Bioinformatic Profiling of Prognosis-Related Genes in Breast Cancer Immune Microenvironment

Abstract

Background: The infiltration of immune and stromal cells in breast cancer microenvironment is strongly related to prognosis. Mining potential immune-related prognostic markers or therapeutic targets is significant. Methods: We had applied multiple approaches to search for the lymphocyte-specific kinase (LCK) metagene scores and LCK metagenes were used as the objects of study to screen the representative genes using weighted gene co-expression network analysis (WGCNA); besides, the differentially ex-pressed genes (DEGs) in samples with high or low LCK metagenes scores were analyzed, and the co-expressed genes markedly correlated with the genes in LCK metagenes were mined jointly. Finally, survival analyses were applied to search for the genes, and gene functions were mined through en-richment analysis. Additionally, these genes were further verified using the external dataset and paired tumor and non-tumor tissues. Findings: LCK metagene scores have high correlations with other categories of immune-related scores, clinical stage, prognosis and the mutation levels of multiple tumor suppressor genes (BRCA1, BRCA2, TP53 and PTEN), from the four kinds of breast cancer subtypes. We screened 22 genes re-lated to immunity and prognosis in breast cancer according to the expression of immune metagenes. Interpretation: Potential immune-related therapeutic targets or markers for the diagnosis and prog-nosis of breast cancer were mined. Funding: Science and Technology Commission of Shanghai Municipality (No. 17411961100). Declaration of Interest: The authors declare no conflicts of interest. Ethical Approval: The undertaken of all the samples from patients of this study was approved by the Ethics Committees of Obstetrics and Gynecology Hospital of Fudan University and consent with all the patients.