Identification of a Novel Transcription Factor Prognostic Index for Breast Cancer

Transcription factors (TFs) are the mainstay of cancer and have a widely reported influence on the initiation, progression, invasion, metastasis, and therapy resistance of cancer. However, the prognostic values of TFs in breast cancer (BC) remained unknown. In this study, comprehensive bioinformatics analysis was conducted with data from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) database. We constructed the co-expression network of all TFs and linked it to clinicopathological data. Differentially expressed TFs were obtained from BC RNA-seq data in TCGA database. The prognostic TFs used to construct the risk model for progression free interval (PFI) were identified by Cox regression analyses, and the PFI was analyzed by the Kaplan-Meier method. A receiver operating characteristic (ROC) curve and clinical variables stratification analysis were used to detect the accuracy of the prognostic model. Additionally, we performed functional enrichment analysis by analyzing the differential expressed gene between high-risk and low-risk group. A total of nine co-expression modules were identified. The prognostic index based on 4 TFs (NR3C2, ZNF652, EGR3, and ARNT2) indicated that the PFI was significantly shorter in the high-risk group than their low-risk counterpart (p < 0.001). The ROC curve for PFI exhibited acceptable predictive accuracy, with an area under the curve value of 0.705 and 0.730. In the stratification analyses, the risk score index is an independent prognostic variable for PFI. Functional enrichment analyses showed that high-risk group was positively correlated with mTORC1 signaling pathway. In conclusion, the TF-related signature for PFI constructed in this study can independently predict the prognosis of BC patients and provide a deeper understanding of the potential biological mechanism of TFs in BC