New Method Predicts Glioblastoma Invasion and Progression

Glioblastoma remains one of the most lethal brain cancers, with patients typically surviving only about 15 months after diagnosis. Current treatments, including surgery and radiation, buy limited time but cannot fully eliminate hidden tumor cells that linger beyond the visible cancer margin. These remaining cells often drive the tumor’s return. A research team led by Jennifer Munson at the Fralin Biomedical Research Institute at Virgina Tech has developed a method to help identify where those hidden cells are located and predict how the cancer is likely to spread.

The approach, detailed in npj Biomedical Innovations, integrates magnetic resonance imaging with knowledge of how fluid moves through tissues, and an algorithm designed in Munson’s lab. The method assesses interstitial fluid flow, which passes through spaces between cells in the brain. In glioblastoma, Munson’s group discovered that faster directional flows signal regions where tumor cells invade surrounding tissue, while more random patterns of diffusion are associated with less cancer cell migration.

A new metric created by the team proved especially accurate at predicting invasion. The researchers found that the fluid flow establishes distinct pathways, like streams merging into rivers, which tumor cells follow to infiltrate new regions. “This could tell a surgeon where there's going to be a higher chance of there being more tumor cells, so they might be a little more aggressive, if it's safe to the patient to go after a more invasive region,” Munson explained.

Current surgical planning relies on scans that outline tumor boundaries and on fluorescent dyes used during operations to highlight visible cancer cells. These methods fail to reveal cells that have already migrated further into healthy tissue. Munson emphasized this limitation: “If you can't find the tumor cells, you can't kill the tumor cells, whether that's by cutting them out, hitting them with radiation therapy, or getting drugs to them. This is a method that now we believe can allow us to find those tumor cells.”

The work is also informing the efforts of a new spinoff company , which is adapting the findings into clinical tools. The company aims to provide “probability maps or hotspot maps” predicting where cancer cells are likely to be. According to Munson, this could allow surgeons and radiation oncologists to apply treatment more aggressively in areas with higher invasion risk while sparing healthy tissue when invasion is less likely.

By improving the ability to detect and predict glioblastoma spread, Munson’s research offers a pathway toward more precise strategies for managing this aggressive cancer.