Wei Huang,1 Wei Wang,2 Tuo-zhou Dong1 

1General Surgery, The First People’s Hospital of Jiande, Jiande, Zhejiang, 311600, People’s Republic of China; 2Urology, The First People’s Hospital of Jiande, Jiande, Zhejiang, 311600, People’s Republic of China

Correspondence: Tuo-zhou Dong, General Surgery, The First People’s Hospital of Jiande, No. 599 Yanzhou Avenue, Xin’anjiang Street, Jiande, Zhejiang, 311600, People’s Republic of China, Email dongtuozhou@163.com

Objective: Telomere maintenance mechanism significantly impacts the metastasis, progression, and survival of breast cancer (BC) patients. This study aimed to investigate the role of telomere maintenance-related genes (TMRGs) in BC prognosis and to construct a related prognostic model.
Methods: Differentially expressed genes were identified from the TCGA-BC cohort, and functional enrichment analysis was conducted. TMRGs were sourced from the literature and intersected with DEGs. Candidate genes were selected using machine learning algorithms, including Lasso Cox, Random Forest, and XGBoost. Multivariate Cox regression analysis was conducted to construct a prognostic model and identify hub genes. Subsequent analyses included survival analysis, gene set enrichment analysis (GSEA), immune infiltration analysis, and drug sensitivity analysis of the hub genes. Finally, in vitro experiments were conducted to validate the expression of the hub genes.
Results: A total of 1329 differentially expressed TMRGs were analyzed, with 128 significantly associated with overall survival. Machine learning identified 7 prognosis-related TMRGs: MECP2, PCMT1, PFKL, PTMA, TAGLN2, TRMT5, and XRCC4. These genes were used to construct a prognostic model, with MECP2, PCMT1, PFKL, TAGLN2, and XRCC4 as harmful factors, while PTMA and TRMT5 were protective. The model demonstrated a significant prognostic value (AUC: 0.81, 0.72, 0.69 for 1-, 3-, and 5-year, respectively). Survival analysis confirmed the prognostic relevance of these genes, and GSEA highlighted their roles in oxidative phosphorylation, glycolysis, and PI3K/AKT/mTOR signaling.
Conclusion: The study identified 7 key TMRGs with significant prognostic value in BC. The constructed model effectively stratifies patient risk, providing a foundation for targeted therapies and personalized treatment strategies.

Keywords: breast cancer, telomere, risk score, machine learning, immune infiltration, seven hub genes