Wu Senbin, Li Qiaohong, Huang Lilan, Li Hong, Ma Ying
The Second Affiliated Hospital, Guangzhou Medical University, Guangzhou, Guangdong, China.
The Shenzhen Yantian District People's Hospital, Shenzhen, Guangdong, China.
Sci Rep. 2025 Jul 25;15(1):27069. doi: 10.1038/s41598-025-11704-5.
Glioma, among the most prevalent and heterogeneous intracranial malignancies, exhibit dismal prognoses despite advances in molecular diagnostics. Protein kinases, critical regulators of tumorigenesis, remain underexplored as systematic prognostic markers in glioma. In this study, we identified 29 differentially expressed protein kinases in TCGA (LGG and GBM) and GEO (GSE16011) databases. These dysregulated kinases were enriched in biological functions and signaling pathways related to glioma progression. Through univariate Cox, LASSO, and multivariate Cox regression, we developed a novel protein kinase prognostic risk signature comprising CAMKK2, PAK1, AK5, WEE1, and CDK4 and validated its performance to predict outcomes. We also constructed a nomogram integrating clinical characteristics and the kinase signature, which showed a high accuracy in predicting overall survival (OS) of glioma patients in TCGA, CGGA and GSE43378 cohorts. Furthermore, GSEA analysis based on risk scores revealed that various signaling pathways associated with glioma genesis and progression were enriched in the high-risk group. Immune cell infiltration analysis displayed high-risk group was characterized by a higher infiltration level of stromal and immune cells and upregulation of key immune checkpoint genes. Finally, we confirmed the protein expression of five protein kinases in glioma samples and normal samples. In conclusion, we developed a novel prognostic signature for glioma. The signature not only has potential as a prognostic biomarker but may also serve as a therapeutic target for glioma, offering new insights into glioma treatment and prognosis assessment.
胶质瘤是最常见且异质性最高的颅内恶性肿瘤之一，尽管分子诊断技术有所进步，但其预后仍不容乐观。蛋白激酶作为肿瘤发生的关键调节因子，在胶质瘤中作为系统性预后标志物仍未得到充分研究。在本研究中，我们在TCGA（低级别胶质瘤和胶质母细胞瘤）和GEO（GSE16011）数据库中鉴定出29种差异表达的蛋白激酶。这些失调的激酶在与胶质瘤进展相关的生物学功能和信号通路中富集。通过单变量Cox、LASSO和多变量Cox回归分析，我们开发了一种新的蛋白激酶预后风险特征，包括钙调蛋白依赖性蛋白激酶激酶2（CAMKK2）、p21激活激酶1（PAK1）、腺苷酸激酶5（AK5）、WEE1 G2期检查点激酶（WEE1）和细胞周期蛋白依赖性激酶4（CDK4），并验证了其预测预后的性能。我们还构建了一个整合临床特征和激酶特征的列线图，该列线图在预测TCGA、CGGA和GSE43378队列中胶质瘤患者的总生存期（OS）方面显示出高准确性。此外，基于风险评分的基因集富集分析（GSEA）显示，与胶质瘤发生和进展相关的各种信号通路在高风险组中富集。免疫细胞浸润分析显示，高风险组的特征是基质和免疫细胞浸润水平较高以及关键免疫检查点基因上调。最后，我们证实了胶质瘤样本和正常样本中五种蛋白激酶的蛋白表达。总之，我们开发了一种新的胶质瘤预后特征。该特征不仅有潜力作为预后生物标志物，还可能作为胶质瘤的治疗靶点，为胶质瘤治疗和预后评估提供新的见解。
