Investigation of the Metabolism-Related Prognostic Gene SERPINE1 as a Prognostic Predictor in Colorectal Adenocarcinoma and Its Regulatory Mechanisms Underlying Immune Infiltration

Investigation of the Metabolism-Related Prognostic Gene SERPINE1 as a Prognostic Predictor in Colorectal Adenocarcinoma and Its Regulatory Mechanisms Underlying Immune Infiltration | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Investigation of the Metabolism-Related Prognostic Gene SERPINE1 as a Prognostic Predictor in Colorectal Adenocarcinoma and Its Regulatory Mechanisms Underlying Immune Infiltration Xiaoyun Feng, Siyuan Chen, Qian Xie, Chenjie Jiang, Jianyan Lu, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8077356/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Objective To explore the role of Metabolic-related Genes (MRGs), especially SERPINE1 , in the prognosis and immune infiltration of Colorectal Adenocarcinoma (COAD). Methods This study incorporated the transcriptomic data of colorectal adenocarcinoma (COAD) retrieved from The Cancer Genome Atlas (TCGA) and the Gene Expression Omnibus (GEO) databases. Based on 2752 metabolism-related genes, the Metabolism-Related Gene Prognostic Index (MRGRI) model was constructed and validated. For the SERPINE1 , a comprehensive analysis was performed, covering differential expression analysis, survival analysis, and nomogram construction. Meanwhile, its associations with the tumor microenvironment, immune response, and drug sensitivity were explored. In addition, single-cell analysis was used to verify the heterogeneous expression characteristics of SERPINE1 . Finally, reverse transcription quantitative polymerase chain reaction (RT-qPCR) was employed to validate the expression level of this gene in COAD cells, providing data support for subsequent mechanistic research.. Results A total of 159 differentially expressed related to metabolic genes (DE-MRGs) were screened out, and 18 genes associated with prognosis were identified. Through LASSO regression, a prognostic model containing 13 genes was established. In this model, the survival rate of the high-risk group was relatively low and the risk score demonstrated strong predictive power. Furthermore, a total of 72 differential genes related to the prognosis of MRGs were obtained, among which SERPINE1 was the hub gene. In COAD tissues, high expression of SERPINE1 indicated a poor prognosis and was correlated with disease stages. The nomogram provided accurate predictions. The differentially expressed genes were mainly enriched in pathways related to immune receptor activity. SERPINE1 regulated the immune microenvironment, affected tumor immune escape and immunotherapy responses, and overexpression of it would reduce drug sensitivity.Single-cell analysis reveals heterogeneous expression of SERPINE1 in macrophages. Conclusion In the MRGs risk model, SERPINE1 is a hub gene that plays a crucial role in the prognosis, immune infiltration and drug sensitivity of COAD. Metabolic-related Genes SERPINE1 Prognostic Colorectal Adenocarcinoma Immune Infiltration Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 1 Introduction Colorectal adenocarcinoma (COAD) represents a formidable global health conundrum, imposing a significant burden on morbidity and mortality rates. As one of the most pervasive malignancies, its incidence has been on a consistent upward trajectory [ 1 ]. The prognoses of COAD patients display pronounced heterogeneity, thereby underscoring the imperative need to identify reliable prognostic biomarkers. These biomarkers are instrumental in steering therapeutic strategies and enhancing patient outcomes [ 2 ]. In recent years, there has been a burgeoning interest in the role of metabolism-associated genes (MRGs) within oncological research [ 3 ]. Metabolism, a fundamental biological mechanism that supplies cells with energy and essential molecular constituents, frequently experiences dysregulation in cancer cells. This dysregulation precipitates disruptions in energy metabolism, biosynthesis, and redox equilibrium [ 4 ]. MRGs assume pivotal roles in these processes and hold substantial promise as biomarkers for cancer diagnosis and prognosis [ 5 ]. In the current investigation, we constructed a risk model associated with MRGs and conducted differential gene analysis to identify the key risk gene, SERPINE1 . Also known as plasminogen activator inhibitor-1, SERPINE1 is a serine protease inhibitor that regulates the activity of plasminogen activators [ 6 ]. It is implicated in a plethora of physiological and pathological processes, including blood coagulation, fibrinolysis, angiogenesis, and inflammation [ 7 ]. Within the oncology domain, SERPINE1 has been shown to exert multifarious effects on tumor progression. Firstly, it can impede extracellular matrix degradation by inhibiting the activity of plasminogen activators involved in matrix breakdown, potentially curtailing tumor cell invasion and metastasis. Secondly, SERPINE1 can promote angiogenesis by modulating the activity of vascular endothelial growth factor and other angiogenic mediators [ 8 ]. Given that angiogenesis is indispensable for tumor growth and metastasis, as it provides oxygen and nutrients to neoplastic cells, this aspect is of utmost importance. Additionally, SERPINE1 is intricately linked to inflammation, which plays a crucial role in cancer development and progression. Inflammatory cells can secrete cytokines and growth factors that facilitate tumor growth and dissemination [ 9 ]. Moreover, SERPINE1 has been reported to be associated with poor prognoses in several cancer types, such as breast, lung, and ovarian cancers. In these malignancies, elevated SERPINE1 expression levels correlate with advanced tumor stages, lymph node metastasis, and distant dissemination. However, the prognostic significance of SERPINE1 in COAD remains ambiguous [ 10 ]. Beyond its role in tumor progression, SERPINE1 may also be associated with immune cell infiltration in COAD [ 11 ]. Immune cell infiltrates, consisting of various immune cell populations within the tumor microenvironment, can interact with cancer cells and stromal components, thereby regulating tumor growth, metastasis, and the response to therapy [ 12 ]. Recent studies have revealed that immune cell infiltrates can serve as prognostic biomarkers and therapeutic targets in COAD [ 13 ]. For example, T cells, B cells, natural killer cells, macrophages, and dendritic cells are key immune cell types that can infiltrate the tumor microenvironment [ 19 ]. However, the relationship between SERPINE1 and immune cell infiltrates in COAD remains incompletely understood. SERPINE1 may regulate the recruitment and activation of immune cells in the tumor microenvironment [ 14 ]. For instance, it may influence the expression of chemokines and cytokines that attract immune cells to the tumor site. It may also modulate immune cell function by interacting with immune receptors or signaling pathways. Elucidating this relationship could provide novel insights into the pathogenesis and prognosis of COAD [ 15 , 18 ]. The objective of this study was to explore the prognostic significance of SERPINE1 in COAD and its association with immune cell infiltrates. We analyzed the correlation between SERPINE1 expression and clinicopathological characteristics, including tumor stage, lymph node metastasis, and distant metastasis. Additionally, we evaluated the prognostic value of SERPINE1 expression in COAD patients through survival analysis and assessed the independent prognostic significance of SERPINE1 expression along with other prognostic factors (such as age, gender, and tumor stage) using a multivariate Cox proportional hazards model. This study provides novel insights into the prognostic role of MRGs in COAD and helps to identify SERPINE1 as a potential prognostic biomarker for COAD patients. It also uncovers the relationship between SERPINE1 and immune cell infiltrates in COAD, which may accelerate the development of innovative immunotherapeutic approaches. In conclusion, this study represents a significant advancement in understanding the role of the metabolism - related risk gene SERPINE1 as a prognostic biomarker for COAD and its association with immune cell infiltrates. The findings of this study may have far-reaching implications for the diagnosis, prognosis, and treatment of this disease. 2 Materials and Methods 2.1 Data Download In this study, COAD data (FPKM, Fragments Per Kilobase of exon per Million reads mapped) were downloaded from The Cancer Genome Atlas (TCGA) database ( https://portal.gdc.cancer.gov/ ). This data set included 473 tumor samples and 41 adjacent normal samples. Transcripts Per Million (TPM) values were calculated from the FPKM values, followed by a log2 transformation for data normalization. The gene expression dataset GSE38832 was retrieved from the Gene Expression Omnibus (GEO) database ( https://www.ncbi.nlm.nih.gov/geo/ ), which consisted of 118 COAD tissue samples. During data analysis and collation, Perl language ( http://www.perl.org/ ) and R software (version 4.2.0, https://www.r-project.org ) were employed. The MRGs were derived from 2752 previously published genes related to metabolism[ 16 ], which encode all known human metabolic enzymes and transporters. 2.2 Cell Culture Technique Both the human colon epithelial cell line FHC and the colon adenocarcinoma cell line Caco-2 adopted in this study were purchased from Wuhan Procell Life Technology Co., Ltd. The cultivation of both FHC cells and Caco-2 cells was conducted in a composite medium composed of 90% DMEM basal medium, 10% fetal bovine serum, and 1% penicillin and streptomycin. The culture conditions were set as a constant temperature environment of 37°C and a CO₂ concentration of 5% to simulate the internal environment of the human body. 2.3 Quantitative Real-time Reverse Transcription Polymerase Chain Reaction (qRT-PCR) Experiment In this experiment, the FastPure® Cell Total RNA Isolation Kit (RC112, Vazyme, Nanjing, China) was used. According to the operation instructions provided by the manufacturer, total RNA was successfully extracted from the cells. Subsequently, the PrimeScript RT Reagent Kit (RR037A, Takara Bio, Kyoto, Japan) was employed to synthesize cDNA from the RNA. For the RT-PCR reaction, SYBR Green Mix (4309155, Thermo Fisher Scientific, USA) was adopted. The sequences of SERPINE1 and GAPDH were designed and applied (the forward primer sequence of GAPDH is 5’-GCACCGTCAAGGCTGAG-AAC-3’, and its reverse primer sequence is 5’-TGGTGAAGACGCCAGTGGA-3’; the forward primer sequence of SERPINE1 is 5’-CCGCCGCCTCTTCCACAA-ATC-3’, and its reverse primer sequence is 5’-TAGGGCAGTTCCAGGATGTCGT-AG-3’). 2.3 Construction and Validation of the MRGs Prognostic Index (MRGRI) This study focused on constructing a MRGRI and calculating the risk score, with the formula: Risk score = (β₁×Exp₁) + (β₂×Exp₂) + … + (βₙ×Expₙ) (where β represents the gene regression coefficient and Exp represents the gene expression level)[ 20 ]. Firstly, the study retrieved the gene expression data and clinical follow-up information of COAD patients from the TCGA and GEO databases. For the 2,752 MRGs, the limma tool was used to process the expression values of duplicate genes. Using the "limma" R package, 159 Differentially Expressed MRGs (DEMRGs) were screened out with the criteria of |log₂ fold change (FC)| >1.5 and an adjusted P - value ≤ 0.05, and a volcano plot was drawn.Subsequently, the "survival" R package was used to standardize the survival time data. A multivariate COX regression model was constructed with "futime" and "fustat" as the dependent variables. After optimization by the step() function, the risk score was calculated. The high - and low - risk groups were divided based on the median value. After censoring the outliers, the distribution plot of risk scores, the scatter plot of survival status, and the heatmap of gene expression were drawn. Combining univariate Cox regression and LASSO regression analysis using the "glmnet" package, by adjusting the penalty parameter λ, 13 core prognostic genes were screened out and their weights were determined.In the model evaluation stage, the "survminer" and "survival" packages were used to draw Kaplan - Meier survival curves. Univariate and multivariate COX regression analyses (presented as forest plots) were conducted to verify the independent prognostic value of the risk score. The pROC package was used to construct the Receiver Operating Characteristic (ROC) curve and calculate the Area Under the Curve (AUC) to evaluate the predictive efficacy. 2.4 Expression and Prognosis Analysis of SERPINE1 in COAD The "ggplot2" package in R language was utilized to draw box plots, aiming to compare the expression differences of the SERPINE1 gene between COAD tissues and adjacent normal tissues. The "stats" R package was employed to conduct difference statistical tests. Meanwhile, the "ggplot2" package was also used to draw paired sample analysis plots, so as to display the correlation of the SERPINE1 gene expression in tumor tissues and adjacent normal tissues of the same patient.Based on the median of the SERPINE1 gene expression level, patients were divided into a high-expression group and a low-expression group. The "survminer" package and the "survival" package were adopted to draw progression-free survival curves for comparing the progression-free survival status of patients in these two groups. Similarly, the overall survival curves were plotted using the "survminer" package and the "survival" package to compare the overall survival status of the two groups of patients.The "survminer" package was used to conduct univariate Cox regression analysis to evaluate the impacts of factors such as the SERPINE1 gene expression, age, gender, and tumor stage on patient survival. On the basis of the univariate analysis, the "survminer" package was further employed to carry out multivariate Cox regression analysis by incorporating multiple factors for comprehensive analysis, with the aim of determining the independent impact of the SERPINE1 gene expression on patient survival.The "ggplot2" package was used to draw box plots. Patients were grouped according to age (≤ 65 years old and > 65 years old), gender (male and female), M stage (M0 and M1), T stage (T1, T2, T3, T4, T4a), clinical stage (Stage I, Stage II, Stage III, Stage IV), and N stage (N0, N1, N2), and the expression differences of the SERPINE1 gene in different subgroups were compared. 2.5 Identification of MRGs Risk DEGs In this study, a risk model based on DEGs of MRGs was constructed, and the fold change in the expression of MRGs risk genes was calculated using the "limma" package in R software. Based on the preset thresholds (logFC > 1 and q-value < 0.05), the MRGs risk DEGs were screened out, and the "ggplot2" package was adopted to draw a volcano plot for visual display. The screened MRGs risk DEGs were input into the STRING database ( https://string-db.org/ ) to obtain the interaction data among genes. Subsequently, Cytoscape 3.7.2 software was used to draw a gene interaction. 2.6 Construction and Calibration of the Nomogram In terms of nomogram construction, the coxph function is used to build a Cox model based on all variables including the expression level of the SERPINE1 gene for multivariate analysis. The regplot function is utilized to generate the nomogram, which incorporates a scoring system for each variable, consistent with the variable score allocation in the figure. The parameter setting failtime = c(1,3,5) enables the prediction of 1-year, 3-year, and 5-year survival rates, and the risk scores are calculated and output as a file. During the calibration curve validation, the cph and calibrate functions are called three times to generate calibration curves for the corresponding time points respectively. 2.7 Analysis of the Expression Difference of SERPINE1 and Its Enriched Pathways The "limma" software package was utilized to conduct differential gene analysis. The differences in gene expression between the high-expression group and the low-expression group of SERPINE1 were compared, and the DEGs were screened out (with the threshold set as logFC > 1 and adj.P.Val < 0.05). The "pheatmap" software package was employed to draw a heatmap, aiming to display the expression patterns of the DEGs in the two groups.The screened DEGs were then imported into the "clusterProfiler" software package for functional enrichment analysis, which encompassed Gene Ontology (GO) enrichment analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis. 2.8 Tumor Microenvironment (TME) Scores The ESTIMATE algorithm was adopted to calculate the Stromal Score, Immune Score, and ESTIMATE Score for each patient's sample, with the aim of evaluating the abundance and functional activity of stromal cells and immune cells in the TME. The Wilcoxon rank-sum test was employed to analyze the differences in the Stromal Score, Immune Score, and ESTIMATE Score between the low-expression group and the high-expression group of SERPINE1 . This test was executed using the wilcox.test function in the stats package.Furthermore, the ggplot2 package was utilized to draw Violin Plots to showcase the score values under different scoring metrics. 2.9 Immune Correlation Analysis A quantitative analysis was conducted on the association between the expression of the SERPINE1 gene and that of immune-related genes by calculating their correlation coefficients. The "pheatmap" software package was employed to draw a correlation heatmap, aiming to visually display the correlation patterns among gene expressions.Furthermore, through the CIBERSORT algorithm, the infiltration proportions of immune cell subsets in different samples were calculated. Subsequently, the "ggplot2" software package was utilized to draw box plots, which enabled the screening of immune cell subsets that exhibited significant correlations with the expression of the SERPINE1 gene. Additionally, correlation scatter plots were also drawn to further explore the relationship between the expression of the SERPINE1 gene and immune cell subsets.To analyze the expression differences of immune checkpoint genes (PD-1, CTLA-4) in the high-expression group and the low-expression group of the SERPINE1 gene, the corresponding expression data were extracted, and violin plots were drawn using the "ggplot2" software package. 2.10 Drug Sensitivity Analysis Based on the datasets of clinical trials and drugs approved by the FDA, the Pearson correlation coefficients between the expression of the SERPINE1 gene and various drugs were calculated, and the obtained results were then screened. The “pRRophetic” R package was utilized to evaluate the half-maximal inhibitory concentration of drugs in colorectal cancer patients. For each drug, the differences in drug sensitivity between the low-expression group and the high-expression group of the SERPINE1 gene were compared. The Wilcoxon rank-sum test was employed to assess the statistical significance of the differences in drug sensitivity between the two groups. Through the “ggplot2” package, box plots were drawn to display the distribution of drug sensitivity for each drug in the low-expression group and the high-expression group of the SERPINE1 gene. 2.11 Methods for single-cell transcriptome data processing and analysis The Harmony algorithm was adopted to correct batch effects in the gene expression matrix, and the spatial uniformity of sample points was observed to guarantee the reliability of subsequent multi-sample integration analysis. For dimensionality reduction and clustering, UMAP was used for nonlinear dimensionality reduction and visualization to analyze cell distribution patterns and compare cellular heterogeneity between normal and tumor tissues. PCA + UMAP and NMF + UMAP strategies were compared to select a dimensionality reduction method with high biological interpretability. Cell types were manually annotated based on marker genes, validated by averaged scaled gene expression and cell expression percentage. Macrophage subsets were further visualized and subdivided via UMAP to explore their functional heterogeneity in the tumor microenvironment. 2.12 Expression of SERPINE1 in COAD-related cell lines in the CCLE database The gene expression matrix for COAD cell lines was obtained from the CCLE dataset ( https://depmap.org/portal/data_page/?tab=allData).Statistica l analysis was conducted using R software. Results were considered statistically significant when the p-value was less than 0.05. 2.13 Statistical Analysis A variety of statistical methods were employed for data analysis. During the data preprocessing stage, the "limma" package was utilized to perform background correction and normalization on gene expression data, while the "tidyverse" package was used to organize clinicopathological information.In the differential analysis section, the "limma" package was adopted to screen for DEGs. The t.test function was applied to conduct statistical tests on the expression differences of the SERPINE1 gene, and the Wilcoxon rank-sum test was used to evaluate the differences in drug sensitivity.In the model construction and evaluation phase, Cox regression analysis, LASSO regression analysis, survival analysis, ROC curve analysis, and the construction and validation of the nomogram were carried out.Moreover, in other analyses, various functions and tools such as kruskal.test, "forestplot", "pheatmap", and the "CIBERSORT" algorithm were used to conduct in-depth analyses of subgroup expression differences, prognosis, immune correlations, and tumor microenvironment scores. 3 Results 3.1 Construction of the Risk Model A differential analysis of MRGs was performed in TCGA-COAD samples, identifying a total of 159 DEGs (Fig. 1 A). To explore the prognostic significance of metabolism-related DEGs, patients with a follow-up period exceeding 30 days were included, yielding 18 prognosis-associated genes ( P < 0.05). Prognostic ratio association analysis showed that increased CD36 expression ( P = 0.010, HR = 1.358, 95% CI: 1.076–1.714) was associated with an elevated risk of specific outcomes; high MORC2 expression ( P < 0.001, HR = 1.966, 95% CI: 1.181–3.273) was significantly linked to a higher outcome risk; while increased ELVOL6 expression ( P = 0.016, HR = 0.653, 95% CI: 0.462–0.924) was associated with a reduced outcome risk. Other genes such as ENO3 and MORC2 also exhibited varying degrees of association with specific outcomes, as detailed in (Fig. 1 B). Thirteen metabolic genes were identified for inclusion in the prognostic model through LASSO regression analysis (Fig. 1 (C-D)). Patients were stratified into high-risk and low-risk groups based on risk scores, with the number of patients in each group annotated below respective survival curves. Using risk scores of each COAD patient from the TCGA database, a risk prognostic model was constructed (Risk score=(0.292246796935572×CD36 Exp) + (0.81270688717358×ENO3 Exp)+(0.357195157656081×ELOVL3 Exp) + (0.357195157656081×ELOVL3 Exp+(-0.387366053×CPT2 Exp))(S1). Results showed that increased risk scores were associated with upregulated expression of prognostic risk genes ( ELOVL6, CPT2, ELOVL3, CD36 , and ENO3 ) (Fig. 2 A), shortened survival time (Supplementary Materials Fig. 2B), and higher mortality rates (Fig. 2 C). In the GSE38832 dataset, overall survival curves for high-risk and low-risk groups (Fig. 1 E) showed a statistically significant difference in survival rates between the two groups ( P = 0.003), with lower survival in the high-risk group. In the TCGA dataset, overall survival curves (Fig. 1 F) demonstrated a highly significant survival difference ( P < 0.001), with the high-risk group exhibiting significantly lower survival rates than the low-risk group. Progression-free survival curves (Fig. 1 G) also showed lower progression-free survival in the high-risk group ( P < 0.001). Univariate Cox regression analysis (Fig. 1 H) indicated that age ( P = 0.002, HR = 1.029, 95% CI: 1.010–1.048), stage ( P < 0.001, HR = 12.067, 95% CI: 6.128–26.628), and risk score ( P < 0.001, HR = 13.171, 95% CI: 2.218–45.533) were significantly associated with disease risk, while gender ( P = 0.560, HR = 1.131, 95% CI: 0.747–1.711) was not. Multivariate Cox regression analysis (Fig. 1 I) further confirmed that age, stage, and risk score remained significantly associated with disease risk after multivariate adjustment, whereas gender was not. ROC curve analysis (Fig. 1 J) revealed an area under the curve (AUC) of 0.733 for the risk score, higher than that of age (AUC = 0.628), gender (AUC = 0.694), and stage (AUC = 0.675), indicating strong predictive ability of the risk score-based prognostic model. ROC curve analysis at different time points (Fig. 1 K) showed AUC values of 0.668, 0.693, and 0.733 for 1-year, 3-year, and 5-year predictions, respectively, suggesting the model’s predictive capacity improves over time. Table 1 Concordance Index(C-index) Gene Coefficient HR HR.95L HR.95H pvalue CD36 0.292246796935572 1.33943354494535 1.06373433948394 1.68658861026828 0.0129397840429795 ENO3 0.81270688717358 2.25400106288746 1.54182381008084 3.29513706967038 2.73305473762638e-05 ELOVL3 0.357195157656081 1.42931478435996 1.1064174242589 1.84644665566286 0.00625700559069411 ELOVL6 -0.309335456 0.733934526014851 0.510101027056577 1.05598667696252 0.0956164499974209 CPT2 -0.387366053 0.678842556862494 0.444142817395086 1.03756539329033 0.073518942445064 3.2 Identification of DEGs in Metabolic Risk Prognosis Based on the constructed MRG risk prognostic model for COAD, this study analyzed MRGs associated with risk prognosis and ultimately screened out 72 genes with significantly differential expression. By constructing a protein-protein interaction (PPI) network diagram, the complex interaction relationships among these DEGs related to risk prognosis were revealed. The results showed that the SERPINE1 gene was closely connected with other genes, suggesting that SERPINE1 may play a key regulatory role in MRG-related risks of COAD and participate in important biological processes or signal transduction pathways. Thirteen genes were identified to construct the risk score model, among which SERPINE1 made a significant contribution. Functional enrichment analysis indicated that the related genes were enriched in pathways such as immune-related pathways, with SERPINE1 being significantly enriched. In the PPI network, SERPINE1 exhibited a high degree of connectivity and extensive interactions, outperforming other genes in multiple aspects. Due to its high expression in COAD tissues, association with poor prognosis, critical role in regulating the tumor microenvironment, and extensive interactions in the PPI network to regulate downstream signaling pathways and influence tumor progression, SERPINE1 was regarded as a "hub gene" (Fig. 3 ). 3.3 Association between Clinicopathological Features and Prognosis of SERPINE1 In COAD tissues, the expression level of the SERPINE1 gene was significantly higher than that in adjacent normal tissues (Fig. 4 A). In the same patient, there was a certain correlation in the expression of the SERPINE1 gene between the tumor tissue and the adjacent normal tissue (Fig. 4 B). Survival analysis revealed that the progression-free survival rate of patients in the high-expression group of SERPINE1 was lower than that of the low-expression group, with a statistically significant difference ( P = 0.007). The progression-free survival curve clearly demonstrated the survival difference between the two groups of patients (Fig. 4 C). In addition, the overall survival rate of patients in the high-expression group of SERPINE1 was significantly lower than that of the low-expression group ( P < 0.001). The overall survival curve intuitively reflected this difference, suggesting that the high expression of the SERPINE1 gene might be associated with a poor prognosis (Fig. 4 D).The results of univariate and multivariate Cox regression analyses showed that in the univariate Cox regression analysis, the expression of the SERPINE1 gene ( P = 0.016, Hazard ratio = 1.205, 95% CI: 1.036–1.403), age ( P = 0.002, Hazard ratio = 1.029, 95% CI: 1.010–1.048), and tumor stage ( P < 0.001, Hazard ratio = 2.068, 95% CI: 1.628–2.628) were significantly associated with patient survival, while gender ( P = 0.564, Hazard ratio = 1.130, 95% CI: 0.747–1.709) was not significantly associated with patient survival. In the multivariate Cox regression analysis, after adjusting for other factors, the expression of the SERPINE1 gene ( P = 0.137, Hazard ratio = 1.132, 95% CI: 0.961–1.333), age ( P < 0.001, Hazard ratio = 1.040, 95% CI: 1.021–1.060), and tumor stage ( P < 0.001, Hazard ratio = 2.190, 95% CI: 1.714-2.800) were still significantly associated with patient survival, and gender ( P = 0.886, Hazard ratio = 0.970, 95% CI: 0.636–1.477) was not significantly associated with patient survival (Fig. 4 E).Subgroup analysis showed that there was no significant difference in the expression of the SERPINE1 gene among patients in different age groups (≤ 65 years old and > 65 years old) ( P = 0.32) (Fig. 4 F). There was no significant difference in the expression of the SERPINE1 gene between male and female patients ( P = 0.62) (Fig. 4 G). The expression of the SERPINE1 gene in patients at stage M1 was significantly higher than that in patients at stage M0 ( P = 0.044) (Fig. 4 H). With the progression of the T stage (T1, T2, T3, T4, T4a), there were significant differences in the expression of the SERPINE1 gene between (T2 and T3, T3 and T4) ( P < 0.01) (Fig. 4 I). With the progression of the clinical stage (Stage I, Stage II, Stage III, Stage IV), the expression of the SERPINE1 gene gradually increased, and there were significant differences among different stages (comparison between Stage I and Stage IV, P = 0.00061) (Fig. 4 J). With the progression of the N stage (N0, N1, N2), the expression of the SERPINE1 gene gradually increased, and there were significant differences among different stages (comparison between N0 and N2, P = 0.022) (Fig. 4 K). 3.4 Construction of the Nomogram A Cox proportional hazards model was constructed using the coxph function, incorporating variables such as SERPINE1 gene expression level, gender, T stage, N stage, and clinical stage for multivariate analysis. The model was visualized as a nomogram via the regplot function, which assigned specific scores to each variable (e.g., high SERPINE1 expression and advanced T stages (T4a) corresponded to higher scores, while early T stages (T1) corresponded to lower scores). The parameter failtime = c(1,3,5) was used to predict 1-year, 3-year, and 5-year survival rates. Patients with a total score of 207 had an estimated 5-year survival rate of approximately 0.75 (Fig. 5 A). Risk scores for patients were calculated and exported to a file (nomoRisk.txt) for clinical stratification, where higher scores indicated poorer prognosis. For calibration curve validation, repeated calls to the cph and calibrate functions generated calibration curves for each time point. Results showed minimal differences between predicted and observed 1-year survival rates (85% vs 83%), with the calibration curve closely aligning with the 45° diagonal line, demonstrating excellent short-term predictive accuracy. Although slight deviations were observed in 3-year and 5-year survival rate predictions (70% vs. 65%), the curves generally remained close to the diagonal line with consistent trends, indicating the model’s reliability in medium- and long-term prognosis (Fig. 5 B). 3.5 Differential Expression and Enrichment Analysis of SERPINE1 The heatmap displayed the expression patterns of DEGs in the sample groups with high and low expressions of SERPINE1 , and a total of 2,138 DEGs were identified (Fig. 6 A). Functional enrichment analysis indicated that SERPINE1 -DEGs were significantly enriched in biological processes such as immune receptor activity, carbohydrate binding, and receptor-ligand activity; in cellular components, they were significantly enriched in structural components of the extracellular matrix, integral components of the postsynaptic membrane, etc.; in molecular functions, they were significantly enriched in receptor activation activity, collagen binding, etc.; in KEGG pathways, they were significantly enriched in pathways such as cytokine-cytokine receptor interaction and neuroactive ligand-receptor interaction. These findings suggest that the SERPINE1 gene may play a role in the development of colon adenocarcinoma by influencing these biological processes and pathways (Fig. 6 B). 3.6 Relationship between SERPINE1 Gene Expression Level and TME Score To investigate the differences in the contents of immune cells and stromal cells between the high-expression and low-expression groups of the SERPINE1 gene and explore its correlation with the tumor immune microenvironment, this study found that in the high-expression group of SERPINE1 , the StromalScore, ImmuneScore, and ESTIMATE Score were all significantly higher than those in the low-expression group. These differences were statistically highly significant ( P < 0.001). The results suggest that the high expression of the SERPINE1 gene may be associated with the increase in the contents and activities of stromal cells and immune cells in the tumor microenvironment of colon adenocarcinoma, indicating that SERPINE1 may play a key role in regulating the tumor microenvironment (Fig. 7 ). 3.7 Analysis of the Role of SERPINE1 in Immune Cell Infiltration The heatmap revealed the correlations between the SERPINE1 gene and a series of immune-related genes, indicating that the SERPINE1 gene was mainly positively correlated with genes of the TNFRSF family and cluster of differentiation(CD) molecules (Fig. 8 A).The box plot showed that there were significant differences in the infiltration proportions of certain immune cell subsets between the high-expression group and the low-expression group of the SERPINE1 gene. Among them, the infiltration proportion of M0 macrophages was higher in the high-expression group, while the infiltration proportion of the memory B cell subset was relatively higher in the low-expression group, suggesting that the SERPINE1 gene might affect the infiltration pattern of immune cells in colon adenocarcinoma tissues (Fig. 8 B).The correlation scatter plot further confirmed that the expression of the SERPINE1 gene was mainly positively correlated with the infiltration proportion of NK cells (R = 0.25, P = 3.9e-05) and negatively correlated with plasma cells (R=-0.2, P = 0.001), indicating that the SERPINE1 gene might regulate the tumor immune microenvironment by influencing the infiltration of these immune cells (Fig. 8 C).The violin plot showed that there were differences in the expressions of immune checkpoint genes between the high-expression group and the low-expression group of the SERPINE1 gene. The expression of PD-1 was higher in the high-expression group ( P = 1.3e-05), and the expression of CTLA-4 was relatively higher in the low-expression group ( P = 0.0008), implying that the SERPINE1 gene might be related to the expression regulation of immune checkpoints, thereby affecting tumor immune escape and the response to immunotherapy(Fig. 8 D). 3.8 Drug Sensitivity Analysis When studying the drug TKI-258 ( P = 3e-05), it was observed that the IC50 value of the SERPINE1 high-expression group was significantly higher than that of the low-expression group, suggesting that the high expression of SERPINE1 may lead to a decrease in the sensitivity of cells to TKI-258. Similarly, for LAQ824 ( P = 0.00017), the IC50 value in the SERPINE1 high-expression group was also significantly higher than that in the low-expression group, revealing the correlation between the high expression of SERPINE1 and the decreased sensitivity to LAQ824. In the studies of KN04-2965 ( P = 0.00001) and CP-724714 ( P = 3.7e-05), the IC50 values of the SERPINE1 high-expression group were also significantly higher than those of the low-expression group, indicating that the high expression of SERPINE1 may increase the drug resistance of cells to these two drugs(Fig. 9 A). For the drug Saracatinib ( P = 1.8e-05), the IC50 value of the SERPINE1 high-expression group was significantly higher than that of the low-expression group, showing the correlation between the high expression of SERPINE1 and the drug resistance to Saracatinib. In the studies of drugs such as TAE684, Sunitinib, and TGX221, similar results were also observed, that is, the IC50 values of the SERPINE1 high-expression group were significantly higher than those of the low-expression group, indicating that the high expression of SERPINE1 may reduce the sensitivity of cells to these drugs. For other drugs, such as Dasatinib, CGP − 60474, and Lapatinib, significant differences in drug sensitivity were also shown between the SERPINE1 high-expression group and the low-expression group, further confirming the close relationship between the expression level of the SERPINE1 gene and the sensitivity to various drugs (Fig. 9 B). 3.9 Characterization of Single-Cell Transcriptome This study systematically analyzed the single-cell transcriptomic characteristics of tumor and normal tissues. First, the Harmony algorithm was used to eliminate batch effects (Fig. 10 A). Cells from different samples showed a uniformly mixed state, effectively avoiding technical biases and laying a foundation for quality control. Comparison via UMAP technology (Fig. 10 (B-C)) revealed that the topological structure of tumor tissue cells (SMC-T) was more dispersed than that of normal tissue cells (SMC-N), confirming higher cellular heterogeneity in the tumor microenvironment. Subsequently, both “Principal Component Analysis (PCA) + Uniform Manifold Approximation and Projection (UMAP)” (Fig. 10 D) and “Non-negative Matrix Factorization (NMF) + Uniform Manifold Approximation and Projection (UMAP)” (Fig. 10 E) could effectively separate core cell populations, with methodological differences observed. A panoramic framework of cell composition was constructed through integrated annotation (Fig. 10 F). Analysis of the SERPINE1 gene (Fig. 10 G) showed its high expression in macrophages, which not only supported the reliability of the annotation results but also provided a potential research target. Finally, subpopulation analysis of macrophages (Fig. 10 H) identified multiple differentiated subpopulations, reflecting the functional heterogeneity of macrophages and offering a research direction for exploring the pro-tumor and anti-tumor mechanisms of their subtypes. 3.9 Analysis of SERPINE1 Expression Based on the CCLE database, the expression of SERPINE1 in different colon adenocarcinoma cell lines is shown. The abscissa represents the median expression level of SERPINE1 , and the ordinate lists the names of different cell lines. It can be seen that the expression levels of SERPINE1 vary among different colon adenocarcinoma cell lines. For example, in cell lines such as HCT 116 and SW480, SERPINE1 is expressed to a certain extent, with some cell lines having relatively high expression levels and others having lower levels, reflecting the heterogeneity of SERPINE1 expression in colon adenocarcinoma cell lines. This suggests that the role of SERPINE1 in the occurrence and development of colon adenocarcinoma may be complex, and its functions may differ under different cellular backgrounds (Fig. 11 A).Through qRT-PCR experiments and analysis of the HPA database, the expression of SERPINE1 in COAD was confirmed at both the gene and protein levels. Immunohistochemical images in the HPA database showed that in COAD tissues (patient number 4453), the SERPINE1 protein exhibited strong positive staining, mainly located in the cytoplasm and intercellular matrix, with a relatively high staining intensity, indicating that the SERPINE1 protein is highly expressed in colon adenocarcinoma tissues. In contrast, in normal colon tissues (patient numbers 1958, 1857, 1960), the positive staining of the SERPINE1 protein was weak, and the staining intensity was significantly lower than that in colon adenocarcinoma tissues, suggesting that the expression level of the SERPINE1 protein in normal colon tissues is low. These results are consistent with the trend of high expression of the SERPINE1 gene in colon adenocarcinoma cell lines observed in the qPCR experiment, further supporting the view that SERPINE1 may play an important role in colon adenocarcinoma (Fig. 11 B).The expression level of SERPINE1 in the colon adenocarcinoma cell line Caco-2 was significantly higher than that in the human normal colon epithelial cell line FHC (Fig. 11 C). This indicates that the SERPINE1 gene is highly expressed in colon adenocarcinoma cells and may be associated with the occurrence and development of colon adenocarcinoma. 4 Discussion This study has delved deeply into the pivotal significance of MRGs in COAD. We have successfully constructed a risk model of MRGs and identified SERPINE1 as a crucial risk gene, thus opening up a novel perspective for the prognostic assessment and formulation of treatment strategies for COAD [ 17 ]. The MRG risk model we established has been validated across multiple datasets. Among them, SERPINE1 has emerged prominently as a key risk gene. Its expression level in COAD tissues is significantly higher than that in normal tissues, and it is closely associated with the M (metastasis), T (tumor), and N (node) stages of the tumor. Survival analysis reveals that high expression of SERPINE1 portends a remarkable shortening of both the progression-free survival and overall survival of patients. Univariate and multivariate Cox regression analyses have further corroborated that SERPINE1 is an independent factor influencing the prognosis of COAD patients. This finding is in consonance with previous research results. For instance, Zhang L et al. also discovered that the expression of SERPINE1 in colorectal cancer tissues is distinctly higher than that in normal tissues, and it is closely related to the clinicopathological characteristics and prognosis of patients, thereby further validating the potential of SERPINE1 as a prognostic biomarker for COAD [ 18 ]. In terms of expression differences and prognostic values, our research outcomes are highly consistent with those observed in databases such as TCGA and the GEO. The tumor microenvironment plays an indispensable role in tumor development, and immune infiltration within it exerts a pivotal regulatory function [ 19 , 21 ]. Our study demonstrates that high expression of SERPINE1 is positively correlated with stromal scores, immune scores, and ESTIMATE scores. Moreover, SERPINE1 exhibits significant associations with a diverse array of immune-related genes, particularly those within the Tumor Necrosis Factor Receptor Superfamily (TNFRSF) and Cluster of Differentiation (CD) molecules. There exist notable disparities in the infiltration patterns of immune cells between the high-expression and low-expression groups of SERPINE1 . Zhang Y et al. comprehensively analyzed the expression, prognostic value, and relationship with the tumor microenvironment of SERPINE1 in colorectal cancer and proposed that SERPINE1 holds promise as a prognostic biomarker and therapeutic target for colorectal cancer. Building upon this, we have further explored the mechanism of action of SERPINE1 within the tumor microenvironment, with a particular focus on its impact on immune cell infiltration and the expression of immune checkpoints. The results of our immune infiltration analysis have underscored the potential application value of SERPINE1 in immunotherapy for COAD [ 22 , 23 ]. In this study, Kaplan-Meier survival analysis and Receiver Operating Characteristic (ROC) curve analysis have indicated that SERPINE1 is a potent prognostic biomarker. Nevertheless, the hazard ratio in the Cox regression model only exhibits a moderate level of significance. Through analysis, this may be attributed to the limitations of Kaplan-Meier survival analysis and ROC curve analysis in handling confounding factors and multivariate relationships, while the Cox regression model is susceptible to various factors, such as the interaction of confounding factors, insufficient sample size, and abnormal variable distribution. Notwithstanding, sensitivity analysis has shown that the hazard ratio of SERPINE1 maintains a certain degree of stability across different models. This study has certain limitations. In the future, we intend to augment the sample size, optimize the model, and explore its mechanism of action to contribute to the development of treatment strategies [ 24 ]. During the treatment of COAD, the drug sensitivity of tumor cells directly impacts the treatment efficacy. Our study represents the first systematic analysis of the relationship between the expression of SERPINE1 and drug sensitivity. It has been found that high expression of SERPINE1 is significantly correlated with an increase in the half-maximal inhibitory concentration (IC50) values of multiple drugs, such as TKI-258, LAQ824, and KN04-2965, which implies that it diminishes the sensitivity of COAD cells to these drugs. Although the specific mechanism of SERPINE1 in COAD remains elusive, drawing on relevant studies of other tumors (such as lung cancer), high expression of SERPINE1 may influence drug uptake, metabolism, or intracellular drug targets, which holds significant guiding implications for the clinical treatment of COAD. When comparing our study with previous research, it is essential to recognize that, despite the disparities in sample sources, analytical methods, and research focuses among various studies, they all commonly acknowledge the significant role of SERPINE1 in COAD. For example, the research by Zhang L et al. focused on the impact of SERPINE1 knockdown on the proliferation and migration of human colorectal cancer cells. Through cell-based experiments, we have also observed the functional role of SERPINE1 in Caco-2 cells and explored that SERPINE1 may affect the development process of COAD by regulating signaling pathways related to cell proliferation and migration. We have conducted a comprehensive investigation of SERPINE1 in colorectal cancer and expanded upon this basis by integrating the relationships among SERPINE1 , metabolism-related genes, immune infiltration, and drug sensitivity. These previous studies have facilitated our understanding of the intricate mechanism of SERPINE1 in COAD. They have not only laid the foundation for our research but also provided a direction for future studies. Conversely, our study has not only verified some of the important findings in the existing literature but also, through a comprehensive analysis of these relationships, provided a novel perspective and potential therapeutic targets for precision medicine in COAD [ 25 ]. Nevertheless, this study predominantly relies on bioinformatics analysis and cell experiments. Due to the absence of validation using large-scale clinical samples, the generalizability of the research findings may be restricted to a certain extent. Additionally, although we have identified the associations between SERPINE1 and the tumor microenvironment as well as immune infiltration, the specific molecular mechanisms remain unclear. This deficiency in understanding limits our in-depth comprehension of the functions and potential applications of SERPINE1 . Despite these limitations, this study has still elucidated the significant role of SERPINE1 in the prognostic assessment, regulation of the tumor microenvironment, and drug sensitivity of COAD, laying a theoretical foundation for the precision treatment of COAD and proposing novel potential therapeutic targets. 5 Conclusion This study was designed to comprehensively explore the underlying mechanism of MRGs in COAD. We meticulously constructed and rigorously validated a risk-assessment model, through which SERPINE1 was successfully pinpointed as a pivotal hub gene for risk-prognosis evaluation. SERPINE1 exhibits a profound correlation with tumor staging. Its expression level significantly influences patient prognosis, thereby demonstrating great potential as a reliable biomarker for COAD. Moreover, our findings reveal intricate associations between SERPINE1 and multiple aspects, including the tumor microenvironment, immune response dynamics, and drug sensitivity profiles.While this research is inevitably circumscribed by the inherent limitations of bioinformatics-based analyses, it has nonetheless unequivocally elucidated the critical role of SERPINE1 in COAD. Declarations Competing interests The authors declare no competing interests.Ethics, Consent to Participate, and Consent to Publish declarations: not applicable. Clinical trial number not applicable. Funding We appreciate the funding:Supported by Guizhou Provincial Science and Technology Projects (Grant NO.QianKeHe Basic-[2024] Youth 269) . Author Contribution X.F. and S.C. designed the study and performed the analysis. Q.X., A.G., C.J.,Z.Z.and J.L. performed the validation in the independent cohort. Y.Q. and L.S. and revised the manuscript. All authors reviewed the manuscript All authors read and approved the fnal manuscript. Data Availability The datasets generated during and/or analyzed during the current study are available from the corresponding author upon reasonable request. References Liu Y, Li X, Zhang Y, et al. Metabolic reprogramming in colorectal cancer: mechanisms and therapeutic implications[J]. Mol Cancer. 2020;19(1):123. Wang Y, Chen H, Zhang H, et al. The role of metabolism-related genes in predicting prognosis of colorectal cancer patients[J]. Front Oncol. 2020;10:1358. Hu J, Yang M, Xia Y, et al. Metabolic genes as biomarkers for cancer diagnosis and prognosis: A systematic review and meta-analysis[J]. J Cell Physiol. 2020;235(10):6937–48. Chen X, Zhang X, Li X, et al. The prognostic value of immune infiltration in colorectal adenocarcinoma[J]. Aging. 2020;12(19):19013–28. Xu H, Wang Z, Liu J, et al. Role of plasminogen activator inhibitor-1 in cancer progression and its potential as a therapeutic target[J]. Front Pharmacol. 2020;11:587716. Li X, Huang Y, Wang L, et al. Association between the expression of SERPINE1 and clinicopathological features in colorectal cancer[J]. BMC Cancer. 2021;21(1):308. Zhang Y, Liu X, Yang Y, et al. Immune microenvironment in colorectal cancer: Insights into prognosis and immunotherapy[J]. Front Immunol. 2021;12:631886. Zhao Y, Jiang Y, Wang X, et al. Metabolic gene signature predicts prognosis and guides personalized treatment in colorectal cancer[J]. Front Genet. 2021;12:652100. Guo X, Li Y, Chen Z, et al. The role of SERPINE1 in regulating tumor angiogenesis and metastasis[J]. Oncol Rep. 2021;46(1):1–12. Yang X, Wu J, Liu Y, et al. Comprehensive analysis of immune infiltration and its prognostic value in colorectal adenocarcinoma[J]. Front Cell Dev Biol. 2021;9:668386. Wang H, Chen X, Li X, et al. SERPINE1 as a potential biomarker for predicting prognosis and immunotherapy response in colorectal cancer[J]. J Immunother Cancer. 2022;10(1):e004180. Liu Y, Zhang Y, Chen X, et al. Novel insights into the role of metabolism-related genes in colorectal cancer metastasis[J]. Front Oncol. 2022;12:852086. Li X, Wang Y, Zhang X, et al. The impact of immune cell infiltration on the prognosis of colorectal adenocarcinoma patients. J Transl Med. 2022;20(1):100. Han Y, Zhang Y, Li X, et al. The role of SERPINE1 in the crosstalk between tumor cells and immune cells in colorectal cancer[J]. Front Immunol. 2022;13:887830. Peng Y, Liu X, Zhang Y, et al. Metabolic gene expression profiling and its clinical significance in colorectal cancer[J]. Front Genet. 2022;13:896456. Yang C, Huang X, Liu Z, Qin W, Wang C. Metabolism-associated molecular classification of hepatocellular carcinoma[J]. Mol Oncol. 2020;14(4):896–913. Chen HJ, Li HQ, Liu Z et al. SERPINA1 is a direct estrogen receptor target gene and a predictor of survival in breast cancer patients[J]. Oncotarget 2015,6(28):25815–27. Pan Y, Zhang X, Wang Y, et al. The role of SERPINE1 in colorectal cancer: A bioinformatics analysis[J]. Mol Biol Rep. 2023;50(12):7127–39. Zhang L, Liang J, Chen Y, et al. Knockdown of SERPINE1 inhibits proliferation and migration in human colorectal cancer cells[J]. Oncol Lett. 2019;18(6):6663–9. Tang Y, Xu L, Ren Y, et al. Identification and Validation of a Prognostic Model Based on Three MVI-Related Genes in Hepatocellular Carcinoma. Int J Biol Sci. 2022;18(1):261–75. Zhang Y, Li X, Wang Y, et al. SERPINE1: A potential prognostic biomarker and therapeutic target for colorectal cancer[J]. J Cancer Res Ther. 2022;18(3):458–63. Hu BW, Chen ZW, Wang XG et al. MicroRNA-148a-3p Directly Targets SERPINE1 to Suppress EMT - Mediated Colon Adenocarcinoma Progression[J]. Cancer Manag Res 2021,13:6349–62. Zhang X, Li X, Wang Y, et al. The role of immune infiltrates in predicting prognosis and guiding immunotherapy in colorectal adenocarcinoma[J]. Front Oncol. 2023;13:1117718. Liu X, Peng Y, Zhang Y, et al. Metabolic reprogramming and its potential as a therapeutic target in colorectal cancer[J]. Front Pharmacol. 2023;14:1133806. Huang X, Li X, Yang X, et al. Comprehensive analysis of the relationship between metabolism-related genes and immune infiltration in colorectal cancer[J]. Front Cell Dev Biol. 2023;11:1174180. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8077356","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":551442507,"identity":"69c6659a-1aab-4575-9843-66c0a576ccbd","order_by":0,"name":"Xiaoyun Feng","email":"","orcid":"","institution":"Shizhen College of Guizhou University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Xiaoyun","middleName":"","lastName":"Feng","suffix":""},{"id":551442508,"identity":"563f35d5-56f3-4232-9ded-2ee7b19a92f2","order_by":1,"name":"Siyuan Chen","email":"","orcid":"","institution":"Shizhen College of Guizhou University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Siyuan","middleName":"","lastName":"Chen","suffix":""},{"id":551442509,"identity":"fe673696-a6d2-4dee-9dd5-9366f9899b17","order_by":2,"name":"Qian Xie","email":"","orcid":"","institution":"Shizhen College of Guizhou University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Qian","middleName":"","lastName":"Xie","suffix":""},{"id":551442510,"identity":"e5db7bcd-609a-43dc-b523-a14f222ebda4","order_by":3,"name":"Chenjie Jiang","email":"","orcid":"","institution":"Shizhen College of Guizhou University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Chenjie","middleName":"","lastName":"Jiang","suffix":""},{"id":551442511,"identity":"8a6f6c9d-809e-4480-8338-b5d740ad7805","order_by":4,"name":"Jianyan Lu","email":"","orcid":"","institution":"Shizhen College of Guizhou University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Jianyan","middleName":"","lastName":"Lu","suffix":""},{"id":551442512,"identity":"21777660-4017-4bc1-99de-8297812c07f4","order_by":5,"name":"Ai Gao","email":"","orcid":"","institution":"Shizhen College of Guizhou University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Ai","middleName":"","lastName":"Gao","suffix":""},{"id":551442513,"identity":"1c9b4f40-7ade-4db2-8796-9b757c5189f4","order_by":6,"name":"Zhourui Zheng","email":"","orcid":"","institution":"Shizhen College of Guizhou University of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Zhourui","middleName":"","lastName":"Zheng","suffix":""},{"id":551442514,"identity":"b1909675-2f0b-4995-b091-66c49094ab1e","order_by":7,"name":"Yufeng Qin","email":"","orcid":"","institution":"Guizhou Medical University","correspondingAuthor":false,"prefix":"","firstName":"Yufeng","middleName":"","lastName":"Qin","suffix":""},{"id":551442515,"identity":"39d5043f-7217-42b2-8b3a-70d0c3f92f39","order_by":8,"name":"Lijuan Shao","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAzElEQVRIiWNgGAWjYBACefnDBx9+MLCp72dvIFKL4Qy2ZGOJgjTGmT0HiLXmBo+ZBM+Hw4wbbiQQqYNxdo+ZhITBYWaGm4833mCosYkmqIVd5lixRYFBOhvj7LRiC4ZjabkNBG1pSN54Q8LAmodZOsdMgrHhMGEtDAcSDCR4DJgl2CTPEKvlRooRUIuzAY8ED5FaDHuOAQPZIC1BggfolwRi/CLP3gyMyj82CfbHD2+88aHGhgiHIQEDiQRSlEO0kKpjFIyCUTAKRgYAAAbzPnwf+r2TAAAAAElFTkSuQmCC","orcid":"","institution":"The Jinyang Hospital Affiliated to Guizhou Medical University, The Second People's Hospital of Guiyang","correspondingAuthor":true,"prefix":"","firstName":"Lijuan","middleName":"","lastName":"Shao","suffix":""}],"badges":[],"createdAt":"2025-11-10 13:08:11","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8077356/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8077356/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":96944610,"identity":"e1e1765c-ba3c-49aa-b322-1f975369a039","added_by":"auto","created_at":"2025-11-27 19:34:23","extension":"doc","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8590545,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.doc","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/a892ca8efc2ca3bd8aa73315.doc"},{"id":97136111,"identity":"17290c07-a817-44c6-a3a8-2d530f2f603e","added_by":"auto","created_at":"2025-12-01 09:55:35","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":10276,"visible":true,"origin":"","legend":"","description":"","filename":"832dfa9d68854714b26917a38c2ed14d.json","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/2c10dc8ba15845042cfbab1f.json"},{"id":97137330,"identity":"d6e77908-85fe-4b96-a3f2-6216a69167f9","added_by":"auto","created_at":"2025-12-01 09:57:35","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":111089,"visible":true,"origin":"","legend":"","description":"","filename":"832dfa9d68854714b26917a38c2ed14d1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/30e3434712c68f6b485e8615.xml"},{"id":96944606,"identity":"afe9b5c7-bbbb-4fb9-a8db-642c6d0706e8","added_by":"auto","created_at":"2025-11-27 19:34:23","extension":"jpeg","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6744712,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/37f73bf93f17c7ddf6c47179.jpeg"},{"id":97137346,"identity":"5fee1a30-1599-4205-ad08-6a3ef18f243b","added_by":"auto","created_at":"2025-12-01 09:57:36","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5405704,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage10.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/a5691f4bac2a5725dee08842.jpeg"},{"id":96944614,"identity":"2d9926c4-aa58-4f02-bd9c-a732b6a5e641","added_by":"auto","created_at":"2025-11-27 19:34:23","extension":"jpeg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6676240,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage11.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/a0300e8080848323e94605e4.jpeg"},{"id":97136577,"identity":"63f57d8b-643c-4171-b8e3-38b36ae11c74","added_by":"auto","created_at":"2025-12-01 09:56:47","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5489392,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/a03a1ef0ea47332c7e74841e.jpeg"},{"id":97136706,"identity":"4d7334c0-5513-430e-a6f8-392b0d650454","added_by":"auto","created_at":"2025-12-01 09:56:55","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4213674,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/e28fa3fc1fdd1c17ae05c666.jpeg"},{"id":96944624,"identity":"5e1c6832-fb92-467c-9623-3b8837e6c701","added_by":"auto","created_at":"2025-11-27 19:34:23","extension":"jpeg","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5542752,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/0546c83ee553d9286f7814bc.jpeg"},{"id":96944638,"identity":"e137cea7-2504-4769-89f3-9503a681a335","added_by":"auto","created_at":"2025-11-27 19:34:23","extension":"jpeg","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2312888,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/30e2dad00eccf8cc6043e3e9.jpeg"},{"id":96944626,"identity":"31080980-9f19-4620-a188-594962cd2eae","added_by":"auto","created_at":"2025-11-27 19:34:23","extension":"jpeg","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6832204,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/de273bfffe1bbc999909e0d1.jpeg"},{"id":97136822,"identity":"a6259585-a946-4a52-aad8-57dc2d6a5f49","added_by":"auto","created_at":"2025-12-01 09:57:02","extension":"jpeg","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":957272,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage7.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/3df3bbe42f362a25bc1f226c.jpeg"},{"id":97136078,"identity":"84a3c6c3-a6ce-4138-a879-4fa6ba919f63","added_by":"auto","created_at":"2025-12-01 09:55:25","extension":"jpeg","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5975392,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage8.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/2cbcb966ed9af44ce361071a.jpeg"},{"id":96944627,"identity":"34ca0e9a-5800-41c1-8128-0abb2708d2ca","added_by":"auto","created_at":"2025-11-27 19:34:23","extension":"jpeg","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5539304,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage9.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/04046cfcd6c5a289ea56c226.jpeg"},{"id":97136836,"identity":"1f637c75-ad42-4b3c-aa3c-f6ff676dea99","added_by":"auto","created_at":"2025-12-01 09:57:02","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":128579,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/3f0315ed14b4fa3a5cfdf889.png"},{"id":97135962,"identity":"c11bd80b-2930-4298-ad99-8ae66f2b292a","added_by":"auto","created_at":"2025-12-01 09:54:43","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":119782,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage10.png","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/13313f4f146d412ad3a007eb.png"},{"id":97136056,"identity":"dc284a88-3afa-462c-972a-500bc90cff32","added_by":"auto","created_at":"2025-12-01 09:55:15","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":235686,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage11.png","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/e9b93e9c4a609dbc63ccacd1.png"},{"id":96944623,"identity":"db1f178e-78e4-4b5f-97fb-806cca12fe06","added_by":"auto","created_at":"2025-11-27 19:34:23","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":46704,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/dfa3eb6f31c401cb37a723f0.png"},{"id":96944620,"identity":"862984a5-3b59-4ee6-b640-7c6bdee42d64","added_by":"auto","created_at":"2025-11-27 19:34:23","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":46376,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/6be9e9af4d1c188e15ca89e7.png"},{"id":96944637,"identity":"2287d789-eb4e-4c76-8036-082a1f2b0dce","added_by":"auto","created_at":"2025-11-27 19:34:23","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1608075,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/73775650774c3b54ed806d43.png"},{"id":97135801,"identity":"4ffca904-d3f7-4cd7-a7f9-8332a0c585ab","added_by":"auto","created_at":"2025-12-01 09:53:44","extension":"png","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":734059,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/c7ff48f8ed96cc671207cc0d.png"},{"id":96944632,"identity":"5dd2d926-da1a-4563-bd78-1b4ebef6685f","added_by":"auto","created_at":"2025-11-27 19:34:23","extension":"png","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":115598,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/6db9aae6b05c5cab90b4d51b.png"},{"id":96944629,"identity":"46577486-0442-422a-af57-dee9ff7c48e1","added_by":"auto","created_at":"2025-11-27 19:34:23","extension":"png","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":282995,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/6dbf4e910db4f3c938f73792.png"},{"id":97136611,"identity":"b8bd2dbf-4263-455a-8969-48594dc2974d","added_by":"auto","created_at":"2025-12-01 09:56:49","extension":"png","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":145317,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/e00fcd6c9cd4b3880754f75e.png"},{"id":96944634,"identity":"cda999a2-9f64-425e-ab1d-bb1127d6dc77","added_by":"auto","created_at":"2025-11-27 19:34:23","extension":"png","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1473867,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/b31f847f6d7ade948c6d0433.png"},{"id":97137051,"identity":"418a1833-4906-495e-9dcc-fb1778ac0042","added_by":"auto","created_at":"2025-12-01 09:57:19","extension":"xml","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":111826,"visible":true,"origin":"","legend":"","description":"","filename":"832dfa9d68854714b26917a38c2ed14d1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/ad5fbffcc240bdb9ddd1255d.xml"},{"id":96944636,"identity":"bce26e7e-7814-4f24-9a5e-46237cb713d4","added_by":"auto","created_at":"2025-11-27 19:34:23","extension":"html","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":125325,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/8662a64b6aa1b16cc036e864.html"},{"id":96944600,"identity":"c723ce43-8a56-4f63-87f2-55512649b872","added_by":"auto","created_at":"2025-11-27 19:34:22","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":931622,"visible":true,"origin":"","legend":"\u003cp\u003eExpression of genes related to the prognosis of COAD and risk assessment.\u003cstrong\u003eA\u003c/strong\u003e Volcano Plot.\u003cstrong\u003eB\u003c/strong\u003e Forest Plot.\u003cstrong\u003eC\u003c/strong\u003e Relationship graph between Partial Likelihood Deviance and Log Lambda.\u003cstrong\u003eD\u003c/strong\u003e Coefficients Path.\u003cstrong\u003eE \u003c/strong\u003eOverall Survival Analysis of the GSE38832 dataset.\u003cstrong\u003eF \u003c/strong\u003eOverall Survival Analysis of the TCGA dataset.\u003cstrong\u003eG \u003c/strong\u003eProgression Free Survival Analysis of the TCGA dataset.\u003cstrong\u003eH\u003c/strong\u003e Univariate Forest Plot.\u003cstrong\u003eI \u003c/strong\u003eMultivariate Forest Plot.\u003cstrong\u003eJ\u003c/strong\u003eTime-dependent ROC Curve.\u003cstrong\u003eK \u003c/strong\u003eROC Curve of the Random Forest Model.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/08f821980f34e52eec7972df.png"},{"id":97135776,"identity":"c53b77e3-946c-438c-9a43-638c08f6ca45","added_by":"auto","created_at":"2025-12-01 09:53:35","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":646549,"visible":true,"origin":"","legend":"\u003cp\u003eThe risk score of each patient with COAD\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/5fe9b46858336e2f1d65c536.png"},{"id":97136820,"identity":"4d386d93-fa6d-4045-9596-20c996c5cc0b","added_by":"auto","created_at":"2025-12-01 09:57:02","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":286663,"visible":true,"origin":"","legend":"\u003cp\u003eIdentification of DEGs in Metabolic Risk Prognosis\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/4d31a973b202d69d35296c77.png"},{"id":97136575,"identity":"ab8e9dbe-76c5-4155-a299-4083dd59c40c","added_by":"auto","created_at":"2025-12-01 09:56:47","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":4814141,"visible":true,"origin":"","legend":"\u003cp\u003eDifferential expression of the \u003cem\u003eSERPINE1\u003c/em\u003e gene in normal tissues and tumor tissues and its prognostic analysis.\u003cstrong\u003eA\u003c/strong\u003e The difference in the expression of the \u003cem\u003eSERPINE1\u003c/em\u003egene between normal tissues and tumor tissues.\u003cstrong\u003eB\u003c/strong\u003e Paired analysis of the expression of the \u003cem\u003eSERPINE1\u003c/em\u003e gene in normal tissues and tumor tissues.\u003cstrong\u003eC-D\u003c/strong\u003eProgression-free survival and overall survival analyses of patients with high and low expressions of the \u003cem\u003eSERPINE1\u003c/em\u003e gene are presented.E Univariate and multivariate forest plots.\u003cstrong\u003eF-K\u003c/strong\u003e Some parts have analyzed in detail the expression of the \u003cem\u003eSERPINE1\u003c/em\u003e gene in different age groups (\u0026lt; 65 years old and ≥ 65 years old), genders (female and male), M stages (M0 and M1), T stages (T1, T2, T3, T4, T4a), different clinical stages (Stage I, Stage II, Stage III, Stage IV), and N stages (N0, N1, N2).\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/141a3fc509eb85b7e745cc74.png"},{"id":96944604,"identity":"ce47afd5-a494-489c-8e5b-fccbc5e48e24","added_by":"auto","created_at":"2025-11-27 19:34:22","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1979672,"visible":true,"origin":"","legend":"\u003cp\u003eConstruction and Validation.A Nomogram used for predicting patient prognosis.B Calibration curve of the OS rate predicted by the nomogram and the actually observed overall survival rate.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/9f71e839019bf5828f9c4798.png"},{"id":97135763,"identity":"1843f2a7-61f2-4c30-9dc1-ae1995967549","added_by":"auto","created_at":"2025-12-01 09:53:28","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1271354,"visible":true,"origin":"","legend":"\u003cp\u003eDifferential Expression and Enrichment Analysis of \u003cem\u003eSERPINE1\u003c/em\u003e.\u003cstrong\u003eA\u003c/strong\u003e Heatmap of DEGs.\u003cstrong\u003eB\u003c/strong\u003eGO-KEGG analysis.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/14a8e4aa1c7b3e4ed9afe922.png"},{"id":97136912,"identity":"00d1a8a1-d1e3-4a70-a99f-6cc0395cf5fc","added_by":"auto","created_at":"2025-12-01 09:57:06","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":845564,"visible":true,"origin":"","legend":"\u003cp\u003eRelationship between \u003cem\u003eSERPINE1\u003c/em\u003e Gene Expression Level and TME Score\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/bfc6cd9bf169600f56487856.png"},{"id":96944611,"identity":"fc242f67-489d-499c-99d8-a9b3b3ce0398","added_by":"auto","created_at":"2025-11-27 19:34:23","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":1093243,"visible":true,"origin":"","legend":"\u003cp\u003eAnalysis of the Relationship between \u003cem\u003eSERPINE1\u003c/em\u003e and Immune Cells and Related Indicators.\u003cstrong\u003eA \u003c/strong\u003eHeatmap of the correlation of immune cells.\u003cstrong\u003eB\u003c/strong\u003e Proportional distribution of different immune cells in the low-expression and high-expression groups of \u003cem\u003eSERPINE1\u003c/em\u003e.\u003cstrong\u003eC \u003c/strong\u003eScatter plot of the correlation between \u003cem\u003eSERPINE1\u003c/em\u003e expression and immune cells and related indicators.\u003cstrong\u003eD\u003c/strong\u003e Differences in the expression of immune-related genes in the low-expression and high-expression groups of \u003cem\u003eSERPINE1\u003c/em\u003e.\u003c/p\u003e","description":"","filename":"floatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/4cd5ae4d3c2a6b5e589dd15a.png"},{"id":97137850,"identity":"42b21cb5-46fa-4900-81f7-fd4d2288d12b","added_by":"auto","created_at":"2025-12-01 09:58:14","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":4962928,"visible":true,"origin":"","legend":"\u003cp\u003eRelationship between drug sensitivity and the expression level of \u003cem\u003eSERPINE1\u003c/em\u003e.\u003c/p\u003e","description":"","filename":"floatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/f93eec7db8098c3d739c34dd.png"},{"id":96944616,"identity":"d75a7ca7-3dff-41b5-a534-3b2f27d6e557","added_by":"auto","created_at":"2025-11-27 19:34:23","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":583339,"visible":true,"origin":"","legend":"\u003cp\u003eAnalysis of Single-Cell transcriptomic features.\u003cstrong\u003eA \u003c/strong\u003eBatch effect correction of samples (harmony dimensionality reduction).\u003cstrong\u003eB \u003c/strong\u003eCell distribution in normal/tumor tissues (umap dimensionality reduction).\u003cstrong\u003eC \u003c/strong\u003eCellular heterogeneity among tissue types (umap dimensionality reduction)\u003cstrong\u003e.D \u003c/strong\u003eCell type clustering via pca-umap.\u003cstrong\u003ee \u003c/strong\u003ecell type clustering via nmf-umap.\u003cstrong\u003eF \u003c/strong\u003eFinal Cell Type Annotation (UMAP).\u003cstrong\u003eG \u003c/strong\u003eCell type-specific expression characteristics of the \u003cem\u003eSERPINE1\u003c/em\u003e gene.\u003cstrong\u003eH \u003c/strong\u003eUMAP-Based subpopulation subdivision of macrophages.\u003c/p\u003e","description":"","filename":"floatimage10.png","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/0018e20a54be6bb2af5e8d25.png"},{"id":97136871,"identity":"ca3d394a-ca3d-41f8-9d08-7b68f931a500","added_by":"auto","created_at":"2025-12-01 09:57:04","extension":"png","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":1027480,"visible":true,"origin":"","legend":"\u003cp\u003eExpression of \u003cem\u003eSERPINE1\u003c/em\u003ein Different Cell Lines and Tissues.\u003cstrong\u003eA \u003c/strong\u003eExpression of \u003cem\u003eSERPINE1\u003c/em\u003e in COAD-related cell lines in the CCLE database.\u003cstrong\u003eB\u003c/strong\u003e Immunohistochemical staining results of \u003cem\u003eSERPINE1\u003c/em\u003e in different tissues.\u003cstrong\u003eC\u003c/strong\u003e Expression levels of \u003cem\u003eSERPINE1\u003c/em\u003e in FHC and Caco-2 cell lines.\u003c/p\u003e","description":"","filename":"floatimage11.png","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/18d5b678b81bce560497953b.png"},{"id":99310787,"identity":"d8b1bc9b-7f80-45bd-a771-78f6642083e0","added_by":"auto","created_at":"2025-12-31 16:13:22","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":19723189,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8077356/v1/bcafa300-b246-40a8-9994-d8ff66ded5fe.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Investigation of the Metabolism-Related Prognostic Gene SERPINE1 as a Prognostic Predictor in Colorectal Adenocarcinoma and Its Regulatory Mechanisms Underlying Immune Infiltration","fulltext":[{"header":"1 Introduction","content":"\u003cp\u003eColorectal adenocarcinoma (COAD) represents a formidable global health conundrum, imposing a significant burden on morbidity and mortality rates. As one of the most pervasive malignancies, its incidence has been on a consistent upward trajectory [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. The prognoses of COAD patients display pronounced heterogeneity, thereby underscoring the imperative need to identify reliable prognostic biomarkers. These biomarkers are instrumental in steering therapeutic strategies and enhancing patient outcomes [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eIn recent years, there has been a burgeoning interest in the role of metabolism-associated genes (MRGs) within oncological research [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Metabolism, a fundamental biological mechanism that supplies cells with energy and essential molecular constituents, frequently experiences dysregulation in cancer cells. This dysregulation precipitates disruptions in energy metabolism, biosynthesis, and redox equilibrium [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. MRGs assume pivotal roles in these processes and hold substantial promise as biomarkers for cancer diagnosis and prognosis [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eIn the current investigation, we constructed a risk model associated with MRGs and conducted differential gene analysis to identify the key risk gene, \u003cem\u003eSERPINE1\u003c/em\u003e. Also known as plasminogen activator inhibitor-1, \u003cem\u003eSERPINE1\u003c/em\u003e is a serine protease inhibitor that regulates the activity of plasminogen activators [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. It is implicated in a plethora of physiological and pathological processes, including blood coagulation, fibrinolysis, angiogenesis, and inflammation [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eWithin the oncology domain, \u003cem\u003eSERPINE1\u003c/em\u003e has been shown to exert multifarious effects on tumor progression. Firstly, it can impede extracellular matrix degradation by inhibiting the activity of plasminogen activators involved in matrix breakdown, potentially curtailing tumor cell invasion and metastasis. Secondly, \u003cem\u003eSERPINE1\u003c/em\u003e can promote angiogenesis by modulating the activity of vascular endothelial growth factor and other angiogenic mediators [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Given that angiogenesis is indispensable for tumor growth and metastasis, as it provides oxygen and nutrients to neoplastic cells, this aspect is of utmost importance. Additionally, \u003cem\u003eSERPINE1\u003c/em\u003e is intricately linked to inflammation, which plays a crucial role in cancer development and progression. Inflammatory cells can secrete cytokines and growth factors that facilitate tumor growth and dissemination [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. Moreover, \u003cem\u003eSERPINE1\u003c/em\u003e has been reported to be associated with poor prognoses in several cancer types, such as breast, lung, and ovarian cancers. In these malignancies, elevated \u003cem\u003eSERPINE1\u003c/em\u003e expression levels correlate with advanced tumor stages, lymph node metastasis, and distant dissemination. However, the prognostic significance of \u003cem\u003eSERPINE1\u003c/em\u003e in COAD remains ambiguous [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eBeyond its role in tumor progression, \u003cem\u003eSERPINE1\u003c/em\u003e may also be associated with immune cell infiltration in COAD [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Immune cell infiltrates, consisting of various immune cell populations within the tumor microenvironment, can interact with cancer cells and stromal components, thereby regulating tumor growth, metastasis, and the response to therapy [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. Recent studies have revealed that immune cell infiltrates can serve as prognostic biomarkers and therapeutic targets in COAD [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. For example, T cells, B cells, natural killer cells, macrophages, and dendritic cells are key immune cell types that can infiltrate the tumor microenvironment [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. However, the relationship between \u003cem\u003eSERPINE1\u003c/em\u003e and immune cell infiltrates in COAD remains incompletely understood. \u003cem\u003eSERPINE1\u003c/em\u003e may regulate the recruitment and activation of immune cells in the tumor microenvironment [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. For instance, it may influence the expression of chemokines and cytokines that attract immune cells to the tumor site. It may also modulate immune cell function by interacting with immune receptors or signaling pathways. Elucidating this relationship could provide novel insights into the pathogenesis and prognosis of COAD [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eThe objective of this study was to explore the prognostic significance of \u003cem\u003eSERPINE1\u003c/em\u003e in COAD and its association with immune cell infiltrates. We analyzed the correlation between \u003cem\u003eSERPINE1\u003c/em\u003e expression and clinicopathological characteristics, including tumor stage, lymph node metastasis, and distant metastasis. Additionally, we evaluated the prognostic value of \u003cem\u003eSERPINE1\u003c/em\u003e expression in COAD patients through survival analysis and assessed the independent prognostic significance of \u003cem\u003eSERPINE1\u003c/em\u003e expression along with other prognostic factors (such as age, gender, and tumor stage) using a multivariate Cox proportional hazards model.\u003c/p\u003e\u003cp\u003eThis study provides novel insights into the prognostic role of MRGs in COAD and helps to identify \u003cem\u003eSERPINE1\u003c/em\u003e as a potential prognostic biomarker for COAD patients. It also uncovers the relationship between \u003cem\u003eSERPINE1\u003c/em\u003e and immune cell infiltrates in COAD, which may accelerate the development of innovative immunotherapeutic approaches. In conclusion, this study represents a significant advancement in understanding the role of the metabolism - related risk gene \u003cem\u003eSERPINE1\u003c/em\u003e as a prognostic biomarker for COAD and its association with immune cell infiltrates. The findings of this study may have far-reaching implications for the diagnosis, prognosis, and treatment of this disease.\u003c/p\u003e"},{"header":"2 Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1 Data Download\u003c/h2\u003e\u003cp\u003eIn this study, COAD data (FPKM, Fragments Per Kilobase of exon per Million reads mapped) were downloaded from The Cancer Genome Atlas (TCGA) database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://portal.gdc.cancer.gov/\u003c/span\u003e\u003cspan address=\"https://portal.gdc.cancer.gov/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). This data set included 473 tumor samples and 41 adjacent normal samples. Transcripts Per Million (TPM) values were calculated from the FPKM values, followed by a log2 transformation for data normalization. The gene expression dataset GSE38832 was retrieved from the Gene Expression Omnibus (GEO) database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ncbi.nlm.nih.gov/geo/\u003c/span\u003e\u003cspan address=\"https://www.ncbi.nlm.nih.gov/geo/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), which consisted of 118 COAD tissue samples. During data analysis and collation, Perl language (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.perl.org/\u003c/span\u003e\u003cspan address=\"http://www.perl.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and R software (version 4.2.0, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.r-project.org\u003c/span\u003e\u003cspan address=\"https://www.r-project.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) were employed. The MRGs were derived from 2752 previously published genes related to metabolism[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e], which encode all known human metabolic enzymes and transporters.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2 Cell Culture Technique\u003c/h2\u003e\u003cp\u003eBoth the human colon epithelial cell line FHC and the colon adenocarcinoma cell line Caco-2 adopted in this study were purchased from Wuhan Procell Life Technology Co., Ltd. The cultivation of both FHC cells and Caco-2 cells was conducted in a composite medium composed of 90% DMEM basal medium, 10% fetal bovine serum, and 1% penicillin and streptomycin. The culture conditions were set as a constant temperature environment of 37\u0026deg;C and a CO₂ concentration of 5% to simulate the internal environment of the human body.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e2.3 Quantitative Real-time Reverse Transcription Polymerase Chain Reaction (qRT-PCR) Experiment\u003c/h2\u003e\u003cp\u003eIn this experiment, the FastPure\u0026reg; Cell Total RNA Isolation Kit (RC112, Vazyme, Nanjing, China) was used. According to the operation instructions provided by the manufacturer, total RNA was successfully extracted from the cells. Subsequently, the PrimeScript RT Reagent Kit (RR037A, Takara Bio, Kyoto, Japan) was employed to synthesize cDNA from the RNA. For the RT-PCR reaction, SYBR Green Mix (4309155, Thermo Fisher Scientific, USA) was adopted. The sequences of \u003cem\u003eSERPINE1\u003c/em\u003e and GAPDH were designed and applied (the forward primer sequence of GAPDH is 5\u0026rsquo;-GCACCGTCAAGGCTGAG-AAC-3\u0026rsquo;, and its reverse primer sequence is 5\u0026rsquo;-TGGTGAAGACGCCAGTGGA-3\u0026rsquo;; the forward primer sequence of \u003cem\u003eSERPINE1\u003c/em\u003e is 5\u0026rsquo;-CCGCCGCCTCTTCCACAA-ATC-3\u0026rsquo;, and its reverse primer sequence is 5\u0026rsquo;-TAGGGCAGTTCCAGGATGTCGT-AG-3\u0026rsquo;).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003e2.3 Construction and Validation of the MRGs Prognostic Index (MRGRI)\u003c/h2\u003e\u003cp\u003eThis study focused on constructing a MRGRI and calculating the risk score, with the formula: Risk score = (β₁\u0026times;Exp₁) + (β₂\u0026times;Exp₂) + \u0026hellip; + (βₙ\u0026times;Expₙ) (where β represents the gene regression coefficient and Exp represents the gene expression level)[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. Firstly, the study retrieved the gene expression data and clinical follow-up information of COAD patients from the TCGA and GEO databases. For the 2,752 MRGs, the limma tool was used to process the expression values of duplicate genes. Using the \"limma\" R package, 159 Differentially Expressed MRGs (DEMRGs) were screened out with the criteria of |log₂ fold change (FC)| \u0026gt;1.5 and an adjusted P - value\u0026thinsp;\u0026le;\u0026thinsp;0.05, and a volcano plot was drawn.Subsequently, the \"survival\" R package was used to standardize the survival time data. A multivariate COX regression model was constructed with \"futime\" and \"fustat\" as the dependent variables. After optimization by the step() function, the risk score was calculated. The high - and low - risk groups were divided based on the median value. After censoring the outliers, the distribution plot of risk scores, the scatter plot of survival status, and the heatmap of gene expression were drawn. Combining univariate Cox regression and LASSO regression analysis using the \"glmnet\" package, by adjusting the penalty parameter λ, 13 core prognostic genes were screened out and their weights were determined.In the model evaluation stage, the \"survminer\" and \"survival\" packages were used to draw Kaplan - Meier survival curves. Univariate and multivariate COX regression analyses (presented as forest plots) were conducted to verify the independent prognostic value of the risk score. The pROC package was used to construct the Receiver Operating Characteristic (ROC) curve and calculate the Area Under the Curve (AUC) to evaluate the predictive efficacy.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003e2.4 Expression and Prognosis Analysis of \u003cem\u003eSERPINE1\u003c/em\u003e in COAD\u003c/h2\u003e\u003cp\u003eThe \"ggplot2\" package in R language was utilized to draw box plots, aiming to compare the expression differences of the \u003cem\u003eSERPINE1\u003c/em\u003e gene between COAD tissues and adjacent normal tissues. The \"stats\" R package was employed to conduct difference statistical tests. Meanwhile, the \"ggplot2\" package was also used to draw paired sample analysis plots, so as to display the correlation of the \u003cem\u003eSERPINE1\u003c/em\u003e gene expression in tumor tissues and adjacent normal tissues of the same patient.Based on the median of the \u003cem\u003eSERPINE1\u003c/em\u003e gene expression level, patients were divided into a high-expression group and a low-expression group. The \"survminer\" package and the \"survival\" package were adopted to draw progression-free survival curves for comparing the progression-free survival status of patients in these two groups. Similarly, the overall survival curves were plotted using the \"survminer\" package and the \"survival\" package to compare the overall survival status of the two groups of patients.The \"survminer\" package was used to conduct univariate Cox regression analysis to evaluate the impacts of factors such as the \u003cem\u003eSERPINE1\u003c/em\u003e gene expression, age, gender, and tumor stage on patient survival. On the basis of the univariate analysis, the \"survminer\" package was further employed to carry out multivariate Cox regression analysis by incorporating multiple factors for comprehensive analysis, with the aim of determining the independent impact of the \u003cem\u003eSERPINE1\u003c/em\u003e gene expression on patient survival.The \"ggplot2\" package was used to draw box plots. Patients were grouped according to age (\u0026le;\u0026thinsp;65 years old and \u0026gt;\u0026thinsp;65 years old), gender (male and female), M stage (M0 and M1), T stage (T1, T2, T3, T4, T4a), clinical stage (Stage I, Stage II, Stage III, Stage IV), and N stage (N0, N1, N2), and the expression differences of the \u003cem\u003eSERPINE1\u003c/em\u003e gene in different subgroups were compared.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003e2.5 Identification of MRGs Risk DEGs\u003c/h2\u003e\u003cp\u003eIn this study, a risk model based on DEGs of MRGs was constructed, and the fold change in the expression of MRGs risk genes was calculated using the \"limma\" package in R software. Based on the preset thresholds (logFC\u0026thinsp;\u0026gt;\u0026thinsp;1 and q-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05), the MRGs risk DEGs were screened out, and the \"ggplot2\" package was adopted to draw a volcano plot for visual display. The screened MRGs risk DEGs were input into the STRING database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://string-db.org/\u003c/span\u003e\u003cspan address=\"https://string-db.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) to obtain the interaction data among genes. Subsequently, Cytoscape 3.7.2 software was used to draw a gene interaction.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003e2.6 Construction and Calibration of the Nomogram\u003c/h2\u003e\u003cp\u003eIn terms of nomogram construction, the coxph function is used to build a Cox model based on all variables including the expression level of the \u003cem\u003eSERPINE1\u003c/em\u003e gene for multivariate analysis. The regplot function is utilized to generate the nomogram, which incorporates a scoring system for each variable, consistent with the variable score allocation in the figure. The parameter setting failtime\u0026thinsp;=\u0026thinsp;c(1,3,5) enables the prediction of 1-year, 3-year, and 5-year survival rates, and the risk scores are calculated and output as a file. During the calibration curve validation, the cph and calibrate functions are called three times to generate calibration curves for the corresponding time points respectively.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003e2.7 Analysis of the Expression Difference of \u003cem\u003eSERPINE1\u003c/em\u003e and Its Enriched Pathways\u003c/h2\u003e\u003cp\u003eThe \"limma\" software package was utilized to conduct differential gene analysis. The differences in gene expression between the high-expression group and the low-expression group of \u003cem\u003eSERPINE1\u003c/em\u003e were compared, and the DEGs were screened out (with the threshold set as logFC\u0026thinsp;\u0026gt;\u0026thinsp;1 and adj.P.Val\u0026thinsp;\u0026lt;\u0026thinsp;0.05). The \"pheatmap\" software package was employed to draw a heatmap, aiming to display the expression patterns of the DEGs in the two groups.The screened DEGs were then imported into the \"clusterProfiler\" software package for functional enrichment analysis, which encompassed Gene Ontology (GO) enrichment analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003e2.8 Tumor Microenvironment (TME) Scores\u003c/h2\u003e\u003cp\u003eThe ESTIMATE algorithm was adopted to calculate the Stromal Score, Immune Score, and ESTIMATE Score for each patient's sample, with the aim of evaluating the abundance and functional activity of stromal cells and immune cells in the TME. The Wilcoxon rank-sum test was employed to analyze the differences in the Stromal Score, Immune Score, and ESTIMATE Score between the low-expression group and the high-expression group of \u003cem\u003eSERPINE1\u003c/em\u003e. This test was executed using the wilcox.test function in the stats package.Furthermore, the ggplot2 package was utilized to draw Violin Plots to showcase the score values under different scoring metrics.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003e2.9 Immune Correlation Analysis\u003c/h2\u003e\u003cp\u003eA quantitative analysis was conducted on the association between the expression of the \u003cem\u003eSERPINE1\u003c/em\u003e gene and that of immune-related genes by calculating their correlation coefficients. The \"pheatmap\" software package was employed to draw a correlation heatmap, aiming to visually display the correlation patterns among gene expressions.Furthermore, through the CIBERSORT algorithm, the infiltration proportions of immune cell subsets in different samples were calculated. Subsequently, the \"ggplot2\" software package was utilized to draw box plots, which enabled the screening of immune cell subsets that exhibited significant correlations with the expression of the \u003cem\u003eSERPINE1\u003c/em\u003e gene. Additionally, correlation scatter plots were also drawn to further explore the relationship between the expression of the \u003cem\u003eSERPINE1\u003c/em\u003e gene and immune cell subsets.To analyze the expression differences of immune checkpoint genes (PD-1, CTLA-4) in the high-expression group and the low-expression group of the \u003cem\u003eSERPINE1\u003c/em\u003e gene, the corresponding expression data were extracted, and violin plots were drawn using the \"ggplot2\" software package.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003e2.10 Drug Sensitivity Analysis\u003c/h2\u003e\u003cp\u003eBased on the datasets of clinical trials and drugs approved by the FDA, the Pearson correlation coefficients between the expression of the \u003cem\u003eSERPINE1\u003c/em\u003e gene and various drugs were calculated, and the obtained results were then screened. The \u0026ldquo;pRRophetic\u0026rdquo; R package was utilized to evaluate the half-maximal inhibitory concentration of drugs in colorectal cancer patients. For each drug, the differences in drug sensitivity between the low-expression group and the high-expression group of the \u003cem\u003eSERPINE1\u003c/em\u003e gene were compared. The Wilcoxon rank-sum test was employed to assess the statistical significance of the differences in drug sensitivity between the two groups. Through the \u0026ldquo;ggplot2\u0026rdquo; package, box plots were drawn to display the distribution of drug sensitivity for each drug in the low-expression group and the high-expression group of the \u003cem\u003eSERPINE1\u003c/em\u003e gene.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003e2.11 Methods for single-cell transcriptome data processing and analysis\u003c/h2\u003e\u003cp\u003eThe Harmony algorithm was adopted to correct batch effects in the gene expression matrix, and the spatial uniformity of sample points was observed to guarantee the reliability of subsequent multi-sample integration analysis. For dimensionality reduction and clustering, UMAP was used for nonlinear dimensionality reduction and visualization to analyze cell distribution patterns and compare cellular heterogeneity between normal and tumor tissues. PCA\u0026thinsp;+\u0026thinsp;UMAP and NMF\u0026thinsp;+\u0026thinsp;UMAP strategies were compared to select a dimensionality reduction method with high biological interpretability. Cell types were manually annotated based on marker genes, validated by averaged scaled gene expression and cell expression percentage. Macrophage subsets were further visualized and subdivided via UMAP to explore their functional heterogeneity in the tumor microenvironment.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003e2.12 Expression of \u003cem\u003eSERPINE1\u003c/em\u003e in COAD-related cell lines in the CCLE database\u003c/h2\u003e\u003cp\u003eThe gene expression matrix for COAD cell lines was obtained from the CCLE dataset (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://depmap.org/portal/data_page/?tab=allData).Statistica\u003c/span\u003e\u003cspan address=\"https://depmap.org/portal/data_page/?tab=allData).Statistica\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003el analysis was conducted using R software. Results were considered statistically significant when the p-value was less than 0.05.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003e2.13 Statistical Analysis\u003c/h2\u003e\u003cp\u003eA variety of statistical methods were employed for data analysis. During the data preprocessing stage, the \"limma\" package was utilized to perform background correction and normalization on gene expression data, while the \"tidyverse\" package was used to organize clinicopathological information.In the differential analysis section, the \"limma\" package was adopted to screen for DEGs. The t.test function was applied to conduct statistical tests on the expression differences of the \u003cem\u003eSERPINE1\u003c/em\u003e gene, and the Wilcoxon rank-sum test was used to evaluate the differences in drug sensitivity.In the model construction and evaluation phase, Cox regression analysis, LASSO regression analysis, survival analysis, ROC curve analysis, and the construction and validation of the nomogram were carried out.Moreover, in other analyses, various functions and tools such as kruskal.test, \"forestplot\", \"pheatmap\", and the \"CIBERSORT\" algorithm were used to conduct in-depth analyses of subgroup expression differences, prognosis, immune correlations, and tumor microenvironment scores.\u003c/p\u003e\u003c/div\u003e"},{"header":"3 Results","content":"\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\u003ch2\u003e3.1 Construction of the Risk Model\u003c/h2\u003e\u003cp\u003eA differential analysis of MRGs was performed in TCGA-COAD samples, identifying a total of 159 DEGs (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA). To explore the prognostic significance of metabolism-related DEGs, patients with a follow-up period exceeding 30 days were included, yielding 18 prognosis-associated genes (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Prognostic ratio association analysis showed that increased \u003cem\u003eCD36\u003c/em\u003e expression (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.010, HR\u0026thinsp;=\u0026thinsp;1.358, 95% CI: 1.076\u0026ndash;1.714) was associated with an elevated risk of specific outcomes; high MORC2 expression (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, HR\u0026thinsp;=\u0026thinsp;1.966, 95% CI: 1.181\u0026ndash;3.273) was significantly linked to a higher outcome risk; while increased ELVOL6 expression (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.016, HR\u0026thinsp;=\u0026thinsp;0.653, 95% CI: 0.462\u0026ndash;0.924) was associated with a reduced outcome risk. Other genes such as \u003cem\u003eENO3\u003c/em\u003e and \u003cem\u003eMORC2\u003c/em\u003e also exhibited varying degrees of association with specific outcomes, as detailed in (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). Thirteen metabolic genes were identified for inclusion in the prognostic model through LASSO regression analysis (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e(C-D)). Patients were stratified into high-risk and low-risk groups based on risk scores, with the number of patients in each group annotated below respective survival curves. Using risk scores of each COAD patient from the TCGA database, a risk prognostic model was constructed (Risk score=(0.292246796935572\u0026times;CD36 Exp) + (0.81270688717358\u0026times;ENO3 Exp)+(0.357195157656081\u0026times;ELOVL3 Exp) + (0.357195157656081\u0026times;ELOVL3 Exp+(-0.387366053\u0026times;CPT2 Exp))(S1). Results showed that increased risk scores were associated with upregulated expression of prognostic risk genes (\u003cem\u003eELOVL6, CPT2, ELOVL3, CD36\u003c/em\u003e, and \u003cem\u003eENO3\u003c/em\u003e) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA), shortened survival time (Supplementary Materials Fig.\u0026nbsp;2B), and higher mortality rates (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). In the GSE38832 dataset, overall survival curves for high-risk and low-risk groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eE) showed a statistically significant difference in survival rates between the two groups (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.003), with lower survival in the high-risk group. In the TCGA dataset, overall survival curves (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eF) demonstrated a highly significant survival difference (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), with the high-risk group exhibiting significantly lower survival rates than the low-risk group. Progression-free survival curves (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eG) also showed lower progression-free survival in the high-risk group (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Univariate Cox regression analysis (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eH) indicated that age (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.002, HR\u0026thinsp;=\u0026thinsp;1.029, 95% CI: 1.010\u0026ndash;1.048), stage (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, HR\u0026thinsp;=\u0026thinsp;12.067, 95% CI: 6.128\u0026ndash;26.628), and risk score (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, HR\u0026thinsp;=\u0026thinsp;13.171, 95% CI: 2.218\u0026ndash;45.533) were significantly associated with disease risk, while gender (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.560, HR\u0026thinsp;=\u0026thinsp;1.131, 95% CI: 0.747\u0026ndash;1.711) was not. Multivariate Cox regression analysis (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eI) further confirmed that age, stage, and risk score remained significantly associated with disease risk after multivariate adjustment, whereas gender was not. ROC curve analysis (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eJ) revealed an area under the curve (AUC) of 0.733 for the risk score, higher than that of age (AUC\u0026thinsp;=\u0026thinsp;0.628), gender (AUC\u0026thinsp;=\u0026thinsp;0.694), and stage (AUC\u0026thinsp;=\u0026thinsp;0.675), indicating strong predictive ability of the risk score-based prognostic model. ROC curve analysis at different time points (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eK) showed AUC values of 0.668, 0.693, and 0.733 for 1-year, 3-year, and 5-year predictions, respectively, suggesting the model\u0026rsquo;s predictive capacity improves over time.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eConcordance Index(C-index)\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGene\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eCoefficient\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eHR\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eHR.95L\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eHR.95H\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003epvalue\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCD36\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.292246796935572\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.33943354494535\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.06373433948394\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1.68658861026828\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.0129397840429795\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eENO3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.81270688717358\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e2.25400106288746\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.54182381008084\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e3.29513706967038\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e2.73305473762638e-05\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eELOVL3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.357195157656081\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1.42931478435996\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.1064174242589\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1.84644665566286\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.00625700559069411\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eELOVL6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.309335456\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.733934526014851\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.510101027056577\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1.05598667696252\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.0956164499974209\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCPT2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.387366053\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.678842556862494\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.444142817395086\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1.03756539329033\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.073518942445064\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e\u003ch2\u003e3.2 Identification of DEGs in Metabolic Risk Prognosis\u003c/h2\u003e\u003cp\u003eBased on the constructed MRG risk prognostic model for COAD, this study analyzed MRGs associated with risk prognosis and ultimately screened out 72 genes with significantly differential expression. By constructing a protein-protein interaction (PPI) network diagram, the complex interaction relationships among these DEGs related to risk prognosis were revealed. The results showed that the \u003cem\u003eSERPINE1\u003c/em\u003e gene was closely connected with other genes, suggesting that \u003cem\u003eSERPINE1\u003c/em\u003e may play a key regulatory role in MRG-related risks of COAD and participate in important biological processes or signal transduction pathways. Thirteen genes were identified to construct the risk score model, among which \u003cem\u003eSERPINE1\u003c/em\u003e made a significant contribution. Functional enrichment analysis indicated that the related genes were enriched in pathways such as immune-related pathways, with \u003cem\u003eSERPINE1\u003c/em\u003e being significantly enriched. In the PPI network, \u003cem\u003eSERPINE1\u003c/em\u003e exhibited a high degree of connectivity and extensive interactions, outperforming other genes in multiple aspects. Due to its high expression in COAD tissues, association with poor prognosis, critical role in regulating the tumor microenvironment, and extensive interactions in the PPI network to regulate downstream signaling pathways and influence tumor progression, \u003cem\u003eSERPINE1\u003c/em\u003e was regarded as a \"hub gene\" (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec20\" class=\"Section2\"\u003e\u003ch2\u003e3.3 Association between Clinicopathological Features and Prognosis of \u003cem\u003eSERPINE1\u003c/em\u003e\u003c/h2\u003e\u003cp\u003eIn COAD tissues, the expression level of the \u003cem\u003eSERPINE1\u003c/em\u003e gene was significantly higher than that in adjacent normal tissues (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). In the same patient, there was a certain correlation in the expression of the \u003cem\u003eSERPINE1\u003c/em\u003e gene between the tumor tissue and the adjacent normal tissue (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB). Survival analysis revealed that the progression-free survival rate of patients in the high-expression group of \u003cem\u003eSERPINE1\u003c/em\u003e was lower than that of the low-expression group, with a statistically significant difference (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.007). The progression-free survival curve clearly demonstrated the survival difference between the two groups of patients (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC). In addition, the overall survival rate of patients in the high-expression group of \u003cem\u003eSERPINE1\u003c/em\u003e was significantly lower than that of the low-expression group (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). The overall survival curve intuitively reflected this difference, suggesting that the high expression of the \u003cem\u003eSERPINE1\u003c/em\u003e gene might be associated with a poor prognosis (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD).The results of univariate and multivariate Cox regression analyses showed that in the univariate Cox regression analysis, the expression of the \u003cem\u003eSERPINE1\u003c/em\u003e gene (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.016, Hazard ratio\u0026thinsp;=\u0026thinsp;1.205, 95% CI: 1.036\u0026ndash;1.403), age (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.002, Hazard ratio\u0026thinsp;=\u0026thinsp;1.029, 95% CI: 1.010\u0026ndash;1.048), and tumor stage (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, Hazard ratio\u0026thinsp;=\u0026thinsp;2.068, 95% CI: 1.628\u0026ndash;2.628) were significantly associated with patient survival, while gender (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.564, Hazard ratio\u0026thinsp;=\u0026thinsp;1.130, 95% CI: 0.747\u0026ndash;1.709) was not significantly associated with patient survival. In the multivariate Cox regression analysis, after adjusting for other factors, the expression of the \u003cem\u003eSERPINE1\u003c/em\u003e gene (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.137, Hazard ratio\u0026thinsp;=\u0026thinsp;1.132, 95% CI: 0.961\u0026ndash;1.333), age (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, Hazard ratio\u0026thinsp;=\u0026thinsp;1.040, 95% CI: 1.021\u0026ndash;1.060), and tumor stage (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, Hazard ratio\u0026thinsp;=\u0026thinsp;2.190, 95% CI: 1.714-2.800) were still significantly associated with patient survival, and gender (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.886, Hazard ratio\u0026thinsp;=\u0026thinsp;0.970, 95% CI: 0.636\u0026ndash;1.477) was not significantly associated with patient survival (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE).Subgroup analysis showed that there was no significant difference in the expression of the \u003cem\u003eSERPINE1\u003c/em\u003e gene among patients in different age groups (\u0026le;\u0026thinsp;65 years old and \u0026gt;\u0026thinsp;65 years old) (\u003cem\u003eP\u0026thinsp;=\u003c/em\u003e\u0026thinsp;0.32) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eF). There was no significant difference in the expression of the \u003cem\u003eSERPINE1\u003c/em\u003e gene between male and female patients (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.62) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eG). The expression of the \u003cem\u003eSERPINE1\u003c/em\u003e gene in patients at stage M1 was significantly higher than that in patients at stage M0 (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.044) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eH). With the progression of the T stage (T1, T2, T3, T4, T4a), there were significant differences in the expression of the \u003cem\u003eSERPINE1\u003c/em\u003e gene between (T2 and T3, T3 and T4) (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eI). With the progression of the clinical stage (Stage I, Stage II, Stage III, Stage IV), the expression of the \u003cem\u003eSERPINE1\u003c/em\u003e gene gradually increased, and there were significant differences among different stages (comparison between Stage I and Stage IV, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.00061) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eJ). With the progression of the N stage (N0, N1, N2), the expression of the \u003cem\u003eSERPINE1\u003c/em\u003e gene gradually increased, and there were significant differences among different stages (comparison between N0 and N2, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.022) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eK).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec21\" class=\"Section2\"\u003e\u003ch2\u003e3.4 Construction of the Nomogram\u003c/h2\u003e\u003cp\u003eA Cox proportional hazards model was constructed using the coxph function, incorporating variables such as \u003cem\u003eSERPINE1\u003c/em\u003e gene expression level, gender, T stage, N stage, and clinical stage for multivariate analysis. The model was visualized as a nomogram via the regplot function, which assigned specific scores to each variable (e.g., high \u003cem\u003eSERPINE1\u003c/em\u003e expression and advanced T stages (T4a) corresponded to higher scores, while early T stages (T1) corresponded to lower scores). The parameter failtime\u0026thinsp;=\u0026thinsp;c(1,3,5) was used to predict 1-year, 3-year, and 5-year survival rates. Patients with a total score of 207 had an estimated 5-year survival rate of approximately 0.75 (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA). Risk scores for patients were calculated and exported to a file (nomoRisk.txt) for clinical stratification, where higher scores indicated poorer prognosis. For calibration curve validation, repeated calls to the cph and calibrate functions generated calibration curves for each time point. Results showed minimal differences between predicted and observed 1-year survival rates (85% vs 83%), with the calibration curve closely aligning with the 45\u0026deg; diagonal line, demonstrating excellent short-term predictive accuracy. Although slight deviations were observed in 3-year and 5-year survival rate predictions (70% vs. 65%), the curves generally remained close to the diagonal line with consistent trends, indicating the model\u0026rsquo;s reliability in medium- and long-term prognosis (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec22\" class=\"Section2\"\u003e\u003ch2\u003e3.5 Differential Expression and Enrichment Analysis of \u003cem\u003eSERPINE1\u003c/em\u003e\u003c/h2\u003e\u003cp\u003eThe heatmap displayed the expression patterns of DEGs in the sample groups with high and low expressions of \u003cem\u003eSERPINE1\u003c/em\u003e, and a total of 2,138 DEGs were identified (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA). Functional enrichment analysis indicated that \u003cem\u003eSERPINE1\u003c/em\u003e-DEGs were significantly enriched in biological processes such as immune receptor activity, carbohydrate binding, and receptor-ligand activity; in cellular components, they were significantly enriched in structural components of the extracellular matrix, integral components of the postsynaptic membrane, etc.; in molecular functions, they were significantly enriched in receptor activation activity, collagen binding, etc.; in KEGG pathways, they were significantly enriched in pathways such as cytokine-cytokine receptor interaction and neuroactive ligand-receptor interaction. These findings suggest that the \u003cem\u003eSERPINE1\u003c/em\u003e gene may play a role in the development of colon adenocarcinoma by influencing these biological processes and pathways (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec23\" class=\"Section2\"\u003e\u003ch2\u003e3.6 Relationship between \u003cem\u003eSERPINE1\u003c/em\u003e Gene Expression Level and TME Score\u003c/h2\u003e\u003cp\u003eTo investigate the differences in the contents of immune cells and stromal cells between the high-expression and low-expression groups of the \u003cem\u003eSERPINE1\u003c/em\u003e gene and explore its correlation with the tumor immune microenvironment, this study found that in the high-expression group of \u003cem\u003eSERPINE1\u003c/em\u003e, the StromalScore, ImmuneScore, and ESTIMATE Score were all significantly higher than those in the low-expression group. These differences were statistically highly significant (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). The results suggest that the high expression of the \u003cem\u003eSERPINE1\u003c/em\u003e gene may be associated with the increase in the contents and activities of stromal cells and immune cells in the tumor microenvironment of colon adenocarcinoma, indicating that \u003cem\u003eSERPINE1\u003c/em\u003e may play a key role in regulating the tumor microenvironment (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec24\" class=\"Section2\"\u003e\u003ch2\u003e3.7 Analysis of the Role of \u003cem\u003eSERPINE1\u003c/em\u003e in Immune Cell Infiltration\u003c/h2\u003e\u003cp\u003eThe heatmap revealed the correlations between the \u003cem\u003eSERPINE1\u003c/em\u003e gene and a series of immune-related genes, indicating that the \u003cem\u003eSERPINE1\u003c/em\u003e gene was mainly positively correlated with genes of the TNFRSF family and cluster of differentiation(CD) molecules (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eA).The box plot showed that there were significant differences in the infiltration proportions of certain immune cell subsets between the high-expression group and the low-expression group of the \u003cem\u003eSERPINE1\u003c/em\u003e gene. Among them, the infiltration proportion of M0 macrophages was higher in the high-expression group, while the infiltration proportion of the memory B cell subset was relatively higher in the low-expression group, suggesting that the \u003cem\u003eSERPINE1\u003c/em\u003e gene might affect the infiltration pattern of immune cells in colon adenocarcinoma tissues (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eB).The correlation scatter plot further confirmed that the expression of the \u003cem\u003eSERPINE1\u003c/em\u003e gene was mainly positively correlated with the infiltration proportion of NK cells (R\u0026thinsp;=\u0026thinsp;0.25, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;3.9e-05) and negatively correlated with plasma cells (R=-0.2, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.001), indicating that the \u003cem\u003eSERPINE1\u003c/em\u003e gene might regulate the tumor immune microenvironment by influencing the infiltration of these immune cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eC).The violin plot showed that there were differences in the expressions of immune checkpoint genes between the high-expression group and the low-expression group of the \u003cem\u003eSERPINE1\u003c/em\u003e gene. The expression of PD-1 was higher in the high-expression group (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.3e-05), and the expression of CTLA-4 was relatively higher in the low-expression group (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0008), implying that the \u003cem\u003eSERPINE1\u003c/em\u003e gene might be related to the expression regulation of immune checkpoints, thereby affecting tumor immune escape and the response to immunotherapy(Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eD).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec25\" class=\"Section2\"\u003e\u003ch2\u003e3.8 Drug Sensitivity Analysis\u003c/h2\u003e\u003cp\u003eWhen studying the drug TKI-258 (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;3e-05), it was observed that the IC50 value of the \u003cem\u003eSERPINE1\u003c/em\u003e high-expression group was significantly higher than that of the low-expression group, suggesting that the high expression of \u003cem\u003eSERPINE1\u003c/em\u003e may lead to a decrease in the sensitivity of cells to TKI-258. Similarly, for LAQ824 (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.00017), the IC50 value in the \u003cem\u003eSERPINE1\u003c/em\u003e high-expression group was also significantly higher than that in the low-expression group, revealing the correlation between the high expression of \u003cem\u003eSERPINE1\u003c/em\u003e and the decreased sensitivity to LAQ824. In the studies of KN04-2965 (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.00001) and CP-724714 (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;3.7e-05), the IC50 values of the \u003cem\u003eSERPINE1\u003c/em\u003e high-expression group were also significantly higher than those of the low-expression group, indicating that the high expression of \u003cem\u003eSERPINE1\u003c/em\u003e may increase the drug resistance of cells to these two drugs(Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003eA). For the drug Saracatinib (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1.8e-05), the IC50 value of the \u003cem\u003eSERPINE1\u003c/em\u003e high-expression group was significantly higher than that of the low-expression group, showing the correlation between the high expression of \u003cem\u003eSERPINE1\u003c/em\u003e and the drug resistance to Saracatinib. In the studies of drugs such as TAE684, Sunitinib, and TGX221, similar results were also observed, that is, the IC50 values of the \u003cem\u003eSERPINE1\u003c/em\u003e high-expression group were significantly higher than those of the low-expression group, indicating that the high expression of \u003cem\u003eSERPINE1\u003c/em\u003e may reduce the sensitivity of cells to these drugs. For other drugs, such as Dasatinib, CGP \u0026minus;\u0026thinsp;60474, and Lapatinib, significant differences in drug sensitivity were also shown between the \u003cem\u003eSERPINE1\u003c/em\u003e high-expression group and the low-expression group, further confirming the close relationship between the expression level of the \u003cem\u003eSERPINE1\u003c/em\u003e gene and the sensitivity to various drugs (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003eB).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec26\" class=\"Section2\"\u003e\u003ch2\u003e3.9 Characterization of Single-Cell Transcriptome\u003c/h2\u003e\u003cp\u003eThis study systematically analyzed the single-cell transcriptomic characteristics of tumor and normal tissues. First, the Harmony algorithm was used to eliminate batch effects (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003eA). Cells from different samples showed a uniformly mixed state, effectively avoiding technical biases and laying a foundation for quality control. Comparison via UMAP technology (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003e(B-C)) revealed that the topological structure of tumor tissue cells (SMC-T) was more dispersed than that of normal tissue cells (SMC-N), confirming higher cellular heterogeneity in the tumor microenvironment. Subsequently, both \u0026ldquo;Principal Component Analysis (PCA)\u0026thinsp;+\u0026thinsp;Uniform Manifold Approximation and Projection (UMAP)\u0026rdquo; (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003eD) and \u0026ldquo;Non-negative Matrix Factorization (NMF)\u0026thinsp;+\u0026thinsp;Uniform Manifold Approximation and Projection (UMAP)\u0026rdquo; (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003eE) could effectively separate core cell populations, with methodological differences observed. A panoramic framework of cell composition was constructed through integrated annotation (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003eF). Analysis of the \u003cem\u003eSERPINE1\u003c/em\u003e gene (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003eG) showed its high expression in macrophages, which not only supported the reliability of the annotation results but also provided a potential research target. Finally, subpopulation analysis of macrophages (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003eH) identified multiple differentiated subpopulations, reflecting the functional heterogeneity of macrophages and offering a research direction for exploring the pro-tumor and anti-tumor mechanisms of their subtypes.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec27\" class=\"Section2\"\u003e\u003ch2\u003e3.9 Analysis of \u003cem\u003eSERPINE1\u003c/em\u003e Expression\u003c/h2\u003e\u003cp\u003eBased on the CCLE database, the expression of \u003cem\u003eSERPINE1\u003c/em\u003e in different colon adenocarcinoma cell lines is shown. The abscissa represents the median expression level of \u003cem\u003eSERPINE1\u003c/em\u003e, and the ordinate lists the names of different cell lines. It can be seen that the expression levels of \u003cem\u003eSERPINE1\u003c/em\u003e vary among different colon adenocarcinoma cell lines. For example, in cell lines such as HCT 116 and SW480, \u003cem\u003eSERPINE1\u003c/em\u003e is expressed to a certain extent, with some cell lines having relatively high expression levels and others having lower levels, reflecting the heterogeneity of \u003cem\u003eSERPINE1\u003c/em\u003e expression in colon adenocarcinoma cell lines. This suggests that the role of \u003cem\u003eSERPINE1\u003c/em\u003e in the occurrence and development of colon adenocarcinoma may be complex, and its functions may differ under different cellular backgrounds (Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003eA).Through qRT-PCR experiments and analysis of the HPA database, the expression of \u003cem\u003eSERPINE1\u003c/em\u003e in COAD was confirmed at both the gene and protein levels. Immunohistochemical images in the HPA database showed that in COAD tissues (patient number 4453), the \u003cem\u003eSERPINE1\u003c/em\u003e protein exhibited strong positive staining, mainly located in the cytoplasm and intercellular matrix, with a relatively high staining intensity, indicating that the \u003cem\u003eSERPINE1\u003c/em\u003e protein is highly expressed in colon adenocarcinoma tissues. In contrast, in normal colon tissues (patient numbers 1958, 1857, 1960), the positive staining of the \u003cem\u003eSERPINE1\u003c/em\u003e protein was weak, and the staining intensity was significantly lower than that in colon adenocarcinoma tissues, suggesting that the expression level of the \u003cem\u003eSERPINE1\u003c/em\u003e protein in normal colon tissues is low. These results are consistent with the trend of high expression of the \u003cem\u003eSERPINE1\u003c/em\u003e gene in colon adenocarcinoma cell lines observed in the qPCR experiment, further supporting the view that \u003cem\u003eSERPINE1\u003c/em\u003e may play an important role in colon adenocarcinoma (Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003eB).The expression level of \u003cem\u003eSERPINE1\u003c/em\u003e in the colon adenocarcinoma cell line Caco-2 was significantly higher than that in the human normal colon epithelial cell line FHC (Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e11\u003c/span\u003eC). This indicates that the \u003cem\u003eSERPINE1\u003c/em\u003e gene is highly expressed in colon adenocarcinoma cells and may be associated with the occurrence and development of colon adenocarcinoma.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"4 Discussion","content":"\u003cp\u003eThis study has delved deeply into the pivotal significance of MRGs in COAD. We have successfully constructed a risk model of MRGs and identified \u003cem\u003eSERPINE1\u003c/em\u003e as a crucial risk gene, thus opening up a novel perspective for the prognostic assessment and formulation of treatment strategies for COAD [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eThe MRG risk model we established has been validated across multiple datasets. Among them, \u003cem\u003eSERPINE1\u003c/em\u003e has emerged prominently as a key risk gene. Its expression level in COAD tissues is significantly higher than that in normal tissues, and it is closely associated with the M (metastasis), T (tumor), and N (node) stages of the tumor. Survival analysis reveals that high expression of \u003cem\u003eSERPINE1\u003c/em\u003e portends a remarkable shortening of both the progression-free survival and overall survival of patients. Univariate and multivariate Cox regression analyses have further corroborated that \u003cem\u003eSERPINE1\u003c/em\u003e is an independent factor influencing the prognosis of COAD patients. This finding is in consonance with previous research results. For instance, Zhang L et al. also discovered that the expression of \u003cem\u003eSERPINE1\u003c/em\u003e in colorectal cancer tissues is distinctly higher than that in normal tissues, and it is closely related to the clinicopathological characteristics and prognosis of patients, thereby further validating the potential of \u003cem\u003eSERPINE1\u003c/em\u003e as a prognostic biomarker for COAD [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. In terms of expression differences and prognostic values, our research outcomes are highly consistent with those observed in databases such as TCGA and the GEO.\u003c/p\u003e\u003cp\u003eThe tumor microenvironment plays an indispensable role in tumor development, and immune infiltration within it exerts a pivotal regulatory function [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. Our study demonstrates that high expression of \u003cem\u003eSERPINE1\u003c/em\u003e is positively correlated with stromal scores, immune scores, and ESTIMATE scores. Moreover, \u003cem\u003eSERPINE1\u003c/em\u003e exhibits significant associations with a diverse array of immune-related genes, particularly those within the Tumor Necrosis Factor Receptor Superfamily (TNFRSF) and Cluster of Differentiation (CD) molecules. There exist notable disparities in the infiltration patterns of immune cells between the high-expression and low-expression groups of \u003cem\u003eSERPINE1\u003c/em\u003e. Zhang Y et al. comprehensively analyzed the expression, prognostic value, and relationship with the tumor microenvironment of \u003cem\u003eSERPINE1\u003c/em\u003e in colorectal cancer and proposed that \u003cem\u003eSERPINE1\u003c/em\u003e holds promise as a prognostic biomarker and therapeutic target for colorectal cancer. Building upon this, we have further explored the mechanism of action of \u003cem\u003eSERPINE1\u003c/em\u003e within the tumor microenvironment, with a particular focus on its impact on immune cell infiltration and the expression of immune checkpoints. The results of our immune infiltration analysis have underscored the potential application value of \u003cem\u003eSERPINE1\u003c/em\u003e in immunotherapy for COAD [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eIn this study, Kaplan-Meier survival analysis and Receiver Operating Characteristic (ROC) curve analysis have indicated that \u003cem\u003eSERPINE1\u003c/em\u003e is a potent prognostic biomarker. Nevertheless, the hazard ratio in the Cox regression model only exhibits a moderate level of significance. Through analysis, this may be attributed to the limitations of Kaplan-Meier survival analysis and ROC curve analysis in handling confounding factors and multivariate relationships, while the Cox regression model is susceptible to various factors, such as the interaction of confounding factors, insufficient sample size, and abnormal variable distribution. Notwithstanding, sensitivity analysis has shown that the hazard ratio of \u003cem\u003eSERPINE1\u003c/em\u003e maintains a certain degree of stability across different models. This study has certain limitations. In the future, we intend to augment the sample size, optimize the model, and explore its mechanism of action to contribute to the development of treatment strategies [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eDuring the treatment of COAD, the drug sensitivity of tumor cells directly impacts the treatment efficacy. Our study represents the first systematic analysis of the relationship between the expression of \u003cem\u003eSERPINE1\u003c/em\u003e and drug sensitivity. It has been found that high expression of \u003cem\u003eSERPINE1\u003c/em\u003e is significantly correlated with an increase in the half-maximal inhibitory concentration (IC50) values of multiple drugs, such as TKI-258, LAQ824, and KN04-2965, which implies that it diminishes the sensitivity of COAD cells to these drugs. Although the specific mechanism of \u003cem\u003eSERPINE1\u003c/em\u003e in COAD remains elusive, drawing on relevant studies of other tumors (such as lung cancer), high expression of \u003cem\u003eSERPINE1\u003c/em\u003e may influence drug uptake, metabolism, or intracellular drug targets, which holds significant guiding implications for the clinical treatment of COAD.\u003c/p\u003e\u003cp\u003eWhen comparing our study with previous research, it is essential to recognize that, despite the disparities in sample sources, analytical methods, and research focuses among various studies, they all commonly acknowledge the significant role of \u003cem\u003eSERPINE1\u003c/em\u003e in COAD. For example, the research by Zhang L et al. focused on the impact of \u003cem\u003eSERPINE1\u003c/em\u003e knockdown on the proliferation and migration of human colorectal cancer cells. Through cell-based experiments, we have also observed the functional role of \u003cem\u003eSERPINE1\u003c/em\u003e in Caco-2 cells and explored that \u003cem\u003eSERPINE1\u003c/em\u003e may affect the development process of COAD by regulating signaling pathways related to cell proliferation and migration. We have conducted a comprehensive investigation of \u003cem\u003eSERPINE1\u003c/em\u003e in colorectal cancer and expanded upon this basis by integrating the relationships among \u003cem\u003eSERPINE1\u003c/em\u003e, metabolism-related genes, immune infiltration, and drug sensitivity. These previous studies have facilitated our understanding of the intricate mechanism of \u003cem\u003eSERPINE1\u003c/em\u003e in COAD. They have not only laid the foundation for our research but also provided a direction for future studies. Conversely, our study has not only verified some of the important findings in the existing literature but also, through a comprehensive analysis of these relationships, provided a novel perspective and potential therapeutic targets for precision medicine in COAD [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eNevertheless, this study predominantly relies on bioinformatics analysis and cell experiments. Due to the absence of validation using large-scale clinical samples, the generalizability of the research findings may be restricted to a certain extent. Additionally, although we have identified the associations between \u003cem\u003eSERPINE1\u003c/em\u003e and the tumor microenvironment as well as immune infiltration, the specific molecular mechanisms remain unclear. This deficiency in understanding limits our in-depth comprehension of the functions and potential applications of \u003cem\u003eSERPINE1\u003c/em\u003e. Despite these limitations, this study has still elucidated the significant role of \u003cem\u003eSERPINE1\u003c/em\u003e in the prognostic assessment, regulation of the tumor microenvironment, and drug sensitivity of COAD, laying a theoretical foundation for the precision treatment of COAD and proposing novel potential therapeutic targets.\u003c/p\u003e"},{"header":"5 Conclusion","content":"\u003cp\u003eThis study was designed to comprehensively explore the underlying mechanism of MRGs in COAD. We meticulously constructed and rigorously validated a risk-assessment model, through which \u003cem\u003eSERPINE1\u003c/em\u003e was successfully pinpointed as a pivotal hub gene for risk-prognosis evaluation.\u003cem\u003eSERPINE1\u003c/em\u003e exhibits a profound correlation with tumor staging. Its expression level significantly influences patient prognosis, thereby demonstrating great potential as a reliable biomarker for COAD. Moreover, our findings reveal intricate associations between \u003cem\u003eSERPINE1\u003c/em\u003e and multiple aspects, including the tumor microenvironment, immune response dynamics, and drug sensitivity profiles.While this research is inevitably circumscribed by the inherent limitations of bioinformatics-based analyses, it has nonetheless unequivocally elucidated the critical role of \u003cem\u003eSERPINE1\u003c/em\u003e in COAD.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003cp\u003eThe authors declare no competing interests.Ethics, Consent to Participate, and Consent to Publish declarations: not applicable.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003ch2\u003eClinical trial number\u003c/h2\u003e\u003cp\u003enot applicable.\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e\u003cp\u003eWe appreciate the funding:Supported by Guizhou Provincial Science and Technology Projects (Grant NO.QianKeHe Basic-[2024] Youth 269) .\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eX.F. and S.C. designed the study and performed the analysis. Q.X., A.G., C.J.,Z.Z.and J.L. performed the validation in the independent cohort. Y.Q. and L.S. and revised the manuscript. All authors reviewed the manuscript All authors read and approved the fnal manuscript.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe datasets generated during and/or analyzed during the current study are available from the corresponding author upon reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eLiu Y, Li X, Zhang Y, et al. Metabolic reprogramming in colorectal cancer: mechanisms and therapeutic implications[J]. Mol Cancer. 2020;19(1):123.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang Y, Chen H, Zhang H, et al. The role of metabolism-related genes in predicting prognosis of colorectal cancer patients[J]. Front Oncol. 2020;10:1358.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHu J, Yang M, Xia Y, et al. Metabolic genes as biomarkers for cancer diagnosis and prognosis: A systematic review and meta-analysis[J]. J Cell Physiol. 2020;235(10):6937\u0026ndash;48.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChen X, Zhang X, Li X, et al. The prognostic value of immune infiltration in colorectal adenocarcinoma[J]. Aging. 2020;12(19):19013\u0026ndash;28.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXu H, Wang Z, Liu J, et al. Role of plasminogen activator inhibitor-1 in cancer progression and its potential as a therapeutic target[J]. Front Pharmacol. 2020;11:587716.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi X, Huang Y, Wang L, et al. Association between the expression of SERPINE1 and clinicopathological features in colorectal cancer[J]. BMC Cancer. 2021;21(1):308.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang Y, Liu X, Yang Y, et al. Immune microenvironment in colorectal cancer: Insights into prognosis and immunotherapy[J]. Front Immunol. 2021;12:631886.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhao Y, Jiang Y, Wang X, et al. Metabolic gene signature predicts prognosis and guides personalized treatment in colorectal cancer[J]. Front Genet. 2021;12:652100.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGuo X, Li Y, Chen Z, et al. The role of SERPINE1 in regulating tumor angiogenesis and metastasis[J]. Oncol Rep. 2021;46(1):1\u0026ndash;12.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYang X, Wu J, Liu Y, et al. Comprehensive analysis of immune infiltration and its prognostic value in colorectal adenocarcinoma[J]. Front Cell Dev Biol. 2021;9:668386.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang H, Chen X, Li X, et al. SERPINE1 as a potential biomarker for predicting prognosis and immunotherapy response in colorectal cancer[J]. J Immunother Cancer. 2022;10(1):e004180.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLiu Y, Zhang Y, Chen X, et al. Novel insights into the role of metabolism-related genes in colorectal cancer metastasis[J]. Front Oncol. 2022;12:852086.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi X, Wang Y, Zhang X, et al. The impact of immune cell infiltration on the prognosis of colorectal adenocarcinoma patients. J Transl Med. 2022;20(1):100.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHan Y, Zhang Y, Li X, et al. The role of SERPINE1 in the crosstalk between tumor cells and immune cells in colorectal cancer[J]. Front Immunol. 2022;13:887830.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePeng Y, Liu X, Zhang Y, et al. Metabolic gene expression profiling and its clinical significance in colorectal cancer[J]. Front Genet. 2022;13:896456.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYang C, Huang X, Liu Z, Qin W, Wang C. Metabolism-associated molecular classification of hepatocellular carcinoma[J]. Mol Oncol. 2020;14(4):896\u0026ndash;913.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChen HJ, Li HQ, Liu Z et al. SERPINA1 is a direct estrogen receptor target gene and a predictor of survival in breast cancer patients[J]. Oncotarget 2015,6(28):25815\u0026ndash;27.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePan Y, Zhang X, Wang Y, et al. The role of SERPINE1 in colorectal cancer: A bioinformatics analysis[J]. Mol Biol Rep. 2023;50(12):7127\u0026ndash;39.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang L, Liang J, Chen Y, et al. Knockdown of SERPINE1 inhibits proliferation and migration in human colorectal cancer cells[J]. Oncol Lett. 2019;18(6):6663\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTang Y, Xu L, Ren Y, et al. Identification and Validation of a Prognostic Model Based on Three MVI-Related Genes in Hepatocellular Carcinoma. Int J Biol Sci. 2022;18(1):261\u0026ndash;75.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang Y, Li X, Wang Y, et al. SERPINE1: A potential prognostic biomarker and therapeutic target for colorectal cancer[J]. J Cancer Res Ther. 2022;18(3):458\u0026ndash;63.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHu BW, Chen ZW, Wang XG et al. MicroRNA-148a-3p Directly Targets SERPINE1 to Suppress EMT - Mediated Colon Adenocarcinoma Progression[J]. Cancer Manag Res 2021,13:6349\u0026ndash;62.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang X, Li X, Wang Y, et al. The role of immune infiltrates in predicting prognosis and guiding immunotherapy in colorectal adenocarcinoma[J]. Front Oncol. 2023;13:1117718.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLiu X, Peng Y, Zhang Y, et al. Metabolic reprogramming and its potential as a therapeutic target in colorectal cancer[J]. Front Pharmacol. 2023;14:1133806.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHuang X, Li X, Yang X, et al. Comprehensive analysis of the relationship between metabolism-related genes and immune infiltration in colorectal cancer[J]. Front Cell Dev Biol. 2023;11:1174180.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Metabolic-related Genes, SERPINE1, Prognostic, Colorectal Adenocarcinoma, Immune Infiltration","lastPublishedDoi":"10.21203/rs.3.rs-8077356/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8077356/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eObjective\u003c/h2\u003e\u003cp\u003eTo explore the role of Metabolic-related Genes (MRGs), especially \u003cem\u003eSERPINE1\u003c/em\u003e, in the prognosis and immune infiltration of Colorectal Adenocarcinoma (COAD).\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e\u003cp\u003eThis study incorporated the transcriptomic data of colorectal adenocarcinoma (COAD) retrieved from The Cancer Genome Atlas (TCGA) and the Gene Expression Omnibus (GEO) databases. Based on 2752 metabolism-related genes, the Metabolism-Related Gene Prognostic Index (MRGRI) model was constructed and validated. For the \u003cem\u003eSERPINE1\u003c/em\u003e, a comprehensive analysis was performed, covering differential expression analysis, survival analysis, and nomogram construction. Meanwhile, its associations with the tumor microenvironment, immune response, and drug sensitivity were explored. In addition, single-cell analysis was used to verify the heterogeneous expression characteristics of \u003cem\u003eSERPINE1\u003c/em\u003e. Finally, reverse transcription quantitative polymerase chain reaction (RT-qPCR) was employed to validate the expression level of this gene in COAD cells, providing data support for subsequent mechanistic research..\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003eA total of 159 differentially expressed related to metabolic genes (DE-MRGs) were screened out, and 18 genes associated with prognosis were identified. Through LASSO regression, a prognostic model containing 13 genes was established. In this model, the survival rate of the high-risk group was relatively low and the risk score demonstrated strong predictive power. Furthermore, a total of 72 differential genes related to the prognosis of MRGs were obtained, among which \u003cem\u003eSERPINE1\u003c/em\u003e was the hub gene. In COAD tissues, high expression of \u003cem\u003eSERPINE1\u003c/em\u003e indicated a poor prognosis and was correlated with disease stages. The nomogram provided accurate predictions. The differentially expressed genes were mainly enriched in pathways related to immune receptor activity. \u003cem\u003eSERPINE1\u003c/em\u003e regulated the immune microenvironment, affected tumor immune escape and immunotherapy responses, and overexpression of it would reduce drug sensitivity.Single-cell analysis reveals heterogeneous expression of \u003cem\u003eSERPINE1\u003c/em\u003e in macrophages.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e\u003cp\u003eIn the MRGs risk model, \u003cem\u003eSERPINE1\u003c/em\u003e is a hub gene that plays a crucial role in the prognosis, immune infiltration and drug sensitivity of COAD.\u003c/p\u003e","manuscriptTitle":"Investigation of the Metabolism-Related Prognostic Gene SERPINE1 as a Prognostic Predictor in Colorectal Adenocarcinoma and Its Regulatory Mechanisms Underlying Immune Infiltration","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-27 19:34:18","doi":"10.21203/rs.3.rs-8077356/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"3c3a4e6c-b559-4349-83b4-4f4c85cc38ac","owner":[],"postedDate":"November 27th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-12-24T12:24:34+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-27 19:34:18","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8077356","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8077356","identity":"rs-8077356","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}