoperative day 7 CA19-9) in patients with R0 and R1 resections, showing no significant change in CA19-9 between R0 and R1 resections. It can be seen that miRNA index I decreased significantly on postoperative day 7, and the magnitude of this decrease could be used to distinguish between R0 and R1 surgical status (p = 0.043), while CA19-9 did not show a significant difference in this respect, indicating that miRNA index I is more sensitive in reflecting tumor burden and clearance.

[0092] (2) Prognostic prediction

[0093] Figure 5In the figure, A and B represent the correlations between miRNA index I and overall survival and recurrence-free survival after surgery for extrahepatic bile duct cancer, respectively, in the Kaplan-Meier survival analysis. Figure 5 As can be seen, high miRNA index I is significantly associated with shorter overall survival and recurrence-free survival. In Cox multivariate regression, miRNA index I is an independent prognostic factor (OS: HR = 2.97, p = 0.012; RFS: HR = 2.37, p = 0.016), which can be used for risk stratification and adjuvant therapy guidance.

[0094] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. The application of a reagent for detecting the expression level of a biomarker in the preparation of products for the differential diagnosis of extrahepatic cholangiocarcinoma, characterized in that, The biomarkers are a combination of miR-34c-5p, miR-100-5p, miR-193b-5p, miR-122-5p, miR-5588-5p and miR-122-3p; The detection includes in vitro detection of the concentration of the biomarker in a biological sample, wherein the biological sample is plasma.

2. The application of a reagent for detecting the expression level of a biomarker in the preparation of a product for postoperative efficacy evaluation of extrahepatic cholangiocarcinoma, characterized in that, The biomarkers are a combination of miR-34c-5p, miR-100-5p, miR-193b-5p, miR-122-5p, miR-5588-5p and miR-122-3p; The detection includes in vitro detection of the concentration of the biomarker in a biological sample, wherein the biological sample is plasma.

3. The application of a reagent for detecting the expression level of a biomarker in the preparation of a product for predicting the prognosis of extrahepatic cholangiocarcinoma, characterized in that, The biomarkers are a combination of miR-34c-5p, miR-100-5p, miR-193b-5p, miR-122-5p, miR-5588-5p and miR-122-3p; The detection includes in vitro detection of the concentration of the biomarker in a biological sample, wherein the biological sample is plasma.

4. The application as described in any one of claims 1 to 3, characterized in that, The product includes at least one of reagents, reagent kits, test strips, chips, and systems.

5. A system for the diagnosis, postoperative efficacy evaluation, or prognostic assessment of extrahepatic bile duct carcinoma, characterized in that, include: The data acquisition module is used to acquire the expression levels of biomarkers in the biological samples of the subjects; the biomarkers are a combination of miR-34c-5p, miR-100-5p, miR-193b-5p, miR-122-5p, miR-5588-5p and miR-122-3p; the biological sample is plasma; The model prediction module is used to construct a prediction model based on the expression levels of biomarkers obtained by the data acquisition module and output a risk score. The data output module is used to output prediction results based on the risk score.

6. The system as described in claim 5, characterized in that, The prediction model is constructed based on a machine learning algorithm, which is selected from any one of the following algorithms: logistic regression, linear regression, random forest, neural network, support vector machine, Bayesian classification, gradient boosting, K-nearest neighbors, or decision tree.

7. The system as described in claim 5, characterized in that, The formula for calculating the risk score is as follows: Risk score = 1.217 + 0.0264×miR-34c-5p expression level + 0.0412×miR-100-5p expression level + 0.0110×miR-193b-5p expression level + 0.0113×miR-122-5p expression level + 0.0156×miR-5588-5p expression level + 0.0172×miR-122-3p expression level; A risk score of -0.53 or higher is considered high risk; a risk score of -0.53 or lower is considered low risk.

8. The system as described in claim 5, characterized in that, The model prediction module is used to construct a prediction model based on the expression levels of biomarkers and CA19-9 obtained by the data acquisition module, and output a risk score. The formula for calculating the risk score is as follows: Where, joint factor = α 0+ α 1× Composite Index + α 2× CA19-9 expression level; Composite Index = 1.217 + 0.0264×miR-34c-5p expression level + 0.0412×miR-100-5p expression level + 0.0110×miR-193b-5p expression level + 0.0113×miR-122-5p expression level + 0.0156×miR-5588-5p expression level + 0.0172×miR-122-3p expression level; α 0 = -3.08443; α 1 = 0.95618; α 2 = 0.09363; CA19-9 serum concentration is a value in "U / mL"; A risk score of ≥0.35 is considered high risk; a risk score of <0.35 is considered low risk.

9. A computer-readable storage medium, characterized in that, It stores a computer program, which, when executed by a processor, can perform the functions of the system according to any one of claims 5 to 8.

Images