Glioblastoma (GBM) is a highly aggressive primary brain cancer with dismal prognoses despite current standards of care. Immunotherapy is being explored for GBM, given its promising results in other solid malignancies; however, the results from early clinical studies in GBM are disappointing. It has been discovered that GBM has numerous mechanisms of immune resistance, including the physical blood–brain barrier, high intratumoral and intertumoral heterogeneity, and numerous cellular and molecular components in the tumor microenvironment (TME) that promote immunosuppression. Furthermore, GBM utilizes numerous metabolic pathways to establish a survival advantage in the TME. Recently, it has begun to become evident that these complex metabolic pathways that promote GBM growth and invasion also contribute to tumor immune resistance. Aerobic glycolysis provides tumor cells with ample ATP while depleting key glucose and increasing acidity in the TME. Increased glutamine, tryptophan, and arginine metabolism deprives T cells of these necessary amino acids for proper anti-tumor function. Sphingolipid metabolism promotes an immunosuppressive phenotype in the TME and affects immune cell trafficking. This review will discuss, in detail, the key metabolic pathways relevant to GBM pathophysiology which also modulate host immunosuppression.

Glioblastoma (GBM) is a universally fatal cancer for which the standard of care has remained largely unchanged for the last 20 years. Recent work has demonstrated that most therapeutic trials for GBM fail due to complex mechanisms of immunosuppression mediated by both the innate and adaptive immune systems. Various metabolic alterations in the tumor microenvironment help maintain this local and systemic immunosuppression, of which the axis of hypoxia-driven tryptophan degradation has garnered substantial attention over the last decade. This paper synthesizes a much-needed elucidation of the immunometabolic reshaping of glioma, myeloid, endothelial, and lymphoid cell lineages induced by hypoxia. The current paper critically evaluates the role of IDO1/TDO2-mediated breakdown of tryptophan and the consequent accumulation of kynurenine, a metabolite that triggers GCN2- and AHR-mediated CD8+ T-cell exhaustion and supports regulatory T-cell differentiation and expansion. Furthermore, we propose a synthesis of mechanistic evidence that establishes a role for the Trp-GCN2-ATF4-VEGFA axis in hypoxia-induced immunosuppression, supporting that pro-tumoral metabolic dysregulation is directly linked to angiogenesis. In GBM, hypoxia and tryptophan–kynurenine pathway dysregulation operate as an integrated metabolic circuit that drives widespread immunosuppression. These mechanisms can be captured by a metabolic signature shared across nearly every cell type in the GBM microenvironment. Drawing on recent spatial transcriptomic, metabolomic, and pharmacologic studies, we outline how this metabolic axis shapes disease biology and how it can be targeted to restore effective antitumor immunity.

Introduction: The advancement of artificial intelligence (AI) in neurosurgery is dependent on high quality, large, labeled datasets. Labeled neurosurgical datasets are rare, driven by the high expertise required for labeling neurosurgical data. A comprehensive resource overviewing available datasets for AI in neurosurgery is essential to identify areas for potential model building and areas of needed data construction. Methods: We conducted a systematic review according to PRISMA guidelines to identify publicly available neurosurgical datasets suitable for machine learning. A PubMed search on 8 February 2025, yielded 267 articles, of which 86 met inclusion criteria. Each study was reviewed to extract dataset characteristics, model development details, validation status, availability, and citation impact. Results: Among the 86 included studies, 83.7% focused on spine pathology, with tumor (3.5%), vascular (4.7%), and trauma (7.0%) comprising the remaining. The majority of datasets were image-based, particularly X-ray (37.2%), MRI (29.1%), and CT (20.9%). Label types included segmentation (36.0%), diagnosis (26.7%), and detection/localization (20.9%), with only 2.3% including outcome labels. While 97.7% of studies reported training a model, only 22.6% performed external validation, 20.2% shared code, and just 7.1% provided public applications. Accuracy was the most frequently reported performance metric, even for segmentation tasks, where only 60% of studies used the Dice score metric. Studies often lacked task-appropriate evaluation metrics. Conclusions: We conducted a systematic review to capture all publicly accessible datasets that can be applied to build AI models for neurosurgery. Current datasets are heavily skewed towards spine imaging and lack both clinical patient specific and outcomes information. Provided baseline models from these datasets are limited by poor external validation, lack of reproducibility, and reliance on suboptimal evaluation metrics. Future efforts should prioritize developing multi-institutional datasets with outcome labels, validated models, public access, and domain diversity to accelerate the safe and effective integration of AI into neurosurgical care.

Whilst typically benign, a subset of meningiomas displays aggressive and recurrent behavior. There is a paucity of reliable treatment options for this subset of patients and a relative lack of consensus on how to best manage these patients. This clinical challenge reflects underlying molecular complexity, driven by NF2, TRAF7, and CDKN2A/B mutations alongside pervasive epigenetic dysregulation. High-throughput molecular profiling studies have proposed biologically distinct meningioma subgroups with varying clinical trajectories and therapeutic vulnerabilities. Distinct cell lineages of meningeal precursors are now appreciated to be essential in the establishment of the meninges. The numerous cellular lineages involved in meningeal development, the heterogeneity of meningioma location and (epi)genomic behavior, and the variability in its clinical and radiological manifestations raise the question of what critical insights can be gained by understanding meningeal development during embryogenesis to understand meningioma tumorigenicity. The current paper examines this paradigm by highlighting spatially linked mechanisms of anaplasia and treatment resistance, including the role of neural crest-derived convexity meninges in promoting dedifferentiation via YAP/TAZ signaling and mesoderm-derived skull base regions in maintaining TRAF7-mediated vulnerabilities. We further elucidate the emerging synthetic lethal paradigms, CRISPR-enabled target discovery, and PROTAC-mediated degradation strategies that may transform the therapeutic landscape of clinically challenging meningiomas driven by complex oncogenic circuitry. By bridging embryogenesis, spatial genomics, and molecular targeting, we propose a developmentally informed, lineage-stratified model for advancing precision therapeutics in high-grade and recurrent meningiomas.

Introduction: Three-dimensional deep learning offers promise for automated accurate brain tumor segmentation and survival prediction but requires robust validation across multiple MRI modalities to be effectively implemented in clinical practice. Methods: This study presents a comprehensive 3D deep learning framework using 369 cases from the BraTS2020 dataset. A 3D U-Net architecture was developed for tumor segmentation utilizing combined imaging data and optimized for computational efficiency and memory. The final 3D U-Net model segmentations were used to build machine learning 6-month and 12-month survival classifiers. Segmentation models were evaluated using multiple metrics, including the Dice Similarity Coefficient, Hausdorff Distance, and Cohen’s d. The classification models were evaluated using AUC-ROC and balanced accuracy. Results: Segmentation achieved a modest, but promising, performance across 30 epochs and with 295 training patients, achieving the best mean validation Dice = 0.8388 and a final-epoch mean Dice of 0.8263. Survival classification with a hybrid clinical and imaging logistic regression showed promising results, with 12-month prediction achieving AUC = 0.746 and 69% accuracy. The top contributing features for the 12-month prediction classifier were extent of resection, T1 contrast-enhanced tumor median, and FLAIR tumor median. Conclusions: This comprehensive framework demonstrates that a multi-modal approach provides meaningful performance gains, while segmentation-derived features show a promising ability to enable survival prediction.