DKK4 gene-based colorectal tumor risk prediction model and application thereof

Abstract

The invention discloses a colorectal tumor risk prediction model based on a DKK4 gene and application thereof, and a construction method of the prediction model comprises the following steps: (1) determining the DKK4 gene as a candidate target through a differential expression gene screened by a public database; (2) evaluating the expression level and the biological function of DKK4 in colorectal cancer and precancerous lesions and the relevance with a Wnt signal channel by utilizing single gene function analysis and combining with a TCGA database; (3) through LASSO regression analysis, constructing a colorectal cancer risk scoring model by using DKK4 and related genes thereof, and screening candidate genes; and (4) verifying the prediction efficiency of the risk scoring model on the one-year and three-year lifetime of the colorectal cancer patient by using an independent data set, and verifying the expression difference of DKK4 in colorectal adenoma and cancer tissues through immunohistochemistry. The DKK4 can be used as a precancerous lesion specific marker, and the expression level of the DKK4 is in positive correlation with the risk of malignant adenoma transformation. The model can effectively evaluate the colorectal cancer progression risk.

Description

Technical Field

[0001] The present invention relates to the field of biotechnology, and in particular to a colorectal tumor risk prediction model based on the DKK4 gene and its application. Background Art

[0002] Colorectal cancer (CRC) is one of the digestive system malignancies with the highest morbidity and mortality worldwide. Its occurrence and development mostly follow the classic "adenoma-carcinoma" sequence, of which about 80% of colorectal cancers originate from adenoma malignant transformation. Although traditional endoscopic techniques (such as colonoscopy) play an important role in the detection of precancerous lesions, their sensitivity and specificity are limited by operator experience and the easy misdiagnosis of minor lesions, resulting in insufficient early diagnosis rate. In addition, existing molecular marker studies mostly focus on the differential genes between colorectal cancer and adjacent tissues, while there is still a significant gap in the screening of specific markers for precancerous lesions (such as colorectal adenoma with intraepithelial neoplasia). If molecular markers related to precancerous lesions can be screened out, it will help to improve the early diagnosis rate and reduce the risk of colorectal cancer.

[0003] DKK4 (Dickkopf-related protein 4), a key regulator of the Wnt signaling pathway, has been previously suggested to play a dual role in tumorigenesis: on the one hand, it participates in the regulation of cell proliferation by inhibiting the Wnt pathway, and on the other hand, it may influence tumor progression by regulating the immune microenvironment. However, the expression characteristics and biological functions of DKK4 in colorectal precancerous lesions, as well as its application in risk prediction, remain unclear. Existing technologies lack colorectal cancer risk prediction models based on DKK4, particularly tools for the dynamic monitoring and prognostic assessment of the transformation of precancerous lesions to invasive cancer. Summary of the Invention

[0004] In order to solve the problems of the prior art, the purpose of the present invention is to provide a colorectal tumor risk prediction model based on the DKK4 gene and its application.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] In a first aspect, the present application provides an RcWRKY29 gene.

[0007] In a second aspect, the present application provides a protein encoded by the RcWRKY29 gene.

[0008] The first aspect of the present application provides a method for constructing a colorectal tumor risk prediction model based on the DKK4 gene, comprising the following steps:

[0009] (1) Using public databases to screen for differentially expressed genes between colorectal precancerous lesions and healthy controls, we identified the DKK4 gene as a candidate target. Genomic sequence: According to the NCBI database, the genomic sequence of the DKK4 gene is NC_000008.11 (42374063..42391322, complement). mRNA sequence: The mRNA sequence of the major transcript ENST00000220812.2 is 894 bp long and can be found in the NCBI database.

[0010] (2) Using single gene functional analysis and combining with the TCGA database, evaluate the expression level, biological function, and association of DKK4 with the Wnt signaling pathway in colorectal cancer and precancerous lesions;

[0011] (3) LASSO regression analysis was used to construct a colorectal cancer risk score model based on DKK4 and its related genes, and 12 candidate genes including DKK4 were screened as risk factors;

[0012] (4) An independent data set was used to verify the predictive efficacy of the risk score model for the 1-year and 3-year survival of colorectal cancer patients, and the expression differences of DKK4 in colorectal adenoma and cancer tissues were verified by immunohistochemistry.

[0013] Furthermore, in step (1), the public database includes the GSE41657 and GSE37364 datasets of the NCBI-GEO database, and the screening conditions are fold difference (FC) ≥ 2 and significance (P < 0.05).

[0014] Furthermore, in step (3), LASSO regression analysis uses ten-fold cross validation to determine the risk scoring formula by optimizing the lambda value. The risk scoring formula is:

[0015] Risk Score = ∑i = 1n (gene expression level × regression coefficient) Risk Score = i = 1∑n (gene expression level × regression coefficient); the regression coefficient is 0.00413,

[0016] Among them, the genes include DKK4 and 11 immune regulatory genes significantly associated with it.

[0017] Furthermore, in step (4), immunohistochemical validation includes scoring the DKK4 expression levels of normal colorectal tissues, adenomas with low / high-grade intraepithelial neoplasia, and invasive carcinoma tissues. The scoring criteria are: the product of the positive staining rate (≤1% for 0 points, 2%-25% for 1 point, 26%-50% for 2 points, 51%-75% for 3 points, >75% for 4 points) and the staining intensity (colorless for 0 points, light yellow for 1 point, yellow for 2 points, brownish yellow for 3 points). A total score of ≥5 points indicates high risk.

[0018] The second aspect of this application provides a colorectal tumor risk prediction model constructed based on the method, which includes expression data of the DKK4 gene and its related immune regulatory genes, LASSO regression coefficient and risk score threshold, and is used to predict the survival prognosis of colorectal cancer patients within 3 years.

[0019] Furthermore, the model integrates patient age, AJCC stage, and risk score through a nomogram to generate individualized survival probability prediction results.

[0020] In a third aspect, the present application provides a colorectal tumor risk prediction kit based on the DKK4 gene, comprising primers, antibodies or probes for detecting DKK4 gene expression, and an algorithm and threshold standard for a risk scoring model.

[0021] Furthermore, the antibody is a rabbit-derived DKK4 monoclonal antibody (Proteintech, catalog number 27080-1-AP), which is suitable for immunohistochemical detection of DKK4 expression levels in colorectal tissue samples.

[0022] The fourth aspect of the present application provides an application of a risk prediction model in the early diagnosis of colorectal cancer, by detecting the expression level of DKK4 in patient tissue samples and combining the risk scoring model to judge the risk of adenoma transforming into cancer.

[0023] The fifth aspect of this application provides an application of a risk prediction model in predicting the efficacy of immunotherapy for colorectal cancer. The model evaluates the patient's sensitivity to immune checkpoint inhibitors by analyzing the correlation between DKK4 and tumor mutation burden (TMB), macrophage polarization and PD-1 / PD-L1 expression.

[0024] Beneficial effects: This invention, for the first time, through public database data mining and experimental verification, clearly shows that DKK4 is significantly overexpressed in colorectal adenomas with high-grade intraepithelial neoplasia, and its expression level is closely related to the risk of malignant transformation of adenomas. It can be used as a specific marker for precancerous lesions to assist endoscopic detection and improve the detection rate of early lesions.

[0025] Compared with the prior art, the present invention has the following advantages:

[0026] (1) This invention fills the gap in the research of molecular markers of precancerous lesions by integrating bioinformatics analysis and experimental verification, and provides important theoretical support and technical tools for the early prevention, prognosis evaluation and targeted treatment of colorectal cancer.

[0027] (2) The LASSO regression risk prediction model constructed based on DKK4 and 12 candidate genes in the present invention can effectively evaluate the survival prognosis of colorectal cancer patients within 3 years (AUC = 0.695), providing a quantitative basis for clinical individualized treatment and follow-up strategies.

[0028] (3) The DKK4 of the present invention significantly affects the tumor immune microenvironment by regulating macrophage polarization, T cell differentiation, and antigen presentation, providing a new direction for the development of immunotherapy targets.

[0029] (4) The present invention verified through immunohistochemistry that the expression of DKK4 in adenoma tissue is significantly higher than that in normal tissue and invasive cancer. Its dynamic expression pattern can assist in pathological typing and provide a molecular basis for the timing of intervention in precancerous lesions.

[0030] (5) This invention combines multi-omics data (transcriptome, immunohistochemistry, clinical cohort) and machine learning models (LASSO regression) to achieve a full-chain study from marker screening to risk prediction, providing a systematic solution for the precise prevention and control of colorectal cancer. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments or descriptions of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0032] Figure 1 This is a data analysis diagram of differentially expressed mRNA in colorectal adenoma with intraepithelial neoplasia of the present invention. Figure 1 A is a cluster heat map analysis (selected differences) of differentially expressed mRNA TOP100 in colorectal adenoma with intraepithelial neoplasia of the present invention; Figure 1 B is a scatter plot analysis of differentially expressed mRNA in colorectal adenoma with intraepithelial neoplasia of the present invention. The gene names marked in red are the top 5 upregulated genes, and the gene names marked in blue are the top 5 downregulated genes.

[0033] Figure 2 This is a single gene analysis diagram of DKK4 of the present invention. Figure 2 A is the differential expression of DKK4 between colon cancer / rectal cancer patients and adjacent tissues analyzed by TCGA data of the present invention; Figure 2 B is a cluster diagram of the TOP15 up-regulated genes and TOP15 down-regulated genes that are differentially enriched between the DKK4 high-level and low-level groups of the present invention, where red indicates relatively high mRNA expression and blue indicates relatively low mRNA expression; Figure 2C is the DKK4 single gene GSEA enrichment analysis diagram of the present invention; Figure 2 D is a diagram showing the regulatory effect of DKK4 on key molecules in the wnt signaling pathway.

[0034] Figure 3 This is a diagram showing the correlation between DKK4 and immune cells of the present invention. Figure 3 A is a correlation analysis diagram of DKK4 expression in immune cell-related datasets in the TCGA data of the present invention; Figure 3 B is the expression abundance distribution diagram of DKK4 in 22 immune cells analyzed by Cibersort of the present invention; Figure 3 C is a correlation analysis diagram of DKK4 in immune cell interactions of the present invention; Figure 3 D is a box plot of the differential analysis of the DKK4 gene in 22 immune cells of the present invention.

[0035] Figure 4 This is a differential expression diagram of the tumor immune pathway genes of the present invention in the DKK4 high / low level groups. Figure 4 A is a boxplot of the differential expression of tumor immunity-related genes in B cells, CD4+T cells, CD8+T cells, and DC cells in the DKK4 high / low level groups of the present invention; Figure 4 B is a boxplot of the differential expression of macrophage, neutrophil, and NK cell tumor immunity-related genes in the DKK4 high / low level groups of the present invention; Figure 4 C is a box plot of the differential expression of the antigen presentation, cytolytic activity, type I interferon and type II interferon response tumor immunity-related genes in the DKK4 high / low level groups of the present invention.

[0036] Figure 5 This is a diagram showing the differential expression of tumor immunity promoting genes / suppressing genes in the DKK4 high / low level groups of the present invention. Figure 5 A- Figure 5 B is a boxplot of differential expression of tumor immunity promotion-related genes in the DKK4 high / low level groups of the present invention; Figure 5 C is a box plot of the differential expression of tumor immunosuppression-related genes in the DKK4 high / low level groups of the present invention.

[0037] Figure 6 This is a LASSO regression model diagram of DKK4 of the present invention. Figure 6 A is a ten-fold cross-validation graph of 603 colorectal cancer transcriptome sequencing data containing survival results in TCGA of the present invention; Figure 6 B is a graph showing the changes in characteristic coefficients of differentially expressed genes in the transcriptome data of the present invention with λ, and screening of candidate genes for risk assessment. Figure 6 C- Figure 6D is a diagram of the present invention that uses risk scores and patient survival status to draw a risk factor linkage diagram (ggrisk) to evaluate the relationship between risk scores and patient survival rates.

[0038] Figure 7 1 is a univariate and multivariate COX regression analysis diagram of the colorectal cancer patient risk score and clinical information of the present invention. Figure 7 A is a univariate COX regression analysis diagram of the impact of age, gender, TNM stage, AJCC grade, blood vessels, lymph node infiltration data and risk score of the colorectal cancer patient population on patient survival status; Figure 7 B is a multivariate COX regression analysis diagram of the impact of age, gender, TNM stage, AJCC grade, blood vessels, lymph node infiltration data and risk score of the colorectal cancer patient population on patient survival status.

[0039] Figure 8 This is a diagram of the survival prognosis risk prediction model for colorectal cancer patients based on the DKK4 gene of the present invention. Figure 8 A is a survival curve of 603 colorectal cancer patients grouped by risk score of the present invention; Figure 8 B- Figure 8 D is the AUC curve of the risk score of the present invention for predicting 1-, 3-, and 5-year survival of colorectal cancer patients; Figure 8 E is a nomogram constructed by using risk scores and colorectal cancer patient data of the present invention; Figure 8 F is a calibration curve diagram drawn by calibrating the nomogram through 1-, 3-, and 5-year survival analysis of the present invention.

[0040] Figure 9 This is a risk factor linkage diagram of the risk prediction model of the present invention verified in the GSE39582 dataset.

[0041] Figure 10 This is a prediction validation diagram of the risk prediction model of the present invention for the survival prognosis of patients in the GSE39582 dataset. Figure 10 A is a survival curve of the survival of colorectal cancer patients in the GSE39582 dataset based on the risk score grouping of the present invention; Figure 10 B- Figure 10 D is the AUC curve of the risk score of the present invention for predicting 1-, 3-, and 5-year survival of colorectal cancer patients; Figure 10 E is a nomogram constructed by using the risk score and GSE39582 dataset to collect colorectal cancer patient data; Figure 10 F is a calibration curve diagram drawn by calibrating the nomogram through 1-, 3-, and 5-year survival analysis of the present invention.

[0042] Figure 11 This is a diagram showing the expression of DKK4 in colorectal tumor tissues according to the present invention. Figure 11 A, Figure 11 C is HE staining and DKK4 immunohistochemical analysis of normal colorectal tissue of the present invention (5×, 50×); Figure 11 B, Figure 11 D is the adenoma with low-grade intraepithelial neoplasia of the present invention HE staining and DKK4 immunohistochemical analysis (5×, 50×); Figure 11 E, Figure 11 G is the HE staining and DKK4 immunohistochemical analysis of adenoma with high-grade intraepithelial neoplasia of the present invention (5×, 50×); Figure 11 F, Figure 11 H is HE staining and DKK4 immunohistochemical analysis of colorectal cancer of the present invention (5×, 50×); DETAILED DESCRIPTION

[0043] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0044] In this application, the term "and / or" describes the relationship between associated objects, indicating that three possible relationships exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. A and B can be singular or plural. The character " / " generally indicates that the associated objects are in an "or" relationship.

[0045] In this application, "-one or more" means one or more, and "more than one" means two or more. "The following - one or more" or similar expressions refer to any combination of these items, including any combination of single items or plural items. For example, "a, b, or c - one or more", or "a, b, and c - one or more" can all mean: a, b, c, ab (i.e., a and b), ac, bc, or abc, where a, b, and c can be single or multiple.

[0046] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. Some or all of the steps can be executed in parallel or sequentially. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0047] The terms used in the embodiments of the present application are for the purpose of describing specific embodiments only and are not intended to limit the present application. The singular forms "a", "an", "the" and "the" used in the embodiments of the present application and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise.

[0048] The weights of the relevant components mentioned in the examples of this application may not only refer to the specific content of each component, but also represent the weight ratio between the components. Therefore, as long as the content of the relevant components is proportionally enlarged or reduced according to the examples of this application, it is within the scope disclosed in the examples of this application. Specifically, the mass described in the examples of this application may be a mass unit known in the chemical industry, such as μg, mg, g, kg, etc.

[0049] The terms "first" and "second" are used solely for descriptive purposes to distinguish objects, such as substances, from one another and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the technical features being referred to. For example, a first XX could also be referred to as a second XX, and similarly, a second XX could also be referred to as a first XX, without departing from the scope of the embodiments of this application. Thus, features defined as "first" or "second" may explicitly or implicitly include one or more of such features.

[0050] The first aspect of the present application provides a method for constructing a colorectal tumor risk prediction model based on the DKK4 gene, comprising the following steps:

[0051] (1) Using public databases to screen for differentially expressed genes between colorectal precancerous lesions and healthy controls, we identified the DKK4 gene as a candidate target. Genomic sequence: According to the NCBI database, the genomic sequence of the DKK4 gene is NC_000008.11 (42374063..42391322, complement). mRNA sequence: The mRNA sequence of the major transcript ENST00000220812.2 is 894 bp long and can be found in the NCBI database.

[0052] (2) Using single gene functional analysis and combining with the TCGA database, evaluate the expression level, biological function, and association of DKK4 with the Wnt signaling pathway in colorectal cancer and precancerous lesions;

[0053] (3) LASSO regression analysis was used to construct a colorectal cancer risk score model based on DKK4 and its related genes, and 12 candidate genes including DKK4 were screened as risk factors;

[0054] (4) An independent data set was used to verify the predictive efficacy of the risk score model for the 1-year and 3-year survival of colorectal cancer patients, and the expression differences of DKK4 in colorectal adenoma and cancer tissues were verified by immunohistochemistry.

[0055] In some embodiments, in step (1), the public database includes the GSE41657 and GSE37364 datasets of the NCBI-GEO database, and the screening conditions are fold difference (FC) ≥ 2 and significance (P < 0.05).

[0056] In some embodiments, in step (3), LASSO regression analysis uses ten-fold cross validation to determine the risk scoring formula by optimizing the lambda value. The risk scoring formula is:

[0057] Risk Score = ∑i = 1n (gene expression level × regression coefficient) Risk Score = i = 1∑n (gene expression level × regression coefficient); the regression coefficient is 0.00413,

[0058] Among them, the genes include DKK4 and 11 immune regulatory genes significantly associated with it.

[0059] In some embodiments, in step (4), the immunohistochemical validation includes scoring the DKK4 expression levels in normal colorectal tissue, adenoma with low / high grade intraepithelial neoplasia, and invasive cancer tissue, and the scoring criteria are:

[0060] The product of the positive staining rate (≤1% for 0 points, 2%-25% for 1 point, 26%-50% for 2 points, 51%-75% for 3 points, >75% for 4 points) and the staining intensity (colorless for 0 points, light yellow for 1 point, yellow for 2 points, brownish yellow for 3 points) is ≥5 points, indicating high risk.

[0061] The second aspect of the embodiment of the present application provides a colorectal tumor risk prediction model constructed based on the method, which includes expression data of the DKK4 gene and its related immune regulatory genes, LASSO regression coefficient and risk score threshold, and is used to predict the survival prognosis of colorectal cancer patients within 3 years.

[0062] In some embodiments, the model integrates patient age, AJCC stage, and risk score through a nomogram to generate personalized survival probability prediction results.

[0063] A third aspect of the embodiments of the present application provides a colorectal tumor risk prediction kit based on the DKK4 gene, comprising primers, antibodies or probes for detecting DKK4 gene expression, and an algorithm and threshold standard for a risk scoring model.

[0064] In some embodiments, the antibody is a rabbit DKK4 monoclonal antibody (Proteintech, Cat. No. 27080-1-AP), which is suitable for immunohistochemical detection of DKK4 expression levels in colorectal tissue samples.

[0065] The fourth aspect of the embodiments of the present application provides an application of a risk prediction model in the early diagnosis of colorectal cancer, by detecting the expression level of DKK4 in patient tissue samples and combining the risk scoring model to judge the risk of adenoma transforming into cancer.

[0066] The fifth aspect of the embodiment of the present application provides an application of a risk prediction model in predicting the efficacy of immunotherapy for colorectal cancer. The model evaluates the patient's sensitivity to immune checkpoint inhibitors by analyzing the correlation between DKK4 and tumor mutation burden (TMB), macrophage polarization and PD-1 / PD-L1 expression.

[0067] Example 1

[0068] The present invention provides a method for constructing a colorectal tumor risk prediction model based on the DKK4 gene, comprising the following steps:

[0069] (1) The differentially expressed genes between colorectal precancerous lesions and healthy subjects were screened through public databases, and the DKK4 gene was identified as a candidate target. Public databases included the GSE41657 and GSE37364 datasets of the NCBI-GEO database. The screening criteria were a fold difference (FC) ≥ 2 and significance (P < 0.05).

[0070] (2) Using single gene functional analysis and combining with the TCGA database, evaluate the expression level, biological function, and association of DKK4 with the Wnt signaling pathway in colorectal cancer and precancerous lesions;

[0071] (3) LASSO regression analysis was used to construct a colorectal cancer risk score model using DKK4 and its related genes, and 12 candidate genes including DKK4 were screened as risk factors. LASSO regression analysis used ten-fold cross validation and the risk score formula was determined by optimizing the lambda value. The risk score formula is:

[0072] Risk Score = ∑i = 1n(gene expression level × regression coefficient)

[0073] Among them, the genes include DKK4 and 11 immune regulatory genes significantly associated with it.

[0074] (4) The predictive efficacy of the risk score model for 1-year and 3-year survival in colorectal cancer patients was validated using an independent data set, and the differential expression of DKK4 in colorectal adenomas and cancer tissues was verified by immunohistochemistry. The immunohistochemical validation included scoring the DKK4 expression levels in normal colorectal tissue, adenoma with low / high-grade intraepithelial neoplasia, and invasive cancer tissues, using the following scoring criteria:

[0075] The product of the positive staining rate (≤1% for 0 points, 2%-25% for 1 point, 26%-50% for 2 points, 51%-75% for 3 points, >75% for 4 points) and the staining intensity (colorless for 0 points, light yellow for 1 point, yellow for 2 points, brownish yellow for 3 points) is ≥5 points, indicating high risk.

[0076] Example 2

[0077] This method-based method constructs a colorectal cancer risk prediction model. The model incorporates expression data for the DKK4 gene and related immune regulatory genes, LASSO regression coefficients, and risk score thresholds to predict the three-year survival prognosis of colorectal cancer patients. The model integrates patient age, AJCC stage, and risk score via a nomogram to generate personalized survival probability predictions.

[0078] Example 3

[0079] The present invention provides a colorectal tumor risk prediction kit based on the DKK4 gene, comprising primers, antibodies or probes for detecting DKK4 gene expression, and an algorithm and threshold standard for a risk scoring model.

[0080] The antibody is a rabbit DKK4 monoclonal antibody (Proteintech, Cat. No. 27080-1-AP), which is suitable for immunohistochemical detection of DKK4 expression levels in colorectal tissue samples.

[0081] Example 4

[0082] The present invention provides a risk prediction model for the early diagnosis of colorectal cancer. The model detects the expression level of DKK4 in patient tissue samples and combines it with a risk scoring model to determine the risk of adenoma transforming into cancer.

[0083] Example 5

[0084] The present invention provides an application of a risk prediction model in predicting the efficacy of immunotherapy for colorectal cancer. The model evaluates the patient's sensitivity to immune checkpoint inhibitors by analyzing the correlation between DKK4 and tumor mutation burden (TMB), macrophage polarization, and PD-1 / PD-L1 expression.

[0085] Example 6

[0086] Materials and Methods

[0087] 1. Experimental Materials

[0088] 1.1 Main Reagents The main reagents and consumables are shown in Table 1:

[0089] Table 1

[0090]

[0091] 1.2 Main instruments and equipment The main instruments and equipment are shown in Table 2:

[0092] Table 2

[0093]

[0094] 2. Research subjects

[0095] 2.1 Online Datasets

[0096] The original data of the dataset were downloaded from the NCBI-GEO database: (1) GSE41657, 12 cases of normal intestinal tissue, 21 cases of low-grade intraepithelial neoplasia, 30 cases of high-grade intraepithelial neoplasia, and 25 cases of invasive colorectal cancer. (2) GSE37364, 16 cases of low-grade intraepithelial neoplasia, 13 cases of high-grade intraepithelial neoplasia, and 14 cases of invasive colorectal cancer.

[0097] 2.2 In vitro cell lines

[0098] The human primary colon cancer cell line HCT116 and the human in situ rectal adenocarcinoma cell line SW480 were purchased from the Cell Bank of Type Culture Collection Committee of the Chinese Academy of Sciences.

[0099] 2.3 Patients

[0100] Patients undergoing colonoscopy at Ningxia Medical University General Hospital between May and December 2024 were enrolled. All examinations were performed by attending physicians with at least 5 years of colonoscopy experience or higher. This study was approved by the Ethics Committee of Ningxia Medical University General Hospital (KYLL-2023-1429), and all patients provided written informed consent before the procedure.

[0101] 3 Diagnostic criteria and data collection for colorectal cancer

[0102] 3.1 Diagnostic criteria:

[0103] Pathological diagnosis is based on the diagnostic criteria for digestive system tumors and is divided into adenoma and adenocarcinoma. Adenoma is divided into low-grade intraepithelial neoplasia (LGIN) and high-grade intraepithelial neoplasia (HGIN) based on the atypia of the tumor cells.

[0104] 3.2 Inclusion criteria for colorectal adenoma

[0105] ① Patients with polyps found by colonoscopy and confirmed by pathological examination to be colorectal adenoma with low-grade intraepithelial neoplasia or colorectal adenoma with high-grade intraepithelial neoplasia;

[0106] ②Complete clinical case data

[0107] 3.3 Inclusion criteria for colorectal cancer

[0108] ① Patients with colorectal cancer diagnosed by colonoscopy and confirmed by pathological examination as advanced colorectal cancer;

[0109] ②The clinical case data are complete.

[0110] 3.4 Inclusion criteria for normal colorectal epithelial tissue

[0111] ① No abnormalities of the colorectal mucosa were found in colonoscopy;

[0112] ② No other digestive system diseases such as inflammatory bowel disease, celiac disease, intestinal infection, chronic pancreatitis, portal hypertension, etc.

[0113] 3.5 Exclusion criteria

[0114] ① Those who are unable to cooperate due to poor cardiopulmonary function or other reasons;

[0115] ②Preoperative anti-tumor treatment such as chemotherapy, radiotherapy or immunotherapy;

[0116] ③ Unable to cooperate in completing all laboratory tests;

[0117] ④ Those with obvious abnormalities in the digestive tract as shown by endoscopy;

[0118] ⑤ Patients with a history of colorectal adenoma resection or digestive system tumors;

[0119] ⑥Boston scale assessment intestinal score <6 points.

[0120] 3.6 Specimen Collection

[0121] Two specimens each of normal colorectal epithelial tissue, adenoma (including low-grade intraepithelial neoplasia and high-grade intraepithelial neoplasia), and colorectal cancer tissue were collected from the enrolled patients: one tissue specimen was used for pathological examination after H&E staining, and the other was reserved for molecular biology experiments.

[0122] 3.7 Pathological examination, diagnosis and grouping

[0123] All patients' tissue specimens were routinely stained with hematoxylin and eosin and then pathologically examined. Based on the pathological diagnosis, the patients were divided into normal colorectal epithelial tissue group, low-grade intraepithelial neoplasia group, high-grade intraepithelial neoplasia group, and invasive colorectal cancer group.

[0124] 4 Research Methods

[0125] 4.1 Data mining of public databases of colorectal cancer transcriptome data

[0126] The NCBI-PUBMED database was used to retrieve relevant literature on sequencing screening of differentially expressed mRNA between colorectal cancer and adjacent tissues, and between healthy subjects and patients. Sequencing data were downloaded from the NCBI-GEO database. After retrieval, the datasets used for data mining included GSE41657 and GSE37364.

[0127] After quality control of all sequencing results, differential analysis was performed using edgeR. A fold change (FC) ≥ 2 and significance (P < 0.05) were set as the primary screening criteria for differentially expressed genes; biological function and immune infiltration, migration, proliferation, and invasion were used as secondary screening criteria.

[0128] 4.2 Public database data mining of candidate target DKK4

[0129] Data for relative expression analysis of DKK4 were collected using the TCGA database (https: / / portal.gdc.cancer.gov). Cluster heatmaps, receiver operating characteristic (ROC) curves, and survival analysis images were generated using the Lianchuan Bio-Cloud Platform (https: / / www.omicstudio.cn). Genetic analysis of biological functions of genes related to DKK4 was performed using the DAVID database (https: / / david.ncifcrf.gov / ). Cluster analysis of related signaling pathways was performed using the KEGG database. Single-gene enrichment analysis of target genes was performed using the GSEA database (https: / / www.gsea-msigdb.org / ), and immune pathway correlation analysis was performed using the GSVA (gene set variation analysis) method. Immune infiltration analysis of target genes was performed using the CIBERSORT data package. Univariate and multivariate Cox regression analyses of target genes were performed using edgeR using TCGA clinical data as input. After screening for independent risk genes, a nomogram was constructed to predict patient survival. The GSE39582 dataset was used as the validation set to analyze the predictive performance of the risk model.

[0130] Example 7

[0131] DKK4 small interfering RNA construction

[0132] 1. Cell culture and passaging

[0133] Human colon cancer primary lesion cell line HCT116 and human in situ rectal adenocarcinoma cell line SW480 were removed from a -80°C freezer. Thawed rapidly in a 37°C water incubator, transferred to a 15mL centrifuge tube, and 2mL of complete culture medium was added. Centrifuged at 1000 rpm for 5 minutes. The supernatant was discarded and the cells were resuspended in fresh culture medium. The cell suspension was transferred to labeled culture flasks and 5mL of complete culture medium was added. The cells were incubated in a cell culture incubator at 37°C, 5% CO2. When the adherent cells had proliferated to approximately 80-90% of their number, the original culture medium was discarded, and the cells were trypsinized. The culture flasks were placed in a cell culture incubator and incubated to detach the adherent cells. After neutralization, 2mL of culture medium was added, the cells were transferred to a 15mL centrifuge tube, centrifuged at 1000 rpm for 5 minutes, and resuspended in 2mL of complete culture medium. The cells were then passaged and added to new culture flasks at a ratio of 1:2 to 1:3. 5mL of complete DMEM was added to the flasks, shaken, and incubated in a cell culture incubator.

[0134] 2. DKK4 siRNA transfection

[0135] DKK4 small interfering RNA was customized by Shanghai Genema Co., Ltd.

[0136] (1) Cell inoculation: Before transfection, cells were plated into a six-well cell culture plate. The next day, when the cells reached approximately 30% confluence, transfection experiments were performed.

[0137] (2) Transfection of siRNA: Use the following transfection system in a six-well plate. Replace the culture medium in the wells with fresh complete culture medium before transfection.

[0138] The transfection system preparation is shown in Table 3:

[0139] Table 3

[0140]

[0141] The specific steps are as follows:

[0142] Dissolve the small interfering RNA and RNAi Max separately in Opti-MEM medium. Mix gently and let stand at room temperature for 30 minutes. Add the mixed transfection solution to the wells and shake the plate to evenly distribute the transfection solution. Continue incubating in the cell culture incubator for 24 hours before replacing with fresh medium. 48 hours after transfection, evaluate the siRNA knockdown efficiency as needed and proceed to the next step.

[0143] Example 8

[0144] Bulk RNA-seq analysis of in vitro cultured cell lines

[0145] 1 RNA extraction and library construction

[0146] RNA was prepared from cells using the Trizol method. The purity and concentration of total RNA were controlled using a NanoDrop ND-2000. Total RNA was required to meet the following conditions: (1) RNA concentration > 50 ng / mL; (2) RIN value > 7.0; (3) 260 / 280 > 1.8; (4) total RNA volume > 1 μg. Oligo(dT) magnetic beads were used to screen and obtain mRNA with PolyA in total RNA. Finally, a 300 bp ± 50 bp mRNA library was obtained. Illumina Novaseq TM Paired-end sequencing was performed using 6000 DNA sequencing in PE150 mode.

[0147] 2 Bioinformatics analysis of sequencing results

[0148] The raw data were downloaded in fastq format and quality-controlled using fastp (https: / / github.com / OpenGene / fastp) software. Sequencing data were aligned to the genome (Homo sapiens, GRCh38) using HISAT2 (https: / / ccb.jhu.edu / software / hisat2) to generate bam files. Transcript data were assembled using StringTie software (https: / / ccb.jhu.edu / software / hisat2), and mRNA was quantified using FPKM (FPKM = total_exon_fragments / mapped_reads (millions) × exon_length (kB)). Differentially expressed genes between sample groups were analyzed using edgeR (https: / / bioconductor.org / packages / release / bioc / html / edgeR.html). Significant differences were defined when the fold-change factor (FC) was >2-fold or <0.5-fold and the p value was <0.05. Finally, DAVID software (https: / / david.ncifcrf.gov / ) was used to perform GO and KEGG enrichment analysis on the genes.

[0149] Example 9

[0150] Hematoxylin-eosin staining of normal colorectal epithelium, colorectal adenoma, and colorectal cancer tissue sections

[0151] HE staining is a classic histological method for evaluating colorectal tumor pathology. This study referred to the Bancroft histological staining technical specifications and performed appropriate optimization.

[0152] (1) Dewaxing: The sections were placed in environmentally friendly dewaxing solution I (20 min, 37°C) and environmentally friendly dewaxing solution II (20 min, 37°C) for dewaxing treatment, followed by gradient hydration with anhydrous ethanol I (5 min), anhydrous ethanol II (5 min), and 75% ethanol (5 min), and finally rinsed with deionized water.

[0153] (2) Section pretreatment: Place the dewaxed tissue sections in a constant staining pretreatment solution (pH 7.4) and incubate for 1 min to enhance the staining effect.

[0154] (3) Hematoxylin staining: Stain with modified Harris hematoxylin solution (Sigma-Aldrich) for 3-5 min, differentiate with differentiation solution for 5 s, rinse with deionized water, and stain with 1% ammonia solution for 30 s, and rinse with deionized water.

[0155] (4) Eosin staining: Dehydrate the sections in 95% ethanol for 1 min and stain with eosin for 15 s.

[0156] (5) Dehydration and sealing: The sections were placed in anhydrous ethanol I (2 min), anhydrous ethanol II (2 min), and anhydrous ethanol III (2 min) for dehydration, n-butanol I (2 min) and n-butanol II (2 min) for transparency, and xylene I (2 min) and xylene II (2 min) for treatment. Finally, the sections were sealed with neutral gum.

[0157] (6) The whole slide was scanned using the Pannoramic 250Flash digital slide scanning system (3DHISTECH, Hungary). The staining results showed that the cell nucleus was blue and the cytoplasm was red.

[0158] 4.6 Immunohistochemistry of colorectal normal epithelium, colorectal adenoma, and colorectal cancer tissue sections

[0159] The improved immunohistochemical staining technique, combined with microwave antigen retrieval and DAB color development system, was used to localize and quantify DKK4.

[0160] (1) Dewaxing of sections: The sections were sequentially placed in dewaxing solution I (10 min, 60°C), dewaxing solution II (10 min, 60°C), and dewaxing solution III (10 min, 60°C) to achieve the dewaxing effect. Subsequently, the sections were sequentially placed in anhydrous ethanol I (5 min), anhydrous ethanol II (5 min), and anhydrous ethanol III (5 min) for treatment, and rinsed with deionized water three times.

[0161] (2) Antigen retrieval: Microwave retrieval was performed using EDTA buffer with the following parameters: medium heat for 5 min, off heat for 5 min, and low heat for 5 min. After retrieval, the sections were placed in PBS and washed three times (5 min each time) on a shaker.

[0162] (3) Blocking treatment: The sections were placed in hydrogen peroxide at room temperature in the dark for 25 minutes, then washed three times in PBS on a shaker (5 minutes each time), and 3% BSA was added dropwise to the tissue area on the sections for blocking for 30 minutes;

[0163] (4) Antibody incubation: Discard the blocking solution and add the primary antibody, DDK4 (Proteintech, Cat No. 27080-1-AP, rabbit source, 1:100 dilution). Place the slices flat in a humidified chamber and incubate overnight at 4°C. After washing with PBS, add HRP-labeled goat anti-rabbit IgG secondary antibody (1:200) to the tissue area of ​​the slices and incubate at room temperature for 50 min.

[0164] (5) Color development: After washing the sections with PBS, add DAB color development solution and control the color development time under a microscope (usually 3 minutes). When the positive signal appears brownish yellow, rinse with running water to terminate the reaction.

[0165] (6) Counterstaining and sealing: Use hematoxylin to counterstain for about 3 minutes to achieve the effect of cell nucleus staining, rinse with tap water, differentiate with differentiation solution for 5-10 seconds, rinse with tap water, and then use hematoxylin bluing solution to bluing and rinse with running water; then put the slices into 75% alcohol (5 minutes), 85% alcohol (5 minutes), anhydrous ethanol I (5 minutes), anhydrous ethanol II (5 minutes), n-butanol (5 minutes), and xylene (5 minutes) in turn, dry and seal with neutral gum.

[0166] (7) The whole slide was scanned using a Pannoramic 250Flash digital slide scanner. The cell nuclei appeared blue and DKK4 positive expression appeared brown.

[0167] (8) Immunohistochemistry scoring criteria: Staining positivity score: ≤1% 0 point, 2%-25% 1 point, 26%-50% 2 points, 51%-75% 3 points, >75% 4 points; Staining intensity score: colorless 0 point, light yellow 1 point, yellow 2 points, brownish yellow 3 points; Staining positivity rate × staining intensity is the total score: 0 point negative (-), 1-4 points weakly positive (+), 5-8 points positive (++), 9-12 points strongly positive (+++).

[0168] Statistical analysis

[0169] Statistical analysis and histograms were performed using SPSS 25.0 and GraphPad Prism 9.0 software. Data that met normal distribution were further analyzed. Statistical comparisons between two groups were performed using the t-test, and differences between multiple groups were analyzed using one-way analysis of variance. If data did not meet normal distribution, the rank-sum test was used for analysis. Data are presented as mean ± standard deviation, and P < 0.05 was considered statistically significant.

[0170] Test Example 1

[0171] 1. Screening of differentially expressed mRNAs in colorectal adenoma with intraepithelial neoplasia using a public dataset

[0172] The GSE37364 dataset was identified by searching the NCBI-GEO database. The dataset was divided into 38 normal colorectal epithelial samples, 13 low-grade intraepithelial neoplasia samples, and 16 high-grade intraepithelial neoplasia samples. After regrouping, high-grade intraepithelial neoplasia was used as the target group, and other samples were analyzed as the control group. Using FC>2 and p<0.05 as the initial screening conditions, a total of 1752 differentially expressed genes were screened, and the gene functions were positioned as tumor-related, proliferation, migration, differentiation, and immunity. A cluster heat map of the TOP100 differentially expressed mRNAs was obtained, and the TOP10 mRNAs with significant differences were annotated in the volcano map ( Figure 1 A, B). Combined with literature research, the DKK4 gene was finally identified as a candidate target.

[0173] Figure 1 This is a data analysis diagram of differentially expressed mRNA in colorectal adenoma with intraepithelial neoplasia of the present invention.

[0174] Figure 1 A is the cluster heat map analysis of differentially expressed mRNA TOP100 in colorectal adenoma with intraepithelial neoplasia of the present invention (selected differences); Figure 1 B is a scatter plot analysis of differentially expressed mRNA in colorectal adenoma with intraepithelial neoplasia of the present invention. The gene names marked in red are the top 5 upregulated genes, and the gene names marked in blue are the top 5 downregulated genes.

[0175] 2. Analysis of DKK4 Single Gene Biological Function

[0176] Single gene functional analysis was used to preliminarily explore the correlation between DKK4 and colorectal tumors. 275 patients with colon cancer and 349 controls, 92 patients with rectal cancer and 318 controls were collected from the TCGA database. The expression differences of DKK4 were analyzed by bar graph. The results showed that compared with the adjacent adjacent controls, DKK4 was significantly overexpressed in both colon cancer patients and rectal cancer patients (p < 0.05) ( Figure 2A). The correlation analysis between DKK4 key genes and clinical characteristics showed that the differential expression of DKK4 was significantly correlated with the tumor mutation burden (TMB) of colorectal cancer patients (p = 0.008) (Table 4). The relative expression of DKK4 gene was divided into high-level group and low-level group according to the median group. The differential gene enrichment analysis was performed and the cluster diagram and volcano diagram were drawn using TOP15 up / down. Figure 2 B). Using 30 DKK4-related genes as input data, DKK4 single gene GSEA enrichment analysis and KEGG signaling pathway analysis were performed. The results showed that the main function of DKK4 was significantly enriched in the wnt signaling pathway ( Figure 2 C, D). The above results suggest that DKK4 is significantly associated with abnormal tumor proliferation and infiltration, and its differential expression may affect the efficacy of tumor immunotherapy.

[0177] Figure 2 This is a single gene analysis diagram of DKK4 of the present invention. Figure 2 A is the differential expression of DKK4 between colon cancer / rectal cancer patients and adjacent tissues analyzed by TCGA data of the present invention; Figure 2 B is a cluster diagram of the TOP15 up-regulated genes and TOP15 down-regulated genes that are differentially enriched between the DKK4 high-level and low-level groups of the present invention, where red indicates relatively high mRNA expression and blue indicates relatively low mRNA expression; Figure 2 C is the DKK4 single gene GSEA enrichment analysis diagram of the present invention; Figure 2 D represents the regulatory effect of DKK4 of the present invention on key molecules of the wnt signaling pathway.

[0178] The correlation analysis between DKK4 key genes and clinical characteristics of colorectal cancer is shown in Table 4:

[0179] Table 4

[0180]

[0181]

[0182] 3. Correlation between DKK4 and immune cells

[0183] The correlation between DKK4 and immune cells was analyzed by calling TCGA data through GSVA and Cibersort. By calling the data set, the expression of DKK4 in different immune cells in different immune cell data sets was analyzed to evaluate its correlation with immune cells ( Figure 3 A). The expression distribution of DKK4 in 22 immune cells was analyzed by Cibersort ( Figure 3B). Combining expression abundance and correlation, we plotted the correlation of DKK4 in different immune cell interactions. The results showed that the immune regulatory function of DKK4 is mainly concentrated in the interaction between macrophages, T cells, DC cells, NK cells, and B cells ( Figure 3 C). The differential expression of DKK4 in different immune cells was analyzed by grouping high / low DKK4 expression levels. The results showed that the differential expression of DKK4 was mainly enriched in CD4 T cells, NK cells, and macrophages ( Figure 3 D) These results suggest that DKK4 plays an important role in tumor immune regulation. Figure 3 This is a diagram showing the correlation between DKK4 and immune cells of the present invention. Figure 3 A is the correlation analysis of DKK4 expression in immune cell-related datasets in the TCGA data of the present invention; Figure 3 B is the Cibersort analysis of the present invention showing the abundance distribution of DKK4 in 22 immune cells; Figure 3 C is the correlation analysis of DKK4 in immune cell interactions of the present invention; Figure 3 D is a box plot of the differential analysis of the DKK4 gene in 22 immune cells of the present invention.

[0184] 4. Correlation analysis between DKK4 and immune regulatory genes

[0185] The correlation of DKK4 tumor immune pathway gene expression profile was further analyzed by Cibersort, and box plots were drawn with DKK4 high expression level and low expression level as groups. The results showed that the differences in DKK4 differential expression regulation were significantly enriched in macrophages (MMP9, TM4SF19, CD68, CYBB) and neutrophils (HSD17B11, EVI2B, MNDA) ( Figure 4 A, Figure 4 B). Based on immunological function analysis, the differentially expressed genes between the high and low DKK4 levels in colorectal cancer patients were mainly clustered in immunological processes such as antigen presentation, cytolytic activity, type I interferon, and type II interferon response ( Figure 4 C). After focusing on immune-promoting regulatory genes, the differential expression of DKK4 can affect immune-promoting genes such as CD80, ENTPD1, ​​IFNA2, IFNG, and PRF1 ( Figure 5 A, Figure 5 B). The differential expression of DKK4 in the regulation of immunosuppression-related genes was mainly enriched in ARG1, CD274, EDNRB, and VEGFB ( Figure 5 C) These results indicate that DKK4 can play a role in multiple processes, including immune activation and immunosuppression. Its role in the development and progression of colorectal cancer needs further exploration.

[0186] Figure 4 This is a differential expression diagram of the tumor immune pathway genes of the present invention in the DKK4 high / low level groups. Figure 4 A is a boxplot of the differential expression of tumor immunity-related genes in B cells, CD4+T cells, CD8+T cells, and DC cells in the DKK4 high / low level groups of the present invention; Figure 4 B is a boxplot of the differential expression of macrophage, neutrophil, and NK cell tumor immunity-related genes in the DKK4 high / low level groups of the present invention; Figure 4 C is a box plot of the differential expression of the antigen presentation, cytolytic activity, type I interferon and type II interferon response tumor immunity-related genes in the DKK4 high / low level groups of the present invention.

[0187] Figure 5 This is a diagram showing the differential expression of tumor immunity promoting genes / suppressing genes in the DKK4 high / low level groups of the present invention. Figure 5 A- Figure 5 B is a boxplot of differential expression of tumor immunity promotion-related genes in the DKK4 high / low level groups of the present invention; Figure 5 C is a box plot of the differential expression of tumor immunosuppression-related genes in the DKK4 high / low level groups of the present invention.

[0188] 5. Prediction of the correlation between DKK4 and CRC survival and prognosis based on the lasso model

[0189] Through the TCGA database, we retrieved the CRC data set and used the transcriptome data of 603 tumor samples with survival information to train the lasso regression model. Using ten-fold cross validation, we calculated the variation of the characteristic coefficient with λ and obtained the λ value ( Figure 6 A, B). Based on the results of LASSO regression analysis, the optimal lambda value of 0.00413 was selected and 12 candidate genes under this coefficient were obtained, including DKK4. The risk score (riskscore) was calculated based on the characteristic coefficient and the 12 candidate genes obtained by screening. The risk factor linkage diagram (ggrisk) was used to analyze the survival status of the enrolled population as the risk score increased. The results showed that as the risk score increased, the overall survival rate of colorectal cancer patients decreased and the number of deaths increased ( Figure 6 C, D). Univariate and multivariate regression analysis was performed on the clinical information of the enrolled population, including age, gender, TNM stage, AJCC grade, vascular and lymph node infiltration, and risk score. The results showed that AJCC grade and risk score can serve as independent risk factors affecting survival ( Figure 1-7 A, B). That is, 12 candidate genes including DKK4 can be used to construct a risk model for colorectal cancer survival prognosis.

[0190] Figure 6This is a LASSO regression model diagram of DKK4 of the present invention. Figure 6 A is a ten-fold cross-validation graph of 603 colorectal cancer transcriptome sequencing data containing survival results in TCGA of the present invention; Figure 6 B is a graph showing the changes in characteristic coefficients of differentially expressed genes in the transcriptome data of the present invention with λ, and screening of candidate genes for risk assessment. Figure 6 C- Figure 6 D is a diagram of the present invention that uses risk scores and patient survival status to draw a risk factor linkage diagram (ggrisk) to evaluate the relationship between risk scores and patient survival rates.

[0191] Figure 7 1 is a univariate and multivariate COX regression analysis diagram of the colorectal cancer patient risk score and clinical information of the present invention. Figure 7 A is a univariate COX regression analysis of the impact of age, gender, TNM stage, AJCC grade, vascularity, lymph node infiltration data and risk score of the colorectal cancer patient population on patient survival status; Figure 7 B is a multivariate COX regression analysis of the impact of age, gender, TNM stage, AJCC grade, blood vessels, lymph node infiltration data and risk score of the colorectal cancer patients enrolled in the present invention on the survival status of patients.

[0192] Test Example 2

[0193] Construction of a risk prediction model for the survival prognosis of colorectal cancer patients based on DKK4

[0194] Using the median risk score as the dividing line, the survival curves of 603 patients and the ROC curves for 1-year, 3-year, and 5-year survival predictions were drawn. The survival curves showed that the survival rate of the high-risk group was significantly lower than that of the low-risk score group (p < 0.001) ( Figure 8 A). The AUC curve showed that the risk score had a relatively good predictive effect on the 1-year (AUC = 0.676), 3-year (AUC = 0.695), and 5-year (AUC = 0.6577) survival of colorectal cancer patients ( Figure 8 B, Figure 8 C, Figure 8 D). Using the independent risk factors screened out in univariate and multivariate COX regression, a nomogram was drawn in combination with the patient's age, and a calibration curve was drawn for the patient's survival using the observation model. The results showed that the model had a good prediction effect on the patient's 1- and 3-year survival, but a poor prediction effect on the 5-year survival ( Figure 8 E, Figure 8 F) These results suggest that the colorectal cancer risk prediction model based on DKK4 can be used to predict patient survival within 3 years, but is not suitable for long-term survival prediction.

[0195] Figure 8This is a diagram of the survival prognosis risk prediction model for colorectal cancer patients based on the DKK4 gene of the present invention. Figure 8 A is a survival curve of 603 colorectal cancer patients grouped by risk score of the present invention; Figure 8 B- Figure 8 D is the AUC curve of the risk score of the present invention for predicting 1-, 3-, and 5-year survival of colorectal cancer patients; Figure 8 E is a nomogram constructed by using risk scores and colorectal cancer patient data of the present invention; Figure 8 F is a calibration curve diagram drawn by calibrating the nomogram through 1-, 3-, and 5-year survival analysis of the present invention.

[0196] Test Example 3

[0197] Establishment and validation of a DKK4-based survival prognostic risk prediction model for colorectal cancer patients

[0198] We randomly searched for datasets containing transcriptome data of colorectal cancer patients in the database and selected GSE39582 to validate the risk prediction model. We used the risk score to draw a risk factor linkage diagram in GSE39582. The results showed that as the risk score increased, the overall survival rate decreased and the number of deaths increased ( Figure 9 A, Figure 9 B). Based on the risk factor cutoff point grouping of the risk model, the survival curve was drawn, showing that the survival rate of the high-risk group was significantly lower than that of the low-risk group (p = 0.0045) ( Figure 10 A). The AUC curve showed that the risk model had a good effect on predicting 1-year survival (AUC = 0.634), but had a poor effect on predicting 3-year (AUC = 0.596) and 5-year (AUC = 0.561) long-term survival ( Figure 10 B, Figure 10 C, Figure 10 D) Validation using a nomogram and calibration curve revealed that the risk prediction model had good predictive efficacy for patients in the validation set within 1 and 3 years, but poor prediction for patients in the 5-year period. These results further demonstrate that the DKK4-based risk prediction model has a good predictive efficacy for colorectal cancer patients within 3 years.

[0199] Figure 9 This is a risk factor linkage diagram of the risk prediction model of the present invention verified in the GSE39582 dataset. Figure 10 This is a prediction validation diagram of the risk prediction model of the present invention for the survival prognosis of patients in the GSE39582 dataset. Figure 10 A is a survival curve of the survival of colorectal cancer patients in the GSE39582 dataset based on the risk score grouping of the present invention; Figure 10 B- Figure 10 D is the AUC curve of the risk score of the present invention for predicting 1-, 3-, and 5-year survival of colorectal cancer patients; Figure 10 E is a nomogram constructed by using the risk score and GSE39582 dataset to collect colorectal cancer patient data; Figure 10 F is a calibration curve diagram drawn by calibrating the nomogram through 1-, 3-, and 5-year survival analysis of the present invention.

[0200] Test Example 4

[0201] Correlation analysis between DKK4 expression and pathology of colorectal adenoma and adenocarcinoma

[0202] Based on the inclusion and exclusion criteria, the pathological examination results and sections of the patients were collected, and 80 patients were finally included. Based on the pathological diagnosis, a total of 19 cases of normal colorectal epithelial tissue, 31 cases of adenoma with low-grade intraepithelial neoplasia, 9 cases of adenoma with high-grade intraepithelial neoplasia, and 21 cases of invasive colorectal cancer were collected. The results showed that the epithelium of the mucosal layer of normal intestinal tissue is composed of a single layer of columnar epithelial cells and goblet cells. There are a large number of straight tubular and densely arranged intestinal glands in the lamina propria of the intestinal tissue. Immunohistochemical staining showed that DKK4 is distributed in small amounts in the mucosal epithelium and the edge of the intestinal glands ( Figure 11 A, Figure 11 C). Adenoma with low-grade intraepithelial neoplasia tissue shows dark staining, increased number of glands and goblet cells, a small amount of basophils in the glandular cavity, extensive edema, loosely arranged glands, and a large number of scattered lymphocytes. Immunohistochemical staining shows that DKK4 is diffusely distributed and significantly increased in the glands with the scattered lymphocytes ( Figure 11 B, Figure 11 D). Adenoma with high-grade intraepithelial neoplasia tissue showed strong basophilia of tumor cells, high nuclear-cytoplasmic ratio, long spindle-shaped nuclei, densely arranged cells in a glandular duct-like pattern, exfoliated tumor cells and basophils in the glandular cavity, and a large number of scattered lymphocytes. Immunohistochemistry results showed that DKK4 expression was significantly increased and mainly concentrated in the glands ( Figure 11 E, Figure 11 G). In invasive carcinoma tissue, tumor cells are highly basophilic, with a high nuclear-cytoplasmic ratio and nuclear atypia, arranged in a glandular duct pattern. Necrotic cell fragments are visible in the glandular cavity, with a large number of scattered lymphocytes and a small amount of granulocyte infiltration. Immunohistochemistry results show that DKK4 expression and distribution are significantly reduced ( Figure 11 F, Figure 11H). Statistical analysis based on immunohistochemical scores showed that DKK4 expression in adenomas (with low-grade intraepithelial neoplasia and with high-grade intraepithelial neoplasia) was significantly different from that in normal tissue and invasive carcinoma tissue (p < 0.05). DKK4 expression in adenomas with high-grade intraepithelial neoplasia was significantly different from that in the other three groups (p < 0.05) (Tables 1-5). These results suggest that DKK4 expression may be significantly associated with the carcinogenesis of adenomas. It is primarily expressed in colorectal glandular cells, and its expression may be related to the function of intestinal glandular duct-like cells. In invasive carcinoma, as glandular structure is lost and the number of glands decreases, DKK4 expression is significantly reduced. The expression of DKK4 in different tissues is shown in Table 5.

[0203] Figure 11 This is a diagram showing the expression of DKK4 in colorectal tumor tissues according to the present invention. Figure 11 A, Figure 11 C is HE staining and DKK4 immunohistochemical analysis of normal colorectal tissue of the present invention (5×, 50×); Figure 11 B, Figure 11 D is the adenoma with low-grade intraepithelial neoplasia of the present invention HE staining and DKK4 immunohistochemical analysis (5×, 50×); Figure 11 E, Figure 11 G is the HE staining and DKK4 immunohistochemical analysis of adenoma with high-grade intraepithelial neoplasia of the present invention (5×, 50×); Figure 11 F, Figure 11 H is the HE staining and DKK4 immunohistochemical analysis of colorectal cancer of the present invention (5×, 50×).

[0204] Colorectal cancer (CRC) is the digestive system malignancy with the highest morbidity and mortality worldwide. Notably, approximately 80% of CRC cases arise from the malignant transformation of colorectal adenoma (CRA), a process that typically follows a classic "adenoma-carcinoma" progression over 10-15 years. Colorectal adenoma is a benign tumor that develops in the colorectal mucosa and may progress to malignant colorectal cancer over time. As a common precancerous lesion, the development and progression of CRA is closely linked to multiple genetic and environmental factors, including mutations in genes such as APC and KRAS, as well as environmental influences such as diet and the gut microbiome. Recent studies have shown that epigenetic modifications play a significant role in the transformation of CRA to CRC. These lesions typically originate from glandular cells in the mucosa, where mutations due to genetic and environmental factors lead to uncontrolled cell proliferation. Identification and treatment of colorectal precancerous lesions are crucial for preventing the development of colorectal cancer. Regular screening programs, such as colonoscopies, can allow for early detection and removal of these lesions, significantly reducing the risk of cancer. Regular screening is recommended for high-risk patients. However, traditional diagnostic methods have limitations, such as reliance on endoscopist experience and high rates of missed lesions. Therefore, screening for specific markers from precancerous lesions can help identify high-risk individuals early and potentially improve the detection rate of colorectal precancerous lesions. Detecting molecular markers for these lesions is crucial for early diagnosis, prognostic assessment, and the development of personalized treatment strategies. Previous molecular marker screening efforts have primarily focused on differentially expressed genes between colorectal cancer and adjacent tissues, while neglecting precancerous lesions. Screening for specific markers from precancerous lesions could help improve the detection rate of colorectal precancerous lesions and enhance colorectal cancer prevention. In recent years, with the advancement of high-throughput sequencing technology, researchers have begun focusing on screening for specific markers from precancerous lesions. Compared with traditional screening for differentially expressed genes between colorectal cancer and adjacent tissues, the advantages of diagnostic methods based on precancerous lesion markers include: improving the detection rate of CRAs, enabling early prevention of CRC; elucidating the molecular mechanisms underlying CRC development; providing new targets for personalized treatment; and optimizing molecular subtyping of CRC. Therefore, the exploration and screening of colorectal tumor markers and related mechanisms are of great significance. For example, the characteristics of immune cell infiltration in the tumor microenvironment can predict the treatment response of CRC patients. These studies have provided new ideas for the precision treatment of CRC.

[0205] Public data mining, as an emerging bioinformatics research method, provides new perspectives and ideas for tumor research by systematically integrating and analyzing public bioinformatics databases. In the field of colorectal tumor research, although a large number of studies have focused on the screening of molecular markers for colorectal cancer (CRC), relatively little attention has been paid to colorectal precancerous lesions. Based on the GEO database, this study selected a GSE dataset containing samples of adenoma with high-grade intraepithelial neoplasia for in-depth analysis. Through differential expression analysis and functional enrichment analysis, we successfully screened DKK4 (Dickkopf-related protein 4) as a key regulatory factor in the transition from adenoma to cancer. DKK4 is a member of the Dickkopf family. Previous studies have reported that DKK4 plays a dual role in tumor development and progression: as a negative regulator of the Wnt signaling pathway, DKK4 inhibits the nuclear translocation of β-catenin by binding to the LRP5 / 6 receptor. In specific tumor microenvironments, DKK4 can promote epithelial-mesenchymal transition (EMT), enhancing the invasion and metastasis of tumor cells. This finding is consistent with the results of several recently published studies, further confirming the key role of DKK4 in the development and progression of colorectal tumors.

[0206] Research results have shown that DKK4 plays a crucial role in regulating the tumor immune microenvironment. DKK4 is significantly overexpressed in a variety of immune cells, including tumor-associated macrophages, dendritic cells, and regulatory T cells. This is consistent with previous reports, and this broad expression pattern suggests that DKK4 may play multiple roles in tumor immune regulation. Recent studies have shown that DKK4, through the Wnt / β-catenin signaling pathway, plays a key role in the dynamic balance between immune activation and immunosuppression. Specific mechanisms include: regulating TAM polarization, where DKK4 promotes M2 macrophage polarization and inhibits M1 macrophage activation, thereby forming an immunosuppressive microenvironment; affecting DC function, where DKK4 impairs anti-tumor immune responses by inhibiting DC maturation and antigen presentation; and regulating T cell differentiation, where DKK4 promotes Treg differentiation while suppressing both effector and cytotoxic T cell function. In colorectal tumors, elevated DKK4 expression is closely associated with the formation of an immunosuppressive microenvironment. Clinical data analysis shows that patients with high DKK4 expression have significantly reduced numbers of tumor-infiltrating lymphocytes and increased PD-1 / PD-L1 expression levels. Therefore, DKK4 may become a potential biomarker for predicting the efficacy of immunotherapy.

[0207] The application of LASSO regression in oncology has the following advantages: it effectively addresses the "curse of dimensionality" problem, is suitable for integrated analysis of multi-omics data, and provides clear guidance for the screening of potential candidate targets in medical research; it enables variable selection through L1 regularization, improving model interpretability; it ensures predictive accuracy while maintaining model simplicity; and it provides a quantitative tool for personalized treatment and prognostic assessment. In this study, a colorectal cancer risk prediction model was constructed using LASSO regression, focusing on DKK4 and combining it with 12 candidate genes significantly associated with it in the protein-protein interaction (PPI) network (correlation coefficient >0.8). Using 10-fold cross-validation and bootstrap resampling, we identified the most predictive feature variables and constructed a risk prediction model. Both training and validation models demonstrated that the DKK4-based risk prediction model had significant predictive efficacy for patients with short-term colorectal cancer within three years. These results are consistent with previous reports. Patients with high DKK4 expression have significantly higher risk scores than those with low DKK4 expression; DKK4 expression levels are positively correlated with tumor invasion depth; and patients with high DKK4 expression have a significantly increased rate of lymph node metastasis. Therefore, the DKK4-based risk prediction model not only provides a new tool for early diagnosis and prognostic assessment of CRC, but also provides a theoretical basis for the development of DKK4-targeted therapeutic strategies.

[0208] Analysis of patient pathological sections and DKK4 immunohistochemical stacks revealed that DKK4 expression in adenomas was higher than in normal colorectal epithelium and invasive colorectal cancer tissue. DKK4 expression was lowest in normal intestinal epithelium, further increased in adenomas, and decreased in invasive colorectal cancer. This unique expression pattern suggests that DKK4 may play a key role in the progression of colorectal precancerous lesions to invasive cancer. Previous reports have shown that DKK4 is differentially expressed in early-stage colorectal cancer tumors, and that differential expression of DKK4 regulates the Wnt signaling pathway, potentially contributing to tumor progression. The expression pattern and biological consequences of DKK4 in early-stage colorectal cancer require further detailed clinical and molecular studies. The specific mechanisms of DKK4's role in precancerous lesions remain to be elucidated. Future research will focus on clarifying the specific mechanisms by which DKK4 regulates the progression of precancerous lesions.

[0209] The basic principles, main features and advantages of the present invention are shown and described above. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and improvements may be made to the present invention without departing from the spirit and scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims, the description and their equivalents.

Claims

1. A method for constructing a colorectal tumor risk prediction model based on the DKK4 gene, characterized in that The steps include: (1) Screening for differentially expressed genes between colorectal precancerous lesions and healthy subjects using public databases, identifying the DKK4 gene as a candidate target; (2) Using single gene functional analysis and combining with the TCGA database, evaluate the expression level, biological function, and association of DKK4 with the Wnt signaling pathway in colorectal cancer and precancerous lesions; (3) LASSO regression analysis was used to construct a colorectal cancer risk score model based on DKK4 and its related genes, and 12 candidate genes including DKK4 were screened as risk factors; (4) An independent data set was used to verify the predictive efficacy of the risk score model for the 1-year and 3-year survival of colorectal cancer patients, and the expression differences of DKK4 in colorectal adenoma and cancer tissues were verified by immunohistochemistry.

2. The construction method according to claim 1, wherein: In step (1), the public database includes the GSE41657 and GSE37364 data sets of the NCBI-GEO database, and the screening conditions are that the fold difference (FC) is ≥2 and significant (P < 0.05).

3. The construction method according to claim 1, wherein: In step (3), the LASSO regression analysis uses ten-fold cross validation and determines the risk scoring formula by optimizing the lambda value. The risk scoring formula is: Risk Score = ∑i = 1n(gene expression level × regression coefficient) Risk Score = i = 1∑n(gene expression level × regression coefficient) ， The regression coefficient is 0.00413, Among them, the genes include DKK4 and 11 immune regulatory genes significantly associated with it.

4. The construction method according to claim 1, wherein: In step (4), the immunohistochemical validation includes scoring the DKK4 expression levels in normal colorectal tissue, adenoma with low / high grade intraepithelial neoplasia and invasive carcinoma tissue, and the scoring criteria are: The product of the positive staining rate (≤1% for 0 points, 2%-25% for 1 point, 26%-50% for 2 points, 51%-75% for 3 points, >75% for 4 points) and the staining intensity (colorless for 0 points, light yellow for 1 point, yellow for 2 points, brownish yellow for 3 points) is ≥5 points, indicating high risk.

5. A colorectal cancer risk prediction model constructed based on the method according to any one of claims 1 to 4, characterized in that: The model includes expression data of the DKK4 gene and its related immune regulatory genes, LASSO regression coefficient and risk score threshold, and is used to predict the survival prognosis of colorectal cancer patients within 3 years.

6. The risk prediction model according to claim 5, characterized in that: The model integrates patient age, AJCC stage, and risk score through a nomogram to generate individualized survival probability prediction results.

7. A colorectal tumor risk prediction kit based on the DKK4 gene, characterized by: Comprising primers, antibodies or probes for detecting DKK4 gene expression, and the algorithm and threshold standard of the risk scoring model according to claim 5.

8. The kit according to claim 7, wherein: The antibody is a rabbit DKK4 monoclonal antibody (Proteintech, catalog number 27080-1-AP), which is suitable for immunohistochemical detection of the expression level of DKK4 in colorectal tissue samples.

9. Use of the risk prediction model of claim 5 in the early diagnosis of colorectal cancer, characterized in that: The risk of adenoma to carcinoma transformation was determined by detecting the expression level of DKK4 in patient tissue samples and combining it with the risk scoring model.

10. Use of the risk prediction model of claim 5 in predicting the efficacy of immunotherapy for colorectal cancer, characterized in that: The model evaluates patients' sensitivity to immune checkpoint inhibitors by analyzing the correlation between DKK4 and tumor mutation burden (TMB), macrophage polarization, and PD-1 / PD-L1 expression.

Images