基于凋亡和化疗相关基因的乳腺癌预后模型构建及PCDHB2在肿瘤干性中的作用

An apoptosis- and chemotherapy-related gene signature for prognosis prediction in breast cancer and the role of PCDHB2 in cancer stemness

摘要:

背景与目的 乳腺癌是全球女性发病率最高的恶性肿瘤之一，化疗耐药和复发转移仍是影响患者长期生存的主要难题。细胞凋亡异常被认为是化疗耐药形成的重要机制，但凋亡和化疗相关基因在乳腺癌预后评估及耐药形成中的协同作用尚缺乏系统性研究。本研究旨在构建基于凋亡和化疗相关基因的多基因预后风险模型，并进一步探讨关键基因在乳腺癌肿瘤干性及化疗耐药中的潜在作用。方法 基于TCGA数据库获取乳腺癌转录组及临床数据，筛选凋亡和化疗相关差异表达基因。通过单因素Cox回归、LASSO回归及多因素Cox回归构建多基因预后风险模型，并采用时间依赖性ROC曲线、列线图及GEO外部数据集进行验证。同时分析风险评分与临床特征、免疫细胞浸润及化疗药物敏感性的关系。选取模型中的关键基因，通过细胞划痕、Transwell及Western blot实验验证其对乳腺癌细胞迁移、侵袭及肿瘤干性相关蛋白表达的影响。结果 共筛选出10个与乳腺癌预后显著相关的凋亡和化疗关键基因，成功构建预后风险模型。该模型在TCGA队列中对3、5年生存率的预测AUC分别为0.705和0.650，并在GEO外部验证队列中表现出良好的稳定性（3年AUC=0.741）。风险评分为乳腺癌患者的独立预后因素，并与晚期TNM分期、免疫抑制性细胞浸润增加及NU7441、氟达拉滨等化疗药物耐药显著相关。PCDHB2在高风险患者中高表达，其沉默可显著抑制MDA-MB-231细胞的迁移和侵袭能力，并下调干性相关蛋白SOX2的表达。结论 本研究构建了基于凋亡和化疗相关基因的多基因预后风险模型，可用于乳腺癌患者的风险分层和个体化生存预测。PCDHB2可能通过调控肿瘤细胞干性参与化疗耐药过程，具有潜在的临床转化价值，为乳腺癌耐药机制研究及精准治疗策略的制定提供了新的理论依据。

Abstract:

Background and Aims Breast cancer is the most prevalent malignancy among women worldwide, and chemotherapy resistance and tumor recurrence remain major obstacles to long-term survival. Dysregulation of apoptosis is considered a key mechanism underlying chemoresistance; however, the synergistic roles of apoptosis- and chemotherapy-related genes in prognostic stratification and drug resistance have not been fully elucidated. This study aimed to construct a multigene prognostic risk model based on apoptosis- and chemotherapy-related genes and to further investigate the role of the key gene in breast cancer stemness and chemoresistance.Methods Transcriptomic and clinical data of breast cancer patients were obtained from The Cancer Genome Atlas (TCGA). Differentially expressed apoptosis- and chemotherapy-related genes were identified, and a multigene prognostic risk model was constructed using univariate Cox regression, LASSO regression, and multivariate Cox regression analyses. The predictive performance of the model was evaluated by time-dependent ROC curves, nomograms, and an external GEO dataset. Associations between the risk score and clinical characteristics, immune cell infiltration, and chemotherapeutic drug sensitivity were further analyzed. Functional assays, including wound healing, Transwell invasion, and Western blot analyses, were performed to validate the biological role of the key gene in breast cancer cells.Results A prognostic signature comprising 10 apoptosis- and chemotherapy-related genes was established. The model demonstrated favorable predictive performance in the TCGA cohort, with AUC values of 0.705 and 0.650 for 3- and 5-year overall survival rate, respectively, and was further validated in an external GEO dataset (3-year AUC=0.741). The risk score was identified as an independent prognostic factor and was significantly associated with advanced TNM stage, increased infiltration of immunosuppressive cells, and resistance to chemotherapeutic agents, including NU7441 and fludarabine. PCDHB2 was highly expressed in high-risk patients, and its knockdown markedly inhibited migration and invasion of MDA-MB-231 cells while reducing the expression of the stemness-related protein SOX2.Conclusion This study established a robust multigene prognostic model based on apoptosis- and chemotherapy-related genes for risk stratification and personalized survival prediction in breast cancer. PCDHB2 may contribute to chemoresistance by regulating cancer stemness, highlighting its potential as a novel therapeutic target and providing new insights into breast cancer precision therapy.