Disease progression and treatment-induced alterations in glioblastoma

Background & objectives. Glioblastoma constitutes the most common malignant brain tumor in adults and is associated with an aggressive disease course. Since 2005 patients are treated according to a standardized first-line treatment protocol, which has resulted in increased survival and quality-of-life. Still, despite intense treatment efforts early tumor recurrence is almost inevitable with median overall survival times of approximately 10 months. Preclinical models have recently provided important insights into the complex mechanisms of therapy-response and malignant tumor-progression. Crucial factors include hypoxia, angiogenesis, tumor-stroma interactions, inflammatory response, selection of tumor stem-cells, and mechanisms mediating glial-mesenchymal transition. In contrast to numerous preclinical models, virtually no in-vivo data on human glioblastoma samples exist. Thus, the aim of this project is to study morphological and molecular changes associated with tumor progression and therapy-resistance in vivo, in matched pre- and post-therapeutic glioblastoma samples. Patients & methods. A unique cooperation of the Austrian Brain Tumor Registry with the Society of Austrian Neuro-Oncology provides a nationwide platform ensuring ascertainment of a homogenously treated patient cohort, in whom multiple resections are available for tissue-based analyses. Comprehensive matched analyses of pre- and post-therapeutic glioblastoma tissues will be conducted using an integrated morphological and molecular (transcriptomic, DNA methylation and genomic profiling) approach. Originality & patient benefit. Matched analyses of pre- and post-therapeutic glioblastoma samples including morphological and molecular techniques constitute an innovative research approach, which will considerably increase the understanding of therapy-related effects and tumor progression in glioblastoma and will help to develop new treatment strategies.

Glioblastoma is the most common malignant brain tumor in adults that is associated with a highly aggressive disease course and poor overall survival. Project KLI394 focused on the central research question: "What happens with glioblastoma over time?" Which mechanisms are involved in therapeutic resistance and tumor recurrence? How and to what extent do recurrent tumors differ from their primary counterparts? To answer these questions we assembled a large, nation-wide cohort of longitudinal tumor tissues, which we analyzed using a cross-disciplinary approach. Precisely, we combined genome-wide DNA methylation sequencing with detailed clinical phenotypes, digital pathology and neuroimaging data and integrated the different data qualities using advanced biostatistics and machine learning approaches. Thereby we were able to show that epigenetic mechanisms contribute significantly to spatial and temporal tumor heterogeneity in glioblastoma, and recurrent alterations in the WNT pathway might indicate a therapeutic vulnerability of a subset of tumors. Overall, using DNA methylation sequencing as a single test platform we were able to infer multiple, clinically relevant markers such as IDH mutation, MGMT methylation, 1p19q codeletion, or transcriptional subtypes. This means that DNA methylation screening might constitute an efficient and cost-effective test platform in the routine clinical setting. In addition to the gain of scientific knowledge and valuable resource of this glioblastoma cohort, the research project started a fruitful collaboration between MedUni Vienna, the Austrian Brain Tumor Registry Network, patient representatives and the Research Center of Molecular Medicine of the Austrian Academy of Sciences.