Bendok, Bernard R.

Abstract High-grade gliomas (HGGs) are aggressive, infiltrative brain tumors with poor survival outcomes and limited therapeutic options. Their spatial and molecular heterogeneity necessitates large, well-curated datasets integrating imaging, clinical, and genomic information to enable precision medicine and drive novel treatment development. Using the Medtronic StealthStation neuronavigation system, we collected over 1,000 spatially localized intraoperative biopsy samples from more than 250 patients undergoing resection for glioma at Mayo Clinic and partnering sites. Each biopsy site was recorded with 3D coordinates, labeled in real time, flash-frozen, and annotated with metadata including collection time, anatomical location, and surgical plan. Samples were processed and entered into a secure database along with the corresponding sampling coordinates, frozen pathology, whole-exome and RNA sequencing data, 5-ALA fluorescence status, patient demographics, and T1, T1Gd, T2, and T2-FLAIR MRI. This multimodal dataset supports radiogenomic research by enabling precise mapping of imaging features to molecular characteristics. A subset of this data, including 268 biopsies from 52 patients (25 male, 27 female) with reported intraoperative 5-ALA fluorescence, was used to develop a radiomics model predicting locoregional patterns of 5-ALA positivity based on preoperative MRI. Image features were extracted from the biopsy sample coordinates across multiple MRI sequences and paired with corresponding 5-ALA fluorescence status to train the model for regional fluorescence prediction. The final model achieved 83% accuracy on the training set and 86% on the validation set using a 70-15-15 data split; this included 208 samples for training and 42 for validation, stratified by patient to ensure no overlap of samples from the same individual across sets. These findings demonstrate the feasibility of predicting 5-ALA fluorescence using radiomic features from standard preoperative MRI, offering a promising noninvasive strategy for improving intraoperative decision-making and enabling more precise, personalized resections in patients with high-grade gliomas.

Abstract BACKGROUND Glioblastoma (GBM) is a molecularly and phenotypically heterogeneous tumor, with regional invasion patterns driving treatment resistance and universal recurrence. In infiltrative regions, GBM cells adopt at least two predominant invasive phenotypes: Neuronal-enriched (NEU), with synaptic and neurodevelopment programs, and Glycolytic/Plurimetabolic (GPM), with heightened immune and metabolic activity. These subtypes differ in therapeutic vulnerability and microenvironmental interactions; features not captured by conventional contrast-enhanced (T1+C) MRI due to minimal blood-brain barrier (BBB) disruption. Our group sought to develop predictive models to distinguish variability of NEU/GPM subtypes within infiltrative GBM, utilizing graph convolutional networks (GCN), advanced multiparametric MRI (mpMRI), and spatially-matched image-localized biopsies with transcriptomic profiling. METHODS We analyzed 18 mpMRI features —including conventional, diffusion-weighted, and Dynamic Susceptibility Contrast (DSC) perfusion MRI— of 101 image-localized biopsies from MRI non-enhancing infiltrative margins of 64 GBM patient tumors. Features were hierarchically clustered per modality; each cluster was represented as a node in a Neo4j graph. Edges connected nodes sharing overlapping samples. Nodes were annotated with NEU or GPM subtype proportion. GCNs were trained to classify subtype-enriched nodes using graph topology and node features. RESULTS GCNs achieved high predictive performance on held-out test data (AUC = 0.95 for NEU, 0.88 for GPM; 70:30 train-test split) with similar 5-fold cross-validation performance and generalized well to an independent cohort (AUC > 0.85, n = 45). Importantly, performance was robust without relying on T1+C imaging, enabling prediction beyond BBB-disrupted regions. GCNs outperformed traditional classifiers (logistic regression, random forest). CONCLUSION Our results demonstrate that GCN-modeling of mpMRI features enables non-invasive prediction of invasive GBM subtypes at the tumor margin. Unlike convolutional neural networks (CNN), which rely on spatial continuity, GCNs leverage relational structure between heterogeneous imaging features, making them ideal for modeling sparse, non-grid data. This may support non-invasive regional vulnerability detection, surgical targeting, and improved response assessment.

Abstract We describe a 46-year-old male with Neurofibromatosis type 1 (NF1) and a previously known parenchymal medullary mass identified in 2022, initially managed with surveillance. In late 2023, he developed a progressive brainstem syndrome prompting presentation to our center. MRI showed a T2/FLAIR hyperintense, heterogeneously enhancing ventral medullary mass with leptomeningeal involvement, a non-enhancing right pontine lesion, right optic nerve glioma, and nodular leptomeningeal foci at T4 and L4–L5. He received IMRT to the brainstem prior to tissue diagnosis with good local response. Due to progression at T4, biopsy revealed a hypercellular, mitotically active glioma without necrosis or microvascular proliferation. Molecular profiling demonstrated IDH1/2-wildtype, H3K27M-negative, retained ATRX, intact 1p/19q, and CDKN2A/B homozygous deletion. DNA methylation profiling classified the tumor as high-grade astrocytoma with piloid features (HGAP). Spinal lesions responded to proton therapy, but new metastatic foci emerged in the subependymal lateral ventricles, leptomeninges, and cervical spinal cord. He underwent whole-brain radiation, oral temozolomide, and intrathecal cytarabine/etoposide. In June 2024, cell-free DNA analysis showed a similar molecular profile with newly identified 1p/19q codeletion, diverging from tissue findings. The patient remains alive >3 years from initial presentation. HGAP is a novel molecular class of IDH-wildtype gliomas defined by DNA methylation profiling and MAPK pathway alterations (commonly NF1, BRAF, FGFR1), CDKN2A/B codeletion, and ATRX alterations. These tumors often resemble glioblastoma or diffuse midline gliomas on imaging and are typically infratentorial or spinal, complicating biopsy. Prognostic data are limited, with <50% 5-year overall survival and no clear prognostic value of MGMT promoter methylation. This case illustrates multifocal HGAP in a patient with NF1 with favorable survival. The acquired 1p/19q codeletion raises questions about tumor mosaicism, subclonal evolution, and prognostic implications.

BACKGROUND AND OBJECTIVES: While many trials of the last decade have shifted the paradigm of stroke care toward endovascular therapy, a lack of timely access to centers with neurointerventional capabilities for many patients has remained a significant barrier to care. Endovascular robotic systems are poised to address this need for expanded access through telerobotic procedures, while also benefiting clinical team safety and raising the possibility of improved procedural precision. METHODS: A telerobotic setup consisting of a magnetic field generator and a mechanical advancer was placed with a silicone vascular phantom in an interventional suite in Switzerland and connected by wireless internet to an operator console in Arizona. From the operator console, the neurosurgeon navigated a guidewire and aspiration catheter to simulated middle cerebral artery (MCA) (n = 10) and basilar artery (n = 10) clots in consecutive timed procedural trials in evaluation of system feasibility and operator learning curve. RESULTS: From over 5700 miles away, the neurosurgeon navigated an aspiration catheter to 10 MCA and 10 basilar artery simulated large vessel occlusions. Clot aspiration was attempted in 4 trials and successful in each. Endovascular navigational time over 10 procedural trials to the M1 segment of the MCA and 10 trials to the basilar artery were completed on average in 114 seconds (SD 44.4 seconds) and 134 seconds (SD 65.8 seconds), respectively. Sufficient operator mastery was demonstrated in 20 procedural iterations with an average 77% reduction in navigational time from aortic arch initial to final trials. CONCLUSION: To our knowledge, we present the first intercontinental aspiration thrombectomy in the preclinical setting in comparable time to the clinical practice and demonstrate the early potential of telerobotic navigation. A lack of haptic feedback did not pose a hindrance, and the potential for intuitive operation offers to drive future research and adoption of this telesurgery system.

BACKGROUND AND OBJECTIVES: Aneurysm risk prediction remains an imprecise science that places patients at risk for either over or undertreatment. Machine learning (ML) models may improve clinical practice by adding precision to risk assessment. This study aims to comprehensively assess the current landscape of ML applications in predicting the risk of aneurysm rupture and compare the performance with the widely used PHASES score. METHODS: A systematic review of PubMed, Scopus, and Web of Science was conducted. All studies using ML tools to predict the rupture risk of intracranial aneurysms were included. Meta-analysis was conducted with consideration to the ML algorithms and compared with the PHASES score. RESULTS: Thirty-six studies involving 22 462 patients were included in the final analysis. ML techniques, including 124 models using 25 algorithms, were employed. Among various ML models, while they had comparable diagnostic performance, deep learning exhibited a slightly better performance profile (sensitivity = 0.792, specificity = 0.788, and accuracy = 0.778 in external validation). Based on our analysis, ML, regardless of the algorithm, provides comparable sensitivity (0.743 vs 0.771, P = .60) and higher specificity (0.763 vs 0.507, P < .01) compared with the PHASES score. Consistently, pooling the area under the receiver operating characteristic curve (AUC) for 60 ML models and 5 PHASES score data, ML models exhibited higher AUC (0.84 vs 0.64, P < .01). Using hemodynamic parameters as input for models improved specificity ( P < .01) in the test sets without any significant changes in the sensitivity. The later improvement was not observed in the external validation sets. CONCLUSION: ML techniques have the potential to enhance the prediction of intracranial aneurysm rupture compared with traditional approaches, like the PHASES score. Incorporating hemodynamic parameters may further enhance the accuracy of ML models. Feature prospective studies are required to validate the utility of ML models for clinical integration.