Radiomics and Radiogenomics in Differentiating Progression, Pseudoprogression, and Radiation Necrosis in Gliomas
Need to claim your poster? Find the KiKo table at the conference and they'll help
you get set up.
Presented at: ACRO Summit 2025
Date: 2025-03-12 00:00:00
Views: 1
Summary: This study investigates the use of radiomics and radiogenomics to distinguish true progression (TP), pseudoprogression (Psp), and radiation necrosis (RN) in gliomas. By combining radiomic imaging features with genomic markers, we hypothesize improved accuracy in differentiating glioma progression states. Our goal is to enhance non-invasive diagnostic capabilities in glioma management, potentially advancing precision neuro-oncology through more accurate treatment planning and monitoring. A review of recent studies assessed the effectiveness of radiomic and radiogenomic techniques for differentiating glioma progression states. Texture and shape features from MRI, combined with genomic data, were analyzed using machine learning (ML) models like support vector machines (SVM) and neural networks. Key metrics such as sensitivity, specificity, and AUC scores were compared to identify the most effective models, aiming to reduce invasive biopsies in glioma diagnostics and improve clinical outcomes. Radiomics models incorporating genomic data demonstrated high accuracy in differentiating TP from Psp and RN, with sensitivity and specificity rates above 80% and some models achieving AUC scores over 0.9. Integration of genomic markers such as MGMT methylation and 1p/19q codeletion status significantly improved diagnostic precision. SVM and neural network models outperformed other ML techniques, supporting radiomics and radiogenomics as a powerful diagnostic tool in glioma care. Radiomics and radiogenomics show promise for non-invasively differentiating TP, Psp, and RN in gliomas. The high diagnostic performance of integrated computational models may reduce the need for invasive biopsies. However, further multicenter studies are needed for clinical standardization. This research highlights the potential of radiogenomics to refine glioma management, optimizing treatment strategies and patient outcomes. Sohil N. Reddy (he/him/his), B.S. (Presenting Author) - The Ohio State University College of Medicine; Tyler Lung, B.S. (Co-Author) - The Ohio State University College of Medicine; Shashank Muniyappa, B.S. (Co-Author) - The Ohio State University College of Medicine; Christine Hadley, B.S. (Co-Author) - The Ohio State University; Benjamin Templeton`, B.S. (Co-Author) - The Ohio State University; Keshav Shah (he/him/his), Undergraduate Student (Co-Author) - The Ohio State University; Jennifer Matsui, B.S., PhD, MD (Co-Author) - Stanford Medicine; Joshua D.. Palmer, MD (Co-Author) - Department of Radiation Oncology, The James Cancer Hospital and Solove Research Institute at The Ohio State University Wexner Medical Center