The aim of the present project is to increase our fundamental understanding of GBM pathobiology and find novel tumor biomarker signatures to address the major challenges in clinical management of glioblastoma: invasiveness of GBM cells and tumor heterogeneity.

Specific objectives are, to:

Objective I: Determine transcriptome profile and biomarkers of invasive and therapy-resistant patient-derived GBM cells that survive standard treatment protocol with irradiation and chemotherapeutic temozolomide in vitro using 3D GBM cellular models.

Objective II: Identify individual biomarkers or their signatures with prognostic and/or predictive value in a cohort of 300 GBM patients by exploring associations between molecular signatures, radiographic features (MRI) and clinical data using artificial intelligence approaches. Molecular signature will combine information on biomarker expression at gene and protein levels, composition of the tumor microenvironment and immune cell infiltration.

Objective III: Develop and validate computational models to predict molecular signatures and subtypes of GBM, the survival and treatment responses of GBM patients using non-invasive material.

Project will be implemented in 5 work packages.