Zheng Yifan, Li Li, Cai Wenqian, Li Lin, Zhang Rongxin, Huang Wenbin, Cao Yulun
Department of General Surgery I, The First Affiliated Hospital of Guangdong Pharmaceutical University, Guangzhou, Guangdong, China.
Guangdong Provincial Key Laboratory for Biotechnology Drug Candidates, Institute of Basic Medical Sciences and Department of Biotechnology, School of Life Sciences and Biopharmaceutics, Guangdong Pharmaceutical University, Guangzhou, China.
Front Oncol. 2024 Nov 27;14:1488137. doi: 10.3389/fonc.2024.1488137. eCollection 2024.
The TGF-β signaling pathway (TSP) is pivotal in tumor progression. Nonetheless, the connection between genes associated with the TSP and the clinical outcomes of breast cancer, as well as their impact on the tumor microenvironment and immunotherapeutic responses, remains elusive.
Breast cancer transcriptomic and single-cell sequencing data were obtained from the The Cancer Genome Atlas (TCGA) and the Gene Expression Omnibus (GEO) databases. We identified 54 genes associated with the TSP from the Molecular Signatures Database (MSigDB) and analyzed both data types to evaluate TSP activity. Using weighted gene co-expression network analysis (WGCNA), we identified modules linked to TSP activity. To assess patient risk, we used 101 machine learning algorithms to develop an optimal TGF-β pathway-related prognostic signature (TSPRS). We then examined immune activity and response to immune checkpoint inhibitors and chemotherapy in these groups. Finally, we validated ZMAT3 expression levels clinically and confirmed its relevance in breast cancer using CCK-8 and migration assays.
At the single-cell level, TSP activity was most notable in endothelial cells, with higher activity in normal tissues compared to tumors. TSPRS was developed. This signature's accuracy was confirmed through internal and external validations. A nomogram incorporating the TSPRS was created to improve prediction accuracy. Further studies showed that breast cancer patients categorized as low-risk by the TSPRS had higher immune phenotype scores and more immune cell infiltration, leading to better prognosis and enhanced immunotherapy response. Additionally, a strong link was found between the TSPRS risk score and the effectiveness of anti-tumor agents. Silencing the ZMAT3 gene in the TSPRS significantly reduced the proliferation and invasiveness of breast cancer cells.
Our study developed a TSPRS, which emerges as a potent predictive instrument for the prognosis of breast cancer, offering novel perspectives on the immunotherapeutic approach to the disease.
转化生长因子-β信号通路（TSP）在肿瘤进展中起关键作用。然而，与TSP相关的基因与乳腺癌临床结局之间的联系，以及它们对肿瘤微环境和免疫治疗反应的影响，仍不清楚。
从癌症基因组图谱（TCGA）和基因表达综合数据库（GEO）获取乳腺癌转录组和单细胞测序数据。我们从分子特征数据库（MSigDB）中鉴定出54个与TSP相关的基因，并分析这两种数据类型以评估TSP活性。使用加权基因共表达网络分析（WGCNA），我们确定了与TSP活性相关的模块。为评估患者风险，我们使用101种机器学习算法开发了一种最佳的转化生长因子-β通路相关预后特征（TSPRS）。然后我们检查了这些组中的免疫活性以及对免疫检查点抑制剂和化疗的反应。最后，我们在临床上验证了ZMAT3的表达水平，并使用CCK-8和迁移实验证实了其在乳腺癌中的相关性。
在单细胞水平上，TSP活性在内皮细胞中最为显著，与肿瘤相比，正常组织中的活性更高。开发了TSPRS。通过内部和外部验证证实了该特征的准确性。创建了一个纳入TSPRS的列线图以提高预测准确性。进一步研究表明，被TSPRS分类为低风险的乳腺癌患者具有更高的免疫表型评分和更多的免疫细胞浸润，从而导致更好的预后和增强的免疫治疗反应。此外，在TSPRS风险评分与抗肿瘤药物疗效之间发现了密切联系。在TSPRS中沉默ZMAT3基因显著降低了乳腺癌细胞的增殖和侵袭性。
我们的研究开发了一种TSPRS，它成为一种强大的乳腺癌预后预测工具，为该疾病的免疫治疗方法提供了新的视角。
