Survival Outcome Prediction in Glioblastoma: Insights from MRI Radiomics

Area of Science:

• Neuro-oncology
• Medical Imaging
• Radiomics

Background:

• Predicting survival in isocitrate dehydrogenase (IDH) wild-type glioblastoma (GBM) is crucial.
• Multiparametric MRI and radiomics offer a promising avenue for non-invasive prognostic assessment.

Purpose of the Study:

• To investigate the utility of multiregional radiomic features from multiparametric MRI in predicting pretreatment survival for IDH-wildtype GBM patients.
• To identify specific radiomic signatures associated with survival outcomes.

Main Methods:

• Radiomic features were extracted from multiparametric MRI (T1W, diffusion, perfusion) in 49 IDH-wildtype GBM patients.
• Feature selection identified 25 significant features, and a Cox proportional-hazards model with LASSO regularization was used for survival analysis.
• Internal and external validation were performed on training (38 patients) and validation (11 patients) cohorts.

Main Results:

• Age and six radiomic features (shape, first/second order) from T1W, diffusion, and perfusion maps were significant predictors of survival.
• Poor survival was linked to small necrotic subregions, inhomogeneous vascularization, and edema-related tissue damage.
• The predictive model achieved a C-Index of 0.66 and demonstrated good external validation accuracy (AUC > 0.65).

Conclusions:

• Radiomics analysis of segmented diffusion and perfusion MRI maps can predict survival in IDH-wildtype GBM.
• The findings highlight the association between microstructural and vascular heterogeneity within the tumor and patient survival.
• Radiomics provides valuable prognostic indicators for GBM patients.