Identification and Validation of a Novel Glycolysis-Related Gene Signature for Predicting the Prognosis and Therapeutic Response in Triple-Negative Breast Cancer

Abstract

Introduction: A high malignancy rate and poor prognosis are common problems with triple-negative breast cancer (TNBC). There is increasing evidence that glycolysis plays vital roles in tumorigenesis, tumor invasion, immune evasion, chemoresistance, and metastasis. However, a comprehensive analysis of the diagnostic and prognostic significance of glycolysis in TNBC is lacking.

Methods: Transcriptomic and clinical data of TNBC patients were obtained from The Cancer Genome Atlas (TCGA) and Molecular Taxonomy of Breast Cancer International Consortium (METABRIC) databases, respectively. Glycolysis-related genes (GRGs) were collected from the Molecular Signatures Database (MSigDB). Differential comparative analysis was performed to obtain the differentially expressed (DE)-GRGs associated with TNBC. Based on the DE-GRGs, a glycolysis-related risk signature was established using Least Absolute Shrinkage and Selector Operation (LASSO) and multivariable Cox regression analyses. The prognostic value, tumor microenvironment, mutation status, and chemotherapy response of different risk groups were analyzed. An independent cohort from the METABRIC database was used for external validation. Furthermore, the expression patterns of five genes derived from the prognostic model were validated by quantitative real-time polymerase chain reaction (RT-qPCR).

Results: The glycolysis-related prognostic signature included five genes (IFNG, ACSS2, IRS2, GFUS, and GAL3ST1) and predicted the prognosis of TNBC patients independent of clinical factors (p < 0.05). Patients were divided into high- and low-risk groups based on the median risk score. Compared to low-risk TNBC patients, high-risk patients had significantly decreased overall survival (HR = 2.718, p < 0.001). Receiver operating characteristic and calibration curves demonstrated that the model had high performance in terms of predicting survival and risk stratification. The results remained consistent after external verification. Additionally, the tumor immune microenvironment significantly differed between the risk groups. Low-risk TNBC patients had a better immunotherapy response than high-risk patients. High-risk TNBC patients with a poor prognosis may benefit from targeted therapy.

Conclusions: This study developed a novel glycolysis and prognosis-related (GRP) signature based on GRGs to predict the prognosis of TNBC patients, and may aid clinical decision-making for these patients.

Keywords: Computational biology; Glycolysis; Nomogram; Prognosis; Triple-negative breast cancer; Tumor microenvironment.