Predisposition of Wingless Subgroup Medulloblastoma for Primary Tumor Hemorrhage

Are the Radiological and Molecular Features of Pediatric Medulloblastomas Valuable Prognostic Indicators? A 10-Year Retrospective Review in the Middle East

BACKGROUND

Medulloblastomas are the most common malignant brain tumors in the pediatric population. Based on the idea that tumors with identical radio-genomic features should behave similarly, the 4 molecular subtypes are now widely accepted as a guide for the management and prognosis. The radiological features of medulloblastomas can predict the molecular subtype; thus, anticipating the subsequent disease progression. However, this has not been evaluated comprehensively. We aim to thoroughly study the association between the molecular subtypes and radiological features of medulloblastomas. Moreover, we aim to investigate the efficacy of this correlation with the use of progression-free survival and 5-year survival rates.

METHODS

A retrospective analysis was conducted for all histopathological confirmed medulloblastomas in pediatric patients (<16 years old) that were operated on in Kuwait over the past ten years (n = 44). The radiological, histological, and molecular characteristics were justifiably evaluated and analyzed in our sample.

RESULTS

The overall progression-free survival after one year was noticed among 27 cases (≈44%) and the nonspecific 5-year survival was seen in 31 cases (≈70%) after a 5-year follow-up. Sonic Hedgehog and Wingless had the best outcomes, while group 3 showed the worst outcomes.

CONCLUSIONS

Our findings did not support the association between most of the typical magnetic resonance imaging characteristics and survival rate. We further established that Sonic Hedgehog and Wingless biological types have a better prognosis. There was no association observed between the radiographic features, specifically the location, and the molecular subtype.

Morphology-based molecular classification of spinal cord ependymomas using deep neural networks

Based on DNA-methylation, ependymomas growing in the spinal cord comprise two major molecular types termed spinal (SP-EPN) and myxopapillary ependymomas (MPE(-A/B)), which differ with respect to their clinical features and prognosis. Due to the existing discrepancy between histomorphogical diagnoses and classification using methylation data, we asked whether deep neural networks can predict the DNA methylation class of spinal cord ependymomas from hematoxylin and eosin stained whole-slide images. Using explainable AI, we further aimed to prospectively improve the consistency of histology-based diagnoses with DNA methylation profiling by identifying and quantifying distinct morphological patterns of these molecular ependymoma types. We assembled a case series of 139 molecularly characterized spinal cord ependymomas (nMPE  = 84, nSP-EPN  = 55). Self-supervised and weakly-supervised neural networks were used for classification. We employed attention analysis and supervised machine-learning methods for the discovery and quantification of morphological features and their correlation to the diagnoses of experienced neuropathologists. Our best performing model predicted the DNA methylation class with 98% test accuracy and used self-supervised learning to outperform pretrained encoder-networks (86% test accuracy). In contrast, the diagnoses of neuropathologists matched the DNA methylation class in only 83% of cases. Domain-adaptation techniques improved model generalization to an external validation cohort by up to 22%. Statistically significant morphological features were identified per molecular type and quantitatively correlated to human diagnoses. The approach was extended to recently defined subtypes of myxopapillary ependymomas (MPE-(A/B), 80% test accuracy). In summary, we demonstrated the accurate prediction of the DNA methylation class of spinal cord ependymomas (SP-EPN, MPE(-A/B)) using hematoxylin and eosin stained whole-slide images. Our approach may prospectively serve as a supplementary resource for integrated diagnostics and may even help to establish a standardized, high-quality level of histology-based diagnostics across institutions-in particular in low-income countries, where expensive DNA-methylation analyses may not be readily available.

Imaging Characteristics of Wingless Pathway Subgroup Medulloblastomas: Results from the German HIT/SIOP-Trial Cohort

BACKGROUND AND PURPOSE

In addition to the 4 histopathologically defined entities of medulloblastoma, 4 distinct genetically defined subgroups have been included in the World Health Organization classification of 2016. The smallest subgroup is the medulloblastoma with activated wingless pathway. The goal of this study was to identify a typical MR imaging morphology in a larger number of pediatric patients with wingless pathway medulloblastoma.

MATERIALS AND METHODS

From January 2001 to October 2017, of 75 patients with histologically confirmed and molecularly subgrouped wingless pathway medulloblastomas recruited to the German Pediatric Brain Tumor (HIT) trials, 38 patients (median age, 12.8 ± 4.6 years at diagnosis; 24 [63.2%] female) had preoperative imaging that passed the entry criteria for this study. Images were rated by the local standardized imaging criteria of the National Reference Center of Neuroradiology. Additionally, a modified laterality score was used to determine tumor localization and extension.

RESULTS

Twenty-eight of 38 (73.7%) were primary midline tumors but with a lateral tendency in 39.3%. One extensively eccentric midline tumor was rated by the laterality score as in an off-midline position. Five tumors were found in the cerebellopontine angle; 3, in the deep white matter; and 2, in a cerebellar hemisphere. Leptomeningeal dissemination was rare (11.5%). In 60.5%, intratumoral blood-degradation products were found, and 26.3% showed cysts with blood contents.

CONCLUSIONS

According to our observations, wingless pathway medulloblastomas are not preferentially off-midline tumors as postulated in previous studies with smaller wingless pathway medulloblastoma cohorts. Dense intratumoral blood-degradation products and cysts with blood contents are frequently found and might help to differentiate wingless pathway medulloblastoma from other medulloblastoma subtypes.