Glioblastoma multiforme (GBM) is the most frequent malignant brain tumor. No cure exists, and GBM patients’ median survival is <24 months. Therefore, a therapeutic strategy to perturb GSCs invasion in the human brain is critical. Since GBMs are highly heterogeneous and are unique from patient to patient, “one therapy for all” is not practical. Furthermore, invasive behavior is unpredictable in the human brain. The “Glioma-PerMed” is establishing an interdisciplinary research consortium aiming to develop personalised glioma invasion assays in the preclinical models of human brain organoids and zebrafish brains. We will quantitatively determine the GBM invasion behaviors with unbiased machine learning algorithms. As our “personalised glioma invasion assay” platform allows drug screenings, we will screen an FDA-approved drug library and identify molecules that can be transferred to clinical patients. Ultimately, these would be the first step towards personalised GBM medicine.