Home›Oncology› ERS-related DNA methylation profiles stratify glioblastoma patients into four subtypes with 92.4% prediction accuracy
ERS-related DNA methylation profiles stratify glioblastoma patients into four subtypes with 92.4% prediction accuracyNew AI Tool Could Help Doctors Choose Better Brain Tumor Treatments
Frontiers in MedicinePublished April 22, 2026DOI ↗Editorial oversight: Dr. Julia Lee, PhD · Oncology, Genomics & Drug Development
AI-generated summary of the cited source, checked by automated accuracy review.
How we work
Share
Key Takeaway
Note that MEK inhibitors are preliminary candidates for specific GBM subtypes identified via epigenetic profiling.
This cohort analysis examined molecular and clinical characteristics across glioblastoma subtypes using a TCGA cohort. The study integrated ERS-related DNA methylation profiles with non-negative matrix factorization (NMF) and a random forest (RF) model to refine molecular stratification. No specific sample size was reported for this analysis.
The model stratified GBM patients into four distinct subtypes with 92.4% accuracy in subtype prediction. Subtype 2 was characterized by an immune-inflamed phenotype, lower tumor purity, and poorer prognosis. The study identified MEK inhibitors as preliminary candidate compounds for further exploration of subtype-related therapeutic strategies.
Safety data, including adverse events, discontinuations, and tolerability, were not reported as this was a retrospective analysis of existing data rather than a clinical trial. The study design does not support causal conclusions regarding the efficacy of MEK inhibitors.
Limitations include the lack of reported sample size and the observational nature of the data. These findings provide an epigenetic framework for refined molecular stratification and further exploration of subtype-related therapeutic strategies, but clinical application requires validation in prospective trials.
Glioblastoma is a very aggressive cancer. It grows fast and is hard to stop. Many patients get the same treatment. But the tumors are not all the same. Doctors need better ways to tell them apart.
The surprising shift
We used to look at cells under a microscope. Now we look at chemical signals inside the cells. This study focuses on a specific stress process inside the cell. This new approach looks at the DNA tags that control this stress.
Think of a cell like a busy factory. Sometimes the machines get clogged. This clog is called cellular stress. It changes how the cell behaves.
The AI looks at DNA tags that control this stress. These tags act like switches that turn genes on or off. When the stress is high, the cell might die. When it is low, the cell might survive.
Researchers used data from thousands of patients. They built a computer model to sort them. The model achieved high accuracy in testing. It successfully grouped the tumors into four distinct categories.
What scientists didn’t expect
The model found four unique groups of tumors. One group had a poor outlook. Another group might respond to specific drugs.
This doesn’t mean this treatment is available yet.
The study also pointed to drugs that might block the stress signals. These drugs could stop the tumor from growing.
Experts say this helps us understand the disease better. It opens doors for new drugs. We are moving toward a time when treatment fits the person. This is a step toward precision medicine for brain cancer.
You cannot get this test today. It is not ready for hospitals. But it shows where medicine is heading. Talk to your doctor if you have concerns about your care.
The study used existing data. It needs more testing in real patients. The results are promising but not final. We need to confirm these findings in larger groups.
Scientists will test these drugs in future trials. Approval takes time and safety checks. We are waiting for more evidence before changes happen. Research takes time, but every step brings us closer to better care.
IntroductionGlioblastoma (GBM) is a highly aggressive brain tumor with significant heterogeneity, leading to poor prognosis and limited treatment options. Developing innovative molecular subtyping approaches is important for gaining deeper insights into disease pathogenesis and optimizing treatment strategies. DNA methylation has been implicated in the regulation of endoplasmic reticulum stress (ERS), which disrupts protein folding and activates the unfolded protein response (UPR), ultimately determining cellular survival or apoptotic outcomes.MethodsERS-related DNA methylation profiles were integrated with non-negative matrix factorization (NMF) to establish a molecular classification framework for GBM. An ERS-based signature was further developed using recursive feature elimination with cross-validation (RFECV), and a random forest (RF) model was constructed for subtype prediction. The model was then applied to an external TCGA cohort for validation and downstream characterization.ResultsThe NMF-based framework stratified GBM patients into four distinct subtypes. The RF model achieved an accuracy of 92.4% in the independent test set. Application of the model to the TCGA cohort revealed distinct molecular and clinical characteristics across subtypes. In particular, Subtype 2 was associated with an immune-inflamed phenotype, lower tumor purity, and poorer prognosis. Connectivity Map (CMap) analysis further identified MEK inhibitors as preliminary candidate compounds for specific subtypes.DiscussionThese findings support an association between ERS-related epigenetic modifications and GBM heterogeneity, and provide an epigenetic framework for refined molecular stratification and further exploration of subtype-related therapeutic strategies.
