Synthesizing Contrast-Enhanced MR Images from Noncontrast MR Images Using Deep Learning.

BACKGROUND AND PURPOSE: Recent developments in deep learning methods offer a potential solution to the need for alternative imaging methods due to concerns about the toxicity of gadolinium-based contrast agents. The purpose of the study was to synthesize virtual gadolinium contrast-enhanced T1-weighted MR images from noncontrast multiparametric MR images in patients with primary brain tumors by using deep learning. MATERIALS AND METHODS: We trained and validated a deep learning network by using MR images from 335 subjects in the Brain Tumor Segmentation Challenge 2019 training data set. A held out set of 125 subjects from the Brain Tumor Segmentation Challenge 2019 validation data set was used to test the generalization of the model. A residual inception DenseNet network, called T1c-ET, was developed and trained to simultaneously synthesize virtual contrast-enhanced T1-weighted (vT1c) images and segment the enhancing portions of the tumor. Three expert neuroradiologists independently scored the synthesized vT1c images by using a 3-point Likert scale, evaluating image quality and contrast enhancement against ground truth T1c images (1 = poor, 2 = good, 3 = excellent). RESULTS: The synthesized vT1c images achieved structural similarity index, peak signal-to-noise ratio, and normalized mean square error scores of 0.91, 64.35, and 0.03, respectively. There was moderate interobserver agreement between the 3 raters, regarding the algorithm's performance in predicting contrast enhancement, with a Fleiss kappa value of 0.61. Our model was able to accurately predict contrast enhancement in 88.8% of the cases (scores of 2 to 3 on the 3-point scale). CONCLUSIONS: We developed a novel deep learning architecture to synthesize virtual postcontrast enhancement by using only conventional noncontrast brain MR images. Our results demonstrate the potential of deep learning methods to reduce the need for gadolinium contrast in the evaluation of primary brain tumors.

Alterations in the RB Pathway With Inactivation of RB1 Characterize Glioblastomas With a Primitive Neuronal Component.

A primitive neuronal component is a feature of some glioblastomas but defining molecular alterations of this histologic variant remains uncertain. We performed next-generation sequencing of 1500 tumor related genes on tissue from 9 patients with glioblastoma with a primitive component (G/PN) and analyzed 27 similar cases from the Cancer Genome Atlas (TCGA) dataset. Alterations in the RB pathway were identified in all of our patients' tumors and 81% of TCGA tumors with the retinoblastoma tumor suppressor gene (RB1) commonly affected. Although RB1 mutations were observed in some conventional glioblastomas, the allelic fractions of these mutations were significantly higher in tumors with a primitive neuronal component in both our and TCGA cohorts (median, 72% vs 25%, p < 0.001 and 80% vs 40%, p < 0.02, respectively). Further, in 78% of patients in our cohort, RB expression was lost by immunohistochemistry. Our findings indicate that alterations in the RB pathway are common in G/PNs and suggest that inactivation of RB1 may be a driving mechanism for the phenotype.