Advancing Breast Cancer Biomarkers: A Centromere-Related Gene Signature Integrated with Single-Cell Analysis for Prognostic Prediction

Lu Ye¹Pei Shengbin¹Wenxiang Zhang¹Zheng Qu¹Xiangyu Wang¹Siqing Liu¹Hao Dong¹Kan Yonemori²Yi Fang¹Xiangyi Kong¹Jing Wang¹GAO JIDONG^1*

• ¹Department of Breast Surgery, Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China
• ²National Cancer Center Hospital, tokyo, Japan

Abstract Breast cancer (BC) is the most common malignancy among women and exhibits significant heterogeneity. Among its subtypes, triple-negative breast cancer (TNBC) has an inferior prognosis. Despite the use of molecular stratification tools like Oncotype DX and MammaPrint, prognostic models based on chromosomal instability remain inadequate. The centromere protein (CENP) family, key regulators of genomic stability, are linked to tumor progression through aberrant expression. In this study, we integrated multi-omics data, including RNA transcriptomic profiles and single-cell RNA sequencing, and used weighted gene co-expression network analysis (WGCNA) to identify gene modules associated with CENPA. A prognostic model was developed using Cox regression and the LASSO algorithm. Validation in independent cohorts demonstrated that the model effectively stratified patients into high-and low-risk groups, with the high-risk group showing significantly reduced five-year survival (p < 0.001). Additionally, single-cell analysis revealed that CENPA-high subpopulations were enriched in proliferative tumor cells and associated with an immunosuppressive tumor microenvironment. This study is the first to construct a CENP-based prognostic model for BC, providing novel biomarkers and potential therapeutic targets for personalized treatment. The biological function of the key molecule MMP1 was further validated through in vitro and in vivo experiments.