Detailed MRI patterns predicted outcomes better than tumor size
Original title: Radiomics Outperforms Conventional Tumor Volume in Prognostic Stratification Across Independent Glioma Cohorts.
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.
This was a look back at scans already collected, and the scoring tool is not something a clinic can use yet.
The short version
Researchers looked back at old MRI scans and found that a detailed scan pattern score predicted survival better than simply measuring tumor size.
What was studied. Researchers reviewed MRI scans and records from 1,447 people with glioma: A tumor that starts in the glial cells, the support cells of the brain and spinal cord. Gliomas are graded 1 to 4 by how fast they tend to grow. See the glossary, using four public collections of scans. They compared two ways of reading the scans: measuring the tumor's volume, and a score built from many fine details in the image.
What they found. Tumor volume did not predict outcomes reliably from one group to the next. In one group of 194 patients, 152 had their tumor grow, and larger active tumor volume was linked to faster growth. But in the other groups, tumor volume was not clearly linked to survival. The detailed image score was linked to worse survival, and it held up even when tumor volume was included in the same comparison.
What this means, and what it doesn't
What it could mean: This is about reading scans, not about a new treatment. If it holds up, doctors may one day get more information from the same MRI you already have, which could help them judge how a tumor is likely to behave. It would not change the scan itself or add a new test.
What it doesn't mean: This does not mean a new treatment, and it is not a cure. It does not mean your own scan can be scored this way today. The work was done by looking back at scans that were already collected, not by testing anything in patients going forward. Tools like this need to be tested in real clinics before doctors can use them to guide care. A score that predicts what may happen does not change what happens.
Source: PubMed, July 8, 2026 · Read the original
This plain-language summary was written by AI and published automatically after passing our automatic safety checks. How we write.