Grant: Team Grant: ERA-Net PerMed

Glioblastoma (GBM) is one of the most aggressive brain tumors with a grim prognosis and median survival times of less than 15 months. Despite multimodal therapy, including surgical resection and radio-chemotherapy, recurrence is inevitable. However, time to recurrence is an individual parameter in patients and depends on genetic factors and on clinical parameters such as extent of resection. In recurrent GBM, repeated surgery can substantially prolong the lifespan if recurrent tumors are detected at early stages. Until now, recurrence monitoring is based on radiologic imaging but suffers from limited sensitivity and confounding effects. Therefore, circulating "liquid biopsy" markers predicting the time-window of recurrence are an unmet clinical need. We propose to use a spectrum of "Omics" methods to dissect the individual parameters that determine recurrence and therapy resistance in GBM. We will use a retrospective cohort of 100 patients clinically fully annotated GBM primary and recurrent samples. Abundance-based biomarkers associated with time-to-recurrence and therapy resistance will be obtained by analyzing extracellular vesicles (EVs) with respect to proteolytic fragments (mass-spectrometry based), protease actvitivies, and miRNA content. In addition, a full NGS analysis will be performed to characterise recurrent tumors in each individual patient to combine all "Omics" with clinical and imaging data. From these analyses, we propose to identify personalised easy-to-access biopsy markers so that investigation of pre- and post-surgery blood samples (at time of initial surgery) allows to faithfully discern tumor-originating molecules through their decreased abundance in post-surgery samples. We further hypothesize that multimodal monitoring, e.g. combining imaging data with molecular markers allows to establish a molecular signature of GBM recurrence which is particularly powerful for reliable prediction of GBM recurrence in a prospective patient cohort.