Glioma pathophysiology is robustly regulated by interactions with neurons. Key to these interactions is the role of neuroligin-3 (NLGN3), a synaptic adhesion molecule shed in response to neuronal activity that functions as a paracrine factor crucial for glioma growth. Here we elucidate the mechanistic pathway whereby shed NLGN3 interacts with glioma and their normal glial counterparts. NLGN3 binds to chondroitin sulfate proteoglycan 4 (CSPG4, also known as NG2) on both glioma and healthy oligodendrocyte precursor cells (OPCs), facilitating CSPG4 shedding by ADAM10. NLGN3-CSPG4 interactions alter membrane tension, thereby activating mechanotransducers, primarily PIEZO1, leading to membrane depolarization and subsequent ADAM10-mediated CSPG4 shedding. The NLGN3-CSPG4-PIEZO1 pathway maintains OPCs in an undifferentiated, stem-like state and promotes glioma proliferation, underscoring its dual roles in healthy and malignant contexts.

Glioblastoma is a highly aggressive primary brain tumor in adults with limited treatment options. Although there has been continuing interest in employing the immune system to combat this disease, monoclonal antibody therapies, such as checkpoint modulating agents, have yet to demonstrate clinically transformative outcomes. Conversely, adoptive cell therapies including chimeric antigen receptor (CAR)-T cell therapies have shown early indications of efficacy in select GBM patients. In a recent report, a first-in-human trial of CARv3-TEAM-E T cells engineered to target the epidermal growth factor receptor variant III tumor-specific antigen (EGFRvIII) as well as wild-type EGFR through secretion of a T-cell-engaging antibody molecule (TEAM) demonstrated preliminary evidence of CAR activity. We now report the findings of single-cell RNA sequencing (scRNA-seq) of cells isolated from the cerebrospinal fluid (CSF) of CARv3-TEAM-E treated GBM patients. Longitudinal CSF sampling in these patients was performed through an Ommaya reservoir. Using scRNA-seq, we demonstrate the ability to detect transduced CARv3-TEAM-E expressing T cells in the CSF from these patients. In this cohort, we assess the expression of cytotoxic effector gene expression programs over the course of longitudinal sampling. Unsupervised analysis shows promising early evidence of the expression of these cytotoxic effector gene expression programs. We further compare expression signatures in the CSF-sampled T cells to those present in the CARv3-TEAM-E infusion products. In summary, our results elucidate the longitudinal dynamics of T cell gene expression programs in the setting of this first-in-human trial of CARv3-TEAM-E T cells in GBM patients.

Transcription factors are frequent cancer driver genes, exhibiting noted specificity based on the precise cell of origin. We demonstrate that ZIC1 exhibits loss-of-function (LOF) somatic events in group 4 (G4) medulloblastoma through recurrent point mutations, subchromosomal deletions and mono-allelic epigenetic repression (60% of G4 medulloblastoma). In contrast, highly similar SHH medulloblastoma exhibits distinct and diametrically opposed gain-of-function mutations and copy number gains (20% of SHH medulloblastoma). Overexpression of ZIC1 suppresses the growth of group 3 medulloblastoma models, whereas it promotes the proliferation of SHH medulloblastoma precursor cells. SHH medulloblastoma ZIC1 mutants show increased activity versus wild-type ZIC1, whereas G4 medulloblastoma ZIC1 mutants exhibit LOF phenotypes. Distinct ZIC1 mutations affect cells of the rhombic lip in diametrically opposed ways, suggesting that ZIC1 is a critical developmental transcriptional regulator in both the normal and transformed rhombic lip and identifying ZIC1 as an exquisitely context-dependent driver gene in medulloblastoma.

Writing in Neuron, Zhang et al. identify a subpopulation of glioblastoma cells from patient tumor samples with progenitor-like features that expresses the potassium ion channel KCND2.1 In mouse and organoid models, these cells enhance neural activity at the glioma-neural interface.

Glioblastoma (GBM) is a devastating primary brain tumor of adults with few treatment options. While there has been long-standing interest in engaging the immune system to combat this disease, monoclonal antibody therapies – including checkpoint modulating agents – have yet to result in widespread clinically meaningful outcomes. By contrast, adoptive cell therapies including chimeric antigen receptor (CAR)-T cell therapies have demonstrated promising early indications of efficacy in select GBM patients. However, heterogeneity of target molecule expression remains a significant barrier to long-term disease control. In a recent report, a first-in-human trial of CARv3-TEAM-E T cells engineered to target the epidermal growth factor receptor variant III tumor-specific antigen (EGFRvIII) as well as wild-type EGFR through secretion of a T-cell-engaging antibody molecule (TEAM) showed preliminary evidence of CAR activity. We now report the results of single cell RNA-sequencing (scRNA seq) of lymphocytes isolated from the cerebrospinal fluid (CSF) of CARv3-TEAM-E treated GBM patients. Longitudinal CSF sampling in these patients was performed via an Ommaya reservoir. Sampled cells were interrogated by scRNA Seq. We demonstrate the ability to detect transduced CARv3-TEAM-E expressing T cells in the CSF from these patients. Unsupervised analysis revealed evidence of the expression of cytotoxic effector gene expression programs, which we explore over the course of longitudinal sampling in this initial cohort. Additionally, we compare expression signatures in the CSF-sampled T cells to those present in the CARv3-TEAM-E infusion products. Taken together, our findings offer insight into the longitudinal dynamics of T cell gene expression programs in the setting of this first-in-human trial of CARv3-TEAM-E T cells in GBM patients. Christopher Mount, Demi Gerovasilis, Sophia Kovatsis, Jun Zhong, Md Raihan Chowdhury, Maxx King, William T. Curry, Elizabeth R. Gerstner, Kathleen M. Gallagher, Bryan D. Choi, Mario Suva, Marcela V. Maus. Single cell RNA sequencing of cerebrospinal fluid lymphocytes in CARv3 TEAM E treated glioblastoma patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_2):Abstract nr LB358.

Despite decades of concerted research and clinical efforts, patient outcomes in glioblastoma (GBM) remain dismal. The emergence of chimeric antigen receptor (CAR) T cell therapies offer great promise as immunotherapies in this disease. To date, efforts to utilize CAR T cell therapies in GBM have resulted in isolated reports of impressive efficacy in individual patients, but these have yet to generalize to larger trial populations. It is thought that inter and intra patient tumor heterogeneity contributes to the variability of these responses. Single cell RNA sequencing has demonstrated that this heterogeneity can be organized along axes representing cellular states. The ability of existing CAR T cell therapies to target these cell states is unclear, and the dynamics of this heterogeneity under the selective pressure of therapy is uncertain. We present the development of a panel of tool CAR T cells targeting cell state associated surface markers with the objective of selective targeting of these cellular states in GBM. We show that these CAR T cells have target specific activity against GBM model systems representing these cellular states. Using single cell RNA sequencing, we model the dynamics of these treatment effects, with the aim of understanding the mechanisms of GBM dynamics in response to the selective pressure of CAR T cell therapies. We demonstrate global shifts in the transcriptional landscape at the single cell level and identify state dependent and state independent effects. We anticipate that these cell state targeted CAR T cells have great potential to elucidate mechanisms of GBM adaptation to the selective pressure of CAR T cell activity and will inform foundational biology necessary for rational design of multi target constructs for therapeutic application. Christopher Mount, Emily Boxer, Sophia Kovatsis, Demi Gerovasilis, Sydney Dumont, Jun Zhong, Jack Lu, Itay Tirosh, Mario Suva. Cell state targeting CAR T cell therapies for glioblastoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 4827.

Poorly differentiated chordoma (PDC) is an aggressive subtype of chordoma characterized by SMARCB1 (INI1) loss and a dismal prognosis. It typically involves the axial skeleton, most commonly the skull base and the cervical spine. To our knowledge, only 5 cases of extraaxial PDC (EAPDC) have been reported, and the natural history of these tumors is not fully understood. We studied 6 cases of EAPDC, with the goal of better understanding these exceptionally rare tumors. The tumors occurred in 4 women and 2 men, ranging from 37 to 68 years of age (median, 57.5 years) and involved or originated in the left knee joint (3 cases), right knee joint (2 cases), and right wrist (1 case). Grossly, all were solid and lobulated, with areas of necrosis. Histologically, the tumors were identical to axial PDC, with sheets and lobules of overtly malignant-appearing epithelioid-to-rhabdoid cells with prominent nucleoli. Mitotic activity and necrosis were present. By immunohistochemistry, all cases expressed keratins and brachyury and were SMARCB1 deficient. Molecular genetic analysis identified SMARCB1 loss-of-function alterations in 4 of the tested cases, including mutations (2 cases) and copy number loss (2 cases). DNA methylation profiling of 4 cases of EAPDC showed clustering with axial PDC. Clinical follow-up (6 patients; median, 11.5 months; range, 1-26 months) showed 4 patients to have received transfemoral amputation and 1 extraarticular resection. None received neoadjuvant radiotherapy; 1 received neoadjuvant chemotherapy and 1 adjuvant chemotherapy/ immunotherapy. Local recurrences were seen in 2 patients at 7 and 8 months; 3 patients developed metastases 7-11 months after surgery. Two patients were alive with metastatic disease (at 7 and 13 months), 1 died of disease (20 months), and 3 were disease free (1-26 months). We conclude that EAPDC are aggressive malignancies with an unusual predilection for the knee joint and unknown pathogenesis. (c) 2024 United States & Canadian Academy of Pathology. Published by Elsevier Inc. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

The locus coeruleus (LC) has been identified as a site that develops phosphorylated tau pathology earlier than cerebral cortex. We present data using high-resolution postmortem MRI and validated tau histopathology in controls and the earliest Braak and Braak (BB) stages (BBI-BBII) in LC. The high-resolution ex vivo MRI provides a 3D volume (quantitative), while the histology reveals tau specificity and severity burden (semi-quantitative). We mapped our highly regionally specific LC data onto high-resolution 3D MRI reconstructions of the same samples used in histology (n = 11). We noted significant structural subatrophy between BB 0 and II (30.0% smaller volumes, p = 0.0381), a trend which primarily affected the rostral-most LC (49.2% smaller average volume, p = 0.0381). We show histopathology data on both the LC and neighboring dorsal raphe caudal (DRc), which were assessed at multiple rostrocaudal levels and mapped with highly sensitive tau severity spatial matrices. We observed significant LC tau accumulation between BB I and II (37.6% increase, p < 0.0001), which may reflect pathology change prior to presumptive cognitive impairment at BB III. Tau pathology was most severe in the middle portion of the LC (11.3% greater compared to rostral LC, p = 0.0289) when including BB III. We noted a significant rostrocaudal gradient of DRc tau severity (58.2% decrease between rostral and caudal DRc, p < 0.0001), suggesting selective regional vulnerabilities of both nuclei. Our study represents a rigorous approach to investigating LC and DRc pathology, having multiple histology sections per sublevel and high-resolution MRI to measure the whole LC, without missing slices in a histological only approach. Taken together, our findings provide novel validated data that demonstrate the tau pathology occurring in the LC and DRc during preclinical AD stages, and alongside spatial reconstructions that will serve as valuable references for in vivo LC imaging.

Advances in image registration and machine learning have recently enabled volumetric analysis of postmortem brain tissue from conventional photographs of coronal slabs, which are routinely collected in brain banks and neuropathology laboratories worldwide. One caveat of this methodology is the requirement of segmentation of the tissue from photographs, which currently requires costly manual intervention. In this article, we present a deep learning model to automate this process. The automatic segmentation tool relies on a U-Net architecture that was trained with a combination of (i)1,414 manually segmented images of both fixed and fresh tissue, from specimens with varying diagnoses, photographed at two different sites; and (ii) 2,000 synthetic images with randomized contrast and corresponding masks generated from MRI scans for improved generalizability to unseen photographic setups. Automated model predictions on a subset of photographs not seen in training were analyzed to estimate performance compared to manual labels – including both inter- and intra-rater variability. Our model achieved a median Dice score over 0.98, mean surface distance under 0.4 mm, and 95% Hausdorff distance under 1.60 mm, which approaches inter-/intra-rater levels. Our tool is publicly available at surfer.nmr.mgh.harvard.edu/fswiki/PhotoTools.

Glioma pathophysiology is robustly regulated by interactions with neurons. Key to these interactions is the role of neuroligin-3 (NLGN3), a synaptic adhesion molecule shed in response to neuronal activity1-5 that functions as a paracrine factor crucial for glioma growth. Here, we elucidate the mechanistic pathway whereby shed NLGN3 interacts with glioma and their normal glial counterpart. NLGN3 interacts with Chondroitin Sulfate Proteoglycan 4 (CSPG4) on both glioma and healthy oligodendrocyte precursor cells (OPCs)6-9, facilitating CSPG4 shedding by ADAM10. NLGN3-CSPG4 interactions and consequent shedding alter membrane tension, thereby activating PIEZO1 mechanosensitive channels and causing membrane depolarization. The NLGN3-CSPG4-PIEZO1 axis maintains OPCs in an undifferentiated, stem-like state and promotes glioma proliferation, underscoring important functional roles for the NLGN3-CSPG4-PIEZO1 axis in both healthy and malignant glial precursors.

The diffusely infiltrative growth pattern of glioblastoma (GBM) is a major obstacle to effective therapy. A complex network of tumor intrinsic and microenvironmental features interact to shape the landscape of this growth pattern, but deciphering the molecular substrates of these interactions remains challenging. While single cell RNA sequencing has reshaped our understanding of GBM heterogeneity and uncovered cellular phenotypes of many cellular populations in these tumors, the spatial context of this heterogeneity could not be assessed with earlier technologies. By contrast, spatial transcriptomics platforms have the potential to map this transcriptome-wide information in situ. We recently reported the use of this technology to develop a layered model of spatial architecture in GBM and showed that hypoxia represents a key organizing feature of GBM heterogeneity. However, the limited resolution of this platform precluded assessment of whole-transcriptome features at truly cellular resolution. In this work, we now present our findings utilizing the next generation of this technology and showcase the ability to map whole transcriptome data in spatial context at true single cell resolution in primary patient GBM samples. We explore heterogeneous GBM phenotypes in the context of GBM spatial microarchitecture. Using copy number inference, we demonstrate the ability to assess clonal heterogeneity in situ and map the pattern of invasion of unique GBM clones. Finally, we discuss our progress to discover organizing principles underlying distinct patterns of GBM cell invasion. Taken together, we demonstrate the potential of single cell spatial transcriptomics to decipher the molecular heterogeneity of GBM microenvironment in situ and uncover the biology driving the aggressive infiltrative growth of this disease.

HGAP was first described in 2018. However, prognostic factors of clinical outcome are not well-understood secondary to recent recognition and paucity of data for this diagnostic entity. Here, we utilize a large HGAP cohort (n=252) to survey the genomic landscape and explore prognostic correlates. NIH DNA methylation profiling was performed to identify HGAP cases and combined with publicly available datasets. Evaluable molecular markers, patient demographics, tumor location, imaging reports, treatment history (e.g. temozolomide, targeted therapy, surgery, and radiation) and survival data were assessed. Kaplan-Meier analysis was performed. The cohort included cases that matched to HGAP on the NIH/Bethesda methylation classifier at ≥0.9 confidence score. More males (60%) than females (40%) were in the cohort. Twenty-one percent of patients were known to have neurofibromatosis type 1. Posterior fossa location was predominant (57%, n=131). Common genomic findings included alterations in NF1 (58%), ATRX (50%), FGFR1 (15%), TP53 (9%) and PIK3CA (8%). CDKN2A/B homozygous deletion was identified in 80% of cases. MGMT promoter methylation was found in 55% of cases. Median progression free survival (mPFS) was 24 months (mo), and median overall survival (mOS) was 108 mo. Central histological review revealed 67% of cases were high-grade. High-grade histopathology (brisk mitotic activity) was not associated with survival. Immunohistochemical ATRX loss was associated with shorter mPFS (18.0 mo vs 35.5 mo, p=0.04) and mOS (93.7 mo vs not reached, p=0.04). The presence of ATRX genetic alterations was associated with shorter mOS (30 mo vs not reached, p=0.008). CDKN2A/B homozygous deletion and MGMT status were not correlative with patient outcome. HGAP is a glial neoplasm that shows frequent tumor recurrence. The majority of HGAP cases are high-grade. Our analysis of correlates with patient outcome suggests immunohistochemical ATRX loss and molecular ATRX alteration may be important poor prognostic markers.

Background Fibroblast growth factor receptor (FGFR) alterations are potential oncogenic drivers that occur in a subset of patients with gliomas, but the natural history of these tumors is not clearly defined. An understanding of outcomes of FGFR-driven glioma has implications for targeted drug development for FGFR inhibitors, which are approved for other cancers. Methods We performed a retrospective cohort study of patients with gliomas who harbored FGFR alterations including fusions, single nucleotide variant (SNV), and copy number variant (CNV) alterations seen at MGH between 2003 and 2023. The electronic medical record was searched to identify additional molecular data, treatment, and MRI scans. Kaplan-Meier analysis was used to assess progression free and overall survival (OS). Results Thirty-one patients with glioblastoma (GBM), diffuse astrocytoma, and glioneuronal tumors with FGFR alterations were identified: 17 with FGFR fusions, 10 with SNV, and 4 with CNV. FGFR3-TACC3 was the most common fusion in patients with GBM or diffuse astrocytoma. Median OS in patients with GBM was 2.75 years despite 55% of tumors having an unmethylated MGMT promoter. There were no clearly co-occurring mutations with an FGFR alteration. Six of 7 patients who underwent subsequent surgery lost the original FGFR alteration. No patient received an FGFR inhibitor. Conclusions While FGFR alterations are rare in glioma, patients with FGFR altered GBM may have prolonged survival which has implications for clinical trial design. We found loss of FGFR alteration at the time of subsequent surgery, raising concern for the therapeutic potential of FGFR targeting agents in recurrent gliomas.

Isocitrate dehydrogenase (IDH)-mutant glioma is the most common primary brain tumor diagnosed in patients younger than 50 years old. While IDH inhibitors have shown promise for patients with low-grade disease, patients with higher-grade tumors still have limited treatment options and face poor clinical outcomes. Chimeric antigen receptor (CAR)-T cell therapies have demonstrated potential in other molecularly-distinct gliomas, but directing this immunotherapeutic strategy towards IDH-mutant glioma remains largely unexplored. Our prior work using single-cell RNA sequencing (scRNAseq) has defined a hierarchical model of transcriptional states in IDH-mutant glioma, including a central ‘stem-like’ population that is enriched for cycling cells and drives overall disease progression. We hypothesize that targeting this stem-like population in IDH-mutant glioma using CAR-T cell therapy will enable effective control of these tumors. To design CAR constructs that target this population, we performed in silico screening of scRNAseq data from IDH-mutant glioma patient samples to identify highly expressed genes associated with the stem-like surfaceome. Candidate targets with publicly available single-chain variable fragments (scFvs) were engineered into tool 2nd generation CARs and screened for antitumor activity against patient-derived IDH-mutant glioma models. Coculture assays demonstrated in vitro cytokine production and antitumor cytotoxicity against these patient-derived models. Additionally, we demonstrate the ability of these CAR-T cells to clear an aggressive patient-derived orthotopic IDH-mutant glioma xenograft model and significantly extend survival. In summary, our findings validate the potential of a cell state-directed strategy to identify CAR-T cell targets in IDH-mutant gliomas that may inform future translational efforts.

INTRODUCTION: Chimeric antigen receptor (CAR) T cells represent a promising approach to cancer and have proven efficacy against hematological malignancies, for which they have become the standard of care. However, the use of CAR T cells in solid tumors has been limited. METHODS: This is a nonrandomized, open-label, single-site Phase I clinical trial. Three patients with EGFRvIII-positive recurrent glioblastoma were enrolled in a safety run-in cohort. Patients were treated with 10 million CARv3-TEAM-E T cells and monitored for toxic effects. Cerebrospinal fluid (CSF) and blood were sampled longitudinally and subjected to correlative analyses. RESULTS: No dose-limiting toxic effects were noted. Radiographic tumor regression occurred in all three patients within days after treatment, but this response was transient in two of the patients. Tumor regression correlated with decreased detection of antigen-specific RNA derived from extracellular vesicles (EVs) in both CSF and peripheral blood. CONCLUSIONS: Early data suggest safety and anti-tumor activity of CARv3-TEAM-E T cells in recurrent glioblastoma. EV-based liquid biopsy may assist in monitoring response to cell therapy. Ongoing enrollment has been modified to enhance durability using lymphodepletive chemotherapy. Additional arms will evaluate this approach in the setting of EGFRvIII-negative tumors and newly-diagnosed disease.

Despite decades of concerted research and clinical efforts, patient outcomes in glioblastoma (GBM) remain dismal. While chimeric antigen receptor (CAR)- T cell therapies in GBM have produced individual reports of efficacy, they have yet to meaningfully alter standard of care. Tumor heterogeneity, both inter- and intra-patient, likely contributes to these variable responses. Single-cell RNA sequencing (scRNA-seq) conducted in our lab has revealed that this transcriptional heterogeneity is organized along distinct cellular state axes. We sought to overcome this heterogeneity by designing CAR-T cells directed against these cellular states. We hypothesize that CAR-T cells targeting a defined GBM cell state will lead to a depletion of those populations. We report the development of a panel of cell state-targeting CAR-T cell candidates engineered for target-specific activity in vitro. To model the three-dimensional tumor environment, we developed patient-derived GBM organoids that recapitulate the cell state landscape seen in primary tumors. We treated GBM organoids with these cell state-targeting CAR-T cells and assessed cell states by scRNA seq. Unexpectedly, we observed convergent transcriptome shifts to a new cell state with overlapping features of mesenchymal (MES) and interferon-induced signatures. To target this dynamic, we designed a novel CAR-T cell directed against this population and observed feedforward target enrichment in response to therapy. We anticipate that our findings will inform rational design of translational CAR-T cell candidates for GBM patients.