Abstract
Background
Triple-negative breast cancer (TNBC) is a highly aggressive subtype characterized by a lack of hormone receptors and specific molecular targets, leading to limited treatment options, high recurrence rates, and poor clinical prognosis. The Warburg effect in cancer cells generates a tumor microenvironment characterized by hypoxia, low glucose, and high lactate levels, which severely promotes chemoresistance, immune escape, and distant metastasis. Nevertheless, the prognostic significance of hypoxia and lactate metabolism-related genes in TNBC remains largely unclear, and more accurate prediction models are urgently needed.Methods
Clinical information and gene expression profiles of patients with TNBC were obtained from online databases. LASSO and multivariate Cox regression analyses were used to establish the hypoxia and lactate metabolism-related prediction model, and the prognostic value was further validated using Kaplan-Meier plotter, receiver operating characteristic curves, and nomogram. Gene set enrichment analysis was performed to evaluate the pathways and molecular functions. Tumor mutation, microsatellite instability, RNA expression-based stemness scores, immune infiltration, and drug susceptibility analyses were performed to identify potential therapeutic targets. Single-cell transcriptome data were processed and quality-controlled using Seurat. Monocle2, CellChat, and ssGSEA analyses were used to investigate the relationship between COL5A3, malignant phenotype, and tumor microenvironment. A series of in vitro experiments was performed to evaluate the biological functions of hub gene COL5A3.Results
Five genes, COL5A3, LRRC8D, IGFL1, SEPTIN3, and PEG10, were identified to establish the hypoxia and lactate metabolism-related risk score. TNBC patients divided into high-risk or low-risk groups according to the risk score exhibited substantial differences in survival outcome, tumor mutation, immune infiltration, and drug sensitivity. The constructed model showed strong predictive performance, with receiver operating characteristic curves showing area under the curve values of 0.82, 0.86, and 0.90 at 365, 1085, and 1825 days, respectively. Single-cell analysis showed that high COL5A3 expression was associated with greater proliferative capacity, drug resistance, and stemness. COL5A3 is also involved in regulating the interaction between TNBC cells and other cells in the tumor microenvironment. In vitro studies have indicated that COL5A3 could promote tumor proliferation, migration, invasion, and drug resistance via the DDR1/FAK/PI3K/AKT pathway in TNBC cells. OF-1 was initially identified as a candidate compound with potential therapeutic relevance for TNBC.Conclusion
A novel hypoxia and lactate metabolism-related gene signature could be used for prognosis prediction in TNBC patients, and the hub gene COL5A3 promotes TNBC progression via the DDR1/FAK/PI3K/AKT pathway. Further research is needed to explore the biological roles of these genes in TNBC to refine the therapeutic approaches.