A predictive model for MGMT promoter methylation status in glioblastoma based on terahertz spectral data.
Journal:
Analytical biochemistry
Published Date:
Mar 29, 2025
Abstract
O-methylguanine-DNA methyltransferase (MGMT) promoter methylation is a crucial biomarker in glioblastoma (GBM) that influences response to temozolomide. Traditional detection methods, such as gene sequencing, are time-consuming and limited to postoperative analysis. This study explores the use of terahertz time-domain spectroscopy (THz-TDS) combined with machine learning to predict MGMT methylation status intraoperatively. By analyzing 180 GBM tissue samples, a Random Forest model was developed, achieving an AUC of 0.862. The findings suggest that THz spectroscopy offers a rapid, intraoperative alternative to traditional MGMT methylation detection methods, potentially enhancing surgical decision-making and personalized treatment strategies in GBM.