Diffusion Kurtosis Imaging Combined With Molecular Markers As A Comprehensive Approach To Predict Overall Survival In Patients With Gliomas

Purpose:The purpose of this study was to explore the usefulness of diffusion kurtosis imaging(DKI)and molecular markers in predicting the prognosis of glioma patients.Method:Fifty-one patients with gliomas were examined by conventional MRI and DKI at 3.0T before operation,and the parameter maps of mean kurtosis(MK),mean diffusivity(MD),axial kurtosis(AK),radial kurtosis(RK)and fractional anisotropy(FA)were obtained by post-processing software.The mean kurtosis(MK),mean diffusivity(MD),axial kurtosis(AK),and radial kurtosis(RK)values of tumors were measured and normalized to the contralateral normal-appearing white matter.The molecular markers of gliomas,including isocitrate dehydrogenase-1(IDH1),a thalassemia/mental retardation syndrome x-linked(ATRX)and 06-methylguanine-DNA methyltransferase(MGMT),were immunohistochemically stained on the resected tumor tissues.The independent sample t-test was used to compare the differences of N-MK,N-MD,N-AK,N-RK and N-FA values between low grade glioma(LGGs)and high grade glioma(HGGs),and to analyze the differences of DKI parameters in different expression status of molecular markers.The chi-square test was used to compare the differences of IDH1,ATRX and MGMT expression status between low grade glioma(LGGs)and high grade glioma(HGGs).Receiver operating characteristic(ROC)was used to evaluate the value of N-MK,N-MD,N-AK,N-RK and N-FA values in the evaluation of glioma grade and prognosis.The Kaplan-Meier survival curves of the DKI parameters and molecular markers were used to evaluate the prognosis of patients.Univariate and multivariate Cox proportional hazards models were used to analyze the impacts of the all investigated factors on prognosis.Results:There were significant differences in the overall survival among glioma patients with different MK,AK,RK,and MD values(P<0.001).Survival was significantly better in patients with IDH1 mutation(P<0.01),ATRX loss(P P<0.05),and MGMT negative expression(P<0.05)in gliomas.Especially,patients with both IDH1 mutation and ATRX loss in gliomas had the best prognosis.However,in the molecular subtypes of gliomas with poor prognosis,the overall survival time of patients with low MK values was significantly improved(P<0.05).Cox multivariate regression analysis demonstrated that MK,RK values and ATRX expression could be used as independent prognostic risk factors,and high MK values had the highest risk for prognosis(HR=65.288).Conclusions:DKI parameter values and expression status of molecular markers showed significant differences between high and low grade gliomas,and MK values had the greatest diagnostic efficacy in the evaluation of glioma grading;Molecular markers and DKI parameters,especially MK values,can be used to effectively evaluate the prognosis of glioma patients.