Value of MRI in prediction of glioma isocitrate dehydrogenase 1 mutation status / 中国医学影像技术

ABSTRACT

Objective: To observe the value of MR in prediction of glioma isocitrate dehydrogenase (IDH) 1 mutation status. Methods: Nineteen-two patients with glioma were divided into IDH mutation positive group and negative group, and their imaging characteristics were retrospectively reviewed, including lesions' site, signal intensity, boundary, growth pattern, degree of enhancement and surrounding edema. Then two-class Logistic model was established. Results: There were significant differences between different grades and location of gliomas between the two groups (both P<0.05). There were no significant differences in tumor signal intensity, boundary and growth pattern (P=0.269, 0.606, 0.139). There were statistically significant difference in degree of enhancement and surrounding edema (all P<0.01). Logistic regression analysis showed that the signal uniformity (X1), boundary (X2) and degree of enhancement (X3) of gliomas were statistically significant (P=0.004, 0.037, 0.001), and the regression equation was: logit (P)=2.668+1.415X1-2.097X2-3.229X3 (χ2=41.583, P<0.001), the sensitivity of the model was 70.70%, and the specificity was 80.40%. Conclusion: MRI can be used to non-invasively predict IDH1 mutation status of gliomas before surgical operation.