Radiomics prediction of MGMT promoter methylation in adult diffuse gliomas: a combination of structural MRI, DCE, and DTI

Yuying Liu ^1,2, Zhengyang Zhu ¹, Jianan Zhou ¹

⁰Department of Radiology, Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, China.

Frontiers in Neurology +

|

February 13, 2025

View abstract on PubMed

Summary

Radiomics analysis of MRI scans shows promise in predicting O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation in diffuse gliomas. Combining structural MRI with dynamic contrast-enhanced (DCE) and diffusion tensor imaging (DTI) improves prediction accuracy.

Area Of Science

Radiology
Oncology
Medical Imaging Analysis

Background

O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation is a crucial biomarker for predicting response to temozolomide in diffuse gliomas.
Accurate prediction of MGMT methylation status is essential for guiding treatment decisions.
Non-invasive methods for predicting MGMT methylation are highly desirable.

Purpose Of The Study

To evaluate the predictive capability of radiomics features from structural MRI, dynamic contrast-enhanced (DCE), and diffusion tensor imaging (DTI) for detecting MGMT promoter methylation.
To assess the added value of DCE and DTI features to structural MRI in predicting MGMT methylation status.
To compare the performance of radiomics models against human expert evaluation.

Main Methods

Retrospective analysis of MRI data from 110 diffuse glioma patients.
Extraction of 2,782 radiomic features from structural MRI, DCE, and DTI.
Feature selection using Pearson correlation and LASSO, followed by dimensionality reduction with PCA.
Support vector machine (SVM) model construction for prediction.
Comparison of model performance against two experienced radiologists.

Main Results

The combined structural MRI + DCE model achieved an AUC of 0.906 on the training dataset.
On the test dataset, the structural MRI + DCE + DTI model demonstrated the highest AUC of 0.868.
The best radiomics model outperformed the diagnostic performance of two experienced radiologists.

Conclusions

Radiomics models show significant potential for non-invasively predicting MGMT promoter methylation status in diffuse gliomas.
Integration of DCE and DTI radiomics features enhances the predictive accuracy of structural MRI.
These findings suggest that radiomics can serve as a valuable tool in personalized glioma treatment strategies.

Keywords:

DCE DTI MGMT glioma radiomics

Related Concept Videos

Magnetic Resonance Imaging

Magnetic resonance imaging (MRI) is a noninvasive medical imaging technique based on a phenomenon of nuclear physics discovered in the 1930s, in which matter exposed to magnetic fields and radio waves was found to emit radio signals. In 1970, a physician and researcher named Raymond Damadian noticed that malignant (cancerous) tissue gave off different signals than normal body tissue. He applied for a patent for the first MRI scanning device in clinical use by the early 1980s. The early MRI...