Identification of lactylation-related biomarkers for diagnosis, prognosis, and treatment responsiveness in triple-negative breast cancer

Triple-negative breast cancer (TNBC), a highly aggressive subtype of invasive breast cancer, lacks effective therapeutic targets and robust prognostic biomarkers. Lactylation, an emerging post-translational modification, is increasingly recognized for its potential role in tumorigenesis and progression. However, the functional significance of lactylation-related genes (LRGs) in TNBC clinicopathology and prognosis remains poorly defined. We employed a combination of differential expression analysis and weighted gene co-expression network analysis (WGCNA) to identify hub LRGs in TNBC. Consensus clustering was used to stratify TNBC samples into distinct molecular subtypes based on hub LRGs. Immune infiltration and functional differences between the two subtypes were further evaluated using single-sample gene set enrichment analysis (ssGSEA) and Gene Set Variation Analysis (GSVA). To screen optimal diagnostic biomarkers, we constructed 117 diagnostic models using nine machine learning algorithms and assessed their efficiency through receiver operating characteristic (ROC) analysis and nomograms. Additionally, Cox regression analysis was conducted to evaluate the prognostic potential of hub LRGs in TNBC. Furthermore, we proposed treatment recommendations for TNBC patients based on immunophenotyping and drug sensitivity analysis. Aberrant expression of biomarkers in TNBC tissues was validated using immunohistochemistry. The impact of key biomarkers on TNBC cell proliferation and migration was investigated via CCK-8, transwell, and wound healing assays. Mechanistically, the role of PGK1 in regulating lactate metabolism and global lactylation was assessed by measuring lactate and lactate modification levels, combined with quantitative lactylome proteomics analysis. We identified 10 differentially expressed hub LRGs through intersection analysis, which effectively classified TNBC samples into two subtypes exhibiting significant heterogeneity in the tumor immune microenvironment. Among 117 diagnostic models, the optimal model achieved 90% accuracy in the detection of TNBC. Cox regression analysis confirmed PGK1 as an independent risk factor and potential prognostic marker for TNBC. Patients with high PGK1 expression showed poor overall survival, and elevated PGK1 contributed to an immunosuppressive microenvironment by modulating immune cell infiltration. In vitro experiments further demonstrated that PGK1 was highly expressed in TNBC and enhanced cell proliferation and migration. Mechanistic studies revealed that PGK1 knockdown reduced intracellular lactate levels, and exogenous lactate rescued the proliferation defect in PGK1-deficient cells. Moreover, PGK1 knockdown significantly decreases overall lysine lactylation levels, an effect that can be restored by lactate supplementation. Furthermore, lactylome analysis identified 227 proteins with altered lactylation downstream of PGK1, enriching pathways critical for cancer progression. Collectively, we identified lactylation-related biomarkers in TNBC and uncovered that PGK1 may promote malignancy partly by modulating the lactate-lactylation axis, providing new perspectives for its diagnosis, prognosis assessment, and treatment strategies.