BB News InternationalBB NEWS INTERNATIONALThe Breaking Bulletin News Agency
Health and Development

AI Breakthrough "GlioTrace" Tracks Brain Cancer Cells Live for Five Days, Revealing How Glioblastoma Shifts Shape to Evade Treatment

A Swedish team's AI method, GlioTrace, live-tracks thousands of glioblastoma cells for five days, capturing how the deadliest brain cancer changes shape to invade healthy tissue and survive standard therapies. The Nature Communications study could reshape how researchers test new treatments.

Microscopy image of a glioblastoma cell spheroid

Full articles are free to read through December 10, 2026 (IST). Subscriptions remain available.

UPPSALA, Sweden — Medical scientists in Sweden have unveiled a powerful new artificial intelligence and live-imaging method capable of tracking thousands of individual brain cancer cells in real time over five days, exposing how deadly glioblastoma tumors adapt their movement to invade healthy tissue and survive standard therapies. Published Wednesday in the peer-reviewed journal Nature Communications and announced by Uppsala University, the breakthrough platform—named GlioTrace—gives researchers an unprecedented window into why glioblastoma remains one of the most aggressive and treatment-resistant human cancers.

Glioblastoma accounts for the majority of malignant primary brain tumors in adults across the United States and Europe, with median survival often measured in months because microscopic cancer cells infiltrate deep into surrounding brain tissue beyond the reach of neurosurgeons. Until now, oncologists have largely relied on static tissue snapshots to infer how tumors spread. Led by Uppsala University researcher Hitesh Mangukiya, doctoral candidate Madeleine Skeppås, and Integrative Cancer Research Professor Sven Nelander alongside computational partners at Chalmers University of Technology and the University of Gothenburg, the team cultivated tumor cells from six human patients inside living mouse brain slices and photographed them via automated microscopy at least every two hours for up to five days.

Feeding those continuous time-lapse sequences into a custom machine-learning pipeline allowed investigators to reconstruct the individual trajectories and morphological shifts of thousands of migrating cancer cells across entire brain slices. The analysis revealed striking behavioral plasticity—while some patient tumors traveled primarily along blood vessels and others dispersed diffusely through neural tissue, individual cancer cells proved capable of changing shape and switching between vascular highways and independent tissue migration within just a few days.

Testing how existing and experimental therapies alter that invasion process yielded critical clinical clues regarding tumor recurrence. While two experimental drug compounds successfully slowed cellular migration in laboratory brain slices, their effectiveness varied sharply depending on the patient tumor and the specific migration mode of the cells. More strikingly, standard combinations of radiation and chemotherapy actually increased the proportion of a distinct, fast-moving cell shape that occupies a central hub in cellular state transitions, suggesting that current frontline treatments may inadvertently select for highly motile cells capable of driving relapse.

To accelerate global brain cancer drug discovery, the Swedish research consortium has made the GlioTrace computational analysis software freely available to laboratories worldwide, noting that it operates using standard laboratory microscopes rather than specialized high-cost imaging rigs. While co-lead author Hitesh Mangukiya emphasized that further preclinical validation is required before the tool can directly guide patient care in hospitals, the open-source platform offers pharmaceutical developers and neuro-oncologists a vital new way to screen drugs designed to lock down tumor invasion before cancer cells escape surgical margins.