Glioblastoma (GBM) is a devastating and incurable brain tumour, with a median overall survival of fifteen months. Identifying the cell of origin that harbours mutations that drive GBM could provide a fundamental basis for understanding disease progression and developing new treatments. Given that the accumulation of somatic mutations has been implicated in gliomagenesis, studies have suggested that neural stem cells (NSCs), with their self-renewal and proliferative capacities, in the subventricular zone (SVZ) of the adult human brain may be the cells from which GBM originates. However, there is a lack of direct genetic evidence from human patients with GBM. Here we describe direct molecular genetic evidence from patient brain tissue and genome-edited mouse models that show astrocyte-like NSCs in the SVZ to be the cell of origin that contains the driver mutations of human GBM. First, we performed deep sequencing of triple-matched tissues, consisting of (i) normal SVZ tissue away from the tumour mass, (ii) tumour tissue, and (iii) normal cortical tissue (or blood), from 28 patients with isocitrate dehydrogenase (IDH) wild-type GBM or other types of brain tumour. We found that normal SVZ tissue away from the tumour in 56.3% of patients with wild-type IDH GBM contained low-level GBM driver mutations (down to approximately 1% of the mutational burden) that were observed at high levels in their matching tumours. Moreover, by single-cell sequencing and laser microdissection analysis of patient brain tissue and genome editing of a mouse model, we found that astrocyte-like NSCs that carry driver mutations migrate from the SVZ and lead to the development of high-grade malignant gliomas in distant brain regions. Together, our results show that NSCs in human SVZ tissue are the cells of origin that contain the driver mutations of GBM.

To investigate whether radiomic features at MRI improve survival prediction in patients with glioblastoma multiforme (GBM) when they are integrated with clinical and genetic profiles. Materials and Methods: Data in patients with a diagnosis of GBM between December 2009 and January 2017 (217 patients) were retrospectively reviewed up to May 2017 and allocated to training and test sets (3:1 ratio). Radiomic features (n = 796) were extracted from multiparametric MRI. A random survival forest (RSF) model was trained with the radiomic features along with clinical and genetic profiles (O-6-methylguanine-DNA-methyltransferase promoter methylation and isocitrate dehydrogenase 1 mutation statuses) to predict overall survival (OS) and progression-free survival (PFS). The RSF models were validated on the test set. The incremental values of radiomic features were evaluated by using the integrated area under the receiver operating characteristic curve (iAUC). Results: The 217 patients had a mean age of 57.9 years, and there were 87 female patients (age range, 22-81 years) and 130 male patients (age range, 17-85 years). The median OS and PFS of patients were 352 days (range, 20-1809 days) and 264 days (range, 21-1809 days), respectively. The RSF radiomics models were successfully validated on the test set (iAUC, 0.652 [95% confidence interval {CI}, 0.524, 0.769] and 0.590 [95% CI: 0.502, 0.689] for OS and PFS, respectively). The addition of a radiomics model to clinical and genetic profiles improved survival prediction when compared with models containing clinical and genetic profiles alone (P = .04 and .03 for OS and PFS, respectively). Conclusion: Radiomic MRI phenotyping can improve survival prediction when integrated with clinical and genetic profiles and thus has potential as a practical imaging biomarker.

Background Glioma prognosis depends on the isocitrate dehydrogenase (IDH) mutation status. We aimed to predict the IDH status of gliomas from preoperative MR images using a fully automated hybrid approach with convolutional neural networks (CNNs) and radiomics. Methods We reviewed 1,166 preoperative MR images of gliomas (grades II-IV) from Severance Hospital (n=856, Severance Set), Seoul National University Hospital (n=107, SNUH set), and The Cancer Imaging Archive (n=203, TCIA set). The Severance set was subdivided into the development (n=727) and internal test (n=129) sets. Based on T1 postcontrast, T2, and fluid-attenuated inversion-recovery images, a fully automated model was developed that comprised a CNN for tumor segmentation (Model 1) and CNN-based classifier for IDH status prediction (Model 2) that uses a hybrid approach based on 2-dimensional tumor images and radiomic features from 3-dimensional tumor shape and loci guided by Model 1. The trained model was tested on internal (a subset of the Severance set) and external (SNUH and TCIA) test sets. Results The CNN for tumor segmentation (Model 1) achieved a dice coefficient of 0.86–0.92 across datasets. Our hybrid model achieved accuracies of 93.8%, 87.9%, and 78.8%; with areas under the receiver operating characteristic curves of 0.96, 0.94, and 0.86; and areas under the precision-recall curves of 0.88, 0.82, and 0.81 in the internal test, SNUH, and TCIA sets, respectively. Conclusions Our fully automated hybrid model demonstrated the potential to be a highly reproducible and generalizable tool across different datasets for the noninvasive prediction of the IDH status of gliomas.