Search

Simple Summary: Meningioma is the most common primary intracranial tumor. DNA methylation-based subtyping, while highly useful for diagnosis and treatment planning, is costly and not widely available. Therefore, the identification of methylation classes based on histological sections would be highly beneficial as it could greatly support and accelerate diagnostic and treatment decisions. We developed and systematically evaluated an AI framework to perform the classification of the most prevalent methylation subtypes based on histological sections. The model achieved a balanced accuracy of 0.870 for benign-1 vs benign-2 and 0.749 for benign-1 vs. intermediate-A in a narrow validation set. Combined with the network's assessed focus on key tumor regions these results provide a promising proof-of-concept of such an AI-driven classification approach in precision medicine. Convolutional neural networks (CNNs) are becoming increasingly valuable tools for advanced computational histopathology, promoting precision medicine through exceptional visual decoding abilities. Meningiomas, the most prevalent primary intracranial tumors, necessitate accurate grading and classification for informed clinical decision-making. Recently, DNA methylation-based molecular classification of meningiomas has proven to be more effective in predicting tumor recurrence than traditional histopathological methods. However, DNA methylation profiling is expensive, labor-intensive, and not widely accessible. Consequently, a digital histology-based prediction of DNA methylation classes would be advantageous, complementing molecular classification. In this study, we developed and rigorously assessed an attention-based multiple-instance deep neural network for predicting meningioma methylation classes using tumor methylome data from 142 (+51) patients and corresponding hematoxylin-eosin-stained histological sections. Pairwise analysis of sample cohorts from three meningioma methylation classes demonstrated high accuracy in two combinations. The performance of our approach was validated using an independent set of 51 meningioma patient samples. Importantly, attention map visualization revealed that the algorithm primarily focuses on tumor regions deemed significant by neuropathologists, offering insights into the decision-making process of the CNN. Our findings highlight the capacity of CNNs to effectively harness phenotypic information from histological sections through computerized images for precision medicine. Notably, this study is the first demonstration of predicting clinically relevant DNA methylome information using computer vision applied to standard histopathology. The introduced AI framework holds great potential in supporting, augmenting, and expediting meningioma classification in the future. [ABSTRACT FROM AUTHOR]

Chronic obstructive pulmonary disease (COPD) is a disease with an inflammatory phenotype with increasing prevalence in the elderly. Expanded population of mutant blood cells carrying somatic mutations is termed clonal hematopoiesis of indeterminate potential (CHIP). The association between CHIP and COPD and its relevant effects on DNA methylation in aging are mainly unknown. Analyzing the deep-targeted amplicon sequencing from 125 COPD patients, we found enhanced incidence of CHIP mutations (~20%) with a predominance of DNMT3A CHIP-mediated hypomethylation of Phospholipase D Family Member 5 (PLD5), which in turn is positively correlated with increased levels of glycerol phosphocholine, pro-inflammatory cytokines, and deteriorating lung function. [ABSTRACT FROM AUTHOR]

Clinicians in molecular tumor boards (MTB) are confronted with a growing amount of genetic high-throughput sequencing data. Today, at German university hospitals, these data are usually handled in complex spreadsheets from which clinicians have to obtain the necessary information. The aim of this work was to gather a comprehensive list of requirements to be met by cBioPortal to support processes in MTBs according to clinical needs. Therefore, oncology experts at nine German university hospitals were surveyed in two rounds of interviews. To generate an interview guideline a scoping review was conducted. For visual support in the second round, screenshot mockups illustrating the requirements from the first round were created. Requirements that cBioPortal already meets were skipped during the second round. In the end, 24 requirements with sometimes several conceivable options were identified and 54 screenshot mockups were created. Some of the identified requirements have already been suggested to the community by other users or are currently being implemented in cBioPortal. This shows, that the results are in line with the needs expressed by various disciplines. According to our findings, cBioPortal has the potential to significantly improve the processes and analyses of an MTB after the implementation of the identified requirements. [ABSTRACT FROM AUTHOR]

About 95% of patients with Glioblastoma (GBM) show tumor relapse, leaving them with limited therapeutic options as recurrent tumors are most often resistant to the first line chemotherapy standard Temozolomide (TMZ). To identify molecular pathways involved in TMZ resistance, primary GBM Stem-like Cells (GSCs) were isolated, characterized, and selected for TMZ resistance in vitro. Subsequently, RNA sequencing analysis was performed and revealed a total of 49 differentially expressed genes (|log2-fold change| > 0.5 and adjusted p-value < 0.1) in TMZ resistant stem-like cells compared to their matched DMSO control cells. Among up-regulated genes, we identified carbonic anhydrase 2 (CA2) as a candidate gene correlated with glioma malignancy and patient survival. Notably, we describe consistent up-regulation of CA2 not only in TMZ resistant GSCs on mRNA and protein level, but also in patient-matched clinical samples of first manifest and recurrent tumors. Co-treatment with the carbonic anhydrase inhibitor Acetazolamid (ACZ) sensitized cells to TMZ induced cell death. Cumulatively, our findings illustrate the potential of CA2 as a chemosensitizing target in recurrent GBM and provide a rationale for a therapy associated inhibition of CA2 to overcome TMZ induced chemoresistance. [ABSTRACT FROM AUTHOR]