Quantitative analysis of permeability for glioma grading using dynamic contrast-enhanced magnetic resonance imaging

The objective of the present study was to quantitatively analyze the permeability of tumor entity and peritumor edema in glioma grading, using dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). In the present retrospective study, 80 patients underwent T1-weighted DCE-MRI examination at 3.0 T and the pathological results (including astrocytoma and oligodendroglioma) were obtained between January 2012 and June 2015. All cases were surgically validated as grade I-IV gliomas. The original DCE-MRI data were analyzed using dual compartment modified Tofts model. The forward volume transfer constant (K^trans), backflux rate (k_ep) and fractional volume (v_e) were calculated with the region of interest selected on the highest permeability area of the tumor entity and peritumor edema. Analysis of variance with the Bonferroni correction was used to compare the values of K^trans, k_ep, and v_e of the tumor entity and peritumor edema in different glioma grades. The results of the present study revealed that the K^trans, k_ep, and v_e values in each stage were associated with the pathological grading (r=0.951, 0.804 and 0.766, respectively). There were significant differences identified between different tumor grades in K^trans, k_ep, with the exception being between grades II and III in k_ep. In addition, there was a significant difference revealed between grade I/II and grade III/IV in v_e. Receiver operator characteristics curve analysis was used to evaluate the diagnosis accuracies of permeability parameters. K^trans was demonstrated to exhibit the highest sensitivity and specificity for evaluating the tumor grade. With the threshold values of 0.160, 0.420 and 0.935 in K^trans on tumor, glioma grades I vs. II, II vs III and III vs. IV may be differentiated with sensitivities of 0.900, 0.950 and 0.950, and specificities of 0.950, 0.950 and 0.850, respectively. Furthermore, associations were observed between the K^trans, k_ep and v_e values of peritumor edema and the pathological grading in glioma (K^trans r=0.438, P<0.001; K_ep r=0.385, P<0.001; V_e r=0.397, P<0.001, respectively). K^trans values in peritumoral edema revealed significant differences between low-grade and high-grade glioma. The sensitivity and specificity for K^trans of peritumor edema were 0.975 and 0.950, with a threshold value of 0.007. Therefore, the DCE-MRI parameters of K^trans of tumor entity and peritumor edema in gliomas may be used to accurately differentiate glioma grades.