DataSheet_1_Molecular subtypes based on centrosome-related genes can predict prognosis and therapeutic responsiveness in patients with low-grade gliomas.docx

BackgroundAbnormalities in centrosome regulatory genes can induce chromosome instability, cell differentiation errors, and tumorigenesis. However, a limited number of comprehensive analyses of centrosome-related genes have been performed in low-grade gliomas (LGG).MethodsLGG data were extracted from The Cancer Genome Atlas (TCGA) and Chinese Glioma Genome Atlas (CGGA) databases. The ConsensusClusterPlus” R package was used for unsupervised clustering. We constructed a centrosome-related genes (CRGs) signature using a random forest model, lasso Cox model, and multivariate Cox model, and quantified the centrosome-related risk score (centS). The prognostic prediction efficacy of centS was evaluated using a Receiver Operating Characteristic (ROC) curve. Immune cell infiltration and genomic mutational landscapes were evaluated using the ESTIMATE algorithm, single-sample Gene Set Enrichment Analysis (ssGSEA) algorithm, and “maftools” R package, respectively. Differences in clinical features, isocitrate dehydrogenase (IDH) mutation, 1p19q codeletion, O6-methylguanine-DNA methyltransferase promoter (MGMTp) methylation, and response to antitumor therapy between the high- and low-centS groups were explored. “pRRophetic” R packages were used for temozolomide (TMZ) sensitivity analysis. qRT-PCR verified the differential expression of the centrosomal gene team, the core of which is CEP135, between LGG cells and normal cells.ResultsTwo distinct CRG-based clusters were identified using consensus unsupervised clustering analysis. The prognosis, biological characteristics, and immune cell infiltration of the two clusters differed significantly. A well-performing centS signature was developed to predict the prognosis of patients with LGG based on 12 potential CRGs. We found that patients in the high-centS group showed poorer prognosis and lower proportion of IDH mutation and 1p19q codeletion compared to those in the low-centS group. Furthermore, patients in the high-centS group showed higher sensitivity to TMZ, higher tumor mutation burden, and immune cell infiltration. Finally, we identified a centrosomal gene team whose core was CEP135, and verified their differential expression between LGG cells and normal glial cells.ConclusionOur findings reveal a novel centrosome-related signature for predicting the prognosis and therapeutic responsiveness of patients with LGG. This may be helpful for the accurate clinical treatment of LGG.

背景：中心体调控基因的异常背景可能导致染色体不稳定、细胞分化错误以及肿瘤发生。然而，在低级别胶质瘤（LGG）中，对中心体相关基因的综合分析数量有限。方法：从癌症基因组图谱（TCGA）和中国胶质瘤基因组图谱（CGGA）数据库中提取LGG数据。使用“ConsensusClusterPlus”R包进行无监督聚类。我们利用随机森林模型、lasso Cox模型和多元Cox模型构建了中心体相关基因（CRGs）特征，并量化了中心体相关风险评分（centS）。通过接收者操作特征（ROC）曲线评估centS的预后预测效能。分别使用ESTIMATE算法、单样本基因集富集分析（ssGSEA）算法和“maftools”R包评估免疫细胞浸润和基因组突变景观。探讨了高centS组和低centS组在临床特征、异柠檬酸脱氢酶（IDH）突变、1p19q共同缺失、O6-甲基鸟嘌呤-DNA甲基转移酶启动子（MGMTp）甲基化以及抗肿瘤治疗反应方面的差异。使用“pRRophetic”R包进行替莫唑胺（TMZ）敏感性分析。qRT-PCR验证了中心体基因团队（核心为CEP135）在LGG细胞与正常细胞之间的差异表达。结果：通过一致性无监督聚类分析确定了两个基于CRG的独立簇。两个簇的预后、生物学特征和免疫细胞浸润存在显著差异。基于12个潜在的CRG，我们开发了一个表现良好的centS特征，用于预测LGG患者的预后。我们发现，与低centS组相比，高centS组的患者预后较差，IDH突变和1p19q共同缺失的比例较低。此外，高centS组的患者对TMZ的敏感性更高，肿瘤突变负荷更高，免疫细胞浸润更严重。最后，我们识别了一个以CEP135为核心的中心体基因团队，并验证了其在LGG细胞与正常胶质细胞之间的差异表达。结论：我们的研究揭示了LGG患者预后和治疗反应预测的一个新的中心体相关特征。这或许有助于LGG的精确临床治疗。