Brain Scans May Help Show Who Does Poorly After Treatment
Original title: PET|MR Open Source Radiomics Models for Treatment Outcome Prediction in Recurrent GBM.
How far along is this research?
- Lab cells
- Animals
- Review
- Tested in people
This was tested in people. That is the most reliable kind of research we share.
Scans from real patients were studied, but the tool is not used in care yet.
The short version
Doctors tested whether two kinds of brain scans can point out who is likely to do worse after treatment.
What was studied. Researchers followed 185 people whose glioblastoma: A glioma that is given grade 4, the highest grade. It grows fast. Treatment usually starts soon after it is found. It often means surgery, then radiation and chemotherapy. See the glossary had come back, from 15 hospitals. They used MRI scans plus a PET scan, then built computer models to predict how long people went before the tumor grew again and how long they lived.
What they found. The models worked best when they used the small area where the MRI and PET scans overlapped. That area gave 15 useful models. The bigger scan areas gave 6 useful models and 0 useful models. Of the 21 useful models, 19 needed the PET scan. Two models put together correctly flagged 82% of the people who did poorly, and correctly cleared 75% of those who did not.
What this means, and what it doesn't
What it could mean: The PET scan may add something the MRI cannot show on its own. If this holds up, doctors might learn early which people are likely to do poorly, and talk about a different plan sooner.
What it doesn't mean: This is not a new treatment, and it is not a cure. It is early work. The team built a computer tool and tested it on scans that were already collected. No one has shown yet that using the tool helps people live longer or feel better. It is not part of everyday care, and your doctor cannot order it today.
Source: PubMed, September 16, 2026 · Read the original
This plain-language summary was written by AI and published automatically after passing our automatic safety checks. How we write.