WO2023128419A1 - Method for screening colorectal cancer and colorectal polyps or advanced adenomas and application thereof - Google Patents

Abstract

The present invention relates to a method for screening colorectal cancer and colorectal polyps and advanced adenomas and an application thereof, and provides a method for screening colorectal cancer and colorectal polyps using 18 genetic markers of the present invention and an application thereof, and a method for screening colorectal cancer and advanced adenoma, using 30 genetic markers of the present invention and an application thereof. The present invention can help screen for colorectal cancer, colorectal polyps, or advanced adenoma by using blood samples that are relatively easy to extract and applying expression patterns of genetic markers expressed in blood to an artificial intelligence algorithm.

Description

Screening method for colorectal cancer and colorectal polyps or advanced adenomas and its applications

The present invention relates to a method for screening colon cancer and colon polyps or advanced adenomas and a kit used for the method.

Colorectal cancer is a malignant tumor that occurs in the colon and rectum, which constitute the large intestine.

According to Globocan data, as of 2020, colorectal cancer is the third most common cancer among all cancers, with 1.9 million new cases worldwide, and 935,000 people died from colorectal cancer, ranking third in the mortality rate due to cancer. It is a major cancer The 5-year relative survival rate of colorectal cancer is significantly lowered according to the degree of progression of the cancer. In Stage I, the 5-year relative survival rate reaches 90%, but in Stage IV, the 5-year survival rate rapidly decreases to 14%. However, only 37% of cases are diagnosed at Stage I because most of them do not show symptoms in Stage I. Therefore, early diagnosis of colorectal cancer through regular screening is important in increasing the survival rate of colorectal cancer.

Most colorectal cancers develop through a long-term adenoma-carcinoma sequence. The selection-cancerization process refers to the process in which normal epithelial cells of the colon progress to colorectal cancer via non-advanced adenoma and advanced adenoma. Progressive adenomas are adenomas that are highly likely to develop into colorectal cancer. Histologically, they are larger than 10 mm, contain more than 25% villous components, have high-grade dysplastic lesions, or have three or more adenomas. A 13-year follow-up study revealed that subjects with advanced adenomas were 2.7 times more likely to develop colorectal cancer and 2.6 times more likely to die from colorectal cancer than subjects with control and non-advanced adenomas. Therefore, in order to lower the incidence of colorectal cancer, it is important to detect and remove advanced adenomas at the stage.

In most countries, colonoscopy and fecal occult blood test are performed for the diagnosis of colorectal cancer. Colonoscopy can examine the entire large intestine at once and can remove adenomas or some early cancers found during the examination. Through a meta-analysis, it has been reported that colonoscopy reduces both the incidence and mortality of colorectal cancer by about 70%. Colonoscopy requires a bowel preparation process before examination, and the degree of bowel preparation has a very important effect on the accuracy and quality of the examination. In addition, intestinal perforation, which can occur as a complication, appears with a frequency of about 0.09%. Therefore, as a regular colorectal cancer screening test targeting a large population, the number of doctors that can be performed is limited, the patient's compliance may be poor due to pain during examination, and discomfort in pretreatment, and complications may occur relatively often. there is

Fecal occult blood test is a method of diagnosis by detecting bleeding from a mass in the large intestine in the stool. It is a non-invasive method, has no special side effects, is relatively easy to implement, and has the advantage of low cost. Compared to those who did not perform fecal occult blood testing, those who underwent fecal occult blood testing had a 10-40% lower mortality rate.

According to a meta-analysis, the sensitivity of the fecal occult blood test for colorectal cancer and advanced adenoma was 56-74% and 23-31%, respectively, and the specificity was 90-95%. Since bleeding from colon tumors is often intermittent, the accuracy of the test may vary depending on whether or not the specimen is properly collected, and the compliance of test subjects in using a stool sample may be low.

Therefore, it is necessary to develop a molecular diagnostic test method with increased sensitivity and specificity to be useful as a screening test using a blood sample that has a low risk during examination and is easy to collect compared to colonoscopy.

[Prior Patent Literature]

US Patent Publication No. 20180238893

The present invention solves the above problems and has been made by the above necessity, and an object of the present invention is to use a quantitative reverse transcription polymerization reaction based on a blood sample that is relatively easy to extract and an artificial intelligence prediction model produced through the result. It is to provide an information provision method for developing a molecular diagnostic test method for colorectal cancer and advanced adenoma or colon polyp with high sensitivity and specificity.

Another object of the present invention is a molecule of colorectal cancer and colorectal polyps or advanced adenomas with high sensitivity and specificity using an artificial intelligence prediction model produced through quantitative reverse transcription polymerization based on a blood sample that is relatively easy to extract and the result. It is to provide a composition for diagnostic testing.

In order to achieve the above object, the present invention is a colorectal cancer group and colorectal cancer high-risk group containing primers or probes involved in measuring the relative expression levels of IL1B, LTF, TNFSF13B, ITIH4, CXCL11, MAPK6, GK, and MCAM genes as active ingredients , It provides a composition capable of distinguishing between a low-risk group and a normal group.

In one embodiment of the present invention,

The primers and probes are SEQ ID NOs: 7 to 9, SEQ ID NOs: 13 to 15, SEQ ID NOs: 16 to 18, SEQ ID NOs: 22 to 24, SEQ ID NOs: 34 to 36, SEQ ID NOs: 40 to 42, SEQ ID NOs: 43 to 45, and sequences It is preferably composed of the sequence shown in Nos. 56 to 58, but is not limited thereto.

In another embodiment of the present invention,

The normal group is a case in which there are no lesions in the colon through colonoscopy, the low-risk group is a case in which there are less than three low-risk adenomas, and the high-risk group is a case in which three or more low-risk adenomas are present through colonoscopy, and a high-risk group is present. It is preferable to include one or more adenomas and carcinoma in situ, but is not limited thereto.

In addition, the present invention is a colorectal cancer group containing primers or probes involved in measuring the relative expression levels of CES1, IL1B, TNFSF13B, ITIH4, CXCL11, MAPK6, GK, and MCAM genes as an active ingredient, and colorectal cancer high-risk, low-risk and normal A composition capable of classifying groups is provided.

In one embodiment of the present invention,

The primers and probes are SEQ ID NOs: 4 to 6, SEQ ID NOs: 7 to 9, SEQ ID NOs: 16 to 18, SEQ ID NOs: 22 to 24, SEQ ID NOs: 31 to 33, SEQ ID NOs: 34 to 36, SEQ ID NOs: 40 to 42, SEQ ID NOs: 43 to 45 and SEQ ID NOs: 56 to 58, but is not limited thereto.

In addition, the present invention provides a composition capable of distinguishing a high-risk group from a colorectal cancer high-risk group, a low-risk group, and a normal group, comprising primers or probes involved in measuring the relative expression levels of IL1B, LTF, TNFSF13B, ITIH4, CXCL11, and MAPK6 genes as an active ingredient to provide.

In one embodiment of the present invention,

The primers and probes preferably consist of the sequences shown in SEQ ID NOs: 7 to 9, SEQ ID NOs: 13 to 15, SEQ ID NOs: 16 to 18, SEQ ID NOs: 22 to 24, SEQ ID NOs: 34 to 36, and SEQ ID NOs: 40 to 42, but therefor Not limited.

In addition, the present invention a) measuring the relative expression levels of IL1B, LTF, TNFSF13B, ITIH4, CXCL11, MAPK6, GK, and MCAM genes using primers and probes through polymerase chain reaction,

b) Relative expression levels of CES1, IL1B, TNFSF13B, ITIH4, PTGS2, CXCL11, MAPK6, GK, and MCAM genes using primers and probes through polymerase chain reaction for samples positive for colon cancer in a) above is measured and if it is positive, it is judged to be colon cancer positive,

c) Samples judged negative in a) and b) were subjected to polymerase chain reaction to measure the relative expression levels of IL1B, LTF, TNFSF13B, ITIH4, CXCL11, and MAPK6 genes using primers and probes. Provided is a method for screening colon cancer and colon polyps, characterized in that colon cancer is judged as positive.

In one embodiment of the present invention, the primers and probes used in a) are SEQ ID NOs: 7 to 9, SEQ ID NOs: 13 to 15, SEQ ID NOs: 16 to 18, SEQ ID NOs: 22 to 24, SEQ ID NOs: 34 to 36, Consisting of the sequences set forth in SEQ ID NOs: 40 to 42, SEQ ID NOs: 43 to 45, and SEQ ID NOs: 56 to 58,

The primers and probes used in b) are SEQ ID NOs: 4 to 6, SEQ ID NOs: 7 to 9, SEQ ID NOs: 16 to 18, SEQ ID NOs: 22 to 24, SEQ ID NOs: 31 to 33, SEQ ID NOs: 34 to 36, and SEQ ID NOs: 40 to 40. 42, SEQ ID NOs: 43 to 45, and SEQ ID NOs: 56 to 58,

The primers and probes used in c) are the sequences shown in SEQ ID NOs: 7 to 9, SEQ ID NOs: 13 to 15, SEQ ID NOs: 16 to 18, SEQ ID NOs: 22 to 24, SEQ ID NOs: 34 to 36, and SEQ ID NOs: 40 to 42. It is preferable that it has been made, but it is not limited thereto.

In addition, the present invention provides a) primers or probe sets involved in measuring the relative expression levels of IL1B, LTF, TNFSF13B, ITIH4, CXCL11, MAPK6, GK, and MCAM genes,

b) primers or probe sets involved in measuring the relative expression levels of CES1, IL1B, TNFSF13B, ITIH4, PTGS2, CXCL11, MAPK6, GK, and MCAM genes, and

c) A kit for screening for colorectal cancer and colorectal polyps including primers or probe sets involved in measuring the relative expression levels of IL1B, LTF, TNFSF13B, ITIH4, CXCL11, and MAPK6 genes is provided.

In one embodiment of the present invention, the primers and probes used in a) are SEQ ID NOs: 7 to 9, SEQ ID NOs: 13 to 15, SEQ ID NOs: 16 to 18, SEQ ID NOs: 22 to 24, SEQ ID NOs: 34 to 36, Consists of the sequences set forth in SEQ ID NOs: 40 to 42, SEQ ID NOs: 43 to 45, and SEQ ID NOs: 56 to 58;

The primers and probes used in b) are SEQ ID NOs: 4 to 6, SEQ ID NOs: 7 to 9, SEQ ID NOs: 16 to 18, SEQ ID NOs: 22 to 24, SEQ ID NOs: 31 to 33, SEQ ID NOs: 34 to 36, and SEQ ID NOs: 40 to 40. 42, SEQ ID NOs: 43 to 45, and SEQ ID NOs: 56 to 58,

The primers and probes used in c) are the sequences shown in SEQ ID NOs: 7 to 9, SEQ ID NOs: 13 to 15, SEQ ID NOs: 16 to 18, SEQ ID NOs: 22 to 24, SEQ ID NOs: 34 to 36, and SEQ ID NOs: 40 to 42. It is preferable that it has been made, but it is not limited thereto.

In addition, the present invention is CCR1, CES1, IL1B, ITGA2, LTF, TNFSF13B, PTGES, ITIH4, TUG1, NME1, PTGS2, CXCL11, MAPK6, GK, KRT19, EpCAM, MCAM, PPARG, ANKHD1-EIF4EBP3, GPR15, MMP23B, TAS2R10, Provided is a composition for screening for colorectal cancer and advanced adenoma, comprising primers or probes involved in measuring the relative expression levels of TYMS, FOXA2, MKi67, ERBB2, NPTN, SNAI2, TERT and VIM genes as active ingredients.

In one embodiment of the present invention,

The primers and probes preferably consist of the sequences shown in SEQ ID NOs: 1 to 91, but all mutant sequences that achieve the effect of the present invention through one or more substitutions, deletions, additions, etc. to the sequences are also included in the scope of the present invention.

In addition, the present invention is CCR1, CES1, IL1B, ITGA2, LTF, TNFSF13B, PTGES, ITIH4, TUG1, NME1, PTGS2, CXCL11, MAPK6, GK, KRT19, EpCAM, MCAM, PPARG, ANKHD1-EIF4EBP3, GPR15, MMP23B, TAS2R10, Provided is a kit for screening for colorectal cancer and advanced adenoma, comprising primers or probes involved in measuring the relative expression levels of TYMS, FOXA2, MKi67, ERBB2, NPTN, SNAI2, TERT and VIM genes as active ingredients.

In one embodiment of the present invention,

The primers and probes preferably consist of the sequences shown in SEQ ID NOs: 1 to 91, but all mutant sequences that achieve the effect of the present invention through one or more substitutions, deletions, additions, etc. to the sequences are also included in the scope of the present invention.

In addition, the present invention is CCR1, CES1, IL1B, ITGA2, LTF, TNFSF13B, PTGES, ITIH4, TUG1, NME1, PTGS2, CXCL11, MAPK6, GK, KRT19, EpCAM, Measuring the relative expression levels of MCAM, PPARG, ANKHD1-EIF4EBP3, GPR15, MMP23B, TAS2R10, TYMS, FOXA2, MKi67, ERBB2, NPTN, SNAI2, TERT and VIM genes A method for providing information for prediction or diagnosis of colorectal cancer or advanced adenoma is provided.

In one embodiment of the present invention, the expression level is performed using primers and probes, and the primers and probes preferably consist of the sequences shown in SEQ ID NOs: 1 to 91, but one or more substitutions, deletions, or additions to the sequences are preferred. All mutant sequences that achieve the effect of the present invention through the like are also included in the scope of the present invention.

In another embodiment of the present invention, when one or more of the following 1) to 3) is confirmed, the subject is determined to have colorectal cancer:

1) CCR1, CES1, GK, IL1B, KRT19, LTF, PPARG, PTGES, PTGS2, TAS2R10, TNFSF13B and TYMS genes, or the protein levels encoded by the genes, were compared with the corresponding genes or those encoded by the genes in normal control subject samples. If expression level changes compared to protein levels are seen;

2) ANKHD1-EIF4EBP3, ITIH4, MCAM, PPARG, TAS2R10, TNFSF13B, TUG1 and TYMS genes or the level of proteins encoded by said genes in a sample of an individual with a non-advanced adenoma or the corresponding gene If expression level changes are seen compared to the level of the protein encoded by; and

3) ANKHD1-EIF4EBP3, CCR1, MCAM, MMP23B, TAS2R10, TNFSF13B, TUG1 and TYMS genes or protein levels encoded by the genes are compared with the level of the gene or the protein encoded by the gene in a sample of an individual with advanced adenoma to show a change in expression level.

In another embodiment of the present invention, the following 1) or 2) is additionally confirmed to determine the pathological characteristics of an individual having colorectal cancer:

1) CXCL11 and PTGS2 genes or the protein levels encoded by the genes are compared with the corresponding genes or the protein levels encoded by the genes of an individual sample having a low TNM stage colorectal cancer, and the change in expression level if present, the subject is judged to have high TNM stage colorectal cancer; or

2) Compare the level of the EpCAM gene or the protein encoded by the gene with the level of the gene or the protein encoded by the gene in a sample of an individual having highly differentiated colorectal cancer. diagnosed as having colorectal cancer.

In addition, the present invention provides blood gene marker combinations (Table 1) for the purpose of screening for colorectal cancer and advanced adenoma composed of the following gene groups.

Serial Number blood genetic markers One ANKHD1-EIF4EBP3 ANKHD1-EIF4EBP3 Readthrough 2 CCR1 C-C Motif Chemokine Receptor 1 3 CES1 Carboxylesterase 1 4 CXCL11 C-X-C Motif Chemokine Ligand 11 5 EpCAM Epithelial Cell Adhesion Molecule 6 ERBB2 Erb-B2 Receptor Tyrosine Kinase 2 7 FOXA2 Forkhead Box A2 8 GK Glycerol Kinase 9 GPR15 G Protein-Coupled Receptor 15 10 IL1B Interleukin 1 Beta 11 ITGA2 Integrin Subunit Alpha 2 12 ITIH4 Inter-Alpha-Trypsin Inhibitor Heavy Chain 4 13 KRT19 Keratin 19 14 LTF Lactotransferrin 15 MAPK6 Mitogen-Activated Protein Kinase 6 16 MCAM Melanoma Cell Adhesion Molecule 17 MKi67 Marker Of Proliferation Ki-67 18 MMP23B Matrix Metallopeptidase 23B 19 NME1 NME/NM23 Nucleoside Diphosphate Kinase 20 NPTN Neuroplastin 21 PPARG Peroxisome Proliferator Activated Receptor Gamma 22 PTGES Prostaglandin E Synthase 23 PTGS2 Prostaglandin-Endoperoxide Synthase 2 24 SNAI2 Snail Family Transcriptional Repressor 2 25 TERT Telomerase Reverse Transcriptase 26 TNFSF13B TNF Superfamily Member 13b 27 TAS2R10 Taste 2 Receptor Member 10 28 TUG1 Taurine Up-Regulated 1 29 TYMS Thymidylate Synthetase 30 VIM Vimentin

In addition, the present invention provides an artificial intelligence algorithm-based classification model for colorectal cancer and advanced adenoma screening tests prepared by substituting the expression levels of the 30 markers.

The present invention will be described below.

In the present invention, primer and probe sequences are provided to indicate the relative expression levels of corresponding biomarkers in blood.

In addition, the present invention provides an artificial intelligence prediction model for colorectal cancer, advanced adenoma, and/or colorectal polyps screening test prepared by substituting the expression levels of the 18 markers.

A method for isolating a commonly used full-length RNA (Total RNA) and a method for synthesizing cDNA therefrom can be performed through a known method, and a detailed description of this process can be found in Joseph Sambrook et al., Molecular Cloning, A Laboratory Manual. , Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. (2001); and Noonan, K.F. etc. are disclosed and may be incorporated by reference into the present invention.

The primers of the present invention can be chemically synthesized using the phosphoramidite solid support method, or other well-known methods. Such nucleic acid sequences can also be modified using a number of means known in the art.

Non-limiting examples of such modifications include methylation, "capping", substitution of one or more homologues of a natural nucleotide, and modifications between nucleotides, such as uncharged linkages (e.g., methyl phosphonates, phosphotriesters, phosphotriesters, phosphoramidates, carbamates, etc.) or to charged associations (eg phosphorothioates, phosphorodithioates, etc.). A nucleic acid can contain one or more additional covalently linked moieties, such as proteins (eg, nucleases, toxins, antibodies, signal peptides, ly-L-lysine, etc.), intercalants (eg, acridine, psoralen, etc.). ), chelating agents (eg, metals, radioactive metals, iron, oxidizing metals, etc.), and alkylating agents.

A nucleic acid sequence of the present invention may also be modified with a label capable of providing, directly or indirectly, a detectable signal. Examples of labels include radioactive isotopes, fluorescent molecules, and biotin.

In the method of the present invention, the amplified target sequence (CCR1, GAPDH gene, etc.) may be labeled with a detectable labeling substance. In one embodiment, the label material may be a material that emits fluorescence, phosphorescence, chemiluminescence, or radioactivity, but is not limited thereto. Preferably, the labeling material may be fluorescein, phycoerythrin, rhodamine, lissamine, Cy-5 or Cy-3. When the target sequence is amplified, by labeling the 5'-end and/or 3'-end of the primer with Cy-5 or Cy-3 and performing RT-PCR, the target sequence can be labeled with a detectable fluorescent labeling material.

In addition, when a radioactive isotope such as 32P or 35 S is added to the PCR reaction solution during RT-PCR, the amplification product is synthesized and radioactive is incorporated into the amplification product, so that the amplification product can be radioactively labeled. . One or more oligonucleotide primer sets used to amplify the target sequence may be used.

Labeling is performed by various methods commonly practiced in the art, such as the nick translation method, the random priming method (Multiprime DNA labeling systems booklet, "Amersham" (1989)), and the kination method (Maxam & Gilbert, Methods in Enzymology, 65:499 (1986)). The label provides a signal that can be detected by fluorescence, radioactivity, chromometry, gravimetry, X-ray diffraction or absorption, magnetism, enzymatic activity, mass analysis, binding affinity, hybridization radiofrequency, nanocrystals.

According to one aspect of the present invention, in the present invention, the expression level is measured at the mRNA level through RT-PCR. To this end, novel primer pairs and fluorescently labeled probes that specifically bind to the CCR1 and GAPDH genes are required, and in the present invention, corresponding primers and probes specified by specific nucleotide sequences can be used, but are not limited thereto , Anything that can specifically bind to these genes to provide a detectable signal to perform RT-PCR can be used without limitation. In the above, FAM and Quen (Quencher) mean fluorescent dyes.

The RT-PCR method applied to the present invention may be performed through a known process commonly used in the art.

The step of measuring the mRNA expression level may be used without limitation as long as it is a method capable of measuring the normal mRNA expression level, and may be performed through radioactivity measurement, fluorescence measurement, or phosphorescence measurement depending on the type of probe label used, but is limited thereto. It doesn't work.

As one of the methods for detecting the amplification product, the fluorescence measurement method is to label the 5'-end of the primer with Cy-5 or Cy-3 and perform real-time RT-PCR to label the target sequence with a detectable fluorescent label. And the fluorescence thus labeled can be measured using a fluorescence meter.

In addition, the radioactivity measurement method is to add a radioactive isotope such as ³² P or ³⁵ S to the PCR reaction solution during RT-PCR to label the amplification product, and then use a radioactivity measurement instrument, for example, a Geiger counter or Radioactivity can be measured using a liquid scintillation counter.

According to a preferred embodiment of the present invention, a fluorescence-labeled probe is attached to the PCR product amplified through the RT-PCR to emit fluorescence of a specific wavelength, and at the same time as amplification, the fluorescence of the genes of the present invention is measured in the fluorescence meter of the PCR device. The mRNA expression level is measured in real time, and the measured value is calculated and visualized through a PC, so that the inspector can easily check the expression level.

According to another aspect of the present invention, the screening kit may be a kit for diagnosing colorectal cancer and colorectal polyps, characterized in that it includes essential elements necessary for carrying out a reverse transcription polymerase reaction. The reverse transcription polymerase reaction kit may include each primer pair specific for the gene of the present invention. The primer is a nucleotide having a sequence specific to the nucleic acid sequence of each marker gene, and may have a length of about 7 bp to 50 bp, more preferably about 10 bp to 30 bp.

Other reverse transcription polymerase reaction kits include a test tube or other suitable container, reaction buffer (with varying pH and magnesium concentration), deoxynucleotides (dNTPs), enzymes such as Taq-polymerase and reverse transcriptase, DNAse, RNAse inhibitors, DEPC-water, sterile water, and the like.

In addition, the kit of the present invention may further include a user guide describing optimal reaction performance conditions.

The guide is a printed matter that explains how to use the kit, eg, how to prepare a buffer solution, suggested reaction conditions, and the like.

The guide may include a brochure in the form of a pamphlet or leaflet, a label affixed to the kit, and instructions on the surface of the package containing the kit. In addition, the guide may include information disclosed or provided through an electronic medium such as the Internet.

In the present invention, the term "information provision method for diagnosing colon cancer and colon polyps" is a preliminary step for diagnosis and provides objective basic information necessary for diagnosis of cancer, and clinical judgment or opinion of a doctor is excluded.

In the present invention, the term "information provision method for screening for colorectal cancer and advanced adenoma" is a preliminary step for diagnosis and provides objective basic information necessary for diagnosis of cancer, and clinical judgment or opinion of a doctor is excluded.

The term "primer" refers to a short nucleic acid sequence having a short free 3-terminal hydroxyl group capable of forming base pairs with a complementary template and serving as a starting point for copying the template strand. Primers can initiate DNA synthesis in the presence of reagents for polymerization (i.e., DNA polymerase or reverse transcriptase) and four different nucleoside triphosphates in an appropriate buffer and temperature. The primers of the present invention are sense and antisense nucleic acids having sequences of 7 to 50 nucleotides specific to each marker gene. A primer may incorporate additional features that do not alter the basic properties of the primer that serve as the starting point of DNA synthesis.

The term “probe” is a single-stranded nucleic acid molecule and contains a sequence complementary to a target nucleic acid sequence.

The term "real-time RT-PCR" refers to reverse transcription of RNA into complementary DNA (cDNA) using reverse transcriptase and using cDNA as a template containing target primers and labels It is a molecular biological polymerization method that amplifies a target using a target probe and quantitatively detects a signal generated from the label of the target probe on the amplified target at the same time.

A data mining method capable of diagnosing colon cancer and colon polyps through information learning can be used for diagnosing or predicting colon cancer and colon polyps of the present invention, and in particular, it can be effectively improved through AI analysis. Therefore, a method capable of measuring the relative expression levels of diagnostic markers for colon cancer and colon polyps and/or an AI analysis method may be preferably used in the method for diagnosing or predicting colon cancer and colon polyps of the present invention.

In the present invention, when AI analysis is used for colorectal cancer and colorectal polyps prediction models, various interpretable models can be used without limitation, and linear regression, logistic regression, neural network analysis, decision tree, decision rule, rule fit, support vector machine A model such as is applicable without limitation, and in a preferred embodiment of the present invention, logistic regression analysis, decision tree, neural network analysis and support vector machine are used in particular.

Meanwhile, the prediction model of the present invention may include a colorectal cancer and colon polyps diagnosis unit, a classification unit, and a weighting unit. Using relative expression level information as input information, the colon-related disease classification unit may perform a process of classifying colon cancer and colon polyps using a neural network as a classifier, and the weighting unit assigns a weight to the classification result, thereby detecting colorectal cancer and colon polyps can be screened.

Neural network analysis according to embodiments of the present invention refers to a system that constructs one or more layers to make a decision based on a plurality of data. For example, in neural network analysis, the input layer is a layer that inputs relative expression level information of gene markers as data into a neural network analysis model, and the output layer determines the presence or absence of colorectal cancer and colon polyp disease patients based on various input information. It is a layer that gives results that can be done. The hidden layer is a layer that proceeds with the process of determining whether or not there is a patient by assigning weights to various criteria (gene mutation information).

The method for predicting colorectal cancer and colorectal polyps using an AI analysis technique according to an embodiment of the present invention estimates a neural network analysis model having the number of hidden nodes using an MLP neural network. In addition, among several neural network models built through various variable transformations of input and output variables, the neural network model with the highest accuracy estimated from each model is determined as the final neural network model for colorectal disease prediction. The AI analysis may be composed of an input layer, a hidden layer, and an output layer, and the neural network analysis model through the neural network analysis step may be a neural network model having several hidden nodes in several hidden layers.

As can be seen through the present invention, the present invention is helpful in screening for colorectal cancer and colorectal polyps by substituting the expression patterns of genetic markers expressed in blood into an artificial intelligence algorithm using blood samples that are relatively easy to extract. can give

In addition, the present invention can help screen for colorectal cancer and advanced adenoma by using a relatively easy-to-extract blood sample and substituting the expression patterns of genetic markers expressed in blood into an artificial intelligence algorithm.

1 is a summary diagram of the present invention;

Figure 2 is the number of samples by group in which the experiment and analysis were performed,

3 is a primer probe nucleotide sequence prepared for detecting a genetic biomarker;

4 to 8 show the relative expression of each gene biomarker for each group derived through RT-qPCR,

Figure 9 shows the results of t-test statistical analysis for each group using all samples (selection of biomarkers for Model A production),

10 shows the results of t-test statistical analysis for each group using positive samples in Model A (selection of biomarkers for Model B production);

Figure 11 shows the results of t-test statistical analysis by group using negative samples in Model A and B (selection of biomarkers for Model C production),

12 shows the accuracy, sensitivity, and specificity results of the four artificial intelligence prediction models finally derived when Model A was produced;

13 shows the accuracy, sensitivity, and specificity results of the four artificial intelligence prediction models finally derived when Model B was produced;

14 shows the accuracy, sensitivity, and specificity results of the four artificial intelligence prediction models finally derived when Model C was produced;

15 is a schematic diagram of the final result of sequentially applying Models A, B, and C; and

16 is a process of building an artificial intelligence algorithm-based classification model and verifying model performance

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. These examples are only for explaining the present invention in more detail, and it will be apparent to those skilled in the art that the scope of the present invention is not limited by these examples according to the gist of the present invention. .

For reference, this patent largely consists of examples of three studies [divided into the following 1), 2) and 3).

1) Examples 1 to 5 are for screening colon cancer and colon polyps, and 18 gene markers [C-C motif chemokine receptor 1 (CCR1), Carboxylesterase 1 (CES1), Interleukin 1 beta (IL1B), Integrin alpha 2 ( ITGA2), Lactotransferrin (LTF), Tumor necrosis factor superfamily 13b (TNFSF13B), Prostaglandin E synthase (PTGES), Inter-alpha-trypsin inhibitor heavy chain H4 (ITIH4), Taurine upregulated gene1 (TUG1), Nucleoside diphosphate kinase 1 (NME1) ), Prostaglandin-endoperoxide synthase 2 (PTGS2), C-X-C motif chemokine 11 (CXCL11), Mitogen-activated protein kinase 6 (MAPK6), Glycerol kinase (GK), Keratin 19 (KRT19), Epithelial cell adhesion molecule (EpCAM), Melanoma Cell Adhesion Molecule (MCAM) and Peroxisome proliferator activated receptor gamma (PPARG)] were used, and colorectal cancer sensitivity: 95.2% through the number of analysis samples consisting of 42 colorectal cancer groups, 156 high-risk adenoma groups, 110 low-risk adenoma groups, and 183 control groups. , high-risk adenoma sensitivity: 91.7% and specificity: 78.5% of the analysis results,

2) Examples 6 to 10 are for diagnosis of colorectal cancer and its precancerous stage, and five new markers ( ANKHD1-EIF4EBP3 Readthrough (ANKHD1-EIF4EBP3), G Protein-Coupled Receptor 15 (GPR15) , Matrix Metallopeptidase 23B (MMP23B), Taste 2 Receptor Member 10 (TAS2R10), and Thymidylate Synthetase (TYMS)] were added to perform experiments on 23 genetic markers,

The number of samples analyzed was 112 in the colorectal cancer group, 178 in the advanced adenoma group, 104 in the non-advanced adenoma group, and 203 in the control group.

3) Examples 11 to 15 are for colorectal cancer and advanced adenoma screening, and 7 new markers (Forkhead box A2 (FOXA2), Marker Of proliferation Ki-67 (MKi67), Erb-B2 Receptor Tyrosine Kinase 2 (ERBB2), Neuroplastin (NPTN), Snail family transcriptional repressor 2 (SNAI2), Telomerase reverse transcriptase (TERT) and Vimentin (VIM)) were added and experiments were performed on 30 genetic markers. , The number of samples analyzed was 148 in the colorectal cancer group, 197 in the advanced adenoma group, and 143 in the control group.

In addition, the relationship between 2) and 3) above is

In the above 2) study (Examples 6 to 10), the model using the genes that were significant in distinguishing between colorectal cancer and advanced adenoma resulted in a sensitivity of 89.3% for colorectal cancer, a sensitivity of 74.5% for advanced adenoma, and a specificity of 72.0%. . Therefore, in order to improve the specificity, we additionally searched for a gene whose relative expression level changes only in colorectal cancer or advanced adenoma, and named this as 3) study.

Circulating tumor cells may exist in the blood in colorectal cancer or advanced adenoma, a precursor of colorectal cancer, and 7 genes ( FOXA2, MKi67, MUC1, NPTN, SNAI2, TERT, VIM ) were used as targets, and the relative expression levels of the corresponding genes were compared by group using blood from normal, advanced adenoma, and colorectal cancer groups.

The relative expression level (2 ^-ΔCq ) of the target gene was calculated using the Cq value of the target gene using the Cq value of the GAPDH gene. In order to compare the relative expression level between the normal group and the colorectal cancer group, the fold change value was obtained from the relative expression level ratio of the normal group to the colorectal cancer group, and the p -value of the difference between the two groups was obtained through Student's t -test analysis. In order to compare the relative expression level between the normal group and the advanced glandular group, the fold change value was calculated as the relative expression level ratio of the normal group to the advanced glandular group, and the p -value of the difference between the two groups was obtained through Student's t -test analysis. In order to compare the relative expression levels between the advanced adenoma group and the colorectal cancer group, the fold change value was calculated as the ratio of the relative expression level of the advanced adenoma group to the colorectal cancer group, and the p -value of the difference between the two groups was obtained through Student's t-test analysis.

As a result, the relative expression levels of FOXA2, MKi67, MUC1, NPTN, SNAI2, TERT, and VIM genes in the advanced adenoma group and colorectal cancer group compared to the normal group showed a statistically significant difference with a p -value of 0.05 or less. When the relative expression level of the hTERT gene was compared between the advanced adenoma group and the colorectal cancer group, a statistically significant difference was confirmed with a p -value of 0.05 or less (Table 2).

gene normal group vs. colorectal cancer group normal group vs. Progress line group Progressive line group vs. colorectal cancer group Fold change p -value Fold change p -value Fold change p -value FOXA2 2.48 <0.001 2.72 <0.001 0.91 0.591 MKi67 0.48 <0.001 0.53 <0.001 0.91 0.392 MUC1 1.42 0.005 1.80 <0.001 0.79 0.021 NPTN 1.74 <0.001 1.89 <0.001 0.92 0.335 SNAI2 2.55 <0.001 2.66 <0.001 0.96 0.790 TERT 2.27 <0.001 3.69 <0.001 0.62 0.006 VIM 1.38 <0.001 1.63 <0.001 0.84 0.061

Table 2 is a comparison of differences in relative expression between groups. Accordingly, 3) genes that were significant in distinguishing between colorectal cancer and advanced adenoma in the study (Examples 11-15) and a total of 30 genes to which the above 7 genes were added. was used to construct a classification model for the purpose of screening for colorectal cancer and advanced adenoma.

For reference, Table 3 is a list of primer and probe sequences for all 30 markers used in the present invention.

Target gene Primers and TaqMan probes sequence number Primer's and Taqman probe's sequences
(5' --> 3') Direction PCR product (bp) CCR1 Forward One TCC CTT GGA ACC AGA GAG AAG 75 Taqman probe 2 FAM-CCG GGA TGG AAA CTC CAA ACA CCA-BHQ1 Reverse 3 CAA ACT CTG TGG TCG TGT CAT AG CES1 Forward 4 GAC CTG TTC CTG GAC TTG ATA G 82 Taqman probe 5 FAM-TGG TGT CCC ATC TGT GAT TGT GGC-BHQ1 Reverse 6 CTC CAG CAT CTC TGT GGT TC IL1B Forward 7 CAC AGA CCT TCC AGG AGA ATG 82 Taqman probe 8 FAM-ACC TGA GCA CCT TCT TTC CCT TCA Reverse 9 CCC ATG TGT CGA AGA AGA TAG G ITGA2 Forward 10 TAT ACA GGA GCC CTC TGA TGT 110 Taqman probe 11 FAM-TGC GTG TGG ACA TCA GTC TGG AAA-BHQ1 Reverse 12 GAC CTT GGC AGT CTC AGA ATA G LTF Forward 13 AAC GAG AGA TAC TAC GGC TAC A 125 Taqman probe 14 FAM-CTG GCT GAG AAT GCT GGA GAC GTT-BHQ1 Reverse 15 GCC CAT GCC TCA TTG TTA TTT C TNFSF13B Forward 16 ACT ACC CAA TAA TTC CTG CTA TTC A 105 Taqman probe 17 FAM-TGC AAG TTG GAG TTC ATC TCC TTC TTC C-BHQ1 (antisense) Reverse 18 CCA TCC AGT GAT ATT TGT GCA TTT PTGES Forward 19 CCA GTA TTG CAG GAG CGAC 101 Taqman probe 20 FAM-CCA CCG GAA CGA CAT GGA GAC CAT C-BHQ1 Reverse 21 AGT AGA CGA AGC CCA GGA AA ITIH4 Forward 22 CAG AAT AAG ACC AAG GCT CAC A 118 Taqman probe 23 FAM-CCC AGA GCA GCA AGA AAC AGT CCT-BHQ1 Reverse 24 GGT CCA CAT CAT AGC GGA TAA T TUG1 Forward 25 CCC TTA CCT AAC AGC ATC TCA C 110 Taqman probe 26 FAM-AAG GCT GCA CCA GAT TCC AGA AA-BHQ1 Reverse 27 TGA TCT AAG AAT AAG TCG GTC ACA A NME1 Forward 28 CAG AAA GGA TTC CGC CTT GT 128 Taqman probe 29 FAM-AAC ACT ACG TTG ACC TGA AGG ACC G-BHQ1 Reverse 30 GGCCCTGAGTGCATGTATTT PTGS2 Forward 31 TGT GTT GAC ATC CAG ATC AC 117 Taqman probe 32 FAM-AGG TTA GAG AAG GCT TCC CAG C-BHQ1 (antisense) Reverse 33 GGA GGA AGG GCT CTA GTA TAA CXCL11 Forward 34 TAC AGT TGT TCA AGG CTT CCC 96 Taqman probe 35 FAM-AGA GGA CGC TGT CTT TGC ATA GGC-BHQ1 Reverse 36 GCT TTC TCA ATA TCT GCC ACT TTC PPARG Forward 37 CCC TTC ACT ACT GTT GAC TTC TC 133 Taqman probe 38 FAM-TCA CAA GAA CAG ATC CAG TGG TTG CA-BHQ1 Reverse 39 CTT TGA TTG CAC TTT GGT ACT CTT MAPK6 Forward 40 TTA GTC GAG AAG CAC TGG ATT T 97 Taqman probe 41 FAM-CCC ATG GAT CGG TTA ACA GCA GAA GA-BHQ1 Reverse 42 CTC ATG TAA GGA TGG GAG AGT G GK Forward 43 TCA GAT TAA TGC GGA GGA AAG T 99 Taqman probe 44 FAM-AGC TGT GAT GAA GTC AAT GGG TTG GG-BHQ1 Reverse 45 CCA CTT TCT GGA GAT TGA GTT GT KRT19 Forward 46 GAT GAG CAG GTC CGA GGT TA 96 Taqman probe 47 FAM-CTG CGG CGC ACC CTT CAG GGT CT-BHQ1 Reverse 48 TCT TCC AAG GCA GCT TTC AT EpCAM Forward 49 GCC AGT GTA CTT CAG TTG GTG CAC 82 Taqman probe 50 FAM-TAC TGT CAT TTG CTC AAA GCT GGC TGC CA-BHQ1 Reverse 51 CAT TTC TGC CTT CAT CAC CAA ACA ERBB2 Forward 52 AAG CAT ACG TGA TGG CTG GTG T 115 Taqman probe1 53 FAM-ATA TGT CTC CCG CCT TCT GGG CAT CT-BHQ1 Taqman probe2 54 FAM-CAT CCA CGG TGC AGC TGG TGA CAC A-BHQ1 Reverse 55 TCT AAG AGG CAG CCA TAG GGC ATA MCAM Forward 56 TTC TGA AGT GCG GCC TCT CC 74 Taqman probe 57 FAM-TCC CAA GGC AAC CTC AGC CAT GTC G-BHQ1 Reverse 58 CGC TTC TCC TTG TGG ACA GAA AAC ANKHD1
-EIF4EBP3 Forward 59 TTCAGTCCCTGCTCTCAAA 108 Taqman probe 60 FAM-ACCGAAGAAGAGAATTGGACGGCC-BHQ1 Reverse 61 ATCCTGGTGCCTCTGGTTA GPR15 Forward 62 CTG TGT CAA CCC TTT CAT TTA C 106 Taqman probe 63 FAM-CAT TGT CCA CTG CTT GTG CCC TTG-BHQ1 Reverse 64 GTG CTA CTC CCA AAG TCA TAG MMP23B Forward 65 ACC TCC GGA TAG GCT TCT A 136 Taqman probe 66 FAM-ATCAACCACACGGACTGCCTGG-BHQ1 Reverse 67 CTG TCG TCG AAG TGG ATG C TAS2R10
Forward 68 CTTGTAAACTGCATTGACTGTG 149 Taqman probe 69 FAM-ACGATTGGCTTTATTCTCACCGGCT-BHQ1 Reverse 70 CGGAGGCATATATATTTGGAGAG TYMS Forward 71 CTGAAGCCAGGTGACTTTATAC 90 Taqman probe 72 FAM-ACCTGAATCACATCGAGCCACTGA-BHQ1 Reverse 73 TTCTCGCTGAAGCTGAATTT FOXA2 Forward 74 CTA CTC CTC CGT GAG CAA CAT GAA C 74 Taqman probe 75 FAM-GCC TGG GGA TGA ACG GCA TGA ACA C-BHQ1 Reverse 76 GCC GCC GAC ATG CTC ATG TA MKI67 Forward 77 TAA TGA GAG TGA GGG AAT ACC TTT G 87 Taqman probe 78 FAM-GGC GTG TGT CCT TTG GTG GGC A-BHQ1 Reverse 79 AGG CAA GTT TTC ATC AAA TAG TTC A NPTN Forward 80 ACC AGT GAA GAG GTC ATT ATT CGA GAC A 88 Taqman probe 81 FAM-CCT GTT CTC CCT GTC ACC CTG CAG TGT AAC-BHQ1 Reverse 82 TAT GTA AGG GTG TGA GAG CTG GAG GT SNAI2 Forward 83 TGT GAC AAG GAA TAT GTG AGC CTG G 81 Taqman probe 84 FAM-CCT GAA GAT GCA TAT TCG GAC CCA CAC ATT-BHQ1 Reverse 85 CGC AGA TCT TGC AAA CAC AAG G TERT Forward 86 TGA CGT CCA GAC TCC GCT TCA T 83 Taqman probe 87 FAM-GCT GCG GCC GAT TGT GAA CAT GGA-BHQ1 Reverse 88 ACG TTC TGG CTC CCA CGA CGT A VIM
Forward 89 ATG TTG ACA ATG CGT CTC TGG CA 99 Taqman probe 90 FAM-TGA CCT TGA ACG CAA AGT GGA ATC TTT GC-BHQ1 Reverse 91 ATT TCC TCT TCG TGG AGT TTC TTC AAA GAPDH Forward 92 CCA TCT TCC AGG AGC GAG ATC C 90 Taqman probe 93 FAM-TCC ACG ACG TAC TCA GCG CCA GCA-BHQ1 Reverse 94 ATG GTG GTG AAG ACG CCA GTG

Table 3 shows the base sequences of each primer and probe used in the present invention.

Hereinafter, the present invention will be described in detail through examples.

Example 1: Materials

From 2017 to 2019, blood samples from people who visited the gastroenterology department at Shinchon Severance Hospital, Gangnam Severance Hospital, and Gangbuk Samsung Hospital were used. Blood was collected using Tempus blood tube (Applied Biosystems®) and transported to Yonsei University Wonju Campus Molecular Diagnostics Laboratory.

As a result of colonoscopy

- 183 normal group without colon lesions,

-110 low-risk group with less than 3 low-risk adenomas,

- 156 high-risk groups including those with 3 or more low-risk adenomas, 1 or more high-risk adenomas, and those with carcinoma in situ

-42 patients were classified as colorectal cancer group.

Example 2: Isolation of Total RNA from Blood Specimens

Total RNA is isolated from a blood sample collected with a Tempus tube using the Tempus blood RNA isolation kit (Applied Biosystems®).

Example 3: Construction of cDNA from isolated total RNA and real-time PCR

i. cDNA synthesis

Separated total RNA 1.5~4.5ug, Random primer (Invitrogen) 0.625ug, dNTP (Intron), MMLV Reverse Transcription Polymerase (Invitrogen) 500 units were added, and DEPC treated DW was added to the final volume of 50μl, mixed well, and then synthesized The reaction solution was reacted in a thermocycler (ABI) at 25°C for 30 minutes - 37°C for 50 minutes - 70°C for 15 minutes to synthesize cDNA.

ii. Perform qPCR

For the composition of the real-time PCR reaction, add 10 μl of THUNDERBIRD® Probe qPCR Mix (TOYOBO), 10 pmoles of Forward / Reverse Primer, and 10 pmoles of the probe, add 15 μg of synthesized cDNA, and add ultra pure water to make the final volume 20 μl. The base sequences of each primer and probe are listed in Table 3.

Real time PCR reaction was performed using CFX96 (Biorad), and the reaction temperature conditions are as follows. After 95°C, 3 minutes, 95°C, 3 seconds - 60°C, 30 seconds were repeated 40 times. Each time the annealing process (60 ° C, 30 seconds) was performed, a process of measuring fluorescence was added to measure the fluorescence value that increased for each number of times.

Table 3 shows the base sequences of each primer and probe used in the present invention.

Example 4: Confirmation of results and calculation of relative expression of target genes

Using the Cq value of the GAPDH gene used as an endogenous control, the relative expression level (2 ^-ΔCq ) of the target gene is calculated using the Cq value of the target gene.

[relational expression]

2 ^-ΔCq = 2 ^{- (target gene Cq - GAPDH gene Cq)}

Example 5: Production of predictive model for diagnosis of colon cancer and colon polyps

Genetic biomarkers for substitution in the diagnosis prediction model for colorectal cancer and colon polyps are selected, and the relative expression levels of the selected genetic biomarkers are substituted to produce a diagnosis prediction model for colon cancer and colon polyps.

In order to select genetic biomarkers for substitution in the diagnosis prediction model for colorectal cancer and colorectal polyps, the SPSS statistical analysis package was used, and for the production of colorectal cancer and colorectal polyps diagnosis prediction models by substituting the relative expression levels of the selected genetic biomarkers. Statistical analysis was performed using the R package.

The production of colorectal cancer and colorectal polyps diagnosis prediction models was performed by decision tree (DT), logistic regression (LR), neural network (NN), and support vector machine (SVM), but is not limited thereto.

The artificial intelligence prediction model is created by substituting the results of a training set composed of a part of the total sample results. After constructing a validation set with samples not included in the training set, the accuracy of the model built with the training set is verified by substituting the results of the validation set. In this case, accuracy means how accurate the prediction of the prediction model is.

A total of four types of models (DT, NR, NN, and SVM) were produced, and the results using the training set and validation set were repeated 1000 times. Among the models created for 1000 times, the model with the highest accuracy confirmed by the Validation set appears as the final result. Among the four types of models finally created, the type with the highest sensitivity or specificity of the total set (491 in total) including all samples was selected.

i. Creation of artificial intelligence prediction model A that can distinguish colorectal cancer group from other groups (high-risk/low-risk/normal group)

Through t-test statistical analysis, genetic markers showing statistical significance in classifying each group are identified. Model A is constructed by substituting the relative expression levels of a total of 8 corresponding genetic markers.

- As a result, an SVM model was selected that differentiates the colorectal cancer group from the high-risk/low-risk/normal group with a sensitivity of 92.9% and a specificity of 65.0%.

ii. Creation of artificial intelligence prediction model B that can distinguish colorectal cancer group from other groups (high-risk/low-risk/normal group)

- Model A has high sensitivity to distinguish between the colorectal cancer group, but up to 40% of the remaining groups are classified as colorectal cancer groups. Therefore, a model that can distinguish between the colorectal cancer group and the rest of the groups should be created again using samples that are positive in model A.

-Confirm the genetic markers that show statistical significance in classifying each group through t-test statistical analysis for samples that are positive in model A. Model B is produced by substituting the relative expression levels of a total of 9 corresponding genetic markers.

- As a result, an SVM model was selected that differentiates the colorectal cancer group from the high-risk/low-risk/normal group with a sensitivity of 94.9% and a specificity of 87.9%.

iii. Creation of artificial intelligence prediction model C that can distinguish high-risk group from low-risk group/normal group

-Since samples that are negative in models A and B include a high-risk group that requires colonoscopy, a model that distinguishes the high-risk group from the low-risk group/normal group must be created.

-Confirm the genetic markers that show statistical significance in classifying each group through t-test statistical analysis for samples that are negative in models A and B. Model C is constructed by substituting the relative expression levels of a total of six corresponding gene markers.

- As a result, an SVM model was selected that distinguished the high-risk group from the low-risk group/normal group with a sensitivity of 91.3% and specificity of 81.9%.

iv. Overall result of sequentially applying artificial intelligence prediction models A - B - C

As a result of sequentially substituting the relative expression levels of gene markers in the entire sample into the artificial intelligence prediction models A - B - C created in i, ii, and iii, 95.2% of the colorectal cancer group and 91.7% of the high-risk group were positive, and 78.5 % of the low-risk group and the normal group were negative.

The biomarkers used in the models A, B and C are as follows.

Model A-　IL1B, LTF, TNFSF13B, ITIH4, CXCL11, MAPK6, GK, MCAM (8 pieces),

Model B-　CES1, IL1B, TNFSF13B, ITIH4, PTGS2, CXCL11, MAPK6, GK, MCAM (9 pieces) and

Model C-　IL1B, LTF, TNFSF13B, ITIH4, CXCL11, MAPK6 (6 pieces)

Example 6; Collection of clinical specimens

From 2017 to 2022, Shinchon Severance Hospital (Approval No. 4-2017-0148), Gangnam Severance Hospital (Approval No. 3-2017-0024), Kangbuk Samsung Hospital (Approval No. 2017-02-022-009) in the Department of Gastroenterology , Blood samples from subjects scheduled for colonoscopy were collected at the Health Examination Center of Wonju Severance Christian Hospital (approval number CR319115) with the approval of the Bioethics Review Board (IRB) of each institution. A total of 3 ml of blood was collected using a Tempus blood tube (Applied Biosystems®). Subjects were classified as follows according to the results of colonoscopy (Table 4).

classification classification criteria No. of samples (persons) colorectal cancer group Colonoscopy results, subjects with cancer in the colon 112 Progress line group Colonoscopy results, subjects with advanced adenoma in the colon 178 non-progressive lineage group Colonoscopy results, subjects with non-advanced adenoma in the colon 104 control group Subjects with no lesions in the large intestine as a result of colonoscopy 203 total 597

Table 4 shows the classification of subjects and the number of samples according to the results of colonoscopy.

Example 7: Isolation of Total RNA from Blood Specimens

Total RNA is isolated from a blood sample collected with a Tempus tube using the Tempus blood RNA isolation kit (Applied Biosystems®).

Example 8: cDNA construction and qPCR from isolated total RNA

i. Complementary DNA (cDNA) synthesis

Isolated total RNA 1.5~4.5 ug, Random primer (3 ug/uL) (Invitrogen) 2.5 uL, dNTP mixture (2.5 mM each) (Intron) 2.5 uL, M-MLV reverse transcription polymerase (200 U/uL) (Invitrogen ) 2.5 uL, 10 μL of 5× First-strand buffer (250 mM Tris-HCl) (Invitrogen), 5 μL of Dithiothreitol (0.1 M) (Invitrogen) were added, and ultrapure water was added to a final volume of 50 μL, and mixed well. The synthesis reaction solution was reacted in a thermocycler (Applied Biosystems) at 25°C, 30 minutes - 37°C, 50 minutes - 70°C, 15 minutes to synthesize cDNA.

ii. Perform quantitative polymerase chain reaction (qPCR)

For the composition of the qPCR reaction, add 10 μl of THUNDERBIRD® Probe qPCR Mix (TOYOBO), Forward / Reverse Primer, and 1 uL of Probe (10 pmole/uL), add 2 μL of synthesized cDNA, and add ultrapure water to make the final volume 20 μl. Mix. The qPCR reaction was performed using CFX96 (Biorad), and the reaction temperature conditions were as follows. After 95°C, 3 minutes, 95°C, 3 seconds - 60°C, 30 seconds were repeated 40 times. Each time the annealing process (60 ° C, 30 seconds) was performed, a process of measuring fluorescence was added to measure the fluorescence value that increased for each number of times. A constant fluorescence value was set as the threshold, and the Cq value, which is the number of cycles at the time of reaching the threshold, was derived.

Example 9: Confirmation of results and analysis of relative expression of target genes

Using the Cq value of the GAPDH gene used as an endogenous control, the relative expression level (2 ^-ΔCq) of the target gene is calculated using the Cq value of the target gene. The list of genes targeted is as follows (Table 5).

[relational expression]

2 ^-ΔCq = 2 ^{- (target gene Cq - GAPDH gene Cq)}

Serial Number blood genetic markers One ANKHD1-EIF4EBP3 ANKHD1-EIF4EBP3 Readthrough 2 CCR1 C-C Motif Chemokine Receptor 1 3 CES1 Carboxylesterase 1 4 CXCL11 C-X-C Motif Chemokine Ligand 11 5 EpCAM Epithelial Cell Adhesion Molecules 6 GK Glycerol Kinase 7 GPR15 G Protein-Coupled Receptor 15 8 IL1B Interleukin 1 Beta 9 ITGA2 Integrin Subunit Alpha 2 10 ITIH4 Inter-Alpha-Trypsin Inhibitor Heavy Chain 4 11 KRT19 Keratin 19 12 LTF Lactotransferrin 13 MAPK6 Mitogen-Activated Protein Kinase 6 14 MCAM Melanoma Cell Adhesion Molecule 15 MMP23B Matrix Metallopeptidase 23B 16 NME1 NME/NM23 Nucleoside Diphosphate Kinase 17 PPARG Peroxisome Proliferator Activated Receptor Gamma 18 PTGES Prostaglandin E Synthase 19 PTGS2 Prostaglandin-Endoperoxide Synthase 2 20 TNFSF13B TNF Superfamily Member 13b 21 TAS2R10 Taste 2 Receptor Member 10 22 TUG1 Taurine Up-Regulated 1 23 TYMS Thymidylate Synthetase

Table 5 is a list of target blood genetic markers

Example 10: Establishment of a classification model for the purpose of screening for colorectal cancer and advanced adenoma by substituting the relative expression level of the target gene

An artificial intelligence algorithm-based classification model was constructed using the H2O package (version 3.32.1.3) of Statistical R software (version 3.6.3). The production of colorectal cancer and advanced adenoma diagnosis prediction models was based on deep neural network (DNN), generalized linear model (GLM), and random forest (RF) algorithms, and additionally several types of models (GLM, RF, DNN, GBM, stacked ensemble (SE)), but is not limited thereto.

By dividing the entire sample into a training set and a test set, and substituting the results of the training set, an artificial intelligence algorithm-based classification model that can distinguish between a normal group and a colorectal cancer group and an advanced cancer group was constructed, and the performance of the built model was evaluated using the test set. do. When building a model using a training set, a 5-fold cross-validation technique is applied so that the training set is divided into 5 areas to learn the model and at the same time verify the performance of the model using each area to provide a high-performance model. wanted to build.

The performance of the artificial intelligence classification model was judged through the AUROC and AUPRC values of the training set and test set based on the AUROC and AUPRC values, which are representative performance indicators of the classification model. Among them, the model with the best performance was selected based on the performance of the new test set that was not used for model learning. The AUROC and AUPRC values of the DNN, GBM, and RF models built based on each algorithm and the SE model built through AutoML are as follows (Table 6). As a result, the AUROC and AUPRC indicators were the highest in the SE model based on the test set.

Model training set Test set AUROC AUPRC AUROC AUPRC DNN 0.87 0.88 0.75 0.73 GLM 0.78 0.79 0.75 0.71 RF 0.80 0.79 0.76 0.76 SE 1.00 1.00 0.80 0.80

Table 6 shows AUROC and AUPRC performance indicators in the training set and test set. As a result, as shown in Table 7, the sensitivity for classifying the colorectal cancer group was 89.3% and the sensitivity for classifying the advanced adenoma group was 74.5%. The specificity to distinguish the control group was 72.0%.

classification Test set result
(total 157) positive (person) voice (person) Sensitivity (%) Specificity (%) Colorectal cancer group + advanced glandular group
(n = 75) 60 15 80.0 colorectal cancer group
(n = 28) 25 3 89.3 Progress line group
(n = 47) 35 12 74.5 Non-progressive species group + control group
(n = 82) 23 59 72.0

Table 7 shows the sensitivity and specificity results for each group of the SE model.

Example 11: Clinical specimen collection

From 2017 to 2022, Shinchon Severance Hospital (Approval No. 4-2017-0148), Gangnam Severance Hospital (Approval No. 3-2017-0024), Gangbuk Samsung Hospital (Approval No. 2017-02-022-009) in the Department of Gastroenterology , Blood samples were collected from subjects scheduled for colonoscopy at Wonju Severance Christian Hospital Health Examination Center (approval number CR319115) with the approval of the Bioethics Review Board (IRB) of each institution. A total of 3 ml of blood was collected using a Tempus blood tube (Applied Biosystems®). The subjects were classified as follows according to the results of colonoscopy (Table 8).

classification classification criteria No. of samples (persons) colorectal cancer group Colonoscopy results, subjects with cancer in the colon 148 Progress line group Colonoscopy results, subjects with advanced adenoma in the colon 289 Normal group No gastrointestinal symptoms and no family history of colorectal cancer
Subjects with no lesions in the large intestine as a result of colonoscopy 142 total 579

Table 8 shows the classification of subjects and the number of specimens according to colonoscopy results.

Example 12: Isolation of Total RNA from Blood Specimens

Total RNA is isolated from a blood sample collected with a Tempus tube using the Tempus blood RNA isolation kit (Applied Biosystems®).

Example 13: cDNA construction and qPCR from isolated total RNA

i. Complementary DNA (cDNA) synthesis

Isolated total RNA 1.5~4.5 ug, Random primer (3 ug/uL) (Invitrogen) 2.5 uL, dNTP mixture (2.5 mM each) (Intron) 2.5 uL, M-MLV reverse transcription polymerase (200 U/uL) (Invitrogen ) 2.5 uL, 10 μL of 5× First-strand buffer (250 mM Tris-HCl) (Invitrogen), and 5 μL of Dithiothreitol (0.1 M) (Invitrogen) were added, and ultrapure water was added to a final volume of 50 μL, and mixed well. The synthesis reaction solution was reacted in a thermocycler (Applied Biosystems) at 25°C, 30 minutes - 37°C, 50 minutes - 70°C, 15 minutes to synthesize cDNA.

ii. Perform quantitative polymerase chain reaction (qPCR)

For the composition of the qPCR reaction, add 10 μl of THUNDERBIRD® Probe qPCR Mix (TOYOBO), Forward / Reverse Primer, and 1 uL of Probe (10 pmole/uL), add 2 μL of synthesized cDNA, and add ultrapure water to make the final volume 20 μl. Mix. The qPCR reaction was performed using CFX96 (Biorad), and the reaction temperature conditions were as follows. After 95°C, 3 minutes, 95°C, 3 seconds - 60°C, 30 seconds were repeated 40 times. Each time the annealing process (60 ° C, 30 seconds) was performed, a process of measuring fluorescence was added to measure the fluorescence value that increased by number of times. A constant fluorescence value was set as the threshold, and the Cq value, which is the number of cycles at the time of reaching the threshold, was derived.

Example 14: Result Confirmation and Analysis of Relative Expression of Target Genes

Using the Cq value of the GAPDH gene used as an endogenous control, the relative expression level (2 ^-*?*Cq ) of the target gene is calculated using the Cq value of the target gene. In order to compare the relative expression amount of each group of 30 genes, the relative expression amount ratio of the colorectal cancer group compared to the normal group, the relative expression amount ratio of the advanced glandular group compared to the normal group, and the relative expression amount ratio of the colorectal cancer group compared to the advanced glandular group were calculated and shown in the table below. can be expressed as (Table 9).

[relational expression]

2 ^-ΔCq = 2 ^{- (target gene Cq - GAPDH gene Cq)}

gene Comparison of relative expression levels between groups normal group vs. colorectal cancer group normal group vs. Progress line group Progressive line group vs. colorectal cancer group ANKHD1-EIF4EBP3 1.215 1.080 1.124 CCR1 1.078 0.960 1.123 CES1 2.179 1.390 1.568 CXCL11 1.219 1.072 1.136 EpCAM 0.863 0.937 0.921 ERBB2 1.097 1.844 0.595 FOXA2 2.768 4.138 0.669 GK 1.055 1.012 1.043 GPR15 1.131 1.215 0.931 IL1B 1.219 1.194 1.021 ITGA2 1.058 1.154 0.917 ITIH4 0.979 1.037 0.944 KRT19 0.605 0.646 0.937 LTF 1.059 1.228 0.863 MAPK6 1.087 1.152 0.944 MCAM 1.101 0.814 1.352 MKi67 0.551 0.698 0.790 MMP23B 1.151 1.367 0.842 NME1 0.886 0.913 0.969 NPTN 1.246 1.539 0.809 PPARG 1.256 1.166 1.077 PTGES 0.969 1.020 0.950 PTGS2 1.212 1.184 1.023 SNAI2 3.455 4.274 0.808 TAS2R10 0.869 1.207 0.720 TERT 1.585 3.502 0.453 TNFSF13B 1.255 1.106 1.134 TUG1 0.975 1.141 0.855 TYMS 1.106 0.800 1.383 VIM 1.069 1.324 0.807

Table 9 compares the relative expression of 30 genes between groups.

Example 15: Establishment of a classification model for the purpose of screening colorectal cancer and advanced adenoma by substituting the relative expression level of target genes

An artificial intelligence algorithm-based classification model was constructed using the H2O package (version 3.32.1.3) of Statistical R software (version 3.6.3). The production of colorectal cancer and advanced adenoma diagnosis prediction models was based on Deep neural network (DNN), Generalized linear model (GLM), Gradient boosting machine (GBM), and Random forest (RF) algorithms, and additionally several types of models (GLM, RF, DNN, GBM, stacked ensemble (SE)) was performed by grafting Automated machine learning (AutoML) method to build a model suitable for data, but is not limited thereto.

By dividing the entire sample into a training set and a test set, and substituting the results of the training set, an artificial intelligence algorithm-based classification model that can distinguish between a normal group and a colorectal cancer group and an advanced cancer group was constructed, and the performance of the built model was evaluated using the test set. (FIG. 12). When building a model using a training set, a 5-fold cross-validation technique is applied so that the training set is divided into 5 areas to learn the model and at the same time verify the performance of the model using each area to provide a high-performance model. It was intended to build (FIG. 16).

The performance of the artificial intelligence classification model was judged through the AUROC and AUPRC values of the training set and test set based on the AUROC and AUPRC values, which are representative performance indicators of the classification model. Among them, the model with the best performance was selected based on the performance of the new test set that was not used for model learning. The AUROC and AUPRC values of the DNN, GBM, GLM, and RF models built based on each algorithm and the SE model built through AutoML are as follows (Table 9). As a result, the AUC and AUPRC indicators were the highest based on the test set in the GBM model and the SE model built through AutoML.

Model training set Test set AUROC AUPRC AUROC AUPRC GLM 0.93 0.98 0.92 0.98 DNN 0.95 0.98 0.91 0.97 RF 0.93 0.98 0.96 0.99 GBM 1.00 1.00 0.97 0.99 SE 1.00 1.00 0.97 0.99

Table 10 shows the results of the AUROC and AUPRC indicators in the training and test sets for each model, the test set results for each group of the GBM model and SE model.

In the GBM model, the sensitivity for distinguishing the colorectal cancer group was 94.6%, the sensitivity for distinguishing the advanced adenoma group was 97.5%, and the specificity for distinguishing the normal group was 80.6% (Table 11). The sensitivity was 91.9%, the sensitivity to distinguish the advanced adenoma group was 95.1%, and the specificity to distinguish the normal group was 80.6% (Table 12). Therefore, the GBM model showing higher sensitivity was finally selected.

classification Test set result positive (person) voice (person) Sensitivity (%) Specificity (%) Colorectal cancer group + advanced glandular group
(n = 118) 114 4 96.6 colorectal cancer group
(n = 37) 35 2 94.6 Progressive gland group (n = 81) 79 2 97.5 Normal group
(n = 36) 7 29 80.6

Table 11 shows the sensitivity and specificity results for each group of the GBM model.

classification Test set result positive (person) voice (person) Sensitivity (%) Specificity (%) Colorectal cancer group + advanced glandular group
(n = 118) 111 7 94.1 colorectal cancer group
(n = 37) 34 3 91.9 Progressive gland group (n = 81) 77 4 95.1 Normal group
(n = 36) 7 29 80.6

Table 12 shows the sensitivity and specificity results for each group of the SE model.

comparative example

In colorectal cancer or advanced adenoma, a precursor of colorectal cancer, circulating tumor cells may exist in the blood, and 10 genes ( EpCAM, ERBB2, FOXA2, KRT19, MCAM, MKi67, NPTN, SNAI2, TERT, VIM ) as a target, the relative expression level by group was calculated, and an artificial intelligence algorithm-based model was constructed to distinguish colorectal cancer or advanced adenoma from the normal group.

1) Specimen collection

From 2017 to 2022, Shinchon Severance Hospital (Approval No. 4-2017-0148), Gangnam Severance Hospital (Approval No. 3-2017-0024), Kangbuk Samsung Hospital (Approval No. 2017-02-022-009) in the Department of Gastroenterology , Blood samples from subjects scheduled for colonoscopy were collected at the Wonju Severance Christian Hospital Health Examination Center (approval number CR319115) with the approval of the Bioethics Review Board (IRB) of each institution. A total of 3 ml of blood was collected using a Tempus blood tube (Applied Biosystems®). Subjects were classified as follows through the results of colonoscopy (Table 13).

classification classification criteria No. of samples (persons) colorectal cancer group Colonoscopy results, subjects with cancer in the colon 148 Progress line group Colonoscopy results, subjects with advanced adenoma in the colon 289 Normal group No gastrointestinal symptoms and no family history of colorectal cancer
Subjects with no lesions in the large intestine as a result of colonoscopy 142 total 579

Table 13 is the classification of subjects and the number of samples according to the results of colonoscopy

2) Isolation of total RNA from blood samples

Total RNA is isolated from a blood sample collected with a Tempus tube using the Tempus blood RNA isolation kit (Applied Biosystems®).

3) cDNA production and qPCR from isolated total RNA

i. Complementary DNA (cDNA) synthesis

Isolated total RNA 1.5~4.5 ug, Random primer (3 ug/uL) (Invitrogen) 2.5 uL, dNTP mixture (2.5 mM each) (Intron) 2.5 uL, M-MLV reverse transcription polymerase (200 U/uL) (Invitrogen ) 2.5 uL, 10 μL of 5× First-strand buffer (250 mM Tris-HCl) (Invitrogen), 5 μL of Dithiothreitol (0.1 M) (Invitrogen) were added, and ultrapure water was added to a final volume of 50 μL, and mixed well. The synthetic reaction solution was reacted in a thermocycler (Applied Biosystems) at 25°C, 30 minutes - 37°C, 50 minutes - 70°C, 15 minutes to synthesize cDNA.

ii. Perform quantitative polymerase chain reaction (qPCR)

For the composition of the qPCR reaction, add 10 μl of THUNDERBIRD®Probe qPCR Mix (TOYOBO), Forward / Reverse Primer, and 1 uL of Probe (10 pmole/uL), add 2 μL of synthesized cDNA, and add ultrapure water to make the final volume 20 μl. Mix. The qPCR reaction was performed using CFX96 (Biorad), and the reaction temperature conditions were as follows. After 95°C 3 minutes, 95°C 3 seconds - 60°C 30 seconds were repeated 40 times. Each time the annealing process (60 ° C, 30 seconds) was performed, a process of measuring fluorescence was added to measure the fluorescence value that increased for each number of times. A constant fluorescence value was set as the threshold, and the Cq value, which is the number of cycles at the time of reaching the threshold, was derived.

4) Confirmation of results and analysis of relative expression of target genes

Using the Cq value of the GAPDH gene used as an endogenous control, the relative expression level (2 ^-ΔCq ) of the target gene is calculated using the Cq value of the target gene. The list of genes targeted is as follows (Table 14).

Serial Number blood genetic markers One EpCAM Epithelial Cell Adhesion Molecules 2 ERBB2 Erb-B2 Receptor Tyrosine Kinase 2 3 FOXA2 Forkhead Box A2 4 KRT19 Keratin 19 5 MCAM Melanoma Cell Adhesion Molecule 6 MKi67 Marker Of Proliferation Ki-67 7 NPTN Neuroplastin 8 SNAI2 Snail Family Transcriptional Repressor 2 9 TERT Telomerase Reverse Transcriptase 10 VIM Vimentin

5) Establishment of a classification model for the purpose of screening for colorectal cancer and advanced adenoma by substituting the relative expression level of target genes

An artificial intelligence algorithm-based classification model was constructed using the H2O package (version 3.32.1.3) of Statistical R software (version 3.6.3). The production of colorectal cancer and advanced adenoma diagnosis prediction models was based on Deep neural network (DNN), Generalized linear model (GLM), Gradient boosting machine (GBM), and Random forest (RF) algorithms, and additionally several types of models (GLM, RF, DNN, GBM, stacked ensemble (SE)) was performed by grafting Automated machine learning (AutoML) method to build a model suitable for data, but is not limited thereto.

By dividing the entire sample into a training set and a test set, and substituting the results of the training set, an artificial intelligence algorithm-based classification model that can distinguish between a normal group and a colorectal cancer group and an advanced cancer group was constructed, and the performance of the built model was evaluated using the test set. do. When building a model using a training set, a 5-fold cross-validation technique is applied so that the training set is divided into 5 areas to learn the model and at the same time verify the performance of the model using each area to provide a high-performance model. wanted to build.

The performance of the artificial intelligence classification model was judged through the AUROC and AUPRC values of the training set and test set based on the AUROC and AUPRC values, which are representative performance indicators of the classification model. Among them, the model with the best performance was selected based on the performance of the new test set that was not used for model learning. The AUROC and AUPRC values of the DNN, GBM, and RF models built based on each algorithm and the GBM model built through AutoML are as follows (Table 3). As a result, the AUROC and AUPRC indicators were the highest in the GBM model based on the test set.

Model training set Test set AUROC AUPRC AUROC AUPRC GLM 0.91 0.96 0.86 0.96 DNN 0.99 1.00 0.92 0.97 RF 0.90 0.96 0.94 0.98 GBM 1.00 1.00 0.94 0.98 GBM (AutoML) 0.98 0.99 0.91 0.97

Table 15 shows AUROC and AUPRC performance indicators in the training set and test set. As a result,

As a result of confirming the test set results for each group in the GBM model, the sensitivity to distinguish the colorectal cancer group was 78.4%, the sensitivity to distinguish the advanced adenoma group was 88.9%, and the specificity to distinguish the normal group was 80.6%.

classification Test set result
(total 157) positive (person) voice (person) Sensitivity (%) Specificity (%) Colorectal cancer group + advanced glandular group
(n = 118) 101 17 85.6 colorectal cancer group
(n = 37) 29 8 78.4 Progressive gland group (n = 81) 72 9 88.9 Normal group
(n = 36) 7 29 80.6

Table 16 shows the sensitivity and specificity results for each group of the GBM model.

Claims (19)

IL1B, LTF, TNFSF13B, ITIH4, CXCL11, MAPK6, GK, and the colorectal cancer group containing primers or probes involved in measuring the relative expression levels of MCAM genes as active ingredients and colorectal cancer high-risk group, low-risk group, and normal group can be distinguished composition in. According to claim 1, The primers and probes are SEQ ID NOs: 7 to 9, SEQ ID NOs: 13 to 15, SEQ ID NOs: 16 to 18, SEQ ID NOs: 22 to 24, SEQ ID NOs: 34 to 36, SEQ ID NOs: 40 to 42, SEQ ID NOs: 43 to 45, and sequences A composition characterized in that it consists of the sequences set forth in numbers 56 to 58. According to claim 1, The normal group is a case in which there are no lesions in the colon through colonoscopy, the low-risk group is a case in which there are less than three low-risk adenomas, and the high-risk group is a case in which three or more low-risk adenomas are present through colonoscopy, and a high-risk group is present. A composition comprising one or more adenomas and carcinoma in situ. CES1, IL1B, TNFSF13B, ITIH4, PTGS2, CXCL11, MAPK6, GK, and MCAM genes involved in measuring the relative expression levels of primers or probes as active ingredients, and colorectal cancer high-risk, low-risk, and normal groups can be distinguished. possible composition. According to claim 4, The primers and probes are SEQ ID NOs: 4 to 6, SEQ ID NOs: 7 to 9, SEQ ID NOs: 16 to 18, SEQ ID NOs: 22 to 24, SEQ ID NOs: 31 to 33, SEQ ID NOs: 34 to 36, SEQ ID NOs: 40 to 42, SEQ ID NOs: 43 to 45, and SEQ ID NOs: 56 to 58. IL1B, LTF, TNFSF13B, ITIH4, CXCL11 and MAPK6 genes involved in measuring the relative expression levels of primers or probes as an active ingredient a composition capable of distinguishing between high-risk, low-risk and normal groups of colorectal cancer. According to claim 6, The primers and probes are characterized by consisting of the sequences shown in SEQ ID NOs: 7 to 9, SEQ ID NOs: 13 to 15, SEQ ID NOs: 16 to 18, SEQ ID NOs: 22 to 24, SEQ ID NOs: 34 to 36, and SEQ ID NOs: 40 to 42 composition. a) measuring the relative expression levels of IL1B, LTF, TNFSF13B, ITIH4, CXCL11, MAPK6, GK, and MCAM genes using primers and probes through polymerase chain reaction; b) Relative expression levels of CES1, IL1B, TNFSF13B, ITIH4, PTGS2, CXCL11, MAPK6, GK, and MCAM genes using primers and probes through polymerase chain reaction for samples positive for colon cancer in a) above is measured and if it is positive, it is judged to be colon cancer positive, c) Samples judged negative in a) and b) were subjected to polymerase chain reaction to measure the relative expression levels of IL1B, LTF, TNFSF13B, ITIH4, CXCL11, and MAPK6 genes using primers and probes. A method for screening colon cancer and colon polyps, characterized in that the colon cancer is judged as positive. According to claim 8, The primers and probes used in a) are SEQ ID NOs: 7 to 9, SEQ ID NOs: 13 to 15, SEQ ID NOs: 16 to 18, SEQ ID NOs: 22 to 24, SEQ ID NOs: 34 to 36, SEQ ID NOs: 40 to 42, and SEQ ID NOs: 43 to 45, and the sequences set forth in SEQ ID NOs: 56 to 58; The primers and probes used in b) are SEQ ID NOs: 4 to 6, SEQ ID NOs: 7 to 9, SEQ ID NOs: 16 to 18, SEQ ID NOs: 22 to 24, SEQ ID NOs: 31 to 33, SEQ ID NOs: 34 to 36, and SEQ ID NOs: 40 to 40. 42, SEQ ID NOs: 43 to 45, and SEQ ID NOs: 56 to 58, The primers and probes used in c) are the sequences shown in SEQ ID NOs: 7 to 9, SEQ ID NOs: 13 to 15, SEQ ID NOs: 16 to 18, SEQ ID NOs: 22 to 24, SEQ ID NOs: 34 to 36, and SEQ ID NOs: 40 to 42. A method for screening colon cancer and colon polyps, characterized in that made. a) a primer or probe set involved in measuring the relative expression levels of IL1B, LTF, TNFSF13B, ITIH4, CXCL11, MAPK6, GK, and MCAM genes; b) primers or probe sets involved in measuring the relative expression levels of CES1, IL1B, TNFSF13B, ITIH4, PTGS2, CXCL11, MAPK6, GK, and MCAM genes, and c) A screening kit for colorectal cancer and colorectal polyps comprising primers or probe sets involved in measuring the relative expression levels of IL1B, LTF, TNFSF13B, ITIH4, CXCL11, and MAPK6 genes. According to claim 10, The primers and probes used in a) are SEQ ID NOs: 7 to 9, SEQ ID NOs: 13 to 15, SEQ ID NOs: 16 to 18, SEQ ID NOs: 22 to 24, SEQ ID NOs: 34 to 36, SEQ ID NOs: 40 to 42, and SEQ ID NOs: 43 to 45, and the sequences set forth in SEQ ID NOs: 56 to 58; The primers and probes used in b) are SEQ ID NOs: 4 to 6, SEQ ID NOs: 7 to 9, SEQ ID NOs: 16 to 18, SEQ ID NOs: 22 to 24, SEQ ID NOs: 31 to 33, SEQ ID NOs: 34 to 36, and SEQ ID NOs: 40 to 40. 42, SEQ ID NOs: 43 to 45, and SEQ ID NOs: 56 to 58, The primers and probes used in c) are the sequences shown in SEQ ID NOs: 7 to 9, SEQ ID NOs: 13 to 15, SEQ ID NOs: 16 to 18, SEQ ID NOs: 22 to 24, SEQ ID NOs: 34 to 36, and SEQ ID NOs: 40 to 42. Colon cancer and colon polyps screening kit, characterized in that consisting of. CCR1, CES1, IL1B, ITGA2, LTF, TNFSF13B, PTGES, ITIH4, TUG1, NME1, PTGS2, CXCL11, MAPK6, GK, KRT19, EpCAM, MCAM, PPARG, ANKHD1-EIF4EBP3, GPR15, MMP23B, TAS2R10, TYMS, FOXA2, A composition for screening for colorectal cancer and advanced adenoma, comprising primers and probes involved in measuring relative expression levels of MKi67, ERBB2, NPTN, SNAI2, TERT and VIM genes as active ingredients. According to claim 12, Wherein the primers and probes are composed of the sequences set forth in SEQ ID NOs: 1 to 91. CCR1, CES1, IL1B, ITGA2, LTF, TNFSF13B, PTGES, ITIH4, TUG1, NME1, PTGS2, CXCL11, MAPK6, GK, KRT19, EpCAM, MCAM, PPARG, ANKHD1-EIF4EBP3, GPR15, MMP23B, TAS2R10, TYMS, FOXA2, A screening test kit for colorectal cancer and advanced adenoma comprising primers and probes involved in measuring the relative expression levels of MKi67, ERBB2, NPTN, SNAI2, TERT and VIM genes as active ingredients. According to claim 14, The primers and probes kit, characterized in that consisting of the sequences shown in SEQ ID NOs: 1 to 91. Samples were subjected to polymerase chain reaction using primers and probes to detect CCR1, CES1, IL1B, ITGA2, LTF, TNFSF13B, PTGES, ITIH4, TUG1, NME1, PTGS2, CXCL11, MAPK6, GK, KRT19, EpCAM, MCAM, PPARG, Measuring the relative expression levels of ANKHD1-EIF4EBP3, GPR15, MMP23B, TAS2R10, TYMS, FOXA2, MKi67, ERBB2, NPTN, SNAI2, TERT and VIM genes A method for providing information for the prediction or diagnosis of colorectal cancer or advanced adenoma. 17. The method according to claim 16, wherein the expression level is performed using primers and probes, and the primers and probes consist of the sequences shown in SEQ ID NOs: 1 to 91. The method of claim 16, wherein the subject is determined to have colon cancer when at least one of the following 1) to 3) is confirmed: 1) CCR1, CES1, GK, IL1B, KRT19, LTF, PPARG, PTGES, PTGS2, TAS2R10, TNFSF13B and TYMS genes, or the protein levels encoded by the genes, were compared with the corresponding genes or those encoded by the genes in normal control subject samples. If expression level changes compared to protein levels are seen; 2) ANKHD1-EIF4EBP3, ITIH4, MCAM, PPARG, TAS2R10, TNFSF13B, TUG1 and TYMS genes or the level of proteins encoded by said genes in a sample of an individual with a non-advanced adenoma or the corresponding gene If expression level changes are seen compared to the level of the protein encoded by; and 3) ANKHD1-EIF4EBP3, CCR1, MCAM, MMP23B, TAS2R10, TNFSF13B, TUG1 and TYMS genes or protein levels encoded by the genes are compared with the level of the gene or the protein encoded by the gene in a sample of an individual with advanced adenoma to show a change in expression level. The method for providing information for prediction or diagnosis of colon cancer or advanced adenoma according to claim 16, wherein the following 1) or 2) is additionally confirmed to determine the pathological characteristics of the individual having colon cancer: 1) CXCL11 and PTGS2 genes or the protein levels encoded by the genes are compared with the corresponding genes or the protein levels encoded by the genes of an individual sample having a low TNM stage colorectal cancer, and the change in expression level If present, the subject is judged to have high TNM stage colorectal cancer; or 2) Compare the level of the EpCAM gene or the protein encoded by the gene with the level of the gene or the protein encoded by the gene in a sample of an individual having highly differentiated colorectal cancer. diagnosed as having colorectal cancer.

Images