Search

Aim: Analysis of cerebrospinal fluid (CSF) is essential for diagnostic workup of patients with neurological diseases and includes differential cell typing. The current gold standard is based on microscopic examination by specialised technicians and neuropathologists, which is time-consuming, labour-intensive and subjective., Methods: We, therefore, developed an image analysis approach based on expert annotations of 123,181 digitised CSF objects from 78 patients corresponding to 15 clinically relevant categories and trained a multiclass convolutional neural network (CNN)., Results: The CNN classified the 15 categories with high accuracy (mean AUC 97.3%). By using explainable artificial intelligence (XAI), we demonstrate that the CNN identified meaningful cellular substructures in CSF cells recapitulating human pattern recognition. Based on the evaluation of 511 cells selected from 12 different CSF samples, we validated the CNN by comparing it with seven board-certified neuropathologists blinded for clinical information. Inter-rater agreement between the CNN and the ground truth was non-inferior (Krippendorff's alpha 0.79) compared with the agreement of seven human raters and the ground truth (mean Krippendorff's alpha 0.72, range 0.56-0.81). The CNN assigned the correct diagnostic label (inflammatory, haemorrhagic or neoplastic) in 10 out of 11 clinical samples, compared with 7-11 out of 11 by human raters., Conclusions: Our approach provides the basis to overcome current limitations in automated cell classification for routine diagnostics and demonstrates how a visual explanation framework can connect machine decision-making with cell properties and thus provide a novel versatile and quantitative method for investigating CSF manifestations of various neurological diseases., (© 2022 The Authors. Neuropathology and Applied Neurobiology published by John Wiley & Sons Ltd on behalf of British Neuropathological Society.)

Aims: Although inactivation of the von Hippel-Lindau gene (VHL) on chromosome 3p25 is considered to be the major cause of hereditary endolymphatic sac tumours (ELSTs), the genetic background of sporadic ELST is largely unknown. The aim of this study was to determine the prevalence of VHL mutations in sporadic ELSTs and compare their characteristics to VHL-disease-related tumours., Methods: Genetic and epigenetic alterations were compared between 11 sporadic and 11 VHL-disease-related ELSTs by targeted sequencing and DNA methylation analysis., Results: VHL mutations and small deletions detected by targeted deep sequencing were identified in 9/11 sporadic ELSTs (82%). No other cancer-related genetic pathway was altered except for TERT promoter mutations in two sporadic ELST and one VHL-disease-related ELST (15%). Loss of heterozygosity of chromosome 3 was found in 6/10 (60%) VHL-disease-related and 10/11 (91%) sporadic ELSTs resulting in biallelic VHL inactivation in 8/10 (73%) sporadic ELSTs. DNA methylation profiling did not reveal differences between sporadic and VHL-disease-related ELSTs but reliably distinguished ELST from morphological mimics of the cerebellopontine angle. VHL patients were significantly younger at disease onset compared to sporadic ELSTs (29 vs. 52 years, p < 0.0001, Fisher's exact test). VHL-disease status was not associated with an increased risk of recurrence, but the presence of clear cells was found to be associated with shorter progression-free survival (p = 0.0002, log-rank test)., Conclusion: Biallelic inactivation of VHL is the main mechanism underlying ELSTs, but unknown mechanisms beyond VHL may rarely be involved in the pathogenesis of sporadic ELSTs., (© 2021 The Authors. Neuropathology and Applied Neurobiology published by John Wiley & Sons Ltd on behalf of British Neuropathological Society.)

Paragangliomas/pheochromocytomas are rare neuroendocrine tumors that arise from the adrenal gland or ganglia at various sites throughout the body. They display a remarkable diversity of driver alterations and are associated with germline mutations in up to 40% of the cases. Comprehensive molecular profiling of abdomino-thoracic paragangliomas revealed four molecularly defined and clinically relevant subtypes. Paragangliomas of the cauda equina region are considered to belong to one of the defined molecular subtypes, but a systematic molecular analysis has not yet been performed. In this study, we analyzed genome-wide DNA methylation profiles of 57 cauda equina paragangliomas and show that these tumors are epigenetically distinct from non-spinal paragangliomas and other tumors. In contrast to paragangliomas of other sites, chromosomal imbalances are widely lacking in cauda equina paragangliomas. Furthermore, RNA and DNA exome sequencing revealed that frequent genetic alterations found in non-spinal paragangliomas-including the prognostically relevant SDH mutations-are absent in cauda equina paragangliomas. Histologically, cauda equina paragangliomas show frequently gangliocytic differentiation and strong immunoreactivity to pan-cytokeratin and cytokeratin 18, which is not common in paragangliomas of other sites. None of our cases had a familial paraganglioma syndrome. Tumors rarely recurred (9%) or presented with multiple lesions within the spinal compartment (7%), but did not metastasize outside the CNS. In summary, we show that cauda equina paragangliomas represent a distinct, sporadic tumor entity defined by a unique clinical and morpho-molecular profile.