ADC-Based Stratification of Molecular Glioma Subtypes Using High b-Value Diffusion-Weighted Imaging

Nils C Nuessle,Felix Behling,Uwe Klose,Salvador Castaneda Vega,Ulrike Ernemann,Johann-Martin Hempel,Jens Schittenhelm,Ghazaleh Tabatabai

doi:10.3390/jcm10163451

Abstract

Purpose: To investigate the diagnostic performance of in vivo ADC-based stratification of integrated molecular glioma grades. Materials and methods: Ninety-seven patients with histopathologically confirmed glioma were evaluated retrospectively. All patients underwent pre-interventional MRI-examination including diffusion-weighted imaging (DWI) with implemented b-values of 500, 1000, 1500, 2000, and 2500 s/mm2. Apparent Diffusion Coefficient (ADC), Mean Kurtosis (MK), and Mean Diffusivity (MD) maps were generated. The average values were compared among the molecular glioma subgroups of IDH-mutant and IDH-wildtype astrocytoma, and 1p/19q-codeleted oligodendroglioma. One-way ANOVA with post-hoc Games-Howell correction compared average ADC, MD, and MK values between molecular glioma groups. A Receiver Operating Characteristic (ROC) analysis determined the area under the curve (AUC). Results: Two b-value-dependent ADC-based evaluations presented statistically significant differences between the three molecular glioma sub-groups (p < 0.001, respectively). Conclusions: High-b-value ADC from preoperative DWI may be used to stratify integrated molecular glioma subgroups and save time compared to diffusion kurtosis imaging. Higher b-values of up to 2500 s/mm2 may present an important step towards increasing diagnostic accuracy compared to standard DWI protocol.

Highlights

Gliomas are one of the most common primary central nervous system tumors and are, in most cases, associated with poor overall survival [1]
Average Apparent Diffusion Coefficient (ADC) values were significantly higher in IDH-mut gliomas than in oligodendrogliomas and IDH-wt gliomas
diffusionweighted imaging (DWI) may be used to stratify molecular glioma subgroups and save time compared to diffusion kurtosis imaging (DKI)

Summary

Introduction

Gliomas are one of the most common primary central nervous system tumors and are, in most cases, associated with poor overall survival [1]. Treatment options include surgical resection, chemotherapy, and radiation therapy, depending on the histopathological entity [1,2]. The distinction between different glioma subtypes with sufficient sensitivity and specificity remains challenging in preoperative settings and imaging. A reliable pre-interventional glioma stratification based on the expected molecular glioma profile may impact therapeutic options, the extent of planned surgical resection, and prognosis [3,4,5]. Various models have been proposed in previous reports to distinguish non-invasively different tumor entities using MRI. ADC-map-based tumor evaluation from diffusionweighted imaging (DWI) seems to be a promising means of differentiation [6,7]

Objectives

Methods

Results

Discussion

Conclusion