Construction and validation of predictive model based on endoplasmic reticulum stress-related genes for triple-negative breast cancer.

Triple-negative breast cancer (TNBC) poses challenges in treatment due to its inherent biological characteristics. Endoplasmic reticulum stress (ERS) has been associated with the development of TNBC. Hence, identifying ERS-related prognostic biomarkers is crucial for the early diagnosis and treatment of TNBC. In this study, we retrieved gene expression profiles from TNBC patients using The Cancer Genome Atlas (TCGA) database. Differentially expressed genes (DEGs) between TNBC tumor and normal tissues were identified using limma package. Using differential expression analysis, we identified 46 ERS-related DEGs. Through univariate Cox, LASSO, and multivariate COX regression analyses, we constructed a prognostic model consisting of 8 genes (IGFBP1, CFTR, THBS4, CREBRF, CLU, HDGF, DERL3, NCCRP1). This model demonstrated robust prognostic accuracy in TNBC patients, validated by the METABRIC dataset. Among the 8 prognostic genes, NCCRP1 showed the highest expression increase in BT-20 and MDA-MB-468 cells. Functional assays further revealed that NCCRP1 significantly promoted proliferation, migration, and invasion, while suppressing apoptosis and ERS in these TNBC cell lines. Our study highlights a strong association between ERS-related genes and the prognosis of TNBC patients. Moreover, we demonstrated that NCCRP1 exerts oncogenic effects in TNBC cells. It provides new insights and possible treatment targets for TNBC.