'The Reports of My Death Are Greatly Exaggerated'-Evaluating the Effect of Necrosis on MGMT Promoter Methylation Testing in High-Grade Glioma.

(1) Background: MGMT (O-6-methylguanine-DNA methyltransferase) promoter methylation remains an important predictive biomarker in high-grade gliomas (HGGs). The influence of necrosis on the fidelity of MGMT promoter (MGMTp) hypermethylation testing is currently unknown. Therefore, our study aims to evaluate the effect of varying degrees of necrosis on MGMTp status, as determined by pyrosequencing, in a series of primary and recurrent HGGs; (2) Methods: Within each case, the most viable blocks (assigned as 'true' MGMTp status) and the most necrotic block were determined by histopathology review. MGMTp status was determined by pyrosequencing. Comparisons of MGMTp status were made between the most viable and most necrotic blocks. (3) Results: 163 samples from 64 patients with HGGs were analyzed. MGMTp status was maintained in 84.6% of primary and 78.3% of recurrent HGGs between the most viable and necrotic blocks. A threshold of ≥60% tumor cellularity was established at which MGMTp status was unaltered, irrespective of the degree of necrosis. (4) Conclusions: MGMTp methylation status, as determined by pyrosequencing, does not appear to be influenced by necrosis in the majority of cases at a cellularity of at least 60%. Further investigation into the role of intratumoral heterogeneity on MGMTp status will increase our understanding of this predictive marker.

Prognostic and predictive biomarkers in central nervous system tumours: the molecular state of play.

Central nervous system (CNS) tumours were one of the first cancer types to adopt and integrate molecular profiling into routine clinical diagnosis in 2016. The vast majority of these biomarkers, used to discriminate between tumour types, also offered prognostic information. With the advent of The Cancer Genome Atlas (TCGA) and other large genomic datasets, further prognostic sub-stratification was possible within tumour types, leading to increased precision in CNS tumour grading. This review outlines the evolution of the molecular landscape of adult CNS tumours, through the prism of World Health Organization (WHO) Classifications. We begin our journey in the pre-molecular era, where high-grade gliomas were divided into 'primary' and 'secondary' glioblastomas. Molecular alterations explaining these clinicopathological observations were the first branching points of glioma diagnostics, with the discovery of IDH1/2 mutations and 1p/19q codeletion. Subsequently, the rigorous characterisation of paediatric gliomas led to the unearthing of histone H3 alterations as a key event in gliomagenesis, which also had implications for young adult patients. Simultaneously, studies investigating prognostic biomarkers within tumour types were undertaken. Certain genomic phenotypes were found to portend unfavourable outcomes, for example, MYCN amplification in spinal ependymoma. The arrival of methylation profiling, having revolutionised the diagnosis of CNS tumours, now promises to bring increased prognostic accuracy, as has been shown in meningiomas. While MGMT promoter hypermethylation has remained a reliable biomarker of response to cytotoxic chemotherapy, targeted therapy in CNS tumours has unfortunately not had the success of other cancers. Therefore, predictive biomarkers have lagged behind the identification of prognostic biomarkers in CNS tumours. Emerging research from new clinical trials is cause for guarded optimism and may shift our conceptualisation of predictive biomarker testing in CNS tumours.

Glioblastoma biomarkers in urinary extracellular vesicles reveal the potential for a 'liquid gold' biopsy.

BACKGROUND: Biomarkers that reflect glioblastoma tumour activity and treatment response are urgently needed to help guide clinical management, particularly for recurrent disease. As the urinary system is a major clearance route of circulating extracellular vesicles (EVs; 30-1000 nm nanoparticles) we explored whether sampling urinary-EVs could serve as a simple and non-invasive liquid biopsy approach for measuring glioblastoma-associated biomarkers. METHODS: Fifty urine specimens (15-60 ml) were collected from 24 catheterised glioblastoma patients immediately prior to primary (n = 17) and recurrence (n = 7) surgeries, following gross total resection (n = 9), and from age/gender-matched healthy participants (n = 14). EVs isolated by differential ultracentrifugation were characterised and extracted proteomes were analysed by high-resolution data-independent acquisition liquid chromatography tandem mass spectrometry (DIA-LC-MS/MS). RESULTS: Overall, 6857 proteins were confidently identified in urinary-EVs (q-value ≤ 0.01), including 94 EV marker proteins. Glioblastoma-specific proteomic signatures were determined, and putative urinary-EV biomarkers corresponding to tumour burden and recurrence were identified (FC ≥ | 2 | , adjust p-val≤0.05, AUC > 0.9). CONCLUSION: In-depth DIA-LC-MS/MS characterisation of urinary-EVs substantiates urine as a viable source of glioblastoma biomarkers. The promising 'liquid gold' biomarker panels described here warrant further investigation.

Understanding the Epitranscriptome for Avant-Garde Brain Tumour Diagnostics.

RNA modifications are diverse, dynamic, and reversible transcript alterations rapidly gaining attention due to their newly defined RNA regulatory roles in cellular pathways and pathogenic mechanisms. The exciting emerging field of 'epitranscriptomics' is predominantly centred on studying the most abundant mRNA modification, N6-methyladenine (m6A). The m6A mark, similar to many other RNA modifications, is strictly regulated by so-called 'writer', 'reader', and 'eraser' protein species. The abundance of genes coding for the expression of these regulator proteins and m6A levels shows great potential as diagnostic and predictive tools across several cancer fields. This review explores our current understanding of RNA modifications in glioma biology and the potential of epitranscriptomics to develop new diagnostic and predictive classification tools that can stratify these highly complex and heterogeneous brain tumours.

Understanding the extracellular vesicle surface for clinical molecular biology.

Extracellular vesicles (EVs) are lipid-membrane enclosed nanoparticles that play significant roles in health and disease. EVs are abundant in body fluids and carry an array of molecules (proteins, lipids, nucleic acids and glycans) that reflect the identity and activity of their cell-of-origin. While the advent of high throughput omics technologies has allowed in-depth characterisation of EV compositions, how these molecular species are spatially distributed within EV structures is not well appreciated. This is particularly true of the EV surface where a plethora of molecules are reported to be both integral and peripherally associated to the EV membrane. This coronal layer or 'atmosphere' that surrounds the EV membrane contributes to a large, highly interactive and dynamic surface area that is responsible for facilitating EV interactions with the extracellular environment. The EV coronal layer harbours surface molecules that reflect the identity of parent cells, which is likely a highly valuable property in the context of diagnostic liquid biopsies. In this review, we describe the current understanding of the mechanical, electrostatic and molecular properties of the EV surface that offer significant biomarker potential and contribute to a highly dynamic interactome.

An Open-Source AI Framework for the Analysis of Single Cells in Whole-Slide Images with a Note on CD276 in Glioblastoma.

Routine examination of entire histological slides at cellular resolution poses a significant if not insurmountable challenge to human observers. However, high-resolution data such as the cellular distribution of proteins in tissues, e.g., those obtained following immunochemical staining, are highly desirable. Our present study extends the applicability of the PathoFusion framework to the cellular level. We illustrate our approach using the detection of CD276 immunoreactive cells in glioblastoma as an example. Following automatic identification by means of PathoFusion's bifocal convolutional neural network (BCNN) model, individual cells are automatically profiled and counted. Only discriminable cells selected through data filtering and thresholding were segmented for cell-level analysis. Subsequently, we converted the detection signals into the corresponding heatmaps visualizing the distribution of the detected cells in entire whole-slide images of adjacent H&E-stained sections using the Discrete Wavelet Transform (DWT). Our results demonstrate that PathoFusion is capable of autonomously detecting and counting individual immunochemically labelled cells with a high prediction performance of 0.992 AUC and 97.7% accuracy. The data can be used for whole-slide cross-modality analyses, e.g., relationships between immunochemical signals and anaplastic histological features. PathoFusion has the potential to be applied to additional problems that seek to correlate heterogeneous data streams and to serve as a clinically applicable, weakly supervised system for histological image analyses in (neuro)pathology.

Postmortem brain donations vs premortem surgical resections for glioblastoma research: viewing the matter as a whole.

There have been limited improvements in diagnosis, treatment, and outcomes of primary brain cancers, including glioblastoma, over the past 10 years. This is largely attributable to persistent deficits in understanding brain tumor biology and pathogenesis due to a lack of high-quality biological research specimens. Traditional, premortem, surgical biopsy samples do not allow full characterization of the spatial and temporal heterogeneity of glioblastoma, nor capture end-stage disease to allow full evaluation of the evolutionary and mutational processes that lead to treatment resistance and recurrence. Furthermore, the necessity of ensuring sufficient viable tissue is available for histopathological diagnosis, while minimizing surgically induced functional deficit, leaves minimal tissue for research purposes and results in formalin fixation of most surgical specimens. Postmortem brain donation programs are rapidly gaining support due to their unique ability to address the limitations associated with surgical tissue sampling. Collecting, processing, and preserving tissue samples intended solely for research provides both a spatial and temporal view of tumor heterogeneity as well as the opportunity to fully characterize end-stage disease from histological and molecular standpoints. This review explores the limitations of traditional sample collection and the opportunities afforded by postmortem brain donations for future neurobiological cancer research.