| Purpose Cervical cancer is one of the most common causes of cancer-related death among women in developing countries.Currently,the first-line treatment of early-stage cervical cancer,including intraepithelial cancer,is surgery.Radiotherapy and chemotherapy have been used to treat patients with advanced uterine cervical cancer,but with limited success.Due to tumor heterogeneity,different patients may have variable responses to chemotherapy and radiation,and "one size fits all" standard therapies have significant limitations.Precision and personalized medicine is poised to change this “one size fits all” approach.Therefore,precise treatment strategies for different patients are more conducive for prognosis evaluation and treatment management.In recent years,based on high-throughput sequencing technology,including RNA sequencing technology and microarray technologies,accurate treatment for various cancer subtypes have received widespread attention.Using bioinformatics to analyze large-scale global gene expression profiling of cervical cancer patients is helpful to foster the development of precision and personalized medicine.Method In this study,a total of 304 cervical cancer patients from the TCGA database were included.The ESTIMATE algorithm was used to evaluate the tumor microenvironment,and the tumor stromal score and immune score were obtained.They were divided into high and low subgroups according to the median score,and we analyzed the correlation between the score and prognosis of patients.We further Screened differentially expressed genes(DEGs)for two subgroups related to patient prognosis,and performed GO and KEGG enrichment analysis on the obtained differentially expressed genes to explore their possible molecular functions,biological processes,cellular components and participation Signal pathway.For each gene obtained,we plotted K-M survival curves and screened for differentially expressed genes that were significantly related to the patients’ five-year survival rate.To further tap the core regulatory genes,we constructed protein-protein interaction network for the prognosis-related differentially expressed genes,and preformed gene enrichment analysis and module mining.Finally,the most statistically significant genes were obtained.In addition,this study used an individual data set of cervical cancer patients GSE52903(n = 55)in the GEO database and their corresponding sample clinical information to further verify the correlation between the key prognostic genes and the prognosis of cervical cancer.On the other hand,we constructed a TF-lncRNA-miRNA-mRNA multi-factor regulatory network for key prognostic genes using pivot method and screened drugs that may have a regulatory effect on key prognostic genes.Result The analysis of 304 cases of cervical cancer in the TCGA database by the ESTIMATE algorithm found that the immune score was significantly correlated with the patient’s five-year survival rate(p = 0.03),and the prognosis of the high score group was better.Screening of differentially expressed genes in two subgroups of high and low immune scores yielded 1367 differentially expressed genes.GO functional annotation and KEGG pathway enrichment analysis found that 1367 differentially expressed genes were mainly related to immune function.We further plotted the survival curve and screened out 401 differentially expressed genes that were significantly related to the five-year survival rate(p <0.05).For 401 prognosis related differentially expressed genes,we constructed a protein-protein interaction network,and performed GSEA gene enrichment analysis and MCODE module mining,and finally selected a key prognostic gene set containing 79 differentially expressed genes.These key prognostic genes were mainly related to bone marrow leukocyte activation,adaptive immune response regulation and receptor signaling pathways.In addition,by external verification,we found that CCR7,PDCD1,ZAP70,and CD28 in the key prognostic gene set were significantly correlated with the five-year survival rate of patients(p <0.05).On the other hand,we constructed a multi-factor regulatory network containing 148 miRNAs,31 lncRNAs,21 TFs and 75 mRNAs and predicted 39 potential drugs that may significantly regulate the key prognostic genes.Conclusion In this study,we used bioinformatics methods to screen differentially expressed genes related to the immune microenvironment and prognosis of cervical cancer patients,analyze their potential regulatory networks,and explore potential therapeutic drugs.The main conclusions were:(1)Cervical cancer immune microenvironment was related to prognosis of cervical cancer patients.(2)79 DEGs that might be related to tumor immune microenvironment and prognosis of cervical cancer were selected,and a multi-factor regulatory network of the DEGs set was constructed and 39 potential drugs that might regulate this DEGs were predicted. |
PDF Full Text Request