Diagnostic Performance of Microstructural Parameters Derived From Time-dependent Diffusion MRI for Grading...

Area of Science:

• Neuroimaging
• Oncology
• Radiology

Background:

• Preoperative meningioma grading is essential for treatment planning and predicting patient outcomes.
• Accurate grading helps differentiate between low-grade meningiomas (LGMs) and high-grade meningiomas (HGMs).

Purpose of the Study:

• To prospectively evaluate microstructural parameters from time-dependent diffusion MRI (t_d-dMRI) for preoperative meningioma grading.
• To assess the correlation between t_d-dMRI parameters and meningioma proliferative activity.

Main Methods:

• 214 meningioma patients underwent t_d-dMRI.
• Intracellular volume fraction (f_in), cell diameter, cellularity, and extracellular diffusivity (D_ex) were derived.
• A derivation cohort (N=159) and validation cohort (N=55) were used; diagnostic performance was assessed using ROC analysis and logistic regression.

Main Results:

• HGMs exhibited higher f_in and cellularity, and lower diameter and D_ex compared to LGMs.
• f_in demonstrated the highest AUC (0.936) for grading in the derivation cohort; a combined model achieved an AUC of 0.951.
• t_d-dMRI parameters strongly correlated with Ki-67 index and histology-based measurements, with consistent performance in the validation cohort.

Conclusions:

• t_d-dMRI-derived microstructural parameters are promising non-invasive biomarkers for preoperative meningioma grading.
• These parameters offer consistent diagnostic performance for differentiating LGMs and HGMs.
• t_d-dMRI facilitates preoperative assessment of meningioma grade and proliferative activity.