Preoperative Diffusion Tensor Imaging and Neurite Dispersion and Density Imaging in Isocitrate Dehydrogenase-Mutant Grade 2 and 3 Gliomas: Definition of Tumor-Related Epilepsy and Predictive Factors of Seizure Outcomes Based on a Single-Center Retrospective Case Series

BACKGROUND AND OBJECTIVES: Understanding and managing seizure activity is crucial in neuro-oncology, especially for highly epileptogenic lesions like isocitrate dehydrogenase (IDH)-mutant gliomas. Advanced MRI techniques such as diffusion tensor imaging (DTI) and neurite orientation dispersion and density imaging (NODDI) have been used to describe microstructural changes associated with epilepsy. However, their role in tumor-related epilepsy (TRE) remains unclear. This study aims to investigate the role of DTI and NODDI tumor-derived metrics in defining TRE and predicting postoperative seizure outcomes in patients undergoing surgical resection for IDH-mutant grade 2 and 3 gliomas. METHODS: This was a single-center retrospective study. Preoperative DTI parameters included fractional anisotropy and mean diffusivity. NODDI parameters included neurite density index (NDI), orientation dispersion index, and free-water fraction (FWF). These metrics were calculated within three volumes of interest (fluid-attenuated inversion recovery [FLAIR] tumor volume, FLAIR peripheral zone, and FLAIR central zone [Fcz]) and correlated with seizure presentation, type, and postoperative control, which was reported according to the Engel classification system. RESULTS: Fifty-seven patients were included in this study. Increased NODDI-derived FWF-Fcz ( P = .031) and NDI-Fcz ( P = .046) values correlated with preoperative generalized seizure presentation, although only the FWF-Fcz confirmed its statistical significance ( P = .047) in the multivariate analysis. Lower mean diffusivity-FLAIR tumor volume correlated with poor postoperative seizure control both in the univariate ( P = .015, P = .026) and multivariate analyses ( P = .024, P = .036), while a trend toward significance was found between higher NDI-FLAIR peripheral zone and worse seizure control ( P = .055). CONCLUSION: DTI and NODDI tumor-derived quantitative parameters may define TRE and predict postoperative seizure outcomes in patients with IDH-mutant gliomas. Notably, DTI metrics were found to be independent predictors of postoperative seizure outcomes, while preoperative NODDI parameters correlated with seizure presentation. Further research is warranted to validate our findings and to better understand the underlying mechanisms driving TRE.