Zadeh, Gelareh

Abstract Glioblastoma (GBM) is a highly aggressive brain tumor with poor survival outcomes and limited tools for accurate prediction of individual patient prognosis. While many prognostic models rely on advanced molecular testing, such methods are not routinely accessible across most clinical settings. In this study, we identify a practical and clinically applicable approach to estimate prognosis for GBM patients using three objective and readily accessible features: the patient’s age, tumor surface irregularity determined from MRI scans, and the expression of ICAM1 measured by standard immunohistochemistry. We show that higher ICAM1 levels are linked to worse survival and a more inflamed but immunosuppressive tumor microenvironment. These findings are functionally validated in ICAM1-wildtype and knockout mouse models using single-cell proteomics via cytometry by time-of-flight, and an immune-comprehensive panel of antibodies. By combining age, ICAM1, and MRI-derived shape features into a single Clinical-Imaging-Molecular (CIM) score, we developed a model that can reliably separate patients with a GBM into higher- and lower-risk groups. This score was validated across multiple international institutions and patient populations and does not require genomic sequencing or specialized technologies. The CIM model could serve as a supplement for care providers to better tailor treatment plans and follow-up strategies for individual patients, making personalized care more achievable even in resource-limited settings.

Abstract Computer vision now enables deep learning models to link histopathologic features to molecular features in a quantitative manner not previously achievable by human pathologists. However, human interpretability is often limited. We trained a deep neural network to predict gene expression from digital whole slide images (WSIs) using 848 central nervous system (CNS) tumors with paired WSI and RNA-sequencing data. We used inferred RNA expression levels as a surrogate for protein expression of the CNS tumor lineage markers GFAP, OLIG2, and SSTR2, as well as the proliferation marker MKI67 (Ki67). We established thresholds for categorizing the inferred expression levels as positive or negative based on levels observed in cross-validation testing in select tumor types to approximate routine clinical interpretation of immunohistochemical (IHC) staining. We tested the sensitivity and specificity of our ‘virtual’ IHC on an independent, multi-institutional cohort of over 2,000 CNS tumors with objective diagnostic labels derived from DNA methylation-based tumor classification. As a dichotomous variable, inferred GFAP expression showed a sensitivity of 74% and a specificity of 99% as a glial marker in a cohort of gliomas and meningiomas. In the same cohort, OLIG2 showed a sensitivity of 75% and a specificity of 97% as a glial marker, while SSTR2 showed a sensitivity of 83% and a specificity of 97% as a marker of meningioma. In a cohort of ependymomas and gliomas, virtual IHC for OLIG2 showed a sensitivity of 75% and a specificity of 82% in distinguishing ependymomas from gliomas. Finally, in a cohort of gliomas, inferred expression of MKI67 demonstrated a trend consistent with what would be expected by actual IHC, with increasing MKI67 expression with increasing tumor grade. Our model provides the basis for a human interpretable and clinically applicable deep neural network to aid human pathologists in the diagnosis and grading of CNS tumors.

Abstract Molecular testing has become critical components in the diagnostic classification of the central nervous system (CNS) tumors. However, these methods require substantial resources, limiting accessibility for many patients. Recent advances in artificial intelligence (AI) and computer vision empower deep learning models to infer molecular features from histopathology images, which may often be more readily available than molecular testing, to classify CNS tumors. We trained pan-CNS tumor models to predict DNA methylation and gene expression from whole-slide images (WSIs). These molecular predictions were then used in a hierarchical machine learning framework, termed Neuropath-AI, to predict nine broad tumor families and 52 tumor types using a large diverse cohort of 5,715 histopathology samples, of which 2,988 had paired DNA methylation profiling and 848 had paired RNA-sequencing. We evaluated Neuropath-AI on another large independent multi-institutional cohort of 5,516 CNS tumors, for which Neuropath-AI predicted tumor class with an associated confidence score. Neuropath-AI designated 46% of test samples as predictable with high-confidence, which it then accuracy classified in 97% of samples. It made moderate-confidence (and above) predictions for 87% of test samples, for which it achieved a diagnostic accuracy of 80% for the top-1 prediction and 86% accuracy for a top-2 predictions. A comparison with human neuropathologists showed that Neuropath-AI is comparable at classifying tumors. A second test showed that human neuropathologists were more accurate when the Neuropath-AI classifier results were provided to them, testifying to the potential translational benefit of integrating AI classification tools into the pathologist’s workflow. Our model provides the basis for a clinically applicable deep learning assistant to improve human efficiency and accuracy of CNS tumor diagnosis. The model will be made publicly available and can be readily implemented to assist human pathologists in clinical workflows, in further expanded prospective studies.

Abstract Pediatric, adolescent and young adult patients (Ped/AYA; 0–39 years) with meningiomas are underrepresented in neuro-oncology research, despite meningiomas being the second most common primary brain tumor in this age group. While DNA methylation-based models have advanced classification and prognostication in adults, their utility in younger populations remains uncertain. To address this gap, we aimed to evaluate the performance of current DNA methylation-based classification and recurrence prediction models in Ped/AYA patients with meningiomas and identify age-associated molecular differences. As part of this study, we analyzed 1,568 meningioma samples with DNA methylation and clinical data, including subsets with whole-exome and RNA sequencing. Patients were grouped into Ped/AYA (≤39 years, n = 213) and adult (>39 years, n = 1,355) cohorts. We assessed classifier performance, performed subgroup-specific model retraining, and compared molecular profiles across age groups. Our results showed that molecular classification differed significantly between age groups, with Ped/AYA tumors less frequently belonging to the most aggressive Proliferative Molecular Group. The DNA methylation-based recurrence model performed well in adults (AUC = 0.83) but poorly in the Ped/AYA cohort (AUC = 0.57). However, retraining the model on only younger patients improved performance (AUC = 0.75). In addition, Ped/AYA tumors exhibited significantly fewer chromosomal alterations, including lower frequencies of 1p, 6q, and 14q loss (p<0.05). Taken together, these findings indicate that predominantly adult-trained methylation models underperform in younger meningioma patients due to both biological divergence and exclusion from training datasets. Our findings underscore the need for age-specific molecular tools to improve prognostication and support personalized care across the lifespan in neuro-oncology.

Abstract BACKGROUND Post-operative prognostication in glioma patients remains a clinical challenge, particularly given the heterogeneity in tumor biology and patient outcomes. We evaluated whether image-derived features from post-operative imaging improve survival modeling beyond established clinicopathologic factors. METHODS We analyzed a cohort of 71 post-surgical glioma patients treated at Princess Margaret Cancer Centre with known clinicopathologic features including age, histological classification, IDH mutation status, MGMT methylation status, extent of resection, and performance status. A baseline Cox proportional hazards model was fit to clinical variables. Radiomic features were extracted using PyRadiomics from post-operative CT and MRI across three regions of interest: whole brain, tumor bed, and peritumoral ring. Features underwent stepwise unsupervised reduction (variance filtering, exclusion of volume-correlated features, and removal of remaining collinear features), followed by dimensionality reduction via principal component analysis. Combined clinical and imaging models were evaluated in the overall cohort and stratified by IDH mutation status. RESULTS In the overall cohort, the clinical model was highly prognostic (C-index 0.74, p ~ 3.2×10⁻⁶). Adding imaging features improved performance (C-index 0.79) but was not significant (p ~ 0.31). Subset analysis revealed differential effects by IDH status. In IDH-mutant gliomas (n=24), neither clinical nor combined models were significant. In IDH wild-type glioblastoma (n=47), the clinical model achieved C-index 0.75 (p=0.04). The addition of imaging features significantly improved model performance (C-index 0.83, p=0.03), with selected imaging components contributing independent prognostic value after adjustment for clinical variables. Post-hoc analysis of radiomic feature importance (random forest ranking) revealed that top contributing features were drawn from both CT and MRI modalities, with most sourced from the whole brain region. CONCLUSIONS In post-surgical patients with IDH wild-type glioblastoma, image-derived features from post-operative imaging can significantly enhance prognostic modeling beyond known clinicopathologic factors. These preliminary findings support the potential role of radiomic biomarkers in refining risk stratification for this patient subgroup.

Abstract Background Adolescent and young adult (AYA) patients remain underrepresented in neuro-oncology research. Despite being the second most common primary brain tumor in this population, meningiomas have not been studied using age-specific molecular analyses. DNA methylation-based classification and prognostic tools have transformed meningioma care. This study aimed to evaluate the performance of these tools across age groups. Methods We analyzed 1,568 meningiomas with DNA methylation and clinical data, including 18 pediatric patients (<15 years), 195 AYA patients (15-39 years), and 1,355 adult patients (>39 years). Pediatric and AYA (P/AYA) tumors were combined and compared with adult tumors. The performance of established molecular classifiers and recurrence predictors, as well as differences in chromosomal copy number alterations were compared across age groups. Results While histologic grading was comparable between cohorts, P/AYA tumors displayed significantly fewer aggressive molecular groups and lower frequencies of chromosomal arm losses, including 1p, 6q, and 14q. The adult-trained recurrence predictor failed in the P/AYA population (AUC 0.57), despite similar score distributions. Retraining the model on an age-specific cohort using an identical analytic framework improved performance (AUC 0.79) and enabled effective stratification of progression-free survival (P = 0.00054). Importantly, 1p loss retained prognostic significance within the P/AYA group, supporting its clinical utility. Conclusions Molecular tools developed in adult-dominant cohorts do not generalize to younger patients due to both biological divergence and exclusion from model development. These findings underscore the need for age-specific molecular frameworks and highlight the imperative of including P/AYA populations in precision neuro-oncology research to ensure lifespan-equitable care.

Abstract Glioblastoma is the most common type of malignant primary brain tumor and a major cause of morbidity and mortality. In 2021, the World Health Organization updated the classification of Central Nervous System (CNS) tumors to restrict glioblastomas to isocitrate dehydrogenase-wildtype (IDHwt) tumors, improving understanding of the prognosis and optimal therapy for these tumors. This revision also enables more homogeneous populations of patients to be enrolled in clinical trials, facilitating the evaluation of novel therapies. In this updated consensus review from the Society for Neuro-Oncology (SNO) and the European Association of Neuro-Oncology (EANO), the current management of patients with glioblastoma is discussed. In addition, novel therapies such as immunotherapies, viral therapies, targeted molecular therapies, theranostics, and antibody–drug conjugates will be reviewed, as well as the current challenges and future directions for research.

Abstract Cerebrospinal fluid (CSF) has emerged as a valuable liquid biopsy source for glioma biomarker discovery and validation. CSF produced within the ventricles circulates through the subarachnoid space, where the composition of glioma-derived analytes is influenced by the proximity and anatomical location of sampling relative to tumor, in addition to underlying tumor biology. The substantial gradients observed between lumbar and intracranial CSF compartments for tumor-derived analytes underscore the importance of sampling site selection. Moreover, radiographic features, such as tumor-CSF contact and blood-brain barrier disruption, are critical covariates that may affect biomarker detection and the abundance of plasma-derived analytes in CSF, respectively. Longitudinal intracranial CSF sampling, enabled by access devices like Ommaya reservoirs, may offer a window into treatment response and disease progression, though variability in analyte yield, sample volumes, and the dynamic effects of surgical resection pose challenges. This review critically evaluates the anatomic, radiographic, and longitudinal factors, or “time-space continuum,” that impact glioma CSF biomarker abundance. Practical considerations for longitudinal CSF biobanking, including access device placement and collection, are also reviewed. Key takeaways and recommendations for CSF glioma biomarker discovery and validation are provided as a “hitchhiker’s guide” based on our collective experience, along with resources for investigators aiming to develop CSF biobanking at their institutions.

Abstract Background Malignant peripheral nerve sheath tumors (MPNSTs) can arise from atypical neurofibromas (ANF). Loss of the polycomb repressor complex 2 (PRC2) is a common event. Previous studies on PRC2-regulated genes in MPNST used genetic add-back experiments in highly aneuploid MPNST cell lines which may miss PRC2-regulated genes in NF1-mutant ANF-like precursor cells. A set of PRC2-regulated genes in human Schwann cells (SCs) has not been defined. We hypothesized that PRC2 loss has direct and indirect effects on gene expression resulting in MPNST, so we sought to identify PRC2-regulated genes in immortalized human Schwann cells (iHSCs). Methods We engineered NF1-deficient iHSCs with loss of function SUZ12 or EED mutations. RNA sequencing revealed 1327 differentially expressed genes to define PRC2-regulated genes. To investigate MPNST pathogenesis, we compared genes in iHSCs to consistent gene expression differences between ANF and MPNSTs. Chromatin immunoprecipitation sequencing was used to further define targets. Methylome and proteomic analyses were performed to further identify enriched pathways. Results We identified potential PRC2-regulated drivers of MPNST progression. Pathway analysis indicates many upregulated cancer-related pathways. We found transcriptional evidence for activated Notch and Sonic Hedgehog (SHH) signaling in PRC2-deficient iHSCs. Functional studies confirm that Notch signaling is active in MPNST cell lines, patient-derived xenografts, and transient cell models of PRC2 deficiency. A combination of MEK and γ-secretase inhibition shows synergy in MPNST cell lines. Conclusions We identified PRC2-regulated genes and potential drivers of MPNSTs. Our findings support the Notch pathway as a druggable target in MPNSTs. Our identification of PRC2-regulated genes and pathways could result in more novel therapeutic approaches.