A novel super-enhancer-related gene signature predicts prognosis and immune microenvironment for breast cancer

BackgroundThis study targeted at developing a robust, prognostic signature based on super-enhancer-related genes (SERGs) to reveal survival prognosis and immune microenvironment of breast cancer.MethodsRNA-sequencing data of breast cancer were retrieved from The Cancer Genome Atlas (TCGA), 1069 patients of which were randomly assigned into training or testing set in 1:1 ratio. SERGs were downloaded from Super-Enhancer Database (SEdb). After which, a SERGs signature was established based on the training set, with its prognostic value further validated in the testing set. Subsequently, we identified the potential function enrichment and tumor immune infiltration of the model. Moreover, in vitro experiments were completed to further explore the biological functions of ZIC2 gene (one of the risk genes in the prognostic model) in breast cancer.ResultsA risk score system of prognostic value was constructed with 6 SERGs (ZIC2, NFE2, FOXJ1, KLF15, POU3F2 and SPIB) to find patients in high-risk group with significantly worse prognosis in both training and testing sets. In addition, a multivariate regression was established via integrating the 6 genes with age and N stage, indicating well performance by calibration, time-dependent receiver operating characteristic (ROC) analysis and decision curve analysis (DCA). Further analysis demonstrated that tumor-associated pathological processes and pathways were significantly enriched in the high-risk group. In general, the novel SERGs signature could be applied to screen breast cancer with immunosuppressive microenvironment for the risk score was negatively correlated with ESTIMATE score, tumor-infiltration lymphocytes (such as CD4 + and CD8 + T cell), immune checkpoints and chemotactic factors. Furthermore, down-regulation of ZIC2 gene expression inhibited the cell viability, cellular migration and cell cycle of breast cancer cells.ConclusionsThe novel SERGs signature could predict the prognosis of breast cancer; and SERGs might serve as potential therapeutic targets for breast cancer.