Bioengineering Glioblastoma Cell Migration Conduits to Mimic Complex and Biorelevant Tumour Microenvironments

Glioblastoma (GMB) treatment remains a substantial unmet need due to its aggressive and highly infiltrative characteristics, facilitated by microstructures within brain tissue such as capillaries and neuronal projections. These key features are seldom represented in pre-clinical models that attempt to evaluate underpinning migratory pathways, leaving drug screening efforts and clinical translation futile. Here, our innovative approach towards modelling biorelevant GBM microenvironments incorporates migratory tracts that are synonymous with migratory pathways observed in GBM patients. This study utilises glioma cell lines U87 and U251 to identify preferential migration strategies in response to physiochemical and mechanical gradients that are fashioned using extracellular matrix-like components. GBM spheroids were printed into collagen conduits of varying density, all encased in a non-cell adhesive hydrogel with stiffness akin to brain matter. Using our newly developed and physiologically advanced bioprinted system and following from our group’s previous work on investigating the role of RhoGTPase-activating proteins (ARHGAPs) as potential contributors to GBM therapeutic resistance (Cheng et al, Cell Reports, 2025) we then introduced stable knockdowns of ARHGAP 12 and 29 to elucidate whether a synergistic effect on migration could be initiated with a panel of antimigratory drugs (CCG-1423, rhosin or combination). Using confocal and light-sheet microscopy, MTT viability assays and physical cytometry, distinct differences were observed in phenotype, adhesion, migration and actin polymerisation between models, with models containing ARHGAP 29 knockdowns and drug cocktail combinations displaying significantly reduced migratory activity, increased adhesion within spheroids, reduced F-actin signalling and reduced cell viability. When compared with classical 2D migration models, striking differences were seen in migratory behaviours, supporting the need for more biorelevant models in predicting in vivo responses. From this, we propose that these sophisticated systems are more capable of tailoring novel, personalised treatments for GBM to subsequently improve patient prognosis.