Screening Of Immune Regulators Of Tumor Microenvironment Related To Breast Cancer Prognosis Based On TCGA Database

Objective:Breast cancer is the most common cancer for women worldwide and the major cause of cancer-related deaths in women.Although the incidence and mortality rates of breast cancer among women in China are slightly lower than those in Europe and the United States,the large population base and low detection rate have led to a much larger total number of breast cancer cases and deaths than those in other countries.With the improvement of the national economy and the reform of medical and health institutions in China,early diagnosis and oncological treatment of breast cancer have greatly improved the survival rate of early stage patients,which are basically on par with those in Europe and America.However,the prognosis of advanced breast cancer patients is still extremely poor,the five-year survival rate is only 20%,which is a great threat to women’s health.With the rise of research focusing on the tumor microenvironment in the past five decades,it has gradually revealed that tumor microenvironment plays an important role in the process of tumor initiation and progression.The interactions between tumor cells and tumor-associated stromal cells alter the ecology of the tumor microenvironment and also influence the disease progression and clinical outcome.Especially,tumor-infiltrating immune cells play a dual role in tumor progression.Breast cancer was initially considered a non-immunogenic tumor,but as research has revealed the rich immune environment in the breast cancer tumor microenvironment.More and more evidence suggested that the level of immune infiltration has important implications for tumor progression and prognosis of breast cancer patients,especially in some tumor with poor prognosis,such as triple-negative breast cancer.Therefore,a comprehensive assessment of the impact of tumor microenvironment on breast cancer prognosis and the screening of immunomodulators important for tumor progression are important references for clinical research and breast cancer immunotherapy.Methods:We downloaded the transcriptome sequencing data of invasive breast cancer and its corresponding clinical information from TCGA database,and extracted the gene expression data and clinical information to merge into a gene expression matrix.First,we used the ESTIMATE algorithm to estimate the proportion of immune and stromal components in the TME of each breast cancer sample by applying the"ESTIMATE"package in the R software,and presented it as Immune Score,Stromal Score and ESTIMATE Score.The three quantitative scores were combined with the survival information of the breast cancer cases,and log-rank p<0.05 was used as the significance criterion for the"survivor"and"survminer"packages in the R software.Kaplan-Meier survival analysis and correlation analysis of clinicopathological characteristics were performed.Subsequently,we used the"limma"package in R software to obtain the DEGs by comparing high score samples with low score samples in stroma or immune by the Wilcoxon rank sum test with︱log2FC︱>1&FDR<0.05.The DEGs shared in the TME were obtained by taking the intersection of the two groups of DEGs were analyzed for GO and KEGG enrichment.One-way COX regression analysis was performed to screen for significant genes associated with breast cancer prognosis in DEGs.Protein-protein interaction network maps with confidence levels greater than 0.7 were constructed by STRING online database using DEGs,and protein interaction node genes were reconstructed by applying Cytoscape software.One-way COX regression analysis and protein interaction node gene intersection analysis were performed to obtain the most valuable key genes in this study.The expression differences of key genes in tumor samples versus normal samples were analyzed by R software as well as survival analysis and correlation analysis of clinicopathological characteristics.Gene set enrichment（GSEA）analysis was performed to predict the possible biological functions and molecular pathways involved in the key genes.Finally,we applied R software to estimate the proportional distribution of tumor-infiltrating immune cell subpopulations in all tumor samples with the CIBERSORT algorithm and analyzed the correlation between immune cells and the expression levels of key genes.Results:（1）Correlation analysis of tumor microenvironment scores with survival analysis and pathological characteristics of patients of invasive breast cancer revealed that the immune component in breast cancer was significantly correlated with age（p=0.045）and survival（p=0.009）,but not with clinical stage;the stromal component was significantly correlated with age（p=0.033）,T classification（p=0.002）,TNM stage（p=0.015）,but not with patient survival.（2）A total of 224DEGs common to both stromal and immune components were screened in this study,of which 203had up-regulated gene expression and 21 had down-regulated gene expression.（3）GO enrichment analysis revealed that DEGs were involved in biological functions such as differentiation of lymphocytes and activation of T cells;KEGG enrichment analysis showed that DEGs were involved in molecular pathways such as chemokine signaling pathways,cytokine-cytokine receptor interactions and hematopoietic cell lineage.（4）One-way COX regression analysis screened 30prognostic significant genes from DEGs;PPI network diagram was constructed with 116 protein nodes;Intersection analysis obtained 20 key genes including CLEC10A.（5）We revealed that CLEC10A was less expressed in tumor tissues than normal tissues and significantly correlated with patient prognosis;Correlation analysis of clinicopathological characteristics showed that CLEC10A expression was significantly correlated with age(p=2×10^-6),T classification(p=7×10^-6),and TNM stage（p=0.022）of breast cancer patients.（6）Gene set enrichment analysis revealed that high expression of CLEC10A involved molecular pathways such as antigen processing and presentation,B-cell signaling pathway,complement signaling pathway,hematopoietic cell lineage and other immune-related activities;low expression of CLEC10A involved glycosylphosphatidylinositol pathway.（7）The expression level of CLEC10A was positively correlated with eight kinds of immune cells,including memory B cells,naive B cells,resting dendritic cells,M1 macrophages,activated memory CD4⁺T cells,resting memory CD4⁺T cells,CD8⁺T cells,andγδT cells,and was negatively correlated with four kinds of M0 macrophages,M2 macrophages,neutrophils,and resting NK cells immune cells were negatively correlated.Conclusions:We confirmed that the level of immune infiltration is significantly associated with the prognosis of breast cancer patients based on transcriptomic data analysis of breast cancer in the TCGA database,and further investigated that the key gene CLEC10A may be an important factor in the regulation of the immune microenvironment of breast cancer.CLEC10A is a specific recognition ligand for the immune system to recognize tumor-associated antigens,especially in the initiation phase of antitumor immunity It plays an important role in the recognition and presentation of tumor antigens and initiation of the immune response.Therefore,the expression level of CLEC10A may contribute to the assessment of prognosis of breast cancer patients and provide new ideas for future immunotherapy.