Advancing breast cancer biomarkers: a centromere-related gene signature integrated with single-cell analysis for prognostic prediction.

Background: Breast cancer (BC) is the most common malignancy among women and shows significant heterogeneity in its prognosis. Among the subtypes, triple-negative breast cancer (TNBC) has the poorest prognosis. Despite advancements in molecular stratification tools, such as Oncotype DX and MammaPrint, prognostic models based on chromosomal instability are still insufficient. The centromere protein (CENP) family, which plays a crucial role in maintaining genomic stability, is associated with tumor progression due to aberrant expression. Methods: In this study, we integrated multi-omics data, including RNA transcriptomic profiles and single-cell RNA sequencing, to identify gene modules linked to CENPA using weighted gene co-expression network analysis (WGCNA). We developed a prognostic model employing Cox regression and the LASSO algorithm. Validation was performed on independent cohorts, and the model's performance was tested by stratifying patients into high- and low-risk groups based on their five-year survival rates (p < 0.001). Results: The prognostic model effectively identified high- and low-risk patient groups, with the high-risk group showing significantly reduced five-year survival. Single-cell analysis revealed that CENPA-high subpopulations were enriched in proliferative tumor cells and were associated with an immunosuppressive tumor microenvironment. Conclusion: This study is the first to establish a CENP-based prognostic model for BC, offering novel biomarkers and potential therapeutic targets for personalized treatment. Additionally, the biological function of the key molecule MMP1 was validated through both in vitro and in vivo experiments.