NI-83 * DIFFUSION AND CONVENTIONAL MR IMAGING GENOMIC BIOMARKER SIGNATURE PREDICTS IDH-1 MUTATION IN GLIOBLASTOMA PATIENTS

Abstract

PURPOSE: The IDH-1(R132H) mutation is one of the strongest prognostic predictors and a diagnostic hallmark of gliomas that is independent of other known prognostic factors, including age, grade, and O6-methylguanine-DNA methyltransferase (MGMT) methylation status and thus has a major clinical relevance. Given the latter, it will likely be integrated into the new WHO classification for GBM. Currently used methods to determine the IDH-1 mutation status of GBM include immunohistochemical analysis and genotyping of the DNA extracted from the tumor specimens. Thus, we seek to identify a diffusion and conventional MR imaging signature associated with IDH-1 mutated tumors that can be considered as a non-invasive predictor of the IDH1 status in Glioblastoma patients. MATERIALS AND METHODS: We identified 80 GBM patients from TCGA who had both genetic expression profiles of IDH-1 and mutation status and neuroimaging available at The Cancer Imaging Archive (TCIA). All morphological image analyses and segmentation were done using slicer 3.6 and reviewed in consensus by 3 neuroradiologists. Fluid-Attenuated Inversion Recovery (FLAIR) was used for segmentation of the edema and post-contrast T1 weighted imaging (T1WI) for segmentation of enhancement and necrosis. Diffusion was analyzed in Olea Sphere 2.3 and Conventional FLAIR/post-contrast T1WI was registered to DWI/ADC maps. ADC,FLAIR,T1 Gadolinium enhancement values were measured using the ROI based method,in the peri-tumoral edema/tumor infiltration and the enhancing tumor zones, with dividing the peri-tumoral edema/tumor infiltration into 3 zones each of 1 cm width, 3 ROI measurements were taken from each zone. Multiple quantitative imaging features were combined to create the imaging biomarker signature predictive of IDH-1 mutation status. RESULTS: We created a complex imaging biomarker signature using quantitative diffusion and conventional MR imaging features to predict those GBM patients with IDH-1 mutation and was predictive of patient survival. CONCLUSIONS: GBM tumors with IDH-1 mutation hold a specific diffusion and conventional MR imaging biomarker signature that can be used as a predictive and prognostic biomarker and non-invasive surrogate for IDH-1 status detection.