Identification Of Potential Prognosis-related Genes In Endometrial Cancer Based On TCGA Database

The incidence of endometrial cancer is increasing each year,and treatment effects are poor for patients with advanced and specific subtypes.Exploring immune infiltration-related factors in endometrial cancer can aid in the prognosis of patients and provide new immunotherapy targets.We downloaded immune metagene and functional data of patients with different subtypes of endometrial cancer from The Cancer Genome Atlas database and selected the lymphocyte-specific kinase(LCK)metagene as a representative genetic marker of the immune microenvironment in endometrial cancer.The results showed that LCK metagene expression is related to the prognosis of patients with endometrioid endometrial adenocarcinoma subtypes and highly correlated with the PTEN and PIK3 CA mutational status.A search for LCK-related modules returned seven independent genetic predictors of survival in patients with endometrial cancer.The TIMER algorithm showed that the expression of these seven genes was positively correlated with the infiltration levels of six types of immune cells.Our results identified CD74,HLA-DRB5,CD52,HLA-DPB1 and HLA-DRB1 as possible valuable genetic markers for the diagnosis and prognosis of endometrial cancer and provided a theoretical basis for immunotherapy targets for its clinical treatment.Objectives:We analyzed the transcriptome data of endometrial cancer in the TCGA database,combined with epigenetics data and patient clinical information,and discovered potential markers related to the immune microenvironment in endometrial cancer.Methods:1.Data sources and pre-processingTCGA-UCEC standardized FPKM data(https://bioinformatics.mdanderson.org/),somatic mutation,and clinical information were downloaded from the UCSC Xena official website(https://xena.ucsc.edu/).The scores of 13 types of immune metagenes were obtained by calculating the average value of log2(FPKM + 1).Six types of immune cell scores(B＿cell,CD4＿Tcell,CD8＿T cell,neutrophil,macrophage,dendritic)were obtained from mRNA expression data using the TIMER package(https://cistrome.shinyapps.io/timer/).The Stromal Score and Immune Score were calculated using the ESTIMATE R package(https://bioinformatics.mdanderson.org/estimate/index.html).ESTIMATE R package uses expression profile data to predict the scores of stromal cells and immune cells,and then predicts the content of these two cells;Immune Sore: immune cell score,Stromal Score: stromal cell score ESTIMATEScore: comprehensive score.The expression of the CD74,HLA-DRB5,HLA-DRB1,and ACP5 proteins in EC tissues was analyzed using the online tool UALCAN(http://ualcan.path.uab.edu/index.html).2.Screening representative genes in the EC immune microenvironmentThe Spearman correlation coefficient was used to calculate the correlation between different immunerelated scores in different EC subtypes.The results showed that the LCK metagene score was the highest relative to other types of immune-related scores.Among the three subtypes(endometrioid endometrial adenocarcinoma: 0.84,serous endometrial adenocarcinoma: 0.83,mixed serous and endometrioid:0.85),LCK metagene was selected as representative of the EC immune microenvironment gene.The samples were divided into two groups according to the median of LCK metagene mRNA expression level,and the Kaplan-Meier(KM)survival curves of the two groups were drawn.We also analyzed the relationship between the LCK metagene mRNA expression level and PTEN,PTK3 CA,TP53,and KRAS mutations.3.Analysis of LCK metagene-related modules by WGCNAScreening for genes with a median absolute deviation of the top 75% and genes with a MAD greater than 0.01 was performed using the WGCNA R package to construct a gene co-expression network with genetic methods used to generate a dynamic shear module and for cluster analysis of the module.Genes with similar expression levels were divided into the same module.The results showed that the 141 genes in the pink module were highly correlated with the LCK metagene(cor =.0.69).The cluster Profiler R package was used for KEGG analysis of this module(FDR < 0.05).4.Screening immune microenvironment genes related to prognosisAccording to the LCK metagene score,the samples were divided into two groups of LCK high and low,and the differential expression analysis of genes was performed using the limma R package(FDR < 0.1,|log2(foldchange)| > 1),and KEGG analysis was performed on the 58 genes of the intersection.In order to identify genes with prognostic value in the immune microenvironment,we performed univariate cox survival analysis and used the survminer R package to draw KM survival curves.5.Lasso-cox prognostic modelLasso regression analysis was performed on the obtained 11 prognostic-related genes,and the optimal tuning parameter λvalue of Lasso-cox regression was obtained after ten-fold cross-validation.3 genes were screened from the prognostic-related genes by tuning parameter fitting.Multivariate cox regression analysis was performed on their expression levels to construct a prognostic model.The risk score calculated by the model(cut-off = 1)is divided into high-risk groups and low-risk groups according to the threshold.The analysis results showed that there was a significant difference in survival between the high-risk group and the low-risk group(p = 0.0077),and the high-risk group had a worse prognosis.In the time-dependent ROC(Time-dependent ROC,t ROC)analysis,the AUC values of the risk prediction model in 1,3,and 5 years were 64.4,59.1,and 68.8,respectively.The results show that the model has good predictive ability,and has certain reliability and accuracy.6.Construction and verification of Nomogram prediction modelWe performed univariate cox regression including risk score and clinical information to select potential risk factors,and established a nomogram model for candidate factors.The results showed that the predicted AUC values of the model for 1,3,and 5-year patients’ survival were 81,80.3,and 81.7,respectively.Results:1.In the three EC subtypes,except for the neoantigen score,the LCK metagene score showed a significant positive correlation with other types of immune-related scores:endometrioid endometrial adenocarcinoma(cor = 0.84),serous endometrial adenocarcinoma(cor = 0.83),and mixed serous and endometrioid(cor = 0.85).Next,we analyzed the distribution of the LCK metagene levels in three EC subtypes at different clinical stages of EC.The results revealed no significant differences in LCK metagene expression at different clinical stages.The expression of LCK in the PTEN,PIK3 CA,TP53,and KRAS groups and difference between the mutant and wild-type groups were assessed.The results showed that LCK metagene expression was higher in the PTEN and PIK3 CA mutant groups than in the wild-type group,with no significant difference in LCK metagene expression between the TP53 and KRAS mutant and wild-type groups.2.The LCK metagene gene score was highly correlated with the pink module(R =0.69).Next,we chose the pink(R = 0.69)module for Kyoto Encyclopedia of Genes and Genomes(KEGG)analysis.This module was enriched in 20 pathways related to various aspects of immunity,such as antigen processing and presentation,Th1 and Th2 cell differentiation,and cell adhesion molecules.3.The genes in the pink module were further filtered.The results of univariate COX analysis showed that 7 of the genes were related to the overall survival of endometrial cancer patients,and the CD74,HLADRB5,HLA-DRB1 and ACP5 genes showed higher protein expression levels in tumor tissues.4.The established Lasso-cox model of 3 genes has good predictive ability for high- and low-risk EC patients.Conclusions:1.We using ESTIMATE algorithm-based immune scoring and TCGA EC cohort information analysis,immune-related genes of EC were screened and prognostic characteristics were established.Notably,we found 7 genes related to the overall survival of patients with endometrial cancer.Patients with high mRNA expression of related genes have a longer prognostic survival,and we found that CD74,HLA-DRB5 and HLA-DRB1 proteins are highly expressed in early tumor tissues.In sum,the prognostic model we have established for these genes can predict the prognostic survival of EC patients.