Elisa Fazzari

Glioblastoma (GBM) is a devastating primary brain tumor with remarkable inter- and intra-tumoral heterogeneity. GBM cells assume a spectrum of neurodevelopmental-like phenotypes and co-opt normal neurophysiological processes, which include synaptic integration with their neuronal microenvironment. This is mediated by neuron-tumor synapses (NTS) that predominantly involve glutamatergic receptors, which drive calcium elevations that promote tumor proliferation and invasion. The exact relationship between synaptic signaling and tumor cell fate specification, however, remains largely unexplored. Here, we develop and leverage a synapse-optimized human organoid tumor transplantation (so-HOTT) model of GBM to decipher how glutamatergic signaling impacts GBM lineage trajectories. so-HOTT preserves patient tumor heterogeneity, features excitatory NTS, and enables clonal lineage tracing of tumor cells after NTS perturbations. Genetic and pharmacological inhibition of AMPA and kainate receptors in so-HOTT shifts tumor cell composition from neuronal fates toward progenitor-proximal astrocytic/mesenchymal states. This occurs through the attenuation of calcium signaling and reduced plasticity of malignant radial glia (RG)-like progenitors, a previously unrecognized target of NTS. Through the integration of inputs from the neuronal microenvironment into glutamatergic signaling, progenitor populations modulate their transcriptional programs and cell fate, ultimately shaping GBM tumor heterogeneity. Targeting synaptic input may thus constrain the heterogeneity that fuels GBM adaptation and therapeutic escape.

Extensive molecular profiling has revealed profound heterogeneity in glioblastoma (GBM), yet how cellular lineages organize over time to govern tumor propagation and therapeutic response remains poorly understood. Existing single-cell approaches define transcriptional states but provide limited insight into how clonal dynamics shape functional tumor behavior. Here, we integrate high-complexity combinatorial DNA barcoding with single-cell transcriptomics in direct-from-patient IDH1-wild-type GBM, enabling lineage-resolved mapping of progenitor organization in a human microenvironmental context. Across 235,155 malignant cells from nine tumors, clonal relationships form reproducible lineage tracks in which distinct progenitor populations give rise to specific differentiated cell types, revealing that tumor growth is sustained by multiple non-redundant progenitors rather than a single dominant population. These progenitors exhibit distinct propensities for self-renewal, fate restriction, and cross-compartment interactions, collectively accounting for the full spectrum of tumor states. Using this lineage-resolved framework, we identify complementary drug targets in distinct progenitor compartments and demonstrate that hierarchy-informed combination therapies disrupt progenitor-progenitor interactions and reshape lineage output. These findings move beyond descriptive heterogeneity to define functional logic underlying GBM propagation and establish a generalizable framework for rational, cell type-specific combinatorial therapies.

A major obstacle to identifying effective therapies for the aggressive brain tumor glioblastoma is the lack of human-specific, immunocompetent models that reflect the human tumor microenvironment. To address this, we developed the immune-Human Organoid Tumor Transplantation (iHOTT) model. This is an autologous co-culture platform that integrates patient-derived tumor cells and matched peripheral blood mononuclear cells (PBMCs) within human cortical organoids, enabling the study of the patient-specific immune response to the tumor and tumor-immune interactions. This platform preserves tumor and immune populations, immune signaling, and cell-cell interactions observed in patient tumors. Treatment of iHOTT with pembrolizumab, a checkpoint inhibitor, mirrored cell type shifts and cell interactions observed in patients. TCR sequencing further revealed pembrolizumab-driven expansion of stem-like CD4-T-cell clonotypes exhibiting patient-specific repertoires. These findings establish iHOTT as a physiologically relevant platform for exploring autologous tumor–immune interactions and underscore the critical need for antigen-targeted strategies to enhance immunotherapy in glioblastoma.

Glioblastoma (GBM) exhibits developmental programs and marked cellular heterogeneity, yet how these features are organized into connected lineage hierarchies remains unclear. Here we identify a rare tumor-intrinsic population, termed the neurovascular progenitor (NVP), that occupies an intermediate position between the major GBM organizational axes. NVP cells co-express neural progenitor and perivascular transcriptional features, are consistently detected across independent patient cohorts, retain canonical GBM copy-number alterations, and localize in situ in both vessel-associated and parenchymal niches. Using direct-from-patient lineage tracing in a human organoid tumor transplantation system, we show that individual NVP cells clonally generate both neural-like and mesenchymal/vascular-like malignant progeny, providing a concrete lineage link between states that are commonly considered mutually exclusive. Despite comprising ∼1% of tumor cells, NVP-derived lineages account for a majority of observed tumor cell types and disproportionately contribute to cycling compartments. Orthogonally, ablation of NVP-associated programs in an in vivo GBM model remodels tumor composition, elicits compensatory progenitor states, and significantly prolongs survival. Together, these findings position NVP as a fate-restricted yet highly influential lineage intermediate that serves as a functional bridge and organizational nexus within GBM hierarchies, linking population-level lineage architecture to the behavior of a specific progenitor cell type.

Glioblastoma, the most aggressive and deadly form of primary brain cancer, is driven by both intrinsic cellular properties and external factors from the tumor microenvironment. Here, we leverage our novel human organoid tumor transplantation (HOTT) system to explore how extrinsic cues modulate glioblastoma cell type specification, heterogeneity, and migration. We show that HOTT recapitulates the core features of major patient tumor cell types and key aspects of peritumor cell types, while providing a human microenvironment that uniquely enables perturbations in both the patient tumor and its microenvironment. Our exploration of patient tumor – microenvironmental interactions in HOTT highlighted PTPRZ1, a receptor tyrosine phosphatase implicated in tumor migration, as a key player in intercellular communication. We observed that tumor knockdown of PTPRZ1 recapitulated previously described roles in migration and maintaining progenitor identity. Unexpectedly, environmental PTPRZ1 knockdown drove opposite migration and cell fate changes in the tumor, even when the tumor was not manipulated. This previously undiscovered mode of tumor-microenvironmental communication highlights the need to study human glioblastoma in the context of a human microenvironment such as HOTT.

Human brain development requires the generation of hundreds of diverse cell types, a process targeted by recent single-cell transcriptomic profiling efforts. Through a meta-analysis of seven of these published datasets, we have generated 225 meta-modules – gene co-expression networks that can describe mechanisms underlying cortical development. Several meta-modules have potential roles in both establishing and refining cortical cell type identities, and we validated their spatiotemporal expression in primary human cortical tissues. These include meta-module 20, associated with FEZF2+ deep layer neurons. Half of meta-module 20 genes are putative FEZF2 targets, including TSHZ3, a transcription factor associated with neurodevelopmental disorders. Human cortical organoid experiments validated that both factors are necessary for deep layer neuron specification. Importantly, subtle manipulations of these factors drive slight changes in meta-module activity that cascade into strong differences in cell fate – demonstrating how of our meta-atlas can engender further mechanistic analyses of cortical fate specification.