Imaging Characteristics of CNS Neuroblastoma-FOXR2: A Retrospective and Multi-Institutional Description of 25 Cases.

BACKGROUND AND PURPOSE: The 5th edition of the World Health Organization Classification of CNS tumors defines the CNS neuroblastoma FOXR2 in the group of embryonal tumors. Published clinical outcomes tend to suggest a favorable outcome after resection, craniospinal irradiation, and chemotherapy. This multicenter study aimed to describe imaging features of CNS neuroblastoma-FOXR2, which have been poorly characterized thus far. MATERIALS AND METHODS: On the basis of a previously published cohort of tumors molecularly classified as CNS neuroblastoma-FOXR2, patients with available imaging data were identified. The imaging features on preoperative MR imaging and CT data were recorded by 8 experienced pediatric neuroradiologists in consensus review meetings. RESULTS: Twenty-five patients were evaluated (13 girls; median age, 4.5 years). The tumors were often large (mean, 115 [ SD, 83] mL), showed no (24%) or limited (60%) perilesional edema, demonstrated heterogeneous enhancement, were often calcified and/or hemorrhagic (52%), were always T2WI-hyperintense to GM, and commonly had cystic and/or necrotic components (96%). The mean ADC values were low (687.8 [SD 136.3] × 10-6 mm2/s). The tumors were always supratentorial. Metastases were infrequent (20%) and, when present, were of nodular appearance and leptomeningeal. CONCLUSIONS: In our cohort, CNS neuroblastoma FOXR2 tumors showed imaging features suggesting high-grade malignancy and, at the same time, showed characteristics of less aggressive behavior. There are important differential diagnoses, but the results of this study may assist in considering this diagnosis preoperatively.

Radiomics Can Distinguish Pediatric Supratentorial Embryonal Tumors, High-Grade Gliomas, and Ependymomas.

BACKGROUND AND PURPOSE: Pediatric supratentorial tumors such as embryonal tumors, high-grade gliomas, and ependymomas are difficult to distinguish by histopathology and imaging because of overlapping features. We applied machine learning to uncover MR imaging-based radiomics phenotypes that can differentiate these tumor types. MATERIALS AND METHODS: Our retrospective cohort of 231 patients from 7 participating institutions had 50 embryonal tumors, 127 high-grade gliomas, and 54 ependymomas. For each tumor volume, we extracted 900 Image Biomarker Standardization Initiative-based PyRadiomics features from T2-weighted and gadolinium-enhanced T1-weighted images. A reduced feature set was obtained by sparse regression analysis and was used as input for 6 candidate classifier models. Training and test sets were randomly allocated from the total cohort in a 75:25 ratio. RESULTS: The final classifier model for embryonal tumor-versus-high-grade gliomas identified 23 features with an area under the curve of 0.98; the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 0.85, 0.91, 0.79, 0.94, and 0.89, respectively. The classifier for embryonal tumor-versus-ependymomas identified 4 features with an area under the curve of 0.82; the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 0.93, 0.69, 0.76, 0.90, and 0.81, respectively. The classifier for high-grade gliomas-versus-ependymomas identified 35 features with an area under the curve of 0.96; the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 0.82, 0.94, 0.82, 0.94, and 0.91, respectively. CONCLUSIONS: In this multi-institutional study, we identified distinct radiomic phenotypes that distinguish pediatric supratentorial tumors, high-grade gliomas, and ependymomas with high accuracy. Incorporation of this technique in diagnostic algorithms can improve diagnosis, risk stratification, and treatment planning.

Radiomic Phenotypes Distinguish Atypical Teratoid/Rhabdoid Tumors from Medulloblastoma.

BACKGROUND AND PURPOSE: Atypical teratoid/rhabdoid tumors and medulloblastomas have similar imaging and histologic features but distinctly different outcomes. We hypothesized that they could be distinguished by MR imaging-based radiomic phenotypes. MATERIALS AND METHODS: We retrospectively assembled T2-weighted and gadolinium-enhanced T1-weighted images of 48 posterior fossa atypical teratoid/rhabdoid tumors and 96 match-paired medulloblastomas from 7 institutions. Using a holdout test set, we measured the performance of 6 candidate classifier models using 6 imaging features derived by sparse regression of 900 T2WI and 900 T1WI Imaging Biomarker Standardization Initiative-based radiomics features. RESULTS: From the originally extracted 1800 total Imaging Biomarker Standardization Initiative-based features, sparse regression consistently reduced the feature set to 1 from T1WI and 5 from T2WI. Among classifier models, logistic regression performed with the highest AUC of 0.86, with sensitivity, specificity, accuracy, and F1 scores of 0.80, 0.82, 0.81, and 0.85, respectively. The top 3 important Imaging Biomarker Standardization Initiative features, by decreasing order of relative contribution, included voxel intensity at the 90th percentile, inverse difference moment normalized, and kurtosis-all from T2WI. CONCLUSIONS: Six quantitative signatures of image intensity, texture, and morphology distinguish atypical teratoid/rhabdoid tumors from medulloblastomas with high prediction performance across different machine learning strategies. Use of this technique for preoperative diagnosis of atypical teratoid/rhabdoid tumors could significantly inform therapeutic strategies and patient care discussions.

MRI Features of Histologically Diagnosed Supratentorial Primitive Neuroectodermal Tumors and Pineoblastomas in Correlation with Molecular Diagnoses and Outcomes: A Report from the Children's Oncology Group ACNS0332 Trial.

BACKGROUND AND PURPOSE: Supratentorial primitive neuroectodermal tumors and pineoblastomas have traditionally been grouped together for treatment purposes. Molecular profiling of these tumors has revealed a number of distinct entities and has led to the term "CNS-primitive neuroectodermal tumors" being removed from the 2016 World Health Organization classification. The purpose of this study was to describe the MR imaging findings of histologically diagnosed primitive neuroectodermal tumors and pineoblastomas and correlate them with molecular diagnoses and outcomes. MATERIALS AND METHODS: Histologically diagnosed primitive neuroectodermal tumors and pineoblastomas were enrolled in this Children's Oncology Group Phase III trial, and molecular classification was retrospectively completed using DNA methylation profiling. MR imaging features were systematically studied and correlated with molecular diagnoses and survival. RESULTS: Of the 85 patients enrolled, 56 met the inclusion criteria, in whom 28 tumors were in pineal and 28 in nonpineal locations. Methylation profiling revealed a variety of diagnoses, including pineoblastomas (n = 27), high-grade gliomas (n = 17), embryonal tumors (n = 7), atypical teratoid/rhabdoid tumors (n = 3), and ependymomas (n = 2). Thus, 39% overall and 71% of nonpineal tumor diagnoses were discrepant with histopathology. Tumor location, size, margins, and edema were predictors of embryonal-versus-nonembryonal tumors. Larger size and ill-defined margins correlated with poor event-free survival, while metastatic disease by MR imaging did not. CONCLUSIONS: In nonpineal locations, only a minority of histologically diagnosed primitive neuroectodermal tumors are embryonal tumors; therefore, high-grade glioma or ependymoma should be high on the radiographic differential. An understanding of molecularly defined tumor entities and their relative frequencies and locations will help the radiologist make more accurate predictions of the tumor types.