Two Studies Map the Rules Glioblastoma Uses to Resist Therapy

The team combined single-cell measurements with spatial transcriptomics, which locates gene activity within pieces of tissue. It also checked the state arrangement in laboratory-grown tumor organoids, or tumor models built in culture.

Bayraktar said: “By integrating the single-cell data with spatial transcriptomic data, we found that there is a meticulous and very specific spatial architecture of cancer cell states. And this was present in every single glioblastoma tumor in every single patient we studied.”

Takeaway: The state map was consistent across the 12-patient sample and also appeared in laboratory models.

What did the AI tool uncover?

The researchers then examined gene regulation in more than 1 million patient-derived glioblastoma cells. They used scDORI, a deep-learning tool that learns patterns in gene regulation, to identify molecular regulators, or controls that influence state changes. Some regulators act as safeguards, keeping cells in stable states and limiting plasticity.

“A tumor generates an almost endless number of variations on its own regulatory program,” said Stegle. “Instead of seeing this heterogeneity only as an obstacle, we can use AI to learn from it and extract the rules that determine how cells choose their fate.”

Here, a regulatory program means a cell’s system of gene controls, while heterogeneity means variation among cells. The results suggest that manipulating a small number of regulators might push cancer cells toward less aggressive states. The method predicts regulators for all major tumor-cell states, and the researchers tested those predictions in experiments.

Takeaway: The software reduced a vast regulatory search to a handful of testable controls.

Could the rules lead to treatment?

“One of our premises has always been that when we understand these rules, we can not only observe what the cells are doing, but we can also change what they are doing,” said Moritz Mall of DKFZ. “We could perhaps reprogram them, and maybe push them into a state that is less malignant, less dangerous.”

One test increased the activity of MYT1L, one of the regulators identified in the study. The change reduced cellular plasticity and malignant characteristics in patient-derived glioblastoma cells. Mice transplanted with the modified tumor cells survived longer than mice receiving control cells.

“The results with MYT1L were interesting, but the findings go much beyond that,” explained Mall. “Our approach predicts the regulators for all of the major tumor cell states, and we tested those predictions experimentally. We were able to steer tumor cells into specific states by manipulating just a handful of regulators.”

Clinical treatments remain a future possibility; the reported experiments involved cells and mice. The researchers are already applying the framework to other cancers and brain disorders. The framework also identifies regulatory sequences, DNA segments that help determine when genes are active.

“With the help of this atlas and the new AI tool, we identified new regulatory sequences that enable us to potentially drive specific cell states in a tumor-specific setting,” said Mall. “Now that we have this framework, we can apply it to many other questions and cancers.”

Takeaway: The study identifies possible control points, not an established treatment for patients.

Papers

• “A spatiotemporal cancer cell trajectory underlies glioblastoma heterogeneity,” by Grant de Jong and colleagues, Nature, 4 November 2026. DOI: 10.1038/s41586-026-11234-8.
• “Decoding the regulatory programs governing glioblastoma plasticity,” by Manu Saraswat and colleagues, Nature, 4 November 2026. DOI: 10.1038/s41586-026-11177-0.