The Slovenian translational platform GlioBank for brain tumor research: Identification of molecular signatures of glioblastoma progression.

Abstract

Background: Glioblastoma (GB) is one of the most lethal solid tumors in humans, with an average patient life expectancy of 15 months and a 5-year survival rate of 5%-10%. GB is still uncurable due to tumor heterogeneity and invasive nature as well as therapy-resistant cancer cells. Centralized biobanks with clinical data and corresponding biological material of GB patients facilitate the development of new treatment approaches and the search for clinically relevant biomarkers, with the goal of improving outcomes for GB patients. The aim of this study was firstly to establish a Slovenian translation platform, GlioBank, and secondly to demonstrate its utility through the identification of molecular signatures associated with GB progression and patient survival.

Methods: GlioBank contains tissue samples and corresponding tumor models as well as clinical data from patients diagnosed with glioma, with a focus on GB. Primary GB cells, glioblastoma stem cells (GSCs), and organoids have been established from fresh tumor biopsies. We performed expression analyses of genes associated with GB progression and bioinformatics analyses of available clinical and research data obtained from a subset of 91 GB patients. qPCR was performed to determine the expression of genes associated with therapy resistance and cancer cell invasion, including markers of different GB subtypes, GSCs, epithelial-to-mesenchymal transition, and immunomodulation/chemokine signaling in tumor tissues and corresponding cellular models.

Results: GlioBank contains biological material and research, and clinical data collected in the SciNote electronic laboratory notebook. To date, more than 240 glioma tissue samples have been collected and stored in GlioBank, most of which are GB tissues (205) and were further processed to establish primary GB cells (n = 64), GSCs (n = 14), and GB organoids (n = 17). Corresponding blood plasma (n = 103) and peripheral blood mononuclear cells (n = 101) are also stored. GB tumors were classified into 4 different subtypes that differed regarding patient survival; the mixed subtype exhibited the longest patient survival. High DAB2, S100A4, and STAT3 expression were associated with poor overall patient survival, and DAB2 was found to be an independent prognostic marker for GB survival. We analyzed the molecular signatures between different tumor regions (core vs. rim). STMN4, ERBB3, and ACSBG1 were upregulated in the tumor rim, suggesting that these genes are associated with the invasive nature of GB.

Conclusions: GlioBank is a centralized biobank that has been built by a multidisciplinary network with the aim of facilitating disease-oriented basic and clinical research. The advantages of GlioBank include the molecular characterization of GB based on targeted gene expression, the availability of diverse cellular models (eg, GB cells and organoids), and a large number of patient-matched tumor core and rim samples, all with accompanying molecular and clinical data. We report here for the first time an association between DAB2 expression and low overall survival in GB patients, indicative of a prognostic value of DAB2.