Alteration Of Blood-brain Barrier In Glioma Tumor Microenvironment And Its Clinical Significance

Aims: More researches indicate that the tumor microenvironment(TME)is a key factor involved in multiple stages of cancer development,particularly local resistance,immune escape,and distant metastasis.A deeper understanding of TME will facilitate the evaluation and selection of drug candidates to effectively treat malignancies in primary and metastatic settings.Due to the histological and functional specificity of the brain,the central nervous system(CNS)tumors exhibit multifarious microenvironmental features that differ from other solid tumors,such as unique stromal cells infiltration,breaking to the integrity of the blood-brain barrier(BBB),etc.Glioma is the most common CNS tumors,and its high grade subtype is one of the most lethal malignant tumors.During the occurrence and progression of glioma,there are obviously structural and functional changes about BBB,but the biological and clinical significance of these pathological changes is still unclear.In this study,we will use the gene expression profile data of glioma tissue and specific algorithms to analyze the BBB features in glioma microenvironment,and focus on the association between BBB alteration with the clinical phenotype and prognosis of glioma.Methods: Our study collated the RNA-seq data of low grade glioma(LGG)and glioblastoma multiforme(GBM)in the The Cancer Genome Altas(TCGA)database and the corresponding patient clinical information.In addition,we gathered serveral additional glioma cohorts through other databases,totally involving more than 1500 glioma patients cross low grade to high grade.According to GSVA algorithm,we calculated the degree of infiltration of six stromal cells in glioma TME and their correlation with tumor purity,we then developed the BBB-score as the core of this study,which can be regarded as an important feature of BBB changes.Next,we performed statistical analysis of BBB-score in different glioma samples of malignancy,and used COX regression model to study the effect of BBB-score on survival prediction.Finally,based on genes enrichment analysis and other bioinformatics methods,we investigated the association between BBB-score with other TME factors.Results: By GSVA algorithm,we obtained the infiltration proportion of six stromal cells(endothelial cells,astrocytes,pericytes,microglia,neuron,and oligodendrocytes)in glioma TME.We found that the infiltration of the first four cells(endothelial cells,astrocytes,pericytes,and microglia)was negatively correlated with tumor purity,but neurons and oligodendrocytes were positively correlated with tumor purity.There was a collinear relationship between the GSVA scores of the three cells(endothelial cells,astrocytes,and pericytes)that consititute of BBB.Therefore,we defined the BBB-score index based on the average vaule of the three GSVA scores.Importantly,high BBB-score would be apparent in more advanced,more malignant gliomas,and BBB-score could also significantly predict patient survival time,and these results can be verified in multiple independent cohorts.Finally,we observed that BBB-score also linked to other TME factors,such as extracellular matrix(ECM)reorganization,anti-tumor immunophenotype.Taken together,our integrated analysis of pan-glioma suggests that BBB alteration remodeling glioma TME,and exhibited potential value for glioma prognostic stratification.Assessing BBB alteration may help inelucidating the complex role of glioma microenvironment and provide new insights into its clinical management.