A four-gene signature in the tumor microenvironment that significantly associates with the prognosis of patients with breast cancer.

Abstract

Breast cancer (BRCA) is a highly heterogeneous disease due to the complicated microenvironment in the tumor, making the treatment benefits varied. Therefore, this study aims to identify a gene signature in the tumor microenvironment (TME) associated with the prognosis of BRCA patients. We downloaded the immune, stromal, and proliferation (ISP)-associated genes from the literature on BRCA. mRNA expression and clinical information obtained from The Cancer Genome Atlas (TCGA) were performed to identify the initial biomarker. Furthermore, we validated the robustness of the gene signature in the independent validation data set GSE20685. A four-gene signature in TME, including CD74, MMP9, RPA3, and SHCBP1, was constructed to predict the overall survival of BRCA. The survival time of the high-risk group was significantly worse than that of the low-risk group. Univariate and multivariate Cox regression analysis showed that our four-gene ISP signature was an independent prognostic factor in TCGA and GSE20685 data sets. The AUC suggested that our four-gene ISP signature was comparable to TNM classification at predicting the overall survival of BRCA patients. Interestingly, BRCA patients with high-risk scores were more likely to be associated with stromal and proliferation of cancer. In contrast, those with high-risk scores were more likely to be associated with tumor immunity-related pathway. We found an innovative biomarker in TME to predict the prognosis of BRCA. This signal might reflect the imbalance of TME and provide potential biomarkers for the individualized and precise treatment of BRCA.