Assessment of IDH1 Genotype and Cell Proliferation in Gliomas Using Diffusion Magnetic Resonance Imaging

Abstract

Purpose To compare the efficacy of parameters from mono-, bi- and stretch-exponential diffusion weighted imaging (DWI), diffusion tensor imaging (DTI), diffusion kurtosis imaging (DKI) and neurite orientation dispersion and density imaging (NODDI) for prediction of IDH1 genotype and assessment of cell proliferation in gliomas. Methods 91 patients with pathologically confirmed gliomas underwent DWI, multi-b-value DWI and DKI/NODDI on 3.0T MRI. The ROIs were manually placed to obtain measurements including apparent diffusion coefficient (ADC), slow ADC (D), fast ADC (D*), perfusion fraction (f), distributed diffusion coefficient (DDC), heterogeneity index (α), fractional anisotropy (FA), mean diffusivity (MD), mean kurtosis (MK), orientation dispersion index (ODI) and intracellular volume fraction (ICVF). Results In LrGGs, IDH1 wild-type group showed significantly lower ADC, D, f, DDC, α, MD and higher D*, MK, ODI and ICVF values than IDH1-mutant group (P < 0.05). Among them, α has the highest AUC value (0.846). In GBMs, no difference between IDH1-mutant and IDH1 wild-type group was found. For IDH1-mutant group, all parameters, except for FA and D*, significantly discriminated LrGGs from GBMs (P < 0.05). However, for IDH1 wild-type group, only ADC and DDC statistically discriminated LrGGs from GBMs (P = 0.039 and 0.046, respectively). In addition, MK has maximal correlation coefficient (r = 0.612, P < 0.001) and D* has the minimal correlation coefficient (r = 0.146, P = 0.176) with Ki-67 LI. Conclusion The α may be the most effective diffusion MRI marker for predicting IDH1 genotype in LrGGs, and MK has shown great potential in assessing glioma cell proliferation.