Application of combined STAT signaling pathway related genes in colorectal cancer prognosis model

Abstract

The application provides application of a combined STAT signal pathway related gene in a colorectal cancer prognosis model, the combined STAT signal pathway related gene is CAV1, EPO, IL13, LEP and NEUROD1; wherein the establishment method of the colorectal cancer prognosis model comprises the following steps: 1) data collection and arrangement, 2) screening of differentially expressed STAT signal pathway related genes, 3) construction of a prognosis model of the STAT signal pathway related genes, and 4) construction of a nomogram. The prognosis model has the advantages of high accuracy, can provide a new method for disease diagnosis and prognosis of colorectal cancer patients in the clinic, reveals the prediction effect on the tumor microenvironment, and provides important information for the development of targeted therapy.

Description

Technical Field

[0001] This invention belongs to the field of tumor molecular biology technology, specifically relating to the application of a combination of STAT signaling pathway-related genes in a colorectal cancer prognostic model. Background Technology

[0002] In 2020, colorectal cancer (CRC) was ranked as the third most common malignant tumor worldwide, with over 1.8 million new cases and over 900,000 deaths annually, posing a serious threat to human health. Over the past decade, rapid advancements in treatment technologies have offered hope for CRC treatment. Targeted therapy, leveraging genomics, transcriptomics, proteomics, and epigenomics data, is emerging as a new option, but it is still in its early stages compared to traditional treatments. Therefore, inventors urgently need new biomarkers and predictive models to assess patient prognosis.

[0003] In tumor cells and the tumor microenvironment, signal transduction and activating transcription factor (STAT) signaling are essential for tumorigenesis and development. STAT proteins are transcription factors that are primarily activated by direct stimulation of tyrosine phosphorylation and serine residues, mediating the activation of various downstream signaling pathways. While the role of STAT in most cancer types has been studied, the role of STAT signaling pathways, including its regulators and effectors, in colorectal cancer remains largely unknown. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an application of combined STAT signaling pathway-related genes in a colorectal cancer prognostic model, which has the advantage of high accuracy and can provide a new method for disease diagnosis and prognosis for colorectal cancer patients in clinical practice.

[0005] To achieve the above objectives, the specific technical solution of the present invention is as follows:

[0006] Application of a combination of STAT signaling pathway-related genes in a colorectal cancer prognostic model, wherein the combined STAT signaling pathway-related genes are CAV1, EPO, IL13, LEP and NEUROD1.

[0007] As another specific embodiment of the present invention, the method for establishing a prognostic model for colorectal cancer includes the following steps:

[0008] Step 1) Data collection and organization

[0009] Clinical and gene expression data for colorectal cancer were obtained from the Cancer Genome Atlas (TCGA) database and the Gene Expression Organisation (GEO) database.

[0010] Step 2) Screening for differentially expressed genes related to the STAT signaling pathway

[0011] Differential expression analysis was performed on genes associated with multiple known STAT signaling pathways using the Limma package, and the "igraph" package was used to demonstrate gene interaction networks and screen for gene combinations that were differentially expressed compared to normal samples.

[0012] Step 3) Construction of prognostic models for genes related to the STAT signaling pathway

[0013] Univariate Cox regression analysis was used to evaluate each differentially expressed gene related to the prognosis of colorectal cancer patients screened in step 2). Based on the p-value and LASSO Cox regression analysis, combinations of differentially expressed genes related to the STAT signaling pathway were further screened. The calculation method for the evaluation model is as follows:

[0014] Risk index score = ;

[0015] Where Coefi represents the coefficient, and Xi represents the standardized gene expression level;

[0016] Step 4) Construct a column chart

[0017] Based on genetic and clinical characteristics, independent predictive factors were integrated using the RMS, foreign, and survival packages. Nomograms were constructed through corrected and eliminated trend correspondence analysis (DCA) to assess patient survival and evaluate model efficacy.

[0018] In another specific embodiment of the present invention, the differentially expressed gene combination selected in step 2) compared with the normal sample is IFNL3, IFNE, CSF2, IFNL2, IL23A, AGT, IL20, OSM, LIF, IL13, PRL, EPO, HAMP, CENPJ, MGAT5, PIGU, OCS2, TSLP, IL6ST, PTPRC, PTK2B, HCLS1, NEUROD1, CCL5, IL10, CSF1R, IL10RA, PTPRT, KIT, LEP, ERBB4, IGF1, IL2, CAV1, and GHR.

[0019] In another specific embodiment of the present invention, the differentially expressed gene combination related to the STAT signaling pathway obtained in step 3) is CAV1, EPO, IL13, LEP and NEUROD1.

[0020] In another specific embodiment of the present invention, the risk score in step 3) is calculated as follows:

[0021] Risk index score = (0.009 × CAV1 mRNA expression level) + (2.236 × EPO mRNA expression level) + ((-0.600) × IL13 mRNA expression level) + (0.116 × LEP mRNA expression level) + (0.059 × NEUROD1 mRNA expression level).

[0022] In another specific embodiment of the present invention, in step 3), patients are divided into high-risk group and low-risk group according to median risk score, the overall survival rate between the two groups is compared using Kaplan-Meier analysis, and the log-rank test is used.

[0023] As another specific embodiment of the present invention, principal component analysis (PCA) was performed on the high-risk group and the low-risk group to evaluate their separability. Receiver operating characteristic (ROC) curves were constructed using the "survival", "risk regression", "timeROC", and "survminer" packages, and the accuracy of the gene markers was assessed by the area under the curve (AUC).

[0024] The present invention has the following beneficial effects:

[0025] This invention utilizes the TCGA and GEO databases to analyze the whole genome of colorectal cancer, and establishes a prognostic model related to the prognosis of colorectal cancer based on genes associated with five STAT signaling pathways. The prognostic model has the advantage of high accuracy and can provide a new method for disease diagnosis and prognosis for colorectal cancer patients in clinical practice. At the same time, it reveals its predictive role in the tumor microenvironment and provides important information for the development of targeted therapy.

[0026] The present invention will now be described in further detail with reference to the accompanying drawings. Attached Figure Description

[0027] Figure 1 This is a heatmap of 35 differentially expressed genes in colorectal cancer tissue and normal tissue.

[0028] Figure 2 It is a network of 35 differentially expressed STAT signaling-related genes;

[0029] Figure 3 This is a forest plot showing the hazard ratios and 95% confidence intervals from a univariate Cox regression analysis;

[0030] Figure 4 This is a heatmap of clinicopathological features and risk scores of included colorectal cancer patients;

[0031] Figure 5 This is a schematic diagram showing the distribution of risk scores, survival time, and survival status;

[0032] Figure 6 This is a Kaplan-Meier curve showing overall survival (OS) between low-risk and high-risk colorectal cancer patients;

[0033] Figure 7 This is the receiver operating characteristic (ROC) curve of the 3-year overall survival (OS) prognosis model;

[0034] Figure 8 This is the receiver operating characteristic (ROC) curve of the 5-year overall survival (OS) prognosis model;

[0035] Figure 9 It is a univariate analysis to evaluate predictive factors;

[0036] Figure 10 It is a multivariate analysis to evaluate predictive factors;

[0037] Figure 11 This is a nomogram predicting the survival curve of colorectal cancer patients;

[0038] Figure 12 It is a calibration plot of the predicted curve and the observed curve;

[0039] Figure 13 It is the decision curve analysis (DCA) curve of the predictive model;

[0040] Figure 14 The receiver operating characteristic (ROC) curve is used to evaluate the nomogram's predictive value for the 3-year survival rate of colorectal cancer patients.

[0041] Figure 15 The receiver operating characteristic (ROC) curve is used to evaluate the nomogram's predictive value for the 5-year survival rate of colorectal cancer patients.

[0042] Figure 16 It is a distribution chart of survival status, survival time, and risk score;

[0043] Figure 17 The Kaplan-Meier curves show overall survival (OS) between low-risk and high-risk colorectal cancer patients.

[0044] Figure 18 It is a univariate analysis to evaluate predictive factors;

[0045] Figure 19 It is a multivariate analysis to evaluate predictive factors;

[0046] Figure 20 It is a calibration plot of the predicted and observed curves of the nomograph;

[0047] Figure 21 It is the decision curve analysis (DCA) curve of the nomogram;

[0048] Figure 22The recipient operating characteristic (ROC) curve is used to evaluate the nomogram's predictive value for 3-year survival in colorectal cancer patients in the GSE14333 cohort.

[0049] Figure 23 The receiver operating characteristic (ROC) curve is used to evaluate the nomogram's predictive value for 5-year survival in colorectal cancer patients in the GSE14333 cohort.

[0050] Figure 24 It is a GO pathway enrichment analysis of differentially expressed genes (DEGs);

[0051] Figure 25 KEGG pathway enrichment analysis of differentially expressed genes (DEGs)

[0052] Figure 26 This study examines the expression of CAV1 in colon and rectal cancer and its relationship with immune cell infiltration.

[0053] Figure 27 The scores of low-risk immune-related cells (left) and function (right) in the TCGA cohort are shown.

[0054] Figure 28 The scores of immune-related cells (left) and function (right) are shown in the high-risk group of the TCGA cohort;

[0055] Figure 29 This is an overview of the composition of immune cells;

[0056] Figure 30 This is a schematic diagram illustrating the differences in immune cell composition categorized by risk group. Figure 1 ;

[0057] Figure 31 This is a schematic diagram illustrating the differences in immune cell composition categorized by risk group. Figure 2 ;

[0058] Figure 32 This demonstrates the correlation of immune cell infiltration.

[0059] In the attached figures, T represents tumor, N represents normal tissue, HR represents hazard ratio, LL represents lower limit, UL represents upper limit, AUC represents area under the curve, and TME represents tumor microenvironment. Detailed Implementation

[0060] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0061] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0062] method

[0063] 1) Data collection

[0064] Clinical and gene expression data for colorectal cancer were obtained from the Cancer Genome Atlas (TCGA) database and the Gene Expression Omnibus (GEO) database.

[0065] 2) Identification of differentially expressed genes (DEGs) in the STAT signaling pathway

[0066] Genes related to the STAT signaling pathway were obtained from the Gene Attribute Classification (GO) class name "STAT receptor signaling pathway"; a total of 175 genes related to the STAT signaling pathway were identified, and the DEGs between CRC and normal samples were selected using the "limma" package (P<0.001, |log2FC|>1) for further analysis. The "igraph" software package was used to demonstrate the gene interaction network.

[0067] 3) Construction of prognostic gene features

[0068] The predicted values ​​of DEGs were evaluated using a univariate Cox regression model. LASSO Cox regression analysis was employed, and the inventors calculated the risk score using the following formula: Coefi represents the coefficient, and Xi represents the standardized gene expression level.

[0069] Patients were divided into low-risk and high-risk groups. Principal component analysis (PCA) was used to evaluate their separability, and Kaplan-Meier analysis was used to assess differences in overall survival (OS). The log-rank test was employed. The accuracy of the genetic markers was assessed using the "Survival," "Risk Regression," "timeROC," and "survminer" software packages, through receiver operating characteristic (ROC) curves and area under the curve (AUC).

[0070] Further univariate and multivariate Cox regression analyses were performed using clinical data. Independent predictive factors were integrated using the "rms", "foreign", and "survival" R packages, and nomograms were constructed using corrected and eliminated trend correspondence analysis (DCA). For the GSE14333 dataset, risk scores were obtained using the above methods.

[0071] 4) Functional enrichment analysis of features

[0072] To investigate potential pathways and functions associated with the features, the inventors used the "clusterProfiler" package to perform GO and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses, with screening thresholds of |log2FC|≥0.5 and P<0.05, respectively.

[0073] 5) Immune cell infiltration and component analysis

[0074] The relationship between prognostic genes and immune cell infiltration was analyzed using the Tumor Immunology Assessment Resource (TIMER) database. Furthermore, single-sample gene set enrichment (ssGSEA) analysis using the "gsva" package assessed the infiltration scores of 16 immune cell types and 13 immune-related pathways. The inventors further evaluated the differences in the composition of 22 immune cell types using the CIBERSORT algorithm.

[0075] 6) Statistical analysis

[0076] The inventors used R software (version 4.1.0) and SPSS software (version 25.0) for statistical analysis. Student's t-test was used to compare the means of the two groups; differences in categorical variables were assessed using the Pearson chi-square test; the Log-rank test was used to compare survival differences, and Kaplan-Meier curves were generated for visualization; univariate and multivariate Cox regression analyses were used to assess prognostic factors; and the Wilcoxon test was used to evaluate differences in immune cell infiltration and composition.

[0077] result

[0078] 1) Differentially expressed genes associated with the STAT signaling pathway in colorectal cancer

[0079] Transcriptome data of colorectal cancer patients in the TCGA database were analyzed, and 35 differentially expressed genes related to the STAT signaling pathway were screened out (genes that meet the criteria of P<0.001 and |log2(fold change [FC])|>1 are differentially expressed genes).

[0080] Compared with normal tissue, such as Figure 1 As shown, the expression levels of IFNL3, IFNE, CSF2, IFNL2, IL23A, AGT, IL20, OSM, LIF, IL13, PRL, EPO, HAMP, CENPJ, MGAT5, and PIGU were significantly upregulated in colorectal cancer tissues, while the expression levels of SOCS2, TSLP, IL6ST, PTPRC, PTK2B, HCLS1, NEUROD1, CCL5, IL10, CSF1R, IL10RA, PTPRT, KIT, LEP, ERBB4, IGF1, IL2, CAV1, and GHR were significantly downregulated in colorectal cancer. Figure 2The paper presents the interactions of 35 differentially expressed genes. Figure 3 Detailed information on 35 differentially expressed genes is shown.

[0081] 2) Construction of prognostic models for genes related to the STAT signaling pathway

[0082] Univariate Cox regression analysis was used to evaluate each differentially expressed gene associated with the prognosis of colorectal cancer patients, and five differentially expressed genes (CAV1, EPO, IL13, LEP and NEUROD1) associated with STAT signaling were further screened based on p-values ​​and LASSO Cox regression analysis.

[0083] The formula for the risk index scoring in the prognostic feature model is as follows:

[0084] Risk index score = (0.009 × CAV1 mRNA expression level) + (2.236 × EPO mRNA expression level) + ((-0.600) × IL13 mRNA expression level) + (0.116 × LEP mRNA expression level) + (0.059 × NEUROD1 mRNA expression level).

[0085] Colorectal cancer patients were divided into low-risk and high-risk groups based on the median risk score.

[0086] Clinicopathological features of colorectal cancer patients, such as Figure 4 As shown, the distribution of risk scores and survival time is as follows: Figure 5 As shown in the Kaplan-Meier survival curves, patients with higher risk scores had poorer survival compared to those with lower risk scores (P<0.001). Figure 6 (As shown). The areas under the curves for 3-year and 5-year survival rates were 0.644 and 0.668, respectively (as shown). Figure 7 , Figure 8 (As shown).

[0087] Univariate Cox regression analysis showed that the risk score was significantly associated with OS (e.g., Figure 9 As shown in the figure, multivariate Cox regression analysis indicated that the risk score could independently predict the prognosis of colorectal cancer patients (e.g., ...). Figure 10 (As shown).

[0088] 3) Construction of nomogram based on risk score

[0089] To quantitatively assess and predict the survival of CRC patients, a nomogram based on risk scores was established, such as... Figure 11 As shown, the predicted values ​​of the nomogram are verified by the calibration curve, such as... Figure 12 As shown.

[0090] Compared to treating all patients or no treatment, nomograms provide better prognostic value (e.g., Figure 13 As shown in the figure), the AUC values ​​of the ROC curves are 0.781 and 0.812, respectively, used to predict 3-year and 5-year survival rates (as shown in the figure). Figure 14 , Figure 15 (As shown).

[0091] 4) Validation of STAT signal-related gene characteristics using external datasets

[0092] To validate the gene signatures associated with STAT signals, the GSE14333 dataset was used. The GSE14333 dataset contains raw transcriptome sequencing data and clinical information of 226 colorectal cancer patients.

[0093] The distribution of risk scores and survival time is as follows: Figure 16 As shown.

[0094] Compared with patients in the low-risk group, patients with higher risk scores had significantly lower survival rates (e.g. Figure 17 As shown; P<0.0001), the risk score was significantly associated with patient survival (e.g., ...). Figure 18 As shown), and is an independent predictor of multivariate Cox regression analysis (e.g. Figure 19 As shown in the figure, the predicted values ​​of the nomogram are verified by the calibration curve. Figure 20 As shown; nomograms provide better prognostic value (e.g. Figure 21 As shown in the figure), the AUC values ​​of the nomogram ROC curves used to predict 3-year and 5-year survival rates are 0.686 and 0.750, respectively (as shown in the figure). Figure 22 , Figure 23 (As shown).

[0095] 5) Functional analysis of risk groups

[0096] To investigate the biological functions and pathways associated with risk scores, the inventors conducted GO and KEGG pathway enrichment analyses to deduce differentially expressed genes (DEGs).

[0097] like Figure 24 As shown, GO pathway analysis revealed that differentially expressed genes (DEGs) were significantly associated with immune-related pathways, including the regulation of immune effector processes, lymphocyte-mediated immunity, and various immune responses. KEGG pathway analysis indicated significant dysregulation of cytokine-cytokine receptor interactions and chemokine signaling pathways (e.g., KEGG pathway). Figure 25 (As shown).

[0098] 6) Immunogenetic analysis of characteristics of colorectal cancer patients

[0099] The association between each gene in the STAT signaling pathway and immune infiltration was assessed using the TIMER database.

[0100] Of these five genes, CAV1 is significantly associated with the infiltration of B cells, CD8+ T cells, CD4+ T cells, macrophages, neutrophils, and dendritic cells in colorectal cancer (e.g., Figure 26 (as shown in the figure), while the correlation between other genes and immune cell infiltration was not significant.

[0101] To comprehensively assess the immunological characteristics of colorectal cancer, single samples were scored using the ssGSEA system for 29 immune gene sets. In the TCGA cohort, patients with higher risk scores showed significantly reduced immune cell infiltration compared to those with lower risk scores (e.g., ...). Figure 27 (as shown) and downregulation of immune-related pathways (such as Figure 28 (As shown).

[0102] Furthermore, the composition of infiltrating immune cells was assessed using the CIBERSORT algorithm. An overview of the tumor microenvironment cell composition is shown below. Figure 29 As shown, the study indicates that the infiltration levels of activated CD4+ memory T cells, eosinophils, neutrophils, follicular helper T cells, and M0 macrophages differed significantly between the low-risk and high-risk groups (as shown in Figures 1-2). Figure 30 , Figure 31 (As shown).

[0103] The correlation of 22 immune cell types showed that M0 macrophages were negatively correlated with resting dendritic cells (r = -0.41), resting mast cells (r = -0.41), and CD8+ T cells (r = -0.39). Furthermore, resting mast cells were negatively correlated with activated mast cells (r = -0.42). Figure 32 (As shown).

[0104] While the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the scope of the invention. Any person skilled in the art can make modifications without departing from the scope of the invention; all equivalent modifications made in accordance with the invention should be covered by the scope of the invention.

Claims

1. The application of combined STAT signaling pathway-related genes in a prognostic model of colorectal cancer, characterized in that, The STAT signaling pathway-related genes in the combination are CAV1, EPO, IL13, LEP, and NEUROD1; The method for establishing a prognostic model for colorectal cancer includes the following steps: Step 1) Data collection and organization Clinical and gene expression data of colorectal cancer were obtained from the Cancer Genome Atlas Database and the Comprehensive Gene Expression Database. Step 2) Screening for differentially expressed genes related to the STAT signaling pathway Differential expression analysis was performed on genes associated with multiple known STAT signaling pathways using the Limma package, and gene interaction networks were explored using the igraph package to screen for gene combinations that were differentially expressed compared to normal samples. The gene combinations that were differentially expressed compared to normal samples in step 2) were IFNL3, IFNE, CSF2, IFNL2, IL23A, AGT, IL20, OSM, LIF, IL13, PRL, EPO, HAMP, CENPJ, MGAT5, PIGU, OCS2, TSLP, IL6ST, PTPRC, PTK2B, HCLS1, NEUROD1, CCL5, IL10, CSF1R, IL10RA, PTPRT, KIT, LEP, ERBB4, IGF1, IL2, CAV1, and GHR. Step 3) Construction of prognostic models for genes related to the STAT signaling pathway Univariate Cox regression analysis was used to evaluate each differentially expressed gene related to the prognosis of colorectal cancer patients screened in step 2). Based on the p-value and LASSOCox regression analysis, further screening was conducted to obtain differentially expressed gene combinations related to the STAT signaling pathway. The differentially expressed gene combinations related to the STAT signaling pathway screened in step 3) were CAV1, EPO, IL13, LEP, and NEUROD1. The calculation method of the evaluation model was as follows: Risk Index Score = ; Where Coefi represents the coefficient, and Xi represents the standardized gene expression level; The risk index score in step 3) is calculated as follows: Risk index score = (0.009 × CAV1 mRNA expression level) + (2.236 × EPO mRNA expression level) + ((-0.600) × IL13 mRNA expression level) + (0.116 × LEP mRNA expression level) + (0.059 × NEUROD1 mRNA expression level). Step 4) Construct a column chart Based on genetic and clinical characteristics, independent predictive factors were integrated using the RMS, foreign, and survival packages. Nomograms were constructed by adjusting and eliminating trend correspondence analysis to assess patient survival and evaluate model efficacy.

2. The application of the combined STAT signaling pathway-related genes as described in claim 1 in a colorectal cancer prognostic model, characterized in that, In step 3), patients were divided into high-risk and low-risk groups based on the median risk score. Kaplan-Meier analysis was used to compare the overall survival rates between the two groups, and log-rank test was used.

3. The application of the combined STAT signaling pathway-related genes as described in claim 2 in a colorectal cancer prognostic model, characterized in that, Principal component analysis was performed on the high-risk and low-risk groups to evaluate their separability. Receiver operating characteristic curves were constructed using the "survival", "risk regression", "timeROC", and "survminer" packages, and the accuracy of the gene markers was assessed by the area under the curve.

Images