Identification of biomarkers associated with extracellular vesicles based on an integrative pan-cancer bioinformatics analysis

Extracellular vesicle (EV) has received increasing attention over the last decade. However, biomarkers and mechanisms underlying remain largely limited. Three microarray profiles, GSE78718 (K562 leukemia cell line), GSE45301 (U87-MG glioblastoma cell line), and GSE9589 (SW480 colon cancer cell line), were analyzed for the overlapped differentially expressed genes (DEGs). SurvExpress was used for the prognostic analysis of hub genes signature. Predicted transcription factors networks were built by NetworkAnalysis. Characterization between hub genes and immune cells was analyzed by the tumor immune estimation resources (TIMER) and single-sample gene set enrichment analysis (ssGSEA). The most significantly enriched pathway was lysosome. Hub genes included lysosomal-associated membrane protein 1 (LAMP1), heat shock protein family A (Hsp70) member 5 (HSPA5), lysosomal-associated membrane protein 2 (LAMP2), integrin subunit alpha V (ITGAV), and transmembrane protein 30A (TMEM30A). Significant prognostic values of hub genes signature were identified in glioblastoma (P-value = 0.006), but not colon cancer. In colon cancer, ITGAV displayed remarkably high correlation with tumor immune infiltrating cells. In glioblastoma, the highest correlation was found between HSPA5 and dendritic cell. Moreover, distinct association of immune cells between cell and EV were identified via ssGSEA. This study identified biomarkers in EV with potential immunological insights and clinical values.