5hmC intrahepatic cholangiocarcinoma curative effect prediction model based on plasma apparent modification

Abstract

The invention discloses a gene combination for establishing an intrahepatic cholangiocarcinoma target-immunization therapy sensitivity prediction model, the gene combination is composed of the following genes ADAMTS12, PP2D1 and FIGN, the 5hmC levels of the three genes can be used for predicting whether an intrahepatic cholangiocarcinoma patient is sensitive to target-immunization therapy or not, an individualized treatment scheme can be conveniently formulated for the patient, and precise medical treatment can be conveniently implemented.

Description

Technical Field

[0001] The invention belongs to the field of biological detection, and relates to a 5hmC intrahepatic cholangiocarcinoma efficacy prediction model based on plasma epigenetic modification and a kit for detecting the sensitivity of intrahepatic cholangiocarcinoma patients to targeted immunotherapy, comprising the model. Background Art

[0002] Intrahepatic cholangiocarcinoma (iCCA) is a type of primary liver cancer that originates from secondary intrahepatic bile duct epithelial cells and is the second most common primary liver cancer after hepatocellular carcinoma (HCC). It accounts for 10% to 15% of primary liver cancers, and its incidence rate is showing a clear upward trend worldwide. iCCA often has no obvious symptoms in the early stage, and most patients are diagnosed at the advanced stage and lose the opportunity for surgery. For a small number (<20%) of early iCCA, radical surgical resection (including liver transplantation) is the best treatment, but the recurrence rate is as high as 50% one year after surgery. For most patients with advanced iCCA, the target immune multi-drug combination therapy of gemcitabine + oxaliplatin combined with lenvatinib and toripalimab can prolong the median survival time of iCCA to 22.5 months, and the objective response rate can reach 80%, but there are still many patients with poor efficacy. Therefore, constructing an efficacy prediction model for this target immune therapy has important clinical value for patients' treatment selection.

[0003] DNA methylation is a very important research direction in the field of epigenetics. It can cause changes in chromatin structure, DNA conformation, DNA stability, and the way DNA interacts with proteins, thereby controlling gene expression, which has very important biological significance. 5-methylcytosine (5mC) is the most common DNA methylation modification in eukaryotes. 5hmC (5-hydroxymethylcytosine, 5-hmC, 5-hydroxymethylcytosine) generated by hydroxymethylation of 5mC is an important intermediate produced in the demethylation process of 5mC. Studies have shown that 5hmC is involved in important biological processes such as chromosome reprogramming, gene expression regulation, and DNA damage repair. Its modification changes are closely related to the occurrence and development of tumors. 5hmC-Seal technology can be used to capture 5hmC and enrich DNA fragments containing 5hmC modifications in nearly 20,000 genes in the whole genome. Further high-throughput sequencing analysis (such as NGS) can obtain information such as the distribution, length, and abundance of 5hmC in the whole genome. The 5hmC-Seal technology detects 5hmC on the genome through glycosylation modification and APOBEC deaminase deamination. The technology has the advantages of high specificity, no sequence preference, and the ability to conduct throughput studies. In addition, many studies have shown that the detection technology has great application potential in the fields of early screening, early diagnosis, companion diagnosis, and prognosis monitoring of tumors.

[0004] Patent document CN201910026672.2 discloses an early diagnosis model of colorectal cancer based on epigenetic modification of 5hmC multi-molecular markers. The detected gene markers include: GBX2, FAM84A, FAM25B, LCE1F, FBXL7, DBX1, KRTAP27-1, AL353791.1, CEBPD, LTB4R2, RP4-583P15.14, OR5B2, RPRM, RNASE4, INSL5, AURKC, L36A, AC017081.1, SPA17, NBPF12, FABP1, CST8. The content of 5-hmC in colorectal cancer gene markers is detected by high-throughput sequencing to determine whether pancreatic cancer exists. Patent document CN201910583680.7 discloses a gene set for predicting the efficacy of transformation therapy for metastatic colorectal cancer, which includes the following 10 gene loci in the human genome: C18orf42, OR10V1, ATOH1, AC140061.12, TMEM26, KRTAP4-2, PDK4, SPAG11B, SEC31B, LILRA5. The 5hmC epigenetic modification levels of these gene loci predict the efficacy of transformation therapy. The mathematical model established using this gene set can be used to predict the efficacy of transformation therapy for metastatic colorectal cancer with unresectable liver metastases, helping doctors determine transformation therapy plans and implement precision medicine. However, there is currently a lack of effective diagnostic means in the clinic to determine which intrahepatic cholangiocarcinoma patients are sensitive to targeted immunotherapy, which makes it impossible to formulate personalized diagnosis and treatment plans and implement precision medicine. Summary of the invention

[0005] This research group explored biomarkers that may be used to predict the efficacy of targeted immunotherapy for intrahepatic cholangiocarcinoma, including screening proteins and genes. It was finally found that 5hmC of three gene loci, ADAMTS12, PP2D1 and FIGN, was related to the sensitivity of intrahepatic cholangiocarcinoma patients to targeted immunotherapy. The combination of the three can be used to predict whether intrahepatic cholangiocarcinoma patients are sensitive to targeted immunotherapy. We first constructed a prediction model for the efficacy of targeted immunotherapy for intrahepatic cholangiocarcinoma based on plasma epigenetic modification of 5hmC. Similar prediction models have not been reported. Therefore, the present invention includes the following technical solutions:

[0006] The first aspect of the present invention provides a gene combination for establishing a prediction model for the sensitivity of intrahepatic cholangiocarcinoma to targeted immunotherapy, characterized in that it includes the following genes or is composed of the following genes: ADAMTS12, PP2D1 and FIGN.

[0007] These three gene loci seem to be unrelated, but when combined, they can form a biomarker set for evaluating the sensitivity of patients with intrahepatic cholangiocarcinoma to targeted immunotherapy. The 5hmC epigenetic modification levels of these three gene loci can predict the efficacy of targeted immunotherapy.

[0008] In one embodiment, the above-mentioned targeted immunotherapy refers to the combined administration of drugs lenvatinib + PD1 monoclonal antibody + gemcitabine + oxaliplatin.

[0009] Preferably, the above-mentioned genes are 5hmC-modified genes ADAMTS12, PP2D1 and FIGN.

[0010] The second aspect of the present invention provides the use of the gene combination (object) as described above in establishing a prediction model for the sensitivity of intrahepatic cholangiocarcinoma to targeted immunotherapy. That is, the gene combination (object) as described above can be used to evaluate and predict the sensitivity of intrahepatic cholangiocarcinoma patients to targeted immunotherapy.

[0011] The third aspect of the present invention provides a prediction model for the sensitivity of intrahepatic cholangiocarcinoma to targeted immunotherapy, which is a mathematical model shown in the following formula (1):

[0012]

[0013] Where: x=w1y1+w2y2+w3y3+b; w is the weight of each feature, which is the weight of each gene, y is the value of each feature, here is the 5hmc level at each site; Score is the score; b is a constant term, which is 8.21534438; y1, y2 and y3 are the 5hmC levels or expression levels of the three gene sites, that is, y1 corresponds to ADAMTS12, y2 corresponds to PP2D1, and y3 corresponds to FIGN; w1, w2 and w3 are the weight factors corresponding to the three gene sites, that is, w1 is the weight factor of ADAMTS12, w2 is the weight factor of PP2D1, and w3 is the weight factor of FIGN. w1 is 0.059264; w2 is -0.958907, w2 is 0.056951;

[0014] The demarcation criterion for judging the sensitivity of targeted immunotherapy for intrahepatic cholangiocarcinoma is: when the Score is greater than 0.5, the patient is judged to be insensitive, resistant or treatment-resistant to targeted immunotherapy and is not suitable for targeted immunotherapy; on the contrary, when the Score is less than or equal to 0.5, the patient is judged to be sensitive to treatment and is suitable for targeted immunotherapy.

[0015] The above mathematical model / mathematical formula is a calculation scoring formula obtained by using orthogonal validation and elastic network modeling.

[0016] The targeted immunochemotherapy mentioned in the above-mentioned intrahepatic cholangiocarcinoma targeted immunochemotherapy sensitivity prediction model refers to the combined administration of the drugs lenvatinib + PD1 monoclonal antibody + gemcitabine + oxaliplatin.

[0017] Preferably, the 5hmc level in the above-mentioned intrahepatic cholangiocarcinoma targeted immunotherapy sensitivity prediction model refers to the 5hmc level / expression amount of genes ADAMTS12, PP2D1 and FIGN in plasma.

[0018] The above 5hmC levels are the 5hmC levels of three characteristic genes ADAMTS12, PP2D1 and FIGN in the patient's plasma detected by 5hmC-Seal technology.

[0019] Optionally, the above-mentioned intrahepatic cholangiocarcinoma targeted immunotherapy sensitivity prediction model can be input into a computer, a genetic testing device such as the Illumina sequencing platform, a blood testing device or an artificial intelligence (AI) medical system through programming and in the form of a mathematical software package, so as to predict whether the patient is sensitive to targeted immunotherapy, making it easier for doctors to formulate dosing plans.

[0020] A fourth aspect of the present invention provides a kit for detecting the sensitivity of intrahepatic cholangiocarcinoma patients to targeted immunotherapy, comprising: tools and reagents for collecting DNA and / or RNA from a biological sample;

[0021] Gene detection auxiliary reagents for implementing 5hmC-Seal technology to capture 5hmC, including β-GT glycosyltransferase and APOBEC3A deaminase (A3A), and necessary buffers such as TE buffer (TE buffer is prepared from Tris and EDTA, mainly used to dissolve nucleic acids, and can stably store DNA and RNA.); or

[0022] If the 5hmC-Seal technology is not used for whole-gene 5hmC capture, the kit includes tools and reagents such as PCR primers for detecting the genes ADAMTS12, PP2D1, and FIGN, or detecting the 5hmC-modified genes ADAMTS12, PP2D1, and FIGN;

[0023] A vector for predicting the sensitivity of the above-mentioned intrahepatic cholangiocarcinoma target immunotherapy is described.

[0024] As we all know, in addition to gene detection devices such as the Illumina sequencing platform, the reagents used in the 5hmC-Seal technology mainly include β-GT glycosyltransferase and APOBEC3A deaminase (A3A). Among them, the 5hmC-Seal technology uses β-GT glycosyltransferase to protect 5hmC and prevent it from being deaminated by APOBEC3A; the APOBEC3A (A3A) enzyme can specifically deaminate methylation, but cannot deaminate cytosine protected by sugar groups, thereby identifying 5hmC during the sequencing process. Therefore, during the sequencing process, 5hmC will still be identified as C because it is not recognized by A3A, so that 5hmC occurring on the genome can be accurately detected.

[0025] In one embodiment, the biological sample applicable to the above-mentioned kit can be a sample to be analyzed in clinical medicine, for example, selected from body tissue (such as liver tissue), whole blood, plasma, serum, saliva, oral mucosa, nasopharyngeal secretions, tissue fluid, body fluid, urine, etc., preferably plasma.

[0026] The present invention is based on the differentiation of 5hmC levels of numerous genes in the plasma of intrahepatic cholangiocarcinoma patients who have received targeted immunochemotherapy, and screens out the biomarker ADAMTS12, PP2D1 and FIGN gene sets; based on the 5hmC levels of these gene loci, a efficacy prediction model for targeted immunochemotherapy of intrahepatic cholangiocarcinoma patients is established, so that a portable clinical kit can be prepared to detect the 5hmC levels of specific genes in the patient's serum, predict whether the patient is sensitive to targeted immunochemotherapy, and help promote precision individualized medicine. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 Shown are the age (top) and sex (bottom) distributions of 47 patients with intrahepatic cholangiocarcinoma who received targeted immunotherapy.

[0028] Figure 2 The PCA analysis of 5hmC levels of all genes in plasma samples of 47 patients is shown.

[0029] Figure 3 The ROC curve (upper) and the score distribution diagram (lower) of the training set (T1) and the test set (V2) in the embodiment are shown. DETAILED DESCRIPTION

[0030] As we all know, 5hmC is an important epigenetic modification of gene DNA and is known as the "sixth DNA base". The gene distribution of 5hmC can accurately correspond to the regulation of gene activity and can dynamically and sensitively reflect the state of gene expression. Studies have shown that 5hmC is involved in important biological processes such as chromosome reprogramming, gene expression regulation, and DNA damage repair. Its modification changes are closely related to the occurrence, development, and treatment response of tumors.

[0031] This research team used the 5hmC detection technology 5hmC-Seal to capture 5hmC sequences in plasma cfDNA and measure and analyze 5hmC levels. After standardizing the 5hmC modification level, bioinformatics tools were used to search for 5hmC gene sites with significant differences in modification levels between the intrahepatic cholangiocarcinoma patient group that was sensitive to targeted immunotherapy and the intrahepatic cholangiocarcinoma patient group that was resistant to targeted immunotherapy. Subsequently, the obtained differential sites were further screened and optimized using regression models to obtain a set of 5hmC biomarkers with high sensitivity and specificity, and the obtained biomarkers were verified in independent samples.

[0032] After screening and comparing a large number of genes and their epigenetic modifications / mutations, it was found that there were some differences in the levels of some genes and their 5hmC modification between the intrahepatic cholangiocarcinoma patients who were sensitive / effective to targeted immunotherapy and those who were resistant to targeted immunotherapy, especially the 5hmC levels of the three gene loci ADAMTS12, PP2D1 and FIGN, which showed significant differences, suggesting that the three gene loci ADAMTS12, PP2D1 and FIGN and their 5hmC levels can be used as factors to judge the sensitivity of intrahepatic cholangiocarcinoma patients to targeted immunotherapy. These three gene loci seem to be unrelated, but the 5hmC epigenetic modification levels of the three can predict the efficacy of targeted immunotherapy, that is, their combination can be used as a biomarker for the efficacy of targeted immunotherapy in intrahepatic cholangiocarcinoma.

[0033] In this article, the terms "gene locus" or "gene" have the same meaning and can be used interchangeably, especially referring to genes that can reflect 5hmC epigenetic modification (or 5hmC-modified genes).

[0034] In this article, the terms "5hmC level", "5hmC expression amount", "5hmC content", "5hmC expression level", "5hmC apparent modification" and "5hmC apparent modification level" have the same meaning and can be used interchangeably, referring to the degree to which a gene is 5hmCized.

[0035] The 5hmC-Seal technology can be used to detect the 5hmC level of genes. The 5hmC-Seal technology uses β-GT glycosyltransferase to protect 5hmC, while the APOBEC3A (A3A) enzyme can specifically deaminize methylation but cannot deaminize cytosine protected by sugar groups. Therefore, during the sequencing process, 5hmC will still be recognized as C because it is not recognized by A3A, so that 5hmC occurring on the genome can be accurately detected.

[0036] It is easy to understand that the three gene loci ADAMTS12, PP2D1 and FIGN and their 5hmC levels can be used to establish an intrahepatic cholangiocarcinoma efficacy prediction model, such as an intrahepatic cholangiocarcinoma efficacy prediction model based on plasma epigenetic modification of 5hmC.

[0037] Based on the analysis of 5hmC epigenetic modification levels of three characteristic genes ADAMTS12, PP2D1 and FIGN in the plasma of clinical samples of patients with intrahepatic cholangiocarcinoma who had received targeted immunotherapy, orthogonal validation and elastic network modeling were used to obtain a statistically significant prediction model for the sensitivity of intrahepatic cholangiocarcinoma to targeted immunotherapy as shown in formula (1).

[0038] As used in this article, the "intrahepatic cholangiocarcinoma targeted immunotherapy sensitivity prediction model" may also be referred to as the "intrahepatic cholangiocarcinoma targeted immunotherapy (treatment) efficacy prediction model", or simply referred to as the "intrahepatic cholangiocarcinoma efficacy prediction model" or "intrahepatic cholangiocarcinoma efficacy diagnostic model", which have the same meaning and refer to mathematical models (1).

[0039] The score calculated by model (1) can basically accurately determine the dividing line between target immunotherapy sensitivity and target immunotherapy resistance, that is, a score greater than 0.5 indicates target immunotherapy resistance, and a score less than or equal to 0.5 indicates treatment sensitivity.

[0040] It is obvious to those skilled in the art that the prediction model for the sensitivity of intrahepatic cholangiocarcinoma to targeted immunotherapy can also be in the form of other mathematical models, and include the 5hmC levels of genes ADAMTS12, PP2D1 and FIGN as variable elements.

[0041] As a specific application of the above mathematical model, it allows to be input into the information processing module of a gene detection device such as a gene sequencer such as an illumina sequencing platform, or input into the information processing module of a blood detection device, or input into the information processing module of a PCR instrument through programming, in the form of a mathematical software package (computer program). Moreover, the above mathematical model also allows to be input into a cloud server in the form of computer software (computer program) to form a shareable artificial intelligence (AI) medical system, or input into a doctor's computer, to help doctors determine whether patients with intrahepatic cholangiocarcinoma are sensitive to targeted immunotherapy, and to facilitate doctors to formulate medication and treatment plans.

[0042] Based on the mathematical model shown in Formula 1, the present invention also provides a kit for detecting the sensitivity of targeted immunotherapy in patients with intrahepatic cholangiocarcinoma. The kit can be used to conveniently collect DNA from biological samples such as plasma and detect the 5hmC epigenetic modification levels of gene loci such as ADAMTS12, PP2D1 and FIGN.

[0043] It is easy for those skilled in the art to understand that the kit of the present invention may also include at least one of the following items: a carrying tool, whose space is divided into a limited space that can accommodate one or more containers, multi-well plates or strips, such as reagent kits, medicine bottles, test tubes, and the like, each of which contains a separate component for the method of the present invention; instructions, which can be written on bottles, test tubes and the like, or on a separate piece of paper, or on the outside or inside of the container, such as a paper or label with an operation demonstration video APP download window such as a QR code, and the instructions can also be in the form of multimedia, such as a CD, U disk, network disk, etc.

[0044] When using the kit, the 5hmC levels of genes such as ADAMTS12, PP2D1 and FIGN detected by the kit are input into this model (1). Based on the calculation results, it is used for pre-drug screening of patients with potential therapeutic effects of targeted immunotherapy, thereby implementing precision medicine.

[0045] Through clinical practice, it is proved that the intrahepatic cholangiocarcinoma targeted immunotherapy efficacy prediction model (1) of the present invention has the following advantages when used for liquid biopsy:

[0046] 1. The 5hmC detection of ex vivo sample plasma epigenetic modification of the present invention is non-invasive and convenient. It only needs to collect 10 ml of peripheral venous blood from the patient. The collection process is no different from conventional blood tests, and the 5hmC detection of the whole genome can be completed. Compared with the previous whole exome sequencing and human-derived transplanted tumors, it can significantly improve the patient's compliance; through the prediction model, the one-stop prediction of the efficacy of targeted immunotherapy is realized, and the stability is strong.

[0047] 2. The present invention was verified within the modeling group, and the model prediction accuracy was high. The areas under the AUC curve of the test set and validation set were 0.925 and 0.900, respectively, filling the technical gap in the use of liquid biopsy for the prediction of the efficacy of targeted immunotherapy in intrahepatic cholangiocarcinoma research.

[0048] 3. The 5hmC detection method used in the present invention is a mature gene sequencing method such as 5hmC-Seal, and the detection results are highly repeatable.

[0049] The present invention will be further described below in conjunction with specific embodiments. It should be understood that these embodiments are only for illustrative purposes and are not intended to limit the present invention. In addition, it should be understood that after reading the concept of the present invention, various changes or adjustments made by those skilled in the art should fall within the scope of protection of the present invention, and these equivalent forms also belong to the scope defined in the appended claims.

[0050] This article involves the addition amount, content and concentration of various substances, and the percentages mentioned here refer to the mass percentages unless otherwise specified.

[0051] Example 1: Establishment of a prediction model for the efficacy of intrahepatic cholangiocarcinoma based on plasma epigenetic modification of 5hmC

[0052] A total of 47 patients with intrahepatic cholangiocarcinoma who received targeted immunotherapy were screened at Zhongshan Hospital Affiliated to Fudan University. With informed consent, 10 ml of peripheral venous blood was collected from each patient before the start of treatment for detection of gene 5hmC levels. After treatment with gemcitabine + oxaliplatin + lenvatinib + PD1 monoclonal antibody, according to the RECIST 1.1 scoring standard, 18 patients were diagnosed as the treatment-effective group (type 0) and 29 patients were diagnosed as the treatment-resistant group (type 1). There was no significant difference in age and gender between the two groups of patients ( Figure 1 ). The 5hmC expression levels of more than 20,000 gene loci in the whole genome were detected, and the PCA analysis results of the 5hmC expression levels of the two groups showed no significant difference ( Figure 2 ). Subsequently, we randomly divided the above 47 patients into a training set (T1, 31 cases) and a validation set (V2, 16 cases), with the sample size ratio of the two groups being approximately 2:1. The test results of the 31 patients in the training set were input, and orthogonal validation and elastic network modeling were used (Formula 1)

[0053]

[0054] Where: x = w1y1+w2y2+w3y3+…+w i y i +b; b is the intercept, w is the weight of each feature (1, 2, 3, …, i), and y is the value of each feature (1, 2, 3, …, i).

[0055] A series of models were obtained, from which a better model (Formula 1) was selected, with AUC = 0.925 for the training set, including 3 characteristic genes: ADAMTS12, PP2D1, and FIGN. The test results of 16 patients in the validation set were input into the model, with AUC = 0.9 ( Figure 3 ). Finally, a constant term b and three meaningful gene terms w1, w2, w3 are actually output, and the diagnostic / prediction model of sensitivity to targeted immunotherapy for intrahepatic cholangiocarcinoma is obtained, namely, formula (1).

[0056] Example 2: Validation of a model for predicting the efficacy of intrahepatic cholangiocarcinoma based on 5hmC epigenetic modification in plasma

[0057] We input the test results of the above 47 samples into the prediction model and scored the results of each sample. The scores of the 18 patients with effective targeted immunotherapy are as follows: 0.258414355, 0.11565163, 0.788717799, 0.273706653, 0.015915699, 0.39905958, 0.460329338, 0.187698559, 0.578671563, 0.233617623, 0.273137528, 0 .008449276, 0.023925026, 0.329126756, 0.042490542, 0.10178703, 0.773395354, 0.725474062. The scoring results of 29 patients with resistance to targeted immunotherapy are as follows: 0.969796493, 0.605719521, 0.953770047, 0.974132832 , 0.20266721, 0.998781589, 0.994414001, 0.999492884, 0.885393337, 0.968994087, 0.94215115, 0.988058006, 0.994682017, 0.608293418, 0.986687259, 0.984708389, 0.553 492513, 0.929740274, 0.969630329, 0.569785341, 0.991792229, 0.053668473, 0.936515771, 0.982346948, 0.820340222, 0.791770663, 0.8816014, 0.295975995, 0.995706425. The results showed that among the 29 patients with targeted immunotherapy resistance, 26 had a score of >0.50 and were predicted to be resistant to targeted immunotherapy, and 3 were misjudged. Among the 18 patients with targeted immunotherapy efficacy, 14 had a score of ≤0.50 and 4 were misjudged. The sensitivity and specificity of the test were 89.7% and 77.8%, respectively. Therefore, the scoring criteria of the above-mentioned mathematical model for predicting the efficacy of targeted immunotherapy can be determined as follows: when P>0.5, it is resistance to targeted immunotherapy; when P≤0.5, it is effective for targeted immunotherapy.

[0058] The above experimental results confirmed the reliability of the diagnostic / predictive model (1) for the sensitivity of intrahepatic cholangiocarcinoma to targeted immunotherapy.

[0059] The above embodiments are only preferred embodiments for fully illustrating the present invention, and the protection scope of the present invention is not limited thereto. Any equivalent substitution or change made by a person skilled in the art based on the present invention is within the protection scope of the present invention.

Claims

1. A gene composition for establishing a prediction model for the sensitivity of intrahepatic cholangiocarcinoma to targeted immunotherapy, characterized in that: It comprises or consists of the following genes: ADAMTS12, PP2D1 and FIGN.

2. The gene composition according to claim 1, characterized in that The targeted immunotherapy refers to the combined administration of the drugs lenvatinib + PD1 monoclonal antibody + gemcitabine + oxaliplatin.

3. The gene composition according to claim 1, characterized in that The genes are 5hmC-modified genes ADAMTS12, PP2D1 and FIGN.

4. Use of the gene composition according to any one of claims 1 to 3 in establishing a prediction model for the sensitivity of intrahepatic cholangiocarcinoma to targeted immunotherapy.

5. A prediction model for the sensitivity of intrahepatic cholangiocarcinoma to targeted immunotherapy is a mathematical model as shown in the following formula (1): in: x=w1y1+w2y2+w3y3+b; w is the weight of each feature, which is the weight of each gene, y is the value of each feature, here is the 5hmc level at each site, that is, the expression; Score is the score; b is a constant term, which is 8.21534438; y1, y2 and y3 are the 5hmC expression levels of the three gene sites, that is, y1 corresponds to ADAMTS12, y2 corresponds to PP2D1, and y3 corresponds to FIGN; w1, w2 and w3 are the weight factors corresponding to the three gene sites, that is, w1 is the weight factor of ADAMTS12, w2 is the weight factor of PP2D1, and w3 is the weight factor of FIGN. w1 is 0.059264; w2 is -0.958907, w2 is 0.056951; The demarcation criterion for judging the sensitivity of targeted immunochemotherapy for intrahepatic cholangiocarcinoma is: when the Score is greater than 0.5, the patient is judged to be insensitive, resistant or treatment-resistant to targeted immunochemotherapy; on the contrary, when the Score is less than or equal to 0.5, the patient is judged to be sensitive to treatment and is suitable for targeted immunochemotherapy.

6. The prediction model for the sensitivity of intrahepatic cholangiocarcinoma to targeted immunotherapy according to claim 5, characterized in that: The targeted immunotherapy refers to the administration of the drugs lenvatinib + PD1 monoclonal antibody + gemcitabine + oxaliplatin.

7. The prediction model for the sensitivity of intrahepatic cholangiocarcinoma to targeted immunotherapy according to claim 5, characterized in that: The 5hmc level refers to the 5hmc level of genes ADAMTS12, PP2D1 and FIGN in plasma.

8. The prediction model for the sensitivity of intrahepatic cholangiocarcinoma to targeted immunotherapy according to claim 5, characterized in that: The model is programmed and input into a computer, genetic testing device, blood testing equipment or artificial intelligence medical system in the form of a mathematical software package.

9. A kit for detecting the sensitivity of intrahepatic cholangiocarcinoma patients to targeted immunotherapy, characterized in that: include: Tools and reagents for collecting DNA and / or RNA from biological samples; Gene detection auxiliary reagents for implementing 5hmC-Seal technology to capture 5hmC, including β-GT glycosyltransferase and APOBEC3A deaminase (A3A), and necessary buffers; or If the 5hmC-Seal technology is not used for whole-gene 5hmC capture, the kit includes tools and reagents such as PCR primers for detecting the genes ADAMTS12, PP2D1, and FIGN, or detecting the 5hmC-modified genes ADAMTS12, PP2D1, and FIGN; A vector for predicting the sensitivity of the above-mentioned intrahepatic cholangiocarcinoma target immunotherapy is described.

10. The kit according to claim 1, characterized in that The biological sample is selected from body tissue, whole blood, plasma, serum, saliva, oral mucosa, nasopharyngeal secretions, tissue fluid, body fluid, urine, preferably plasma.

Images