NIMG-88. RADIONOMIC ANALYSIS OF WHO GRADE 2 AND 3 GLIOMAS WITH GENETIC SUBGROUP PREDICTION

AbstractPURPOSEGenetic alterations found in WHO grade 2 and 3 gliomas include IDH1/2 and TERT promoter mutations and 1p19q co-deletion, which alterations are known to have great impact on the prognosis of the patient. In this research, the authors attempted to test the hypothesis that genetic alterations could contribute to the locations and heterogeneity of the tumor by analyzing 191 WHO grade 2 and 3 gliomas via MR radionomics.METHODS191 WHO grade 2 and 3 gliomas were retrospectively collected and the genomic DNA of the tumor was sequenced for IDH1/2 and TERT promoter mutations. Treatment naive MR images were also collected. T2 weighted, T1 weighted, FLAIR and Gd-enhanced T1 weighted images were collected for analysis. Voxel-based lesion mapping (VLM), and redionomics analysis was performed for all images and were further challenged to construct a model that predicts genetic alterations within the tumor. 126 patients were allocated as training set and 65 for validation set. Patient cohort was divided into the following 3 groups; IDH mt/TERTp wt, IDH mt/TERTp mt and IDH wt. A multi-regression model was constructed using the training set to predict the 3 genetic subgroups and the validation set was used for model validation.RESULTSVLM revealed that IDH mt/TERTp wt gliomas dominated temporal lobe involvement while IDH mt/TERTp mt occupied the frontal lobe. IDH wt tumors, on the other hand, located at much posterior lobes and centered at the deep white matter. 15 radionomic features were identified for model construction to predict 3 genetic subgroups of WHO grade 2 and 3 gliomas. When these 15 texture elements were used to construct a prediction model of the 3 genetic subgroups, AUCs calculated from the training set ranged from 0.7 to 0.75. Accuracy for prediction was 63% for the training set and 60% for the validation set.