Lactylation-driven gene signatures define breast cancer prognosis: a predictive model and insights into immune microenvironment dynamics.

[BACKGROUND] Breast cancer is a leading cause of cancer-related mortality worldwide, with poor prognosis largely due to its invasive and metastatic nature. Tumor lactylation plays a crucial role in cancer progression by influencing immune modulation and metabolic reprogramming. This study aimed to identify lactylation-related gene signatures associated with breast cancer prognosis and develop a predictive survival model.

[METHODS] Bioinformatics analyses were performed using RNA-seq and clinical data from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) datasets. An unsupervised consensus clustering analysis was applied to classify breast cancer samples into distinct groups based on lactylation-related gene expression. Differentially expressed genes (DEGs) between clusters were identified and subjected to functional enrichment analysis. To assess immune-related differences between groups, the ESTIMATE and CIBERSORT algorithms were used, along with an analysis of human leukocyte antigen (HLA) and immune checkpoint molecule expression levels, to explore the relationship between lactylation and the breast cancer immune microenvironment. A prognostic model was constructed using univariate Cox and Lasso regression analyses, followed by validation. Machine learning techniques identified key biomarkers, which were further analyzed for clinical relevance. Additionally, single-cell clustering was performed to investigate the expression patterns of these genes within the breast cancer microenvironment.

[RESULTS] Consensus clustering identified two distinct groups: high and low lactylation. Differentially expressed genes were enriched in pathways related to cytokine-cytokine receptor interaction, immune response, cell activation, and adhesion. Lactylation-related genes were found to influence immune cell infiltration in the breast cancer microenvironment. Thirty-seven prognostic lactylation-related genes were identified through univariate Cox regression and used to develop a predictive model. The high-risk group was associated with poorer survival, and the model's performance was validated in the GEO cohort. Specific hub genes involved in immune modulation and malignant cell proliferation were also identified.

[CONCLUSION] We successfully developed a lactylation-based prognostic model that can assess breast cancer prognosis and potentially guide personalized treatment strategies.