Glioblastoma (GBM) is the most lethal tumour in the central nervous system (CNS), GBM has a poor prognosis due to treatment tolerance and tumour recurrence; new molecular biomarkers are needed to acquire accurate prognosis and to promote therapeutic strategies. Data from Gene Expression Omnibus (GEO) was analysed to screen differentially expressed genes (DEGs), and 279 DEGs were screened. The protein-protein interaction (PPI) network of DEGs was constructed and visualized, top 10 hub genes were identified by using Cytoscape consequently. The function of DEGs was explored by enrichment analysis, DEGs were enriched in tumour-associated biologic processions and pathways. Gene Expression Profiling Interactive Analysis (GEPIA) database was used to identify prognostic genes; serum amyloid A1 (SAA1) was identified as a critical prognostic gene due to higher SAA1 expression associated with poor overall survival (OS) (HR = 1.5, p < .05) and poor disease-free survival (DFS) (HR = 1.9, p < .01). Dataset from The Chinese Glioma Genome Atlas database validated the prognostic value of SAA1 and reported the relationship between SAA1 expression and clinical characteristics, including age, sex, history of relapse, and the status of IDH. Gene set enrichment analysis (GSEA) identified six SAA1-related pathways; the identification of pathways could provide insight into the therapeutic strategies of GBM. Lastly, the relationship between SAA1 expression and immune infiltration was explored, and the result showed that SAA1 expression negatively correlated with the infiltration level of T cells, and SAA1 expression positively correlated with the infiltration level of Treg cells. The overexpression of SAA1 was associated with poor OS and DFS in GBM, and the expression of the SAA1 gene may affect the infiltration level of immune cells. Therefore, SAA1 could be a promising prognostic biomarker associated with immune infiltration and therapeutic target for GBM.
Keywords: Serum amyloid A1; biomarker; glioblastoma; immunotherapy; prognosis.