NIMG-15. DIFFERENTIATION OF IDH1 MUTANT AND WILD TYPE GLIOMAS USING pH- AND OXYGEN-SENSITIVE MOLECULAR MRI

摘要

The presence of isocitrate dehydrogenase 1 (IDH1) gene mutation (IDH1mut) in glioma patients is associated with longer survival compared to IDH1 wild-type (IDH1wt) patients. Non-invasive detection of IDH1 mutation could provide valuable diagnostic and prognostic information. Studies showed that IDH1mut correlated with decreased hypoxia-inducible-factor 1-alpha (HIF1α) expression and decreased level of glutamate. With a newly developed MRI technique using amine-CEST with spin-and-gradient echo EPI readout (CEST-SAGE-EPI), we simultaneously measure the pH-dependent chemical exchange effects and oxygen-sensitive reversible transverse relaxation rate (R2’). In the current study we propose a novel imaging biomarker, which is sensitive to hypoxia, amine proton concentration, and tissue acidity, providing a potential non-invasive approach to detect IDH1 mutation and monitor related metabolic changes. In this study, 25 patients with glioma (8 WHO grade II: 2 IDH1wt/6 IDH1mut, 11 WHO grade III: 4 IDH1wt/7 IDH1mut, 6 WHO grade IV: 5 IDH1wt/1 IDH1mut) underwent whole brain CEST-SAGE-EPI scanning (7:30 minutes) and anatomic imaging on 3T MRI scanners. Stereotactic MRI-guided biopsies/tissue resections were performed in 10 of the 25 patients (29 samples), and immunohistochemistry (IHC) staining for HIF1a expression was analyzed. The results showed a significantly higher product of magnetization transfer asymmetry (MTRasym) at 3.0ppm and R2’ in IDH1wt gliomas compared to IDH1mut gliomas (p=0.0033, Student’s t-test). The significantly higher MTRasymxR2’ persisted even when excluding grade IV gliomas (p=0.0063, Student’s t-test). The ROC of differentiating IDH1 mutation status with MTRasymxR2’ gave area under the curve (AUC) of 0.857 (sensitivity 85.71%, specificity 72.73%). The IHC staining of HIF1α showed a moderate positive correlation (Pearson’s correlation R=0.439, p=0.0360) with MTRasymxR2’. The results demonstrated the potential of using imaging biomarker, MTRasymxR2’, to detect IDH1 mutation status in gliomas, and to provide non-invasive information of tumor microenvironment.