Reconstructing the Single-Cell Spatiotemporal Dynamics of Glioblastoma Invasion

Glioblastoma invasion into healthy brain tissue remains a major barrier to effective treatment, yet current models fail to capture its full complexity in a scalable and patient-specific manner. Here, we introduce GlioTrace, a novel ex vivo imaging and AI-based analytical framework that enables real-time, spatiotemporal tracking of glioblastoma invasion dynamics in patient-derived glioma cell culture xenograft (PDCX) brain slices. By integrating whole-specimen confocal microscopy, vascular counterstaining, and an advanced computational pipeline combining convolutional neural networks and Hidden Markov Models, GlioTrace identifies distinct invasion modes, including dynamic morphological switching, vessel-guided migration, and immune cell interactions and quantifies patient-specific variations in invasion plasticity. Using GlioTrace, we demonstrate that targeted therapies can selectively modulate invasion phenotypes, revealing spatially and temporally distinct drug responses. This scalable platform provides an unprecedented window into glioblastoma progression and treatment response, offering a powerful tool for precision oncology and anti-invasion therapeutic development.