胶质母细胞瘤是最恶性的胶质瘤类型之一。胶质母细胞瘤肿瘤生长迅速、恶性程度高，具有快速的疾病进展特点。遗憾的是，目前缺乏有效的治疗方法。针对胶质母细胞瘤的有效策略是鉴定与其发生和进展相关的关键生物标志物，并将这些生物标志物发展为治疗靶点。本研究采用综合生物信息学分析方法，鉴定了差异表达基因（DEGs），其中130个基因在胶质母细胞瘤中上调表达，128个基因下调表达。基于基因本体富集分析和京都百科全书基因与基因组通路分析，这些基因与胶质母细胞瘤中肿瘤细胞粘附、分化和形态调控相关，并主要富集于补体和凝血级联通路。使用检索相互作用基因（STRING）数据库构建了蛋白质相互作用网络。鉴定出十个中心基因，包括FN1、CD44、MYC、CDK1、SERPINE1、COL3A1、COL1A2、LOX、POSTN和EZH2，所有这些基因在胶质母细胞瘤中显著上调表达，这些结果通过oncomine数据库探索得到了确认。中心基因的变异分析发现，至少有一个中心基因发生变异的患者，其中位生存时间较短（p = 0.013），无中心基因发生变异的患者的无病生存时间中位数也较短（p = 2.488E-3）。生存分析的多重检验表明，只有LOX的表达与患者生存相关（P < 0.05）。使用GDS4467数据集分析了不同恶性程度的胶质瘤中LOX的表达，发现LOX表达水平与胶质瘤的恶性程度呈正相关。通过分析GDS4535数据集发现，LOX的表达水平与胶质母细胞瘤细胞的分化程度呈正相关。结论：本研究表明FN1、CD44、MYC、CDK1、SERPINE1、COL3A1、COL1A2、LOX、POSTN和EZH2是胶质母细胞瘤的关键基因。然而，只有LOX与患者生存相关，并促进胶质母细胞瘤细胞分化和肿瘤复发。LOX可能是一种潜在的预后生物标志物和治疗靶点。版权所有 © 2023. Elsevier Inc. 发表。

Glioblastoma multiforme (GBM) is the most malignant type of glioma. GBM tumors grow rapidly, have a high degree of malignancy, and are characterized by a fast disease progression. Unfortunately, there is a lack of effective treatments. An effective strategy for the treatment of GBM would be to identify key biomarkers correlating with the occurrence and progression of GBM and developing these biomarkers into therapeutic targets.In this study, using integrated bioinformatics analysis, we identified differentially expressed genes (DEGs), including 130 genes that were upregulated in GBM compared to normal brain tissue, and 128 genes that were downregulated in GBM. Based on Gene Ontology enrichment analysis and Kyoto Encyclopedia of Genes and Genomes pathway analysis, these genes were associated with regulation of tumor cell adhesion, differentiation, morphology in GBM and were mainly enriched in Complement and coagulation cascades pathway. The Search Tool for the Retrieval of Interacting Genes (STRING) database was used to construct a Protein-Protein Interaction network. Ten hub genes were identified, including FN1, CD44, MYC, CDK1, SERPINE1, COL3A1, COL1A2, LOX, POSTN and EZH2, all of which were significantly upregulated in GBM, these results were confirmed by oncomine database exploration. Alteration analysis of hub genes found that patients with alteration in at least one of the hub genes showed shorter median survival times (p = 0.013) and shorter median disease-free survival times (p = 2.488E-3) than patients without alterations in any of the hub genes. Multiple tests for survival analysis showed that among individual hub genes only expression of LOX was correlated with patient survival (P < 0.05).GDS4467 data set was used to analyze the expression of LOX in gliomas with different degrees of malignancy, and it was found that the expression level of LOX was positively correlated with the malignant degree of gliomas.By analyzing GDS 4535 data set showed that the expression level of LOX was positively correlated with the differentiation degree of GBM cells CONCLUSION: This research suggests that FN1, CD44, MYC, CDK1, SERPINE1, COL3A1, COL1A2, LOX, POSTN and EZH2 are key genes in GBM. However, only LOX is correlated with patient survival and promotes glioblastoma cell differentiation and tumor recurrence. LOX may be a candidate prognostic biomarker and potential therapeutic target for GBM.Copyright © 2023. Published by Elsevier Inc.