Discrimination method for intrahepatic cholangiocarcinoma based on serum polypeptide characteristics and application thereof

Abstract

This invention discloses a method for identifying intrahepatic cholangiocarcinoma (ICC) based on serum peptide characteristics and its application. The invention collects serum samples from three groups: individuals with ICC, those with benign liver disease, and healthy controls. Peptide characteristic peaks associated with ICC are screened, and a model is constructed using a logistic regression algorithm. This model demonstrates excellent discriminative ability on both the training and independent test sets, with AUCs reaching 0.986 and 0.963, respectively, significantly outperforming traditional tumor markers CA19-9 and CEA. The peptide characteristic peak combination and the discriminative model constructed based on it can be used for early detection of ICC, differentiation between benign and malignant cases, and screening of high-risk populations. It offers advantages such as rapid detection, standardized procedures, strong early ICC identification capability, and low false positive rate, providing a highly sensitive, specific, and widely applicable non-invasive screening tool for clinical use.

Description

Technical Field

[0001] This invention belongs to the fields of biomedical detection technology, mass spectrometry detection technology, and computer-aided diagnostic technology. Specifically, it relates to a method for identifying intrahepatic cholangiocarcinoma (ICC) based on serum polypeptide characteristics and its application. Background Technology

[0002] Intrahepatic cholangiocarcinoma (ICC) is a highly aggressive malignant tumor originating from the epithelial cells of the intrahepatic bile ducts, second only to hepatocellular carcinoma among primary liver cancers. Due to its insidious onset, rapid progression, and lack of specific early symptoms, most patients are diagnosed at an intermediate or advanced stage, losing the opportunity for radical treatment, resulting in a generally poor prognosis.

[0003] Currently, the clinical diagnosis of ICC mainly relies on imaging examinations (such as CT and MRI) combined with serum tumor marker detection. Among them, CA19-9 and carcinoembryonic antigen (CEA) are the most commonly used serological markers in clinical practice. However, their sensitivity in early-stage ICC is limited, and they are prone to false positives in benign hepatobiliary diseases such as cholangitis, cirrhosis, and bile duct stones, which seriously limits their application value in early screening and differential diagnosis. In addition, imaging methods have limited ability to detect small lesions or very early-stage lesions and are not suitable for large-scale population screening.

[0004] With the development of liquid biopsy technology, the detection of molecular markers based on body fluid samples such as serum and plasma has gradually become an important research direction for the early diagnosis of tumors. Among them, serum peptidomics can reflect the comprehensive changes in protein hydrolysis, immune response, and metabolic networks in the body under disease states, and has advantages such as non-invasive sampling, high throughput, and good reproducibility. Matrix-assisted laser desorption / ionization time-of-flight mass spectrometry (MALDI-TOF MS) technology has shown the characteristics of fast detection speed, stable analysis, and suitability for large-scale clinical applications in serum peptide detection.

[0005] Although existing studies have suggested the potential value of serum peptide profiling in the diagnosis of biliary tract tumors, there is currently a lack of stable peptide combinations and quantitative discrimination models for the early identification of intrabiliary tract tumors (ICCs) based on large-sample, multi-center data systems. In particular, there is a lack of clearly defined characteristic peptide sets, algorithmic procedures, and discrimination thresholds, hindering clinical translation. Therefore, it is necessary to provide a method for constructing an early ICC discrimination model based on serum peptidomics to improve the early detection rate of ICCs and reduce the misdiagnosis rate. Summary of the Invention

[0006] The purpose of this invention is to provide a method for identifying intrahepatic cholangiocarcinoma (ICC) based on serum polypeptide characteristics and its application. By combining serum polypeptide detection with statistical and machine learning algorithms, characteristic polypeptide biomarkers significantly associated with ICC are screened, and a stable and quantifiable early identification model is established. This enables highly sensitive and specific identification of ICC, especially early-stage ICC, providing technical support for non-invasive clinical screening and auxiliary diagnosis.

[0007] In a first aspect, the present invention provides a method for identifying intrahepatic cholangiocarcinoma based on serum polypeptide characteristics, comprising:

[0008] Serum samples were obtained from the subjects and peptides were extracted and mass spectrometrically analyzed to obtain serum peptide spectra.

[0009] Serum polypeptide spectra were preprocessed and characteristic polypeptide peaks corresponding to the ICC polypeptide marker set were extracted;

[0010] The characteristic peaks of the polypeptide are input into a pre-constructed early discrimination model for intrahepatic cholangiocarcinoma. The discrimination score of the subject is calculated based on the polypeptide characteristics. The subject is then judged to be either ICC or non-ICC based on the discrimination score.

[0011] The early detection model for intrahepatic cholangiocarcinoma was constructed through the following steps:

[0012] (1) Collect serum samples from subjects, including serum samples from intrahepatic cholangiocarcinoma, benign liver disease and healthy controls, and determine the final study cohort according to the pre-set inclusion and exclusion criteria; use a unified standardized operating procedure to preprocess serum samples, extract peptides by silicon-based micro-nano particle enrichment method, and use MALDI-TOF mass spectrometry to acquire serum peptide spectra; the characteristic peak range of the initial extraction is 680 to 18600 Da, and batch quality control and instrument calibration quality control are performed on all serum samples;

[0013] (2) Baseline correction, normalization, peak identification and alignment were performed on the obtained serum polypeptide spectrum data, and the polypeptide amino acid sequence was confirmed. The difference fold, P value and genetic algorithm were used to screen polypeptide characteristic peaks that were significantly related to intrahepatic cholangiocarcinoma and had a high repetition frequency, so as to obtain the ICC polypeptide marker set.

[0014] (3) Based on the set of ICC peptide markers obtained by screening, a machine learning discrimination model is constructed and trained to obtain an early discrimination model for intrahepatic cholangiocarcinoma.

[0015] Furthermore, in step (1), the research samples are from multiple centers, and the ratio of serum samples from intrahepatic cholangiocarcinoma, benign liver disease and healthy controls is not less than 16:13:11.

[0016] Further, in step (2), the peptide characteristic peaks screened by difference fold, P-value, and genetic algorithm are shown in Table 2, totaling 15 peaks. Their mass-to-charge ratios are 4058.69, 2009.70, 2724.45, 1477.25, 2270.33, 1467.69, 4106.66, 3931.45, 2080.85, 1629.48, 2112.56, 1279.83, 1627.32, 1451.07, and 1516.21, respectively. The mass-to-charge ratio deviation range of each mass spectrum peak is ±500 ppm. The corresponding peptide sequences are YNPQSRSVPPSASHVAPTETFTYEWTVPKEVGPTNAD, DQTVSDNELQEMSNQGSK, and PGV. LSSRQLGLPGPPDVPDHAAYHPF,TDTGALLFIGKILD,EIVLTQSPATLSLSPGERATLS,ALTNAVAHVDDMPN,KPALEDLRQGLLPVLESFKVSFLSALEEYTKKLNTQ,HGKKVADALTNAVAHVDDM PNALSALSDLHAHKLRVD, SETESPRNPSSAGSWNSGSSG, GSTSYGTGSETESPRN, AALLSPYSYSTTAVVTNPKE, FFSTYDRDND, KNSLFEYQKNNKD, SWNSGSSGPGSTGNR, STFTGFLLYHDTN.

[0017] Furthermore, in step (3), the early detection model for intrahepatic cholangiocarcinoma is represented as follows:

[0018] Score = (-0.90299) * Intensity (M1) + (-0.17492) * Intensity (M2) + (-0.21178) * Intensity (M3) + (0.02580) * Intensity (M4) + (-0.71448) * Intensity (M5) + (0.52736) * Intensity (M6) + (-0.30933) * Intensity (M7) + (-0.25675) * Intensity (M8) + 2.69877, where M1 to M8 are the 8 most discriminative characteristic peptide peaks selected as shown in Table 2, Intensity (*) represents the intensity of the corresponding peptide characteristic peak, Score represents the discriminant score output by the model, and the positive judgment threshold is Score ≥ 2.

[0019] In a second aspect, the present invention provides a system for identifying intrahepatic cholangiocarcinoma based on serum polypeptide characteristics, comprising:

[0020] The peptide detection module is used to acquire serum samples from the subject and perform peptide extraction and mass spectrometry detection to obtain a serum peptide spectrum.

[0021] The data processing module is used to preprocess the serum polypeptide spectrum and extract the polypeptide characteristic peaks corresponding to the ICC polypeptide marker set;

[0022] The discriminant analysis module is used to input the characteristic peaks of peptides into a pre-constructed early discrimination model for intrahepatic cholangiocarcinoma. Based on the peptide characteristics, the discriminant score of the subject is calculated, and the subject is determined to be either ICC or non-ICC based on the discriminant score.

[0023] Furthermore, it also includes:

[0024] The combined analysis module is used to perform combined analysis of the peptide discrimination score and traditional serum biomarkers, wherein the traditional serum biomarkers include CA19-9 and carcinoembryonic antigen.

[0025] Furthermore, when the discrimination score is ≥2, the system determines it as ICC; when the discrimination score is below the threshold, it is determined as non-ICC.

[0026] In a third aspect, the present invention provides a combination of serum polypeptide markers for screening intrahepatic cholangiocarcinoma, wherein the combination of serum polypeptide markers includes any two or more of the 15 characteristic polypeptides shown in Table 2.

[0027] Furthermore, the polypeptide marker combination includes at least polypeptides M1 to M8.

[0028] In a fourth aspect, the present invention also provides the application of a reagent for detecting a set of ICC polypeptide markers in the preparation of a screening kit for intrahepatic cholangiocarcinoma, wherein the set of ICC polypeptide markers includes any two or more of the 15 characteristic polypeptides shown in Table 2.

[0029] Furthermore, the kit includes at least one or more of the following components:

[0030] Serum peptide enrichment reagents are used to enrich endogenous peptides from serum samples and remove high-abundance proteins.

[0031] Mass spectrometry matrix reagents for matrix-assisted laser desorption / ionization time-of-flight mass spectrometry detection of serum peptides;

[0032] Standards or internal reference materials are used for instrument calibration or batch quality control during mass spectrometry detection.

[0033] Sample processing buffers are used for the dilution, reaction, or stabilization of serum samples.

[0034] Instructions for use are provided to guide the handling, testing, and interpretation of test samples.

[0035] The aforementioned reagents may be packaged individually or in combination within the same packaging unit.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] This invention discloses a method for identifying intrahepatic cholangiocarcinoma based on serum peptide characteristics. This method utilizes a pre-constructed early identification model for intrahepatic cholangiocarcinoma based on serum peptide characteristics. The model includes several core serum peptide characteristics screened by peptide mass spectrometry analysis. Genetic algorithms and logistic regression methods are combined to model and analyze these peptide characteristics. A discrimination score for intrahepatic cholangiocarcinoma is obtained through quantitative calculation of the peptide signals in serum samples. This discrimination model can be used for early risk assessment of intrahepatic cholangiocarcinoma in the examined population, helping clinicians to identify intrahepatic cholangiocarcinoma in its early stages, thus providing a basis for subsequent imaging examinations and further diagnosis and treatment. Simultaneously, this invention's method, based on the model's low false positive rate in individuals with benign liver disease, can effectively reduce misdiagnosis and unnecessary examinations caused by elevated traditional tumor markers, avoiding excessive intervention in non-tumor patients. This invention helps improve the early detection rate of intrahepatic cholangiocarcinoma, achieving non-invasive screening and auxiliary identification of intrahepatic cholangiocarcinoma, and has good clinical application value and promising prospects for promotion. Attached Figure Description

[0038] The present invention will be further described below with reference to the accompanying drawings and embodiments;

[0039] Figure 1 This is a flowchart of a method for constructing an early detection model for intrahepatic cholangiocarcinoma based on serum polypeptide characteristics according to an embodiment of the present invention.

[0040] Figure 2 The figure shows the ROC curves of the early intrahepatic cholangiocarcinoma discrimination model constructed according to an embodiment of the present invention in the test set (left) and cross-validation set (right). In the figure, the peptide model is the ROC curve corresponding to the early intrahepatic cholangiocarcinoma discrimination model of the present invention, CEA represents the ROC curve corresponding to the single indicator discrimination model constructed based on the detection results of carcinoembryonic antigen in the serum of the tested subjects, CA19-9 represents the ROC curve corresponding to the single indicator discrimination model constructed based on the detection results of carbohydrate antigen 19-9 in the serum of the tested subjects, and the combined model represents the ROC curve corresponding to the discrimination model constructed after the score output by the early intrahepatic cholangiocarcinoma discrimination model of the present invention is combined with two traditional serum tumor marker indicators.

[0041] Figure 3 This is a schematic diagram of the ROC curve of an early intrahepatic cholangiocarcinoma discrimination model constructed according to an embodiment of the present invention on an external validation set. Detailed Implementation

[0042] This invention provides a method for identifying intrahepatic cholangiocarcinoma based on serum peptide characteristics and its application. This method detects peptides in serum samples, combines statistical analysis with machine learning algorithms to screen for peptide characteristics significantly associated with intrahepatic cholangiocarcinoma, and establishes a stable and quantifiable early detection model, thereby achieving intrahepatic cholangiocarcinoma identification based on serum peptide characteristics. Specifically, the method of this invention is as follows:

[0043] Serum samples were obtained from the subjects and peptides were extracted and mass spectrometrically analyzed to obtain serum peptide spectra.

[0044] Serum polypeptide spectra were preprocessed and characteristic polypeptide peaks corresponding to the ICC polypeptide marker set were extracted;

[0045] The characteristic peaks of the polypeptide are input into a pre-constructed early discrimination model for intrahepatic cholangiocarcinoma. The discrimination score of the subject is calculated based on the polypeptide characteristics, and the subject is judged to be ICC or not based on the discrimination score.

[0046] The early detection model for intrahepatic cholangiocarcinoma was constructed through the following steps:

[0047] (1) Collect serum samples from subjects, including serum samples from intrahepatic cholangiocarcinoma, benign liver disease and healthy controls, and determine the final study cohort according to the pre-set inclusion and exclusion criteria; use a unified standardized operating procedure to preprocess serum samples, extract peptides by silicon-based micro-nano particle enrichment method, and use MALDI-TOF mass spectrometry to acquire serum peptide spectra; the characteristic peak range of the initial extraction is 680 to 18600 Da, and batch quality control and instrument calibration quality control are performed on all serum samples;

[0048] (2) Baseline correction, normalization, peak identification and alignment were performed on the obtained serum polypeptide spectrum data, and the polypeptide amino acid sequence was confirmed. The difference fold, P value and genetic algorithm were used to screen polypeptide characteristic peaks that were significantly related to intrahepatic cholangiocarcinoma and had a high repetition frequency, so as to obtain the ICC polypeptide marker set.

[0049] (3) Based on the peptide characteristic peaks corresponding to the ICC peptide marker set obtained by screening, a machine learning discrimination model is constructed and trained to obtain an early discrimination model for intrahepatic cholangiocarcinoma.

[0050] In one specific embodiment, the present invention first retrospectively collects serum samples from patients with intrahepatic cholangiocarcinoma, patients with benign liver disease, and healthy individuals undergoing physical examinations at multiple medical centers. Based on pre-defined inclusion and exclusion criteria, the final total cohort for inclusion in the study is determined. The total cohort is randomly divided into a training set and a validation set (the ratio of training set sample size to validation set sample size is 3:1). Peptide feature screening and model construction are performed in the training set, while the discriminative power of the model is evaluated in the validation set.

[0051] Both internal and external validation results demonstrate that the early detection model for intrahepatic cholangiocarcinoma constructed based on eight core serum polypeptide characteristics exhibits high accuracy, stability, and clinical applicability in distinguishing patients with intrahepatic cholangiocarcinoma from those without. Stratified analysis of the model's scoring results allows for the classification of subjects into different risk levels, providing clinicians with a reference for further imaging examinations and treatment decisions.

[0052] The effects of the present invention will be further explained below with reference to specific embodiments.

[0053] Example 1: Construction of an early detection model for intrahepatic cholangiocarcinoma based on serum peptide characteristics

[0054] This embodiment provides a method for constructing an early discrimination model for intrahepatic cholangiocarcinoma based on serum polypeptide characteristics, such as... Figure 1 As shown, the specific steps include the following:

[0055] (1) Collection of serum samples and clinical information, including the following sub-steps:

[0056] (1.1) Sample Collection and Grouping A retrospective collection of serum samples from 421 subjects from multiple medical institutions was conducted (clinical baseline characteristics are shown in Table 1). Subjects included patients with clinically and / or pathologically confirmed intrahepatic cholangiocarcinoma, patients with benign liver disease, and healthy controls. All subjects met the pre-defined inclusion and exclusion criteria. Based on the disease diagnosis results, the samples were divided into the ICC group, the benign liver disease group, and the healthy control group.

[0057] Inclusion criteria:

[0058] (1) ICC queue inclusion criteria:

[0059] ① Age ≥ 18 years. ② Patients diagnosed with intrahepatic cholangiocarcinoma by pathological examination and / or clinical diagnosis.

[0060] (2) Inclusion criteria for the interference cohort (benign liver disease):

[0061] ① Age ≥ 18 years. ② No history of hepatobiliary malignancy, and no suspicious related malignant lesions seen clinically and / or on imaging. ③ Meets any of the following criteria: chronic hepatitis B (HBV chronic infection), chronic hepatitis C (HCV chronic infection), cirrhosis of any cause, alcoholic liver disease, or metabolic dysfunction-associated fatty liver disease (MAFLD).

[0062] (3) Inclusion criteria for the healthy population cohort:

[0063] ① Age ≥ 18 years, with no physical discomfort. ② No history of hepatobiliary malignancies, and no suspicious related malignant lesions seen clinically and / or on imaging. ③ No presence of any of the following factors: chronic hepatitis B (HBV chronic infection), chronic hepatitis C (HCV chronic infection), cirrhosis of any cause, alcoholic fatty liver disease, MAFLD. ④ Individuals deemed healthy by physical examination or comprehensive clinical assessment.

[0064] Exclusion criteria:

[0065] ① Pregnancy or lactation. ② Severe bleeding tendency, or a major trauma requiring blood transfusion within the past week. ③ Participation in a clinical intervention study within the past 30 days (e.g., taking an investigational drug). ④ Received surgery, radiotherapy, chemotherapy, targeted therapy, or other related tumor interventions within the past 30 days. ⑤ Concurrently suffering from other malignant tumors.

[0066] Table 1. Clinical characteristics of 421 patients in the modeling cohort.

[0067]

[0068] (1.2) Serum Sample Processing and Peptide Extraction: Peripheral venous blood was collected, and serum samples were obtained by centrifugation and then stored at low temperature. A standardized serum peptide enrichment reagent was used to extract endogenous peptides from the serum to reduce the interference of high-abundance proteins on the detection results.

[0069] (1.3) Peptide mass spectrometry detection: Matrix-assisted laser desorption / ionization time-of-flight mass spectrometry (MALDI-TOF MS) was used to detect the extracted serum peptides, obtaining peptide mass spectral peak information for each sample within a preset mass-to-charge ratio range, forming a serum peptide fingerprint spectrum. The initial extracted characteristic peak range was 680–18,600 Da, and batch quality control and instrument calibration quality control were performed on all serum samples.

[0070] (1.4) Peptide spectrum data preprocessing: Baseline correction, noise reduction, peak identification, peak alignment and normalization are performed on the raw mass spectrometry data to obtain standardized peptide peak intensity data for subsequent analysis.

[0071] (2) Divide the samples into training set and test set in a 3:1 ratio.

[0072] (3) Differential peptide screening and feature determination, followed by the construction of a discriminant model, as follows: (3.1) In the training samples, differential analysis was performed on the peptide peaks of the ICC group and the non-ICC group (benign liver disease group and healthy control group), and peptide characteristic peaks with stable differential expression between the two groups were screened out; further, combined with the feature selection algorithm, the set of ICC peptide biomarkers used for model construction was determined. In this embodiment, 15 peptide characteristic peaks were screened out, and detailed information is shown in Table 2. Among them, the M / Z mass-to-charge ratio deviation range of each mass spectrometry peak is ±500ppm.

[0073] Table 2. Characteristics of core peptides

[0074]

[0075] (3.1) The discriminant model is constructed by using the mass spectrometry intensity of the peptide characteristic peaks corresponding to the set of ICC peptide markers as input variables and using machine learning methods to construct and train an early discriminant model for intrahepatic cholangiocarcinoma. During training, the loss function is to minimize the error between the discriminant result output by the model and the true value until the loss function converges or the set number of training times is reached, and the final trained early discriminant model for intrahepatic cholangiocarcinoma is obtained. The model constructed in this embodiment is represented as follows: Score = (-0.90299) * Intensity (M1) + (-0.17492) * Intensity (M2) + (-0.21178) * Intensity (M3) + (0.02580) * Intensity (M4) + (-0.71448) * Intensity (M5) + (0.52736) * Intensity (M6) + (-0.30933) * Intensity (M7) + (-0.25675) * Intensity (M8) + 2.69877, where M1 to M8 are the 8 most discriminative characteristic peptide peaks obtained through screening, and the positive judgment threshold is Score ≥ 2.

[0076] Example 2: Application of an early detection model for intrahepatic cholangiocarcinoma based on serum peptide characteristics (internal and external validation)

[0077] This embodiment applies and validates the early detection model of intrahepatic cholangiocarcinoma based on serum polypeptide characteristics constructed in Example 1, in order to evaluate its discriminative efficacy and stability.

[0078] (1) Internal Validation (Training Set and Test Set): The 421 samples from Example 1 were divided into a training set and a test set according to a preset ratio. The parameters of the discrimination model were optimized using the training set, and the completed model was applied to the test set to calculate the corresponding discrimination score for each subject. The details are as follows:

[0079] Serum samples were obtained from the subjects and peptides were extracted and mass spectrometrically analyzed to obtain serum peptide spectra.

[0080] Serum polypeptide spectra were preprocessed and characteristic polypeptide peaks corresponding to the ICC polypeptide marker set were extracted;

[0081] The characteristic peaks of the peptides are input into a pre-constructed early discrimination model for intrahepatic cholangiocarcinoma. The discrimination score of the subject is calculated based on the peptide characteristics. The subject is judged as ICC or non-ICC based on the discrimination score. When the discrimination score is ≥2, the system judges it as ICC. When the discrimination score is below the threshold, it is judged as non-ICC.

[0082] The model's ability to distinguish between intrahepatic cholangiocarcinoma (ICC) and non-ICC populations was analyzed using receiver operating characteristic (ROC) curves, and the area under the curve (AUC) was calculated as an evaluation metric for discriminative performance. Results showed that the early-stage intrahepatic cholangiocarcinoma discrimination model constructed from eight peptide peaks had an AUC of 0.986 in the training set and 0.963 in the test set, indicating that the model possessed high discriminative ability in internal validation. Figure 2 As shown in the figure. In contrast, the traditional serum tumor markers CA19-9 and CEA showed significantly lower discriminative power in the same test set, further demonstrating that the peptide discriminant model described in this invention has superior distinguishing ability.

[0083] (2) External independent validation: Further external independent serum samples that did not participate in the model building process were collected as an external validation set. The external validation set samples came from independent subject populations different from the modeling cohort. The serum processing, peptide detection and data analysis process was carried out according to the method of the present invention. The external samples were detected, and the standard intensity of the peptide characteristic peak corresponding to the ICC peptide marker set was extracted. The discriminant score was directly calculated by calling the constructed discriminant model.

[0084] The discriminant performance of the model in external samples was evaluated using ROC curve analysis to verify its stability and generalization ability under different sample sources and testing batch conditions. The results showed that the AUC of the serum peptide discriminant model on the external validation set was 0.971, indicating that the model can maintain good discriminant performance under different sample source conditions. Figure 3 As shown.

[0085] (3) The verification results show that the early discrimination model for intrahepatic cholangiocarcinoma can effectively distinguish patients with intrahepatic cholangiocarcinoma from the control group in both internal and external verification, showing good discrimination accuracy and stability, and is suitable for early auxiliary discrimination of intrahepatic cholangiocarcinoma.

[0086] Furthermore, this invention further combines the output score of the early-stage intrahepatic cholangiocarcinoma discrimination model with the detection results of serum tumor markers such as CA19-9 and CEA to construct a joint discrimination model, which exhibits higher sensitivity and specificity in both internal and external test sets. Specifically, the joint model in this invention uses the output score of the serum peptide model and the detection values ​​of CA19-9 and CEA as joint input variables, and performs fusion modeling using logistic regression to output a comprehensive screening probability for the auxiliary diagnosis of intrahepatic cholangiocarcinoma.

[0087] Specifically, firstly, an early discrimination model for intrahepatic cholangiocarcinoma was constructed based on peptide characteristics to obtain a discrimination score for each subject; secondly, the quantitative detection values ​​of CA19-9 and CEA, along with the discrimination score, were used as independent variables, and logistic regression analysis was introduced to establish a joint discrimination model with whether it is intrahepatic cholangiocarcinoma as the dependent variable.

[0088] During model construction, the regression coefficients of each variable are determined by fitting the training set data, thereby achieving a weighted fusion of peptide characteristics and serum tumor marker information. The joint model ultimately outputs a comprehensive discriminant probability value to determine the subject's disease risk.

[0089] like Figure 2 As shown, the combined model described in this invention was validated on an independent test set and an external validation set. Its discriminative efficacy is superior to that of using a single peptide model or a single tumor marker, demonstrating the technical advantages of multimodal information fusion in the early diagnosis of intrahepatic cholangiocarcinoma.

[0090] Example 3: An early detection system for intrahepatic cholangiocarcinoma based on serum polypeptide characteristics and its application in combination with traditional serum biomarkers

[0091] This embodiment provides an early detection system for intrahepatic cholangiocarcinoma based on serum polypeptide characteristics, which can be used for clinical auxiliary diagnosis and screening of high-risk groups.

[0092] The discrimination system includes:

[0093] (1) Peptide detection module, used to obtain serum samples of the test subjects and perform peptide extraction and mass spectrometry detection to obtain serum peptide spectrum;

[0094] (2) Data processing module, used to preprocess serum polypeptide spectra and extract core polypeptide characteristic peaks;

[0095] (3) Discriminant analysis module, used to call the early discrimination model of intrahepatic cholangiocarcinoma constructed in Example 1, and calculate the discrimination score of the subject based on the core polypeptide characteristics;

[0096] Furthermore, it also includes: a result output module, used to output the risk assessment results of the examinee having intrahepatic cholangiocarcinoma, that is: used to determine whether the examinee is ICC or not based on the discrimination score; specifically, when the discrimination score is ≥ a preset threshold T (e.g., T=2), the system determines it to be ICC; when the discrimination score is lower than the threshold, it determines it to be not ICC.

[0097] Furthermore, it also includes: a joint analysis module, used to perform joint analysis of the polypeptide diagnostic score with traditional serum biomarker indicators, wherein the traditional serum biomarker indicators include, but are not limited to, CA19-9 and carcinoembryonic antigen;

[0098] By combining serum peptide peak characteristics with traditional hematological indicators, this system can further improve the screening accuracy of intrahepatic cholangiocarcinoma, especially early-stage intrahepatic cholangiocarcinoma, and provide auxiliary decision-making basis for clinicians to formulate subsequent examination and treatment plans.

[0099] Example 4: A screening kit for intrahepatic cholangiocarcinoma based on serum peptide characteristics

[0100] This embodiment provides a serum polypeptide detection kit for screening intrahepatic cholangiocarcinoma. The kit is used to obtain polypeptide characteristic information in the serum sample of the subject and, in combination with a discriminant model constructed based on serum polypeptide characteristics, to assess the risk of intrahepatic cholangiocarcinoma in the subject.

[0101] (1) Kit composition

[0102] The kit includes at least one or more of the following components:

[0103] Serum peptide enrichment reagents are used to enrich endogenous peptides from serum samples and remove high-abundance proteins.

[0104] Mass spectrometry matrix reagents for matrix-assisted laser desorption / ionization time-of-flight mass spectrometry detection of serum peptides;

[0105] Standards or internal reference materials are used for instrument calibration or batch quality control during mass spectrometry detection.

[0106] Sample processing buffers are used for the dilution, reaction, or stabilization of serum samples.

[0107] Instructions for use are provided to guide the handling, testing, and interpretation of test samples.

[0108] The aforementioned reagents may be packaged individually or in combination within the same packaging unit.

[0109] (2) How to use the kit

[0110] The following steps are included when using the kit to screen for intrahepatic cholangiocarcinoma:

[0111] 1) Collect peripheral venous blood from the subjects and separate to obtain serum samples;

[0112] 2) Process the serum sample using the serum peptide enrichment reagent to obtain enriched peptide components; 3) Mix the peptide sample with the mass spectrometry matrix reagent and perform MALDI-TOF MS detection to obtain serum peptide mass spectrometry data; 4) Extract peptide characteristic peak information related to intrahepatic cholangiocarcinoma and input the peptide characteristics into the constructed early intrahepatic cholangiocarcinoma discrimination model; 5) Output the risk assessment results of the subject having intrahepatic cholangiocarcinoma.

[0113] (3) Technical effects

[0114] Using the above-mentioned kit and its usage method, stable detection of serum polypeptide characteristics can be achieved, and combined with intrahepatic cholangiocarcinoma-related polypeptide markers, to achieve efficient screening of intrahepatic cholangiocarcinoma, which is especially suitable for early risk assessment of people undergoing physical examinations or high-risk groups.

[0115] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for identifying intrahepatic cholangiocarcinoma based on serum polypeptide characteristics, characterized in that, include: Serum samples were obtained from the subjects and peptides were extracted and mass spectrometrically analyzed to obtain serum peptide spectra. Serum polypeptide spectra were preprocessed and characteristic polypeptide peaks corresponding to the ICC polypeptide marker set were extracted; The characteristic peaks of the polypeptide are input into a pre-constructed early discrimination model for intrahepatic cholangiocarcinoma. The discrimination score of the subject is calculated based on the polypeptide characteristics. The subject is then judged to be either ICC or non-ICC based on the discrimination score. The early detection model for intrahepatic cholangiocarcinoma was constructed through the following steps: (1) Collect serum samples from subjects, including serum samples from intrahepatic cholangiocarcinoma, benign liver disease and healthy controls, and determine the final study cohort according to the pre-set inclusion and exclusion criteria; use a unified standardized operating procedure to preprocess serum samples, extract peptides by silicon-based micro-nano particle enrichment method, and use MALDI-TOF mass spectrometry to acquire serum peptide spectra; the characteristic peak range of the initial extraction is 680 to 18600 Da, and batch quality control and instrument calibration quality control are performed on all serum samples; (2) Baseline correction, normalization, peak identification and alignment were performed on the obtained serum polypeptide spectrum data, and the polypeptide amino acid sequence was confirmed. The difference fold, P value and genetic algorithm were used to screen polypeptide characteristic peaks that were significantly related to intrahepatic cholangiocarcinoma and had a high repetition frequency, so as to obtain the ICC polypeptide marker set. (3) Based on the peptide characteristic peaks corresponding to the ICC peptide marker set obtained by screening, a machine learning discrimination model is constructed and trained to obtain an early discrimination model for intrahepatic cholangiocarcinoma.

2. The method according to claim 1, characterized in that, In step (1), the serum samples of the subjects are from multiple centers, and the ratio of serum samples from intrahepatic cholangiocarcinoma, benign liver disease and healthy controls is not less than 16:13:

11.

3. The method according to claim 1, characterized in that, In step (2), the set of ICC peptide markers selected by difference fold, P-value, and genetic algorithm includes any two or more of the following 15 characteristic peptides: The polypeptide M1 with a mass-to-charge ratio of 4058.69±500ppm has the amino acid sequence YNPQSRSVPPSASHVAPTETFTYEWTVPKEVGPTNAD. The polypeptide M2 ​​with a mass-to-charge ratio of 2009.70±500ppm has the amino acid sequence DQTVSDNELQEMSNQGSK. The polypeptide M3, with a mass-to-charge ratio of 2724.45 ± 500 ppm, has the amino acid sequence PGVLSSRQLGLPGPPDVPDHAAYHPF. The polypeptide M4, with a mass-to-charge ratio of 1477.25 ± 500 ppm, has the amino acid sequence TDTGALLFIGKILD. The polypeptide M5, with a mass-to-charge ratio of 2270.33 ± 500 ppm, has the amino acid sequence EIVLTQSPATLSLSPGERATLS; The polypeptide M6 with a mass-to-charge ratio of 1467.69±500ppm has the amino acid sequence ALTNAVAHVDDMPN. The polypeptide M7, with a mass-to-charge ratio of 4106.66±500ppm, has the amino acid sequence KPALEDLRQGLLPVLESFKVSFLSALEEYTKKLNTQ. The polypeptide M8 with a mass-to-charge ratio of 3931.45±500ppm has the amino acid sequence HGKKVADALTNAVAHVDDMPNALSALSDLHAHKLRVD. The amino acid sequence of peptide M9, with a mass-to-charge ratio of 2080.85±500ppm, is SETESPRNPSSAGSWNSGSSG. The polypeptide M10, with a mass-to-charge ratio of 1629.48 ± 500 ppm, has the amino acid sequence GSTSYGTGSETESPRN. The polypeptide M11 with a mass-to-charge ratio of 2112.56±500ppm has the amino acid sequence AALLSPYSYSTTAVVTNPKE. The polypeptide M12, with a mass-to-charge ratio of 1279.83 ± 500 ppm, has the amino acid sequence FFSTYDRDND. The polypeptide M13, with a mass-to-charge ratio of 1627.32 ± 500 ppm, has the amino acid sequence KNSLFEYQKNNKD. The polypeptide M14, with a mass-to-charge ratio of 1451.07 ± 500 ppm, has the amino acid sequence SWNSGSSGPGSTGNR. The polypeptide M15, with a mass-to-charge ratio of 1516.21 ± 500 ppm, has the amino acid sequence STFTGFLLYHDTN.

4. The method according to claim 3, characterized in that, In step (3), the early detection model for intrahepatic cholangiocarcinoma is represented as follows: Score = (-0.90299) * Intensity (M1) + (-0.17492) * Intensity (M2) + (-0.21178) * Intensity (M3) + (0.02580) * Intensity (M4) + (-0.71448) * Intensity (M5) + (0.52736) * Intensity (M6) + (-0.30933) * Intensity (M7) + (-0.25675) * Intensity (M8) + 2.69877, where Intensity (*) represents the intensity of the corresponding peptide characteristic peak, and Score represents the discriminant score output by the model.

5. The method according to claim 3, characterized in that, Also includes: The discriminant score was combined with traditional serum biomarkers, including CA19-9 and carcinoembryonic antigen.

6. A system for identifying intrahepatic cholangiocarcinoma based on serum polypeptide characteristics, characterized in that, include: The peptide detection module is used to acquire serum samples from the subject and perform peptide extraction and mass spectrometry detection to obtain a serum peptide spectrum. The data processing module is used to preprocess the serum polypeptide spectrum and extract the polypeptide characteristic peaks corresponding to the ICC polypeptide marker set; The discriminant analysis module is used to input the characteristic peaks of peptides into a pre-constructed early discrimination model for intrahepatic cholangiocarcinoma, calculate the discrimination score of the subject based on the peptide characteristics, and determine whether the subject is ICC or not based on the discrimination score. The early detection model for intrahepatic cholangiocarcinoma was constructed through the following steps: (1) Collect serum samples from subjects, including serum samples from intrahepatic cholangiocarcinoma, benign liver disease and healthy controls, and determine the final study cohort according to the pre-set inclusion and exclusion criteria; use a unified standardized operating procedure to preprocess serum samples, extract peptides by silicon-based micro-nano particle enrichment method, and use MALDI-TOF mass spectrometry to acquire serum peptide spectra; the characteristic peak range of the initial extraction is 680 to 18600 Da, and batch quality control and instrument calibration quality control are performed on all serum samples; (2) Baseline correction, normalization, peak identification and alignment were performed on the obtained serum polypeptide spectrum data, and the polypeptide amino acid sequence was confirmed. The difference fold, P value and genetic algorithm were used to screen polypeptide characteristic peaks that were significantly related to intrahepatic cholangiocarcinoma and had a high repetition frequency, so as to obtain the ICC polypeptide marker set. (3) Based on the peptide characteristic peaks corresponding to the ICC peptide marker set obtained by screening, a machine learning discrimination model is constructed and trained to obtain an early discrimination model for intrahepatic cholangiocarcinoma.

7. The intrahepatic cholangiocarcinoma discrimination system based on serum polypeptide characteristics according to claim 6, characterized in that, Also includes: The combined analysis module is used to perform combined analysis of the discrimination score and traditional serum biomarkers, including CA19-9 and carcinoembryonic antigen.

8. The intrahepatic cholangiocarcinoma discrimination system based on serum polypeptide characteristics according to claim 6, characterized in that, When the discrimination score is ≥2, the system determines it as ICC; otherwise, it is determined as non-ICC.

9. The application of a reagent for detecting a set of ICC polypeptide markers in the preparation of a screening kit for intrahepatic cholangiocarcinoma, characterized in that, The ICC peptide biomarker set includes any two or more of the following 15 characteristic peptides: The polypeptide M1 with a mass-to-charge ratio of 4058.69±500ppm has the amino acid sequence YNPQSRSVPPSASHVAPTETFTYEWTVPKEVGPTNAD. The polypeptide M2 ​​with a mass-to-charge ratio of 2009.70±500ppm has the amino acid sequence DQTVSDNELQEMSNQGSK. The polypeptide M3, with a mass-to-charge ratio of 2724.45 ± 500 ppm, has the amino acid sequence PGVLSSRQLGLPGPPDVPDHAAYHPF. The polypeptide M4, with a mass-to-charge ratio of 1477.25 ± 500 ppm, has the amino acid sequence TDTGALLFIGKILD. The polypeptide M5, with a mass-to-charge ratio of 2270.33 ± 500 ppm, has the amino acid sequence EIVLTQSPATLSLSPGERATLS; The polypeptide M6 with a mass-to-charge ratio of 1467.69±500ppm has the amino acid sequence ALTNAVAHVDDMPN. The polypeptide M7, with a mass-to-charge ratio of 4106.66±500ppm, has the amino acid sequence KPALEDLRQGLLPVLESFKVSFLSALEEYTKKLNTQ. The polypeptide M8 with a mass-to-charge ratio of 3931.45±500ppm has the amino acid sequence HGKKVADALTNAVAHVDDMPNALSALSDLHAHKLRVD. The amino acid sequence of peptide M9, with a mass-to-charge ratio of 2080.85±500ppm, is SETESPRNPSSAGSWNSGSSG. The polypeptide M10, with a mass-to-charge ratio of 1629.48 ± 500 ppm, has the amino acid sequence GSTSYGTGSETESPRN. The polypeptide M11 with a mass-to-charge ratio of 2112.56±500ppm has the amino acid sequence AALLSPYSYSTTAVVTNPKE. The polypeptide M12, with a mass-to-charge ratio of 1279.83 ± 500 ppm, has the amino acid sequence FFSTYDRDND. The polypeptide M13, with a mass-to-charge ratio of 1627.32 ± 500 ppm, has the amino acid sequence KNSLFEYQKNNKD. The polypeptide M14, with a mass-to-charge ratio of 1451.07 ± 500 ppm, has the amino acid sequence SWNSGSSGPGSTGNR. The polypeptide M15, with a mass-to-charge ratio of 1516.21 ± 500 ppm, has the amino acid sequence STFTGFLLYHDTN.

10. The application according to claim 9, characterized in that, The kit includes at least one or more of the following components: Serum peptide enrichment reagents are used to enrich endogenous peptides from serum samples and remove high-abundance proteins. Mass spectrometry matrix reagents for matrix-assisted laser desorption / ionization time-of-flight mass spectrometry detection of serum peptides; Standards or internal reference materials are used for instrument calibration or batch quality control during mass spectrometry detection. Sample processing buffers are used for the dilution, reaction, or stabilization of serum samples. Instructions for use are provided to guide the handling, testing, and interpretation of test samples.

Images