A human iPSC-based neural spheroid platform for modeling glioblastoma infiltration using high-content imaging

A human iPSC-based neural spheroid platform for modeling glioblastoma infiltration using high-content imaging | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article A human iPSC-based neural spheroid platform for modeling glioblastoma infiltration using high-content imaging Victoria S K Tsang, Federica Riccio, Hannah Nudds, Jason Coombes, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7536545/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 13 Dec, 2025 Read the published version in Scientific Reports → Version 1 posted 12 You are reading this latest preprint version Abstract Glioblastoma is the most aggressive adult brain tumor, characterized by resistance to therapy and high recurrence due to diffuse infiltration. To mimic glioblastoma migration, we developed a physiologically relevant co-culture model, combining patient-derived glioblastoma cell lines with cortical-like neural spheroids differentiated from human induced pluripotent stem cells. Using high-content imaging, we demonstrate that GBM1 and GBM20 cell lines migrate directionally along axons toward neural spheroids in live imaging assays and infiltrate spheroids extensively in endpoint assays, unlike non-cancerous neural stem cells. A proof-of-principle drug screen identified PF-573228 (FAK inhibitor) and Motixafortide (CXCR4 inhibitor) as potent suppressors of GBM1 and GBM20 infiltration, respectively. Bulk RNA sequencing revealed gene expression profiles correlating with invasive behavior and drug sensitivity. This platform offers a valuable model for studying glioblastoma infiltration along axons and provides proof-of-principle that migration can serve as a measurable and actionable phenotype to screen therapeutic vulnerabilities in glioblastoma. Biological sciences/Cancer Biological sciences/Cell biology Biological sciences/Neuroscience Glioblastoma cancer migration induced-pluripotent stem cells neural spheroid high-content imaging stem cell modelling Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Glioblastoma (GB) is the most common and aggressive form of primary brain tumor in adults, with a median overall survival of just 12 to 15 months following diagnosis and standard-of-care treatment with temozolomide 1 – 8 . GB is defined by its IDH1/IDH2-wildtype status and classified as a World Health Organization (WHO) grade IV tumor, reflecting its high malignancy 3 , 5 . A major barrier to effective treatment is the pronounced inter- and intra-tumoral heterogeneity in genetic mutations and cellular phenotypes, which complicates therapeutic targeting and drives drug resistance 9 – 11 . Moreover, the highly infiltrative nature of GB precludes complete surgical resection and temozolomide treatment, promoting tumor recurrence 6 , 12 , 13 . A key contributor to the aggressive invasiveness of GB is a subpopulation of neoplastic cells called the GB stem-like cells (GSCs) 14 – 17 . GSCs exhibit hallmark features of stemness, including self-renewal and multipotency, and are found at the tumor core and in infiltrative margins, where they play a key role in tumor recurrence and therapy resistance 17 – 20 . Most importantly, GSCs are highly invasive and utilize conserved mechanisms of invasion, emphasizing their critical role in disease progression. Their invasive capacity is mediated by intrinsic processes, including cytoskeletal remodeling and upregulation of surface receptors that regulate adhesion and motility 15 . These mechanisms facilitate GSC migration along two primary routes, the vascular endothelium and axonal tracts within the white matter of the brain, although the mechanisms underlying the route selection remain unclear 14 , 15 , 21 . An increased understanding of the prognostic significance of tumor invasion has driven the development of advanced in vitro models that more accurately recapitulate GB complexity for drug screening purposes 22 . Among these, patient-derived GSC lines have emerged as a valuable research tool. These lines faithfully preserve the stem cell-like properties, genetic landscape, and transcriptional profile of the parental tumor, even after extended passaging, thereby maintaining key features of the in vivo tumor phenotype 23 , 24 . Particularly, patient-specific GSCs have been used to generate physiologically relevant model systems, including glioma spheroids and GB cerebral organoids that better represent the complexity of GB and the tumor microenvironment 25 . High-content imaging combines automated microscopy with quantitative single-cell analysis, allowing detailed, large-scale assessment of cellular phenotypes 26 , 27 . In cancer research, high-content imaging enables dynamic studies of cell behaviors such as migration, especially when using label-free modalities like phase contrast to avoid phototoxicity during long-term live-cell assays 28 , 29 . These advancements, alongside computational and multi-omics approaches, have significantly enhanced drug screening and target discovery in complex tumor models like GB. In this study, we present a novel in vitro co-culture system that combines patient-derived GB cells and axons extending from neural spheroids derived from human iPSCs. This model allows for the interaction between GB cells and neuronal axons, providing a more physiologically relevant microenvironment for the study of GB cells migration and infiltration, which are critical processes driving GB recurrence. Using high-content imaging, we reveal significant differences in the migration patterns of two different GB cell lines and provide insights into the mechanisms underlying GB cell line-specific infiltration strategies, with potential implications for anti-infiltrative therapies. Additionally, we correlate migration phenotypes with drug responses identified through a screening of relevant inhibitors. This drug screen provides proof-of-principle that migration is a measurable phenotype with potential for guiding precision therapies against GB infiltration. Results Patient-derived GB cells migrate along axons toward the neural spheroid in vitro To generate an in vitro model suitable for the characterization of infiltration of GB cells and amenable to drug-screening studies, we combined the culture of patient-derived GB cell lines with human iPSC-derived neural spheroids with radially extending axons. Wildtype human iPSCs are differentiated into excitatory cortical-like neurons using NGN2-mediated forward programming differentiation 30 . Following the induction phase, early neuronal progenitors are placed into non-adherent Pluronic acid-coated V-bottom plates where they self-aggregate into individual spheroids of reproducible size and shape (Figure 1A). Then, spheroids are transferred onto laminin-coated plates where they adhere to the culture substrate and extend long, radially organized neurites (Figure 1B). Immunofluorescence staining performed on the spheroids revealed positive expression of the axonal marker βIII-tubulin (Tubb3) in the projecting neurites (Figure 1B), validating the generation of a neuronal substrate suitable for GB co-culture assays. Having established a co-culture of neural spheroids with emerging axons, we next examined how two different patient-derived GB cell lines, GBM1 and GBM20, behave in this axon-rich microenvironment. GBM1, first described by Wurdak et al., was derived from a primary, treatment-naïve tumor 9,31 , while GBM20, reported by Polson et al., was obtained from a recurrent tumor previously treated with chemoradiotherapy and the IMA950 peptide vaccine 31 . A human neural stem cell line called NS17 was examined in parallel as a non-tumorigenic control with features of self-renewal and multipotency 32 . Live-cell imaging of the GB and NS17 lines revealed that all three cell lines aligned with and migrated along neuronal axons, indicating a direct interaction between the cells and the axons (Figure 2 and supplementary videos). To determine whether the direction of cell migration along axons was random or biased toward the neural spheroid, we quantified the directionality of individual cell tracks (Figure 3A). In the absence of axons, migration was isotropic, as shown by the uniform distribution of the polar histograms. In contrast, cells plated on axons exhibited directional migration along axons, with a clear enrichment of movement toward the neural spheroid positioned on the right side of the imaging field. This directional bias was evident across all cell lines, including the non-cancerous NS17 control (10% of cells moving at an angle 0 o ), and was most pronounced in GBM20 (12% of cells moving at an angle 0 o ). A significant increase in instantaneous velocity was observed when all three cell lines were cultured on axons compared with axon-free conditions (****, p < 0.0001) (Figure 3B). Interestingly, a change in overall mean speed reached significance only for GBM1 cells on axons, with no such effect on the other cell lines. Although migration directionality and velocity were strongly influenced by the presence of neuronal axons, the confinement ratio remained unchanged between axon-present and axon-absent conditions for all cell lines (Figure 3B). These findings suggest that axons function as a permissive, orienting substrate that enhances guidance cues, particularly for GBM1 cells. Moreover, the increased variability in directionality and velocity on axons highlights the heterogeneity of cell responses and supports the need for further investigation into temporal dynamics and GB subtype-specific migration mechanisms. GB cells exhibit cell line-specific infiltration in neural spheroid co-culture We hypothesized that cells migrating toward the neural spheroid would subsequently infiltrate it. To assess the infiltration potential of patient-derived GB cells in comparison to non-tumorigenic neural stem cells, GBM1, GBM20, and NS17 cells were labelled with a permeable cytoplasmic dye, added to neural spheroids in culture, and imaged every day for 5 days. A customized image analysis pipeline was applied to endpoint images acquired from Day 1 to 5 to quantify the number of infiltrating cells across conditions (Figure 4A). All three cell lines showed a progressive increase in the number of infiltrated cells over time (Figure 4B). On Day 5, NS17 cells reached an average infiltration of 6.5 ± 0.9 cells per spheroid, whereas GBM1 and GBM20 demonstrated significantly greater infiltration, with 17.2 ± 0.6 and 11.9 ± 1.6 cells, respectively. Statistically significant differences emerged by Day 2, at which point GBM1 cells had infiltrated the spheroids at a higher rate (5.5 ± 1.0 cells) compared to NS17 (1.6 ± 0.4 cells). Normal distribution analysis of cell infiltration values revealed progressive divergence between patient-derived GB lines and the non-tumorigenic control (Figure 4C). On Day 3, data distribution showed a clear separation between NS17 and GBM1/GBM20, with a calculated Z-factor of 0.701, indicating excellent assay robustness and strong discriminatory power suitable for screening applications. While the Z-factor decreased to 0.218 by Day 4, reflecting increased variability or overlap in infiltration measurements, it improved again on Day 5, reaching 0.575, consistent with a reliable and moderately robust assay window (See also Figure S1A). These data suggest that Day 3 represents the optimal time point for distinguishing differential infiltration phenotypes in this model, with potential applicability for high-throughput anti-invasion compound screening. GB and neural spheroid co-cultures enable identification of tumor-specific infiltration inhibitors Having developed a quantitative pipeline to measure GB cell migration along axons and infiltration into the neural spheroid core, we next used this co-culture platform for a pilot drug screen. To investigate mechanisms regulating GB infiltration, a panel of 18 inhibitors (Table 1) targeting cytoskeletal dynamics, kinase signaling, and axonal guidance pathways that mediate glioma invasion was selected based on literature and clinical relevance 15,17,33 . This included established cytoskeletal disruptors such as inhibitors to actin polymerization (Latrunculin B), myosin II ATPase (Blebbistatin), and ROCK1/2 (Y-27632), commonly used to reduce cell motility, along with targeted canonical GB signaling pathways involved in cell migration and invasion, including inhibitors to TGF-β receptor (SB 431542), PI3K/Akt (API-2), Protein Kinase C (GF 109203X), and integrins (Cilengitide), pathways often targeted in clinical evaluation 34–38 . Additional targets included non-canonical regulators such as hedgehog (HPI-1) and Wnt (IWP-2) pathways, and receptor tyrosine kinases, including FAK (PF 573228), FGFR (PD 173074), PDGFR (AC 710), and EGFR (Iressa) 39–41 . Given the importance of axon-guided migration in GBM, we included inhibitors of Ephrin receptors (ALW-II-41-27 for EphA2, NVP-BHG712 for EphB4, and Ehp-inhibitor-1), and the CXCL12/CXCR4 axis (Motixafortide), known to regulate directed migration in the hypoxic tumor core in immunotherapy 42,43 . Finally, we tested repurposed drugs such as pranlukast (ONO-1078), a leukotriene receptor antagonist shown to impair GB spread along white matter tracts in preclinical models 44 . On Day 0, GB and NS17 cells were labelled and seeded in co-culture with the neural spheroids. Inhibitors were applied at 1 μM on Days 1 to 3, and cultures were imaged each day after 6 hours of incubation. Representative images illustrate the increasing accumulation of GBM1 cells within and around the spheroid from Day 1 to Day 3 in the untreated controls and with Motixafortide treatment (Figure 5A). As expected from previous observations, both GBM1 and GBM20 cells showed progressive infiltration into the neural spheroid over three days, and inhibitor-specific effects became apparent by Day 3. For the GBM20 line, significant reduction of infiltration was observed with PF 573228 (**, p = 0.0043), Y-27632 (*, p = 0.0244), Motixafortide (*, p = 0.0232), and GF 109203X (*, p = 0.0262) (Figure 5B). Conversely, only Latrunculin B (*, p = 0.0386) and Motixafortide (*, p = 0.0293) significantly reduced GBM1 infiltration (Figure 5B). Importantly, none of the tested inhibitors significantly affected infiltration of the non-cancerous NS17 neural stem cells (See also Figure S1B). Their infiltration remained consistently low and comparable to vehicle controls, underscoring the specificity of the drug responses observed in the GB models. GBM1 and GBM20 exhibit distinct transcriptomic signatures that align with differential drug sensitivities To investigate the molecular basis underlying the different infiltrative behaviors and drug responses of GBM1 and GBM20 observed in our co-culture system, we conducted bulk RNA-sequencing followed by differential gene expression analysis. The two GB lines exhibited distinct transcriptomic profiles, as visualized in a volcano plot. GBM20 showed elevated expressions of PCOLCE , GABBR2 , ERAP2 , PCDHB5 , ERAP2 and CCL2 , while GBM1 was characterized by higher expressions of COL1A2 , MGST1 , CDKN2A , MCAM , and RGMB , suggesting distinct biological programs (Figure 6A). Gene Ontology (GO) enrichment analysis of significantly dysregulated genes (adjusted p 1) further highlighted these differences (Figure 6B). GBM1 was enriched for neurodevelopmental programs such as “positive regulation of nervous system development” and “axonogenesis,” as well as cell adhesion and guidance processes (“cell–cell adhesion via plasma membrane adhesion molecules” and “axon guidance”) and cilium-associated programs (“cilium assembly” and “cilium organization”). In contrast, GBM20 showed enrichment for extracellular matrix (ECM) organization pathways (“ECM structural organization” and “external encapsulating structure organization”) and genes involved in neurite outgrowth and synapse formation (“pattern specification process” and “regulation of neuron projection development”). To explore how these transcriptional differences may relate to drug responses, we conducted pre-ranked Gene Set Enrichment Analysis (GSEA) using MSigDB gene sets (GO, KEGG, Reactome, and Hallmark). In the drug screen, infiltration of GBM1 cells was significantly reduced by Latrunculin B (actin polymerization inhibitor) and Motixafortide (CXCR4 antagonist). Correspondingly, GSEA revealed enrichment of actin remodeling and wound healing pathways in GBM1, suggesting a dependence on actin-driven motility. Enrichment of immune-related pathways, including PD-1 signaling and antigen presentation, also supports involvement of the CXCL12–CXCR4 axis, targeted by Motixafortide (Figure 6C). GBM20 infiltration was significantly inhibited by GF 109203X (Protein Kinase C inhibitor) and Y-27632 (ROCK inhibitor). GSEA revealed enrichment of presynaptic organization and oxidative stress–related pathways in GBM20, consistent with ROCK- and PKC-mediated cytoskeletal remodeling and redox signaling dependencies (Figure 6C). Discussion We present a physiologically relevant in vitro co-culture model combining patient-derived GB cells with human iPSC-derived neural spheroids. These spheroids form a dense core of neuronal bodies with radially extending axons, allowing a reproducible and scalable assessment of GB cell infiltration. Using high-content imaging, we established automated pipelines to quantify axon engagement, directional migration in live assays, and spheroid infiltration in endpoint assays. Both GBM1 and GBM20 actively migrated toward neural spheroids, but GBM1 showed a higher infiltration capacity. A focused drug screen revealed line-specific sensitivities, with four compounds reducing GBM1 infiltration and two affecting GBM20, establishing proof-of-principle for our model as a platform for identifying anti-infiltration compounds. Transcriptomic profiling further highlighted distinct gene expression programs underpinning the observed migratory phenotypes and drug responses. This integrative approach correlates cell behavior with molecular signatures, laying the foundation for functional precision-medicine studies in GB. We demonstrated that both GB lines and a non-tumorigenic neural stem cell line (NS17) migrated toward neural spheroids, but only the GB lines infiltrated them. This suggests that while neural-derived cues attract all lines, only GB cells possess the mechanism for infiltration. Consistent with prior work, we observed that GBM1 infiltrated more readily compared to neural progenitors 45 . However, our previous findings reported greater aggressiveness with patient-derived GBM20 spheroids spontaneously fusing and infiltrating human cerebral organoids at a faster rate than GBM1 spheroids 46 . These differences may reflect shifts in phenotype following treatment, as GBM20 is derived from a recurrent tumor that was treated with radiotherapy and chemotherapy. Notably, recurrent GB cells have been shown to adopt less proliferative, pre-oligodendrocyte-like states in response to damage, potentially explaining the reduced invasiveness 44 . Our model also identified compounds that differentially impair GB cell infiltration. GBM20 was sensitive to inhibition of FAK (PF 573228), PKC (GF 109203X), and ROCK (Y-27632), while GBM1 responded to actin polymerization inhibition (Latrunculin B). Both lines were sensitive to inhibition of CXCR4 (Motixafortide), suggesting reliance on CXCL12/CXCR4 signaling, a known target in GB immunotherapy 43 . Transcriptomic analyses revealed expression patterns consistent with drug sensitivity. GBM20 showed upregulation of FAK pathway–related genes ( SRC , ITGA5 , ITGB3 , PPFIA1 and PRKCB ), aligning with sensitivity to PF 573228 and GF 109203X. Previous studies support these findings, as GF 109203X effectively targeted Protein Kinase C in GB 47 and PF 573228 has also reduced the migration of established GB cell lines, T98G and U-118, in other models 39 . High CXCL12 expression in GBM1 matched its higher sensitivity to Motixafortide with an increased reduction in the number of infiltrated cells in GBM1 compared to GBM20. GSEA revealed additional mechanistic differences: GBM1 showed enrichment for actin cytoskeleton remodeling and immune-related pathways, supporting a model of CXCR4- and actin-driven motility. GBM20 exhibited enrichment for synaptic signaling and oxidative stress pathways, consistent with its dependence on ROCK/PKC signaling. These findings reinforce the value of transcriptomics-informed migration assays in uncovering subtype-specific vulnerabilities. Differential drug responses may reflect distinct pathway dependencies or compensatory mechanisms, underscoring the need for tailored therapeutic strategies. The alignment between transcriptional programs and drug responses supports behavior-based functional stratification as a complementary strategy to conventional molecular profiling. This approach offers a powerful tool for evaluating patient-specific vulnerabilities and complements recent advances using machine learning and high-throughput drug screening 48 . Our results not only highlight divergent invasive programs between GBM1 and GBM20 but also demonstrate the utility of patient-derived co-culture systems in identifying functionally relevant pathways. This work provides proof-of-principle that migration can serve as a robust, measurable, and functionally relevant phenotype to screen therapeutic vulnerabilities in GB. Limitations and future direction This pilot study involved only two GB lines and a single control, limiting broader applicability. Expanding the model to include a diverse panel of patient-derived lines will enhance its predictive utility for therapeutic development. Future improvements include incorporating iPSC-derived oligodendrocytes to mimic myelinated tracts, as glioma cells preferentially migrate on myelinated axons 49 , 50 . Moreover, while phase-contrast imaging enabled live tracking of GB migration and pseudopodia formation 50 – 52 , it was not optimal for imaging infiltration into dense neural spheroids. Implementing high-content Z-stack imaging platforms such as the Operetta CLS permitted higher-throughput and higher-resolution quantification of infiltration dynamics. Conclusion Although glioma growth has been widely studied, its invasion and migration remain poorly understood. Our co-culture model, pairing patient-derived GB cells with axon-extending human neural spheroids, provides a physiologically relevant platform for migration-based drug screening. By integrating high-content imaging, drug screening, and transcriptomic data, we captured cell line–specific behaviors and subtype-dependent migratory mechanisms. These findings underscore the heterogeneity of GB invasion and demonstrate the potential of our platform to guide personalized therapeutic strategies. Methods Cell culture GB cells culture The patient-derived GBM1 and GBM20 cell lines were derived under informed consent according to ethical approvals (LREC 115/ES/0094; courtesy of Dr Heiko Wurdak, University of Leeds 9,31 . GBM1 and GBM20 cell lines were derived respectively from a primary tumor of a 58-year-old female patient and a 50-year-old male patient with a recurrent GB, previously treated with radiotherapy, temozolomide, and IMA950. The cell models maintain the stem cell-like characteristics of GB under adherent culture conditions 9,31,45 . Cells were cultured 5 μg/mL poly-L-ornithine (PLO) (Sigma) and 2 μg/mL laminin (Thermofisher) coated plates in Neurobasal medium (Invitrogen) supplemented with 0.5X B27 (Invitrogen), 0.5X N2 (Invitrogen), 40 ng/mL epidermal growth factor (EGF) (R&D Systems), and 40 ng/mL basic-fibroblast growth factor (b-FGF) (Invitrogen). Cells were passaged using 1X TrypLE (Gibco) when approximately 80% confluency was reached, and passages remained between passage numbers 14 to 20. Cell cultures were routinely tested for mycoplasma, and all were negative for contamination. Neural stem cells The cells GCGR.NS17ST_A (shortened to NS17) from the Glioma Cellular Genetics Resource (GCGR) was acquired under informed consent according to local ethical approvals (08/S1101/1) 32 and kindly provided by Dr Steven Pollard (University of Edinburgh). The cells were cultured according to previously published methods DMEM/HAMS-F12 medium (Sigma) supplemented with 1.5 mg/mL of glucose (Sigma), 100 μM MEM NEAA (Gibco), 50 U/mL and 50 mg/mL Pen Strep (Gibco), 0.01% BSA (Gibco), 100 μM beta-mercaptoethanol (Gibco), 0.5X B27, 0.5X N2, EGF to 10 ng/mL, and b-FGF to 10 ng/mL. The cells were passaged using StemPro Accutase Cell Dissociation Reagent (Gibco) when they reached approximately 80% confluency at a split no lower than 1:6. Induced glutamatergic neuron cell culture HPSI1013I-PAMV_1 human iPSCs obtained from the HipSci biobank, where pluripotency characterization, such as Pluritest, was performed as part of their standard quality control pipelines ( www.hipsci.org ). The line was derived from skin fibroblasts from a healthy male donor aged 65-69 from a White British ethnicity. HPSI1013I-PAMV_1 cells were transduced with a doxycycline-inducible Neurogenin-2 (iNGN2) transgene following the protocol from the Ward laboratory 30 . The integration of the transgene was achieved by electroporation and TALEN-mediated integration into the AAVS1 safe-harbor locus. iNGN2 PAMV_1 cells were cultured in feeder-free conditions on 4% Vitronectin XF (Stem Cell Technologies) coated plates in Essential 8 medium (Gibco) supplemented with 50 U/mL and 50 mg/mL Penicillin/Streptomycin respectively (Life Technologies). The medium was changed daily, and cells were passaged using TryplE and resuspended in Essential 8 medium with 10 μM Y-27632 Rho-kinase (ROCK) inhibitor (ENZO Life Sciences) at approximately 1.5 million cells per well. All cell cultures were routinely screened for mycoplasma and confirmed to be negative for contamination. Transcription factor-mediated differentiated CNS-like glutamatergic neurons The transcription factor-mediated differentiation protocol was adapted from the Ward laboratory 30 . On Day 0, iNGN2 PAMV_1 cells were dissociated using Accutase at 37°C for 5 minutes. The cells were resuspended with the Induction Medium - DMEM/F12 Hepes medium (Life Technologies), supplemented with 0.5X N2, 100 μM MEM NEAA, 2 mM L-Glutamine (Gibco), Doxycycline (2 μg/ml) (Sigma), and 10 μM Y-27632 ROCKi (1:100), on GFR Matrigel (BD Biosciences) (1:50) at a density of 1.5 million cells per well. Doxycycline will induce the differentiation into Glutamatergic Excitatory Neurons. On Days 1 and 2, nascent neurites began to be evident, and the media was replaced with fresh induction media with Doxycycline. Cells were dissociated with Accutase for freezing in 10% dimethyl sulfoxide (Sigma) in medium or for neural spheroid generation. Generation of neural spheroids Following dissociation, differentiated cells were resuspended in the induction media and were seeded at a density of 10,000 cells/well in the 5% Pluronic F127 (Sigma-Aldrich) coated V-bottom plate and centrifuged for 2 min at 200 x g to aggregate the cells at the bottom of the well according to the in-house protocol 53 . The cells were left for 48 h with daily half-medium changes. Maturation and generation of axon bundles On Day 6, the neural spheroids were transferred to laminin-coated 24-well or 96-well glass-bottom plates (Cellvis) in Cortical Neuron Culture Medium - BrainPhys neuronal medium (Stemcell Technologies) supplemented with 0.5X B27, 10 ng/mL Brain-Derived Neurotrophic Factor (BDNF) (PeproTech), 10 ng/mL Neurotrophin-3 (NT-3) (PeproTech), and 1 μg/mL Laminin. The cells were checked daily for the presence of cell debris and morphological changes with bi-weekly half-medium changes. Co-culture of GB cells with neural spheroids On Day 11, 1X BioTracker 488 Green Nuclear Dye (Sigma-Aldrich) was added to the cell culture medium to stain the neural spheroids for 20 min at 37°C and washed twice. GB cells were labelled with 1X BioTracker 655 Red Cytoplasmic Membrane Dye (Sigma-Aldrich) in suspension for 20 min at 37°C and washed three times by centrifugation. The GB cells were seeded on top of the neural spheroids at a density of 5,000 cells/well in a 24-well plate or 840 cells/well in a 96-well plate with a 50/50 medium. Staining and Imaging Immunostaining Neural spheroids were fixed at 4% PFA/PBS at RT for 20 min and permeabilized with 3% BSA/0.1% Triton-X-100/PBS for 1 h at RT. Cells were incubated with the primary mouse axonal antibody βIII-tubulin (Tubb3) (R&D Systems) diluted in 3% BSA/0.1% Triton-X-100/PBS (1/1000) overnight at 4°C in the dark. The next day, primary antibodies were rinsed away with 3% BSA/PBS. The cells were incubated with the secondary anti-mouse antibody conjugated to AF 488 (Invitrogen), diluted in 3% BSA/0.1% Triton-X-100/PBS (1:500) at RT for 45 min and rinsed with PBS. To stain the nucleus, the cells are incubated with 10 μM Hoechst (Invitrogen) for 10 min at RT and rinsed with PBS. Finally, the cells were incubated with 80% glycerol as a clearing agent for 1 h min at RT and rinsed with PBS before being placed in fresh PBS at 4°C until imaging. Confocal microscopy Confocal images were acquired using a Leica TCS SP8 Confocal laser scanning microscope, using a 10x dry objective, and viewed with the Leica software. Each fluorochrome was excited with the corresponding laser line (DAPI, UV (355 nm); AF488, Green (530 nm); AF594, Red (639 nm) laser line). Immunofluorescence data were analyzed in Fiji open-source software (2.0.0-rc-64/1.51s version). Images from different fields were tiled and stitched, and the maximum projection was obtained using the Z-stack. Imaging assay Live-cell imaging of non-labelled cells Live-cell imaging of non-labelled cells on laminin-coated glass plates and on axons was performed on the Livecyte quantitative phase imager (Phasefocus), which uses quantitative phase imaging and has an accurate tracking apparatus. ROIs (10 mm x 25 mm) on the right and the top of the neural spheroids were selected, and images were acquired every 15 min for 24 h. Endpoint assay Co-cultures of patient-derived GB cells and human iPSC-derived cortical-like neural spheroids were imaged every 24 h at Day 0, 1, 2, and 3 using the Operetta CLS High Content Analysis (PerkinElmer) with the brightfield, 488 nm, and 655 nm channels. The PreciScan Intelligent Acquisition plug-in for Harmony software was used to locate the neural spheroid within the well. The plug-in allows the accurate targeting of the region of interest (ROI) whilst reducing acquisition and analysis times. Once the PreciScan was performed and the region of the neural spheroid was located, a Z-stack of 10 images over 20 μm (distance 2 μm) was acquired. Inhibitor screen Inhibitors were added to the co-culture at a final concentration of 1 μM. Following a 6 h antagonist treatment, images were acquired using the Operetta CLS Imaging System. This was repeated for the next two days. Image analysis Live-cell imaging assay analysis using the Phasefocus Livecyte Images were analyzed using the built-in Analysis Software with the Motility Assay on the Livecyte quantitative phase imager (Phasefocus). Using the built-in applications, the cells were identified and tracked, and the tracking properties were calculated to generate cell migration parameters. Directionality results were analyzed in a polar histogram using R Studio (version 2025.05.1). Analyzes of instantaneous velocity, speed, and confinement ratio were performed using GraphPad Prism software, and the Welch’s t-test was performed between cells plated with or without axons. A p-value below 0.05 was considered significant and was indicated with an asterisk: ****, p < 0.0001. Endpoint assay analysis using Harmony Images were analyzed using the built-in Harmony HCI and Analysis Software on the Operetta CLS device. The maximum projection of the Z-stack was taken, and the neural spheroid was stained with the BioTracker 488 and identified as an ROI using the green channel. The GB cells stained with the BioTracker 655 in the ROI were identified and segmented using the far-red channel (Segmentation Method C in Harmony Software). All experiments were performed in technical triplicate and were independently repeated at least three times to perform statistical analysis. All statistical analyses were performed using GraphPad Prism software (version 9.2). Results represented as means with standard deviation (SD), and different ANOVA tests with multiple comparisons between control and other conditions were performed to calculate the statistical significance of multiple experimental conditions. A p-value below 0.05 was considered significant and was indicated with an asterisk: *, p ≤ 0.05; **, p ≤0.01; ***, p ≤0.001. The number of independent experiments is indicated in the figure legend as N. RNA sequencing The GBM1 and GBM20 cell lines were prepared for RNA sequencing. They were rinsed once with PBS, detached with TrypLE, and snap-frozen in pellets of 10 6 cells. The cells were submitted to Eurofins Genomics, Germany, under Project ID: NG-29040. Raw sequencing data were pre-processed using the fastp software to generate clean data, termed quality control. This involves checking the quality of the raw sequencing filtering for high-quality reads to remove poor-quality bases (below Phred Quality 20) 54 . High-quality sequence reads were aligned using the STAR (Spliced Transcripts Alignment to a Reference) to the reference genome UCSC Homo sapiens version hg38 55 . Gene-wise quantification was achieved to inspect the transcriptome alignments using the RSEM tool 56 . For the differential gene expression between cell lines, genes with fewer than 10 average reads were removed. Using the R/Bioconductor DESeq2 package, the abundance counts of each gene were then used to perform differential gene expression 57 . Eurofins Genomics provided this pre-processing. RNA-seq data analysis Further analysis was performed on R Studio using the ggplot2 package to generate a volcano plot of differential gene expression. Sample-wise comparison values (log₂ fold change and p-value) provided by Eurofins Genomics were used. Gene Ontology (GO) enrichment and Gene Set Enrichment Analysis (GSEA) were performed using the clusterProfiler and fgsea packages to identify biological processes and pathways associated with differential expression. Significantly enriched terms were visualized using bar plots and enrichment plots. Declarations Acknowledgements We appreciate the support, advice, and reagents provided by everyone at CSCRM. We are grateful to Dr Bronwyn Irving, Dr James Williams, and Dr Léa R’Bibo for their valuable expertise in GB cell culture, stem cell culture, and neural differentiation, respectively. We thank Thomas Williams, Dr Lazaros Fotopoulos, and Erika Wiseman at the Stem Cell Hotel for their expert assistance and management of the imaging facilities. We would also like to thank Ioanna Kourouzidou, Oluwaseun Adegbite, Layla Kadhim for their involvement in preliminary experiments. Furthermore, we acknowledge Dr Andrea Serio, Dr Ciro Chiappini and Dr Steve Pollard for their valuable support and input in the project initiation. HipSci Lines samples were collected from consented research volunteers recruited from the NIHR Cambridge BioResource through https://www.cambridgebioresource.group.cam.ac.uk/. The HipSci consortium obtained ethics approval for a revised consent (REC ref. 09/H0304/77, V3 15/03/2013), under which all data, except for the Y chromosome from males, can be made openly available (Y chromosome data can be used to de-identify men by surname matching). For open access, the author has applied a CC BY public copyright license to any author-accepted manuscript version arising from this submission. Funding V.T. and F.R. were funded by the Wellcome Trust as part of the “Cell Therapies & Regenerative Medicine” PhD Programme (108874/Z/15/Z). The authors acknowledge financial support Rosetrees Trust to D.D., from the Medical Research Council (grant MR/N025865/1) to I.L, and from Innovate UK (TSB/89370) from J.D.C and D.D. Author contributions V.T., D.D., and I.L., conceptualization; V.T. and F.R., data curation; V.T., F.R., I.L., and D.D., investigation; V.T., F.R., I.L., and D.D., funding acquisition; V.T., formal analysis; V.T., F.R., H.N., and J.C., methodology; D.D. and I.L., supervision; V.T., and D.D., visualization; V.T., and D.D., writing – original draft; V.T., F.R., I.L., H.N., H.B., and D.D., writing – review & editing. All authors contributed to the article and approved the submitted version. Data availability Processed data have been deposited in the Gene Expression Omnibus (GEO) under accession number GSE308010 (accessible during review at: https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.ncbi.nlm.nih.gov%2Fgeo%2Fquery%2Facc.cgi%3Facc%3DGSE308010&data=05%7C02%7Cv.tsang%40qmul.ac.uk%7C73893a06442148f1678508ddf46aabc3%7C569df091b01340e386eebd9cb9e25814%7C0%7C0%7C638935461170639673%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=%2BW3x03LMJE%2BRKGccA12v15qzHemSMEIWJ4wYTU4kkpE%3D&reserved=0 Reviewer token: efixgocyffgbpmb). Raw sequencing data (BAM files) are available in the Sequence Read Archive (SRA) under BioProject accession number PRJNA1327912 (accessible view only at: https://dataview.ncbi.nlm.nih.gov/object/PRJNA1327912?reviewer=1p614t1hlcnep52di60emi03d3). Data will be released publicly upon acceptance and publication. Competing interests D.D. is co-founder of Migration Biotherapeutics, and the rest of the authors declare no competing interests. References Ceccarelli, M. et al. Molecular Profiling Reveals Biologically Discrete Subsets and Pathways of Progression in Diffuse Glioma. Cell 164 , 550–563 (2016). Guo, X. et al. Clinical updates on gliomas and implications of the 5th edition of the WHO classification of central nervous system tumors. Front. Oncol. 13 , (2023). Hertler, C. et al. Long-term survival with IDH wildtype glioblastoma: first results from the ETERNITY Brain Tumor Funders’ Collaborative Consortium (EORTC 1419). Eur. J. Cancer 189 , (2023). Louis, D. N. et al. The 2021 WHO Classification of Tumors of the Central Nervous System: a summary. Neuro-Oncol. 23 , 1231 (2021). Melhem, J. M., Detsky, J., Lim-Fat, M. J. & Perry, J. R. Updates in IDH-Wildtype Glioblastoma. Neurotherapeutics 19 , 1705–1723 (2022). Ostrom, Q. T. et al. CBTRUS Statistical Report: Primary Brain and Other Central Nervous System Tumors Diagnosed in the United States in 2011–2015. Neuro-Oncol. 20 , iv1 (2018). Ostrom, Q. T. et al. Editor’s choice: The epidemiology of glioma in adults: a “state of the science” review. Neuro-Oncol. 16 , 896 (2014). World Health Organization Classification of Tumours of the Central Nervous System. Int. Agency Res. Cancer 5th editio , (2021). Wurdak, H. et al. An RNAi Screen Identifies TRRAP as a Regulator of Brain Tumor-Initiating Cell Differentiation. Cell Stem Cell 6 , 37–47 (2010). Gimple, R. C., Bhargava, S., Dixit, D. & Rich, J. N. Glioblastoma stem cells: lessons from the tumor hierarchy in a lethal cancer. Genes Dev. 33 , 591–609 (2019). Dirks, P. B. Brain tumor stem cells: the cancer stem cell hypothesis writ large. Mol. Oncol. 4 , 420–430 (2010). Stupp, R. et al. Effects of radiotherapy with concomitant and adjuvant temozolomide versus radiotherapy alone on survival in glioblastoma in a randomised phase III study: 5-year analysis of the EORTC-NCIC trial. Lancet Oncol. 10 , 459–466 (2009). Ozdemir-Kaynak, E., Qutub, A. A. & Yesil-Celiktas, O. Advances in glioblastoma multiforme treatment: New models for nanoparticle therapy. Front. Physiol. 9 , (2018). de Gooijer, M. C., Guillén Navarro, M., Bernards, R., Wurdinger, T. & van Tellingen, O. An Experimenter’s Guide to Glioblastoma Invasion Pathways. Trends Mol. Med. 24 , 763–780 (2018). Armento, A. et al. Molecular mechanisms of cell motility. in Cancer and Metastasis Review vol. 11 73–93 (Codon Publications, 2017). Robertson, F. L., Marqués-Torrejón, M. A., Morrison, G. M. & Pollard, S. M. Experimental models and tools to tackle glioblastoma. DMM Dis. Models Mech. 12 , (2019). Nishikawa, M. et al. Significance of glioma stem-like cells in the tumor periphery that express high levels of CD44 in tumor invasion, early progression, and poor prognosis in glioblastoma. Stem Cells Int. 2018 , (2018). Darmanis, S. et al. Single-Cell RNA-Seq Analysis of Infiltrating Neoplastic Cells at the Migrating Front of Human Glioblastoma. Cell Rep. 21 , 1399–1410 (2017). Singh, S. K. et al. Identification of human brain tumour initiating cells. Nature 432 , 396–401 (2004). Garnier, D., Renoult, O., Alves-Guerra, M. C., Paris, F. & Pecqueur, C. Glioblastoma stem-like cells, Metabolic strategy to kill a challenging target. Front. Oncol. 9 , 118 (2019). Cha, J. & Kim, P. Biomimetic strategies for the glioblastoma microenvironment. Front. Mater. 4 , (2017). Selvi, H., Brüning-Richardson, A. & Danovi, D. Systematic Review of Pre-Clinical Systems Using Artificial Microenvironments and Anti-Migratory Drugs to Control Migration of Glioblastoma Cells. Expert Rev. Mol. Med. 27 , e6 (2025). Xie, Y. et al. The Human Glioblastoma Cell Culture Resource: Validated Cell Models Representing All Molecular Subtypes. EBioMedicine 2 , 1351–1363 (2015). Lee, J. et al. Tumor stem cells derived from glioblastomas cultured in bFGF and EGF more closely mirror the phenotype and genotype of primary tumors than do serum-cultured cell lines. Cancer Cell 9 , 391–403 (2006). Pine, A. R. et al. Tumor Microenvironment Is Critical for the Maintenance of Cellular States Found in Primary Glioblastomas. Cancer Discov. CD-20-0057 (2020) doi:10.1158/2159-8290.cd-20-0057. Liu, E. & Nolan, J. P. Surface Enhanced Raman Scattering (SERS) Image Cytometry for High-Content Screening. Fluoresc. Microsc. Super-Resolut. Nov. Tech. 93–108 (2014) doi:10.1016/B978-0-12-409513-7.00007-5. Bon, C., Si, Y. & Arimondo, P. B. Targeting DOT1L for mixed-lineage rearranged leukemia. Histone Modif. Ther. 81–99 (2020) doi:10.1016/B978-0-12-816422-8.00005-2. Garrett, A. M., Lastakchi, S. & McConville, C. The Personalisation of Glioblastoma Treatment Using Whole Exome Sequencing: A Pilot Study. Genes 11 , (2020). Lee, S. et al. Targeting tumour-intrinsic neural vulnerabilities of glioblastoma. bioRxiv 2022.10.07.511321 (2022) doi:10.1101/2022.10.07.511321. Fernandopulle, M. S. et al. Transcription Factor–Mediated Differentiation of Human iPSCs into Neurons. Curr. Protoc. Cell Biol. 79 , e51 (2018). Polson, E. S. et al. KHS101 disrupts energy metabolism in human glioblastoma cells and reduces tumor growth in mice. Sci. Transl. Med. 10 , 2718 (2018). Al Shboul, S. et al. Kinomics platform using GBM tissue identifies BTK as being associated with higher patient survival. Life Sci. Alliance 4 , (2021). Liu, X. et al. Overview of the molecular mechanisms of migration and invasion in glioblastoma multiforme. J. Chin. Med. Assoc. 84 , 669–677 (2021). Ströbele, S. et al. A Potential Role for the Inhibition of PI3K Signaling in Glioblastoma Therapy. PLOS ONE 10 , e0131670 (2015). Djuzenova, C. S. et al. Differential effects of the Akt inhibitor MK-2206 on migration and radiation sensitivity of glioblastoma cells. BMC Cancer 2019 191 19 , 1–18 (2019). Bogdahn, U. et al. Targeted therapy for high-grade glioma with the TGF-β2 inhibitor trabedersen: results of a randomized and controlled phase IIb study. Neuro-Oncol. 13 , 132 (2011). Wick, A. et al. Phase 1b/2a study of galunisertib, a small molecule inhibitor of transforming growth factor-beta receptor I, in combination with standard temozolomide-based radiochemotherapy in patients with newly diagnosed malignant glioma. Invest. New Drugs 38 , 1570 (2020). Wen, P. Y. et al. Buparlisib in Patients With Recurrent Glioblastoma Harboring Phosphatidylinositol 3-Kinase Pathway Activation: An Open-Label, Multicenter, Multi-Arm, Phase II Trial. J. Clin. Oncol. 37 , 741 (2019). Anderson, H. J. & Galileo, D. S. Small-molecule inhibitors of FGFR, integrins and FAK selectively decrease L1CAM-stimulated glioblastoma cell motility and proliferation. Cell. Oncol. 39 , 229–242 (2016). Parker, J. J. et al. Gefitinib selectively inhibits tumor cell migration in EGFR-amplified human glioblastoma. Neuro-Oncol. 15 , 1048 (2013). Singh, J., Sharma, K. & Pillai, P. P. PDGFR inhibition mediated intracellular signalling in C6 glioma growth and migration: role of ERK and ROCK pathway. Cytotechnology 70 , 465 (2018). Affinito, A. et al. Targeting Ephrin Receptor Tyrosine Kinase A2 with a Selective Aptamer for Glioblastoma Stem Cells. Mol. Ther. Nucleic Acids 20 , 176–185 (2020). Zhou, W., Guo, S., Liu, M., Burow, M. E. & Wang, G. Targeting CXCL12/CXCR4 Axis in Tumor Immunotherapy. Curr. Med. Chem. 26 , 3026 (2019). Brooks, L. J. et al. The white matter is a pro-differentiative microenvironment for glioblastoma. J. Chem. Inf. Model. 01 , 1689–1699 (2020). da Silva, B. B., Mathew, R. K., Polson, E. S., Williams, J. & Wurdak, H. Spontaneous Glioblastoma Spheroid Infiltration of Early-Stage Cerebral Organoids Models Brain Tumor Invasion. SLAS Discov. 23 , 862–868 (2018). Irving, B. K. Modelling invasion, migration and cellular networks in Glioblastoma stem cells. (University of Leeds, 2020). Geribaldi-Doldán, N. et al. Targeting Protein Kinase C in Glioblastoma Treatment. Biomedicines 9 , (2021). Smer-Barreto, V. et al. Identification of drug candidates against glioblastoma with machine learning and high-throughput screening of heterogeneous cellular models. 2025.03.06.641926 Preprint at https://doi.org/10.1101/2025.03.06.641926 (2025). Giese, A. et al. Migration of Human Glioma Cells on Myelin : Neurosurgery . Neurosurgery vol. 38 755–764 (1996). Zepecki, J. P. et al. Regulation of human glioma cell migration, tumor growth, and stemness gene expression using a Lck targeted inhibitor. Oncogene 38 , 1734–1750 (2019). Bayat, N. et al. The anti-angiogenic effect of atorvastatin in glioblastoma spheroids tumor cultured in fibrin gel: In 3D in vitro model. Asian Pac. J. Cancer Prev. 19 , 2553–2560 (2018). Gudbergsson, J. M. et al. A tumorsphere model of glioblastoma multiforme with intratumoral heterogeneity for quantitative analysis of cellular migration and drug response. Exp. Cell Res. 379 , 73–82 (2019). Alsehli, H. et al. An integrated pipeline for high-throughput screening and profiling of spheroids using simple live image analysis of frame to frame variations. Methods San Diego Calif 190 , 33 (2021). Chen, S., Zhou, Y., Chen, Y. & Gu, J. fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics 34 , i884–i890 (2018). Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29 , 15–21 (2013). Li, B. & Dewey, C. N. RSEM: Accurate transcript quantification from RNA-Seq data with or without a reference genome. BMC Bioinformatics 12 , 1–16 (2011). Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15 , 1–21 (2014). Table Table 1. Table summarizing the inhibitors used in the drug screen, including the compound name, pharmacological target, IC50, provider, and catalogue number Compound name Pharmacological Target IC50 in nM Company Catalogue number API-2 Protein kinase B (PKB/Akt) 50 Tocris 2151 PF 573228 Focal Adhesion Kinase (FAK) 4 Tocris 3239 ONO 1078 (Pranlukast) Cysteinyl leukotriene receptor 1 (CysLT1) 4 Tocris 3026/50 Blebbistatin Myosin II ATPase 2000 Tocris 1760 Latrunculin B actin polymerization 60 Tocris 3974 Y-27632 Rho-associated protein kinase (ROCK) 140 Tocris 1254 PD 173074 Fibroblast Growth Factor Receptor (FGFR) 5 Tocris 3044 HPI 1 Sonic Hedgehog 150 Tocris 3839 IWP 2 Wnt 27 Tocris 3533 ALW II-41-27 (Compound 7) Ephrin receptor EphA2 12 Selleck Chem S6515 NVP-BHG712 S2202 Ephrin receptor EphB4 25 Selleck Chem S2202 Ehp-inhibitor-1 Ephrin receptors (EphB2, EphB4 and related Eph kinases) 10 Selleck Chem S0256 Cilengitide (EMD121974) integrins αvβ3 and αvβ5 4.1 Tocris 5870 Motixafortide (BL-8040) stromal derived factor 1 (SDF-1, CXCL12) 1 Selleck S9665 GF 109203X Protein Kinase C 8.4 Tocris 0741 SB 431542 TGF-β receptors 94 Tocris 1614 AC 710 Platelet-Derived Growth Factor Receptor (PDGFR) 1.2 Tocris 5013 Iressa Epidermal growth factor receptor (EGFR) 23 Tocris 3000 Additional Declarations Competing interest reported. D.D. is co-founder of Migration Biotherapeutics, and the rest of the authors declare no competing interests. Supplementary Files TsangScientificReportsSupplementaryinformation.docx Cite Share Download PDF Status: Published Journal Publication published 13 Dec, 2025 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 20 Oct, 2025 Reviews received at journal 17 Oct, 2025 Reviews received at journal 06 Oct, 2025 Reviewers agreed at journal 29 Sep, 2025 Reviewers agreed at journal 26 Sep, 2025 Reviewers agreed at journal 23 Sep, 2025 Reviewers agreed at journal 23 Sep, 2025 Reviewers invited by journal 23 Sep, 2025 Editor assigned by journal 23 Sep, 2025 Editor invited by journal 23 Sep, 2025 Submission checks completed at journal 21 Sep, 2025 First submitted to journal 21 Sep, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7536545","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":522283814,"identity":"f8cb206b-32a4-4e45-bce8-282658c536b3","order_by":0,"name":"Victoria S K Tsang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAzElEQVRIiWNgGAWjYBADOSBmPABm8hBUzAwmjUEEaVoSG4jWYnCA//CHnztq0/uunTE4wFBjx2Bw5gAhLcxskr1njufOvJ0D1HIsmcHgbAN+LZINzGwMvG3HcjeAtbAdYDA4T8BhQC3MH/+2HUs3AGv5R4QWfqD3pXnbahLAWhjbDhB2GD8zs5m0bNsBw5m30woOJPYl80gS8j4be+Pjj2/b6uT5bidvfPDhm50c35kEAi6DRMthSKQkEBORUFAHi8dRMApGwSgYBZgAAOZiRNEqOtqEAAAAAElFTkSuQmCC","orcid":"","institution":"King’s College London","correspondingAuthor":true,"prefix":"","firstName":"Victoria","middleName":"S K","lastName":"Tsang","suffix":""},{"id":522283815,"identity":"c4ce9117-1863-43b1-89d9-529448c4c0e0","order_by":1,"name":"Federica Riccio","email":"","orcid":"","institution":"King’s College London","correspondingAuthor":false,"prefix":"","firstName":"Federica","middleName":"","lastName":"Riccio","suffix":""},{"id":522283816,"identity":"9f8ae230-0912-4878-82a1-3f1edf784ec5","order_by":2,"name":"Hannah Nudds","email":"","orcid":"","institution":"University College London","correspondingAuthor":false,"prefix":"","firstName":"Hannah","middleName":"","lastName":"Nudds","suffix":""},{"id":522283817,"identity":"f73fd502-b7a3-4628-bfa8-8e7c1bca5bcf","order_by":3,"name":"Jason Coombes","email":"","orcid":"","institution":"Saint Louis University","correspondingAuthor":false,"prefix":"","firstName":"Jason","middleName":"","lastName":"Coombes","suffix":""},{"id":522283818,"identity":"2e21e740-3e3c-4834-81d8-042761212324","order_by":4,"name":"Heiko Wurdak","email":"","orcid":"","institution":"University of Leeds","correspondingAuthor":false,"prefix":"","firstName":"Heiko","middleName":"","lastName":"Wurdak","suffix":""},{"id":522283819,"identity":"26db791a-c54c-4724-8dcd-8a397bd3ab4e","order_by":5,"name":"Harry J C J Bulstrode","email":"","orcid":"","institution":"University of Cambridge","correspondingAuthor":false,"prefix":"","firstName":"Harry","middleName":"J C J","lastName":"Bulstrode","suffix":""},{"id":522283820,"identity":"7c0a9c9a-ec86-4b00-85cc-c34a9b555b48","order_by":6,"name":"Ivo Lieberam","email":"","orcid":"","institution":"King’s College London","correspondingAuthor":false,"prefix":"","firstName":"Ivo","middleName":"","lastName":"Lieberam","suffix":""},{"id":522283821,"identity":"8601f91d-c6e1-404a-a53d-6d98b1c31a72","order_by":7,"name":"Davide Danovi","email":"","orcid":"","institution":"King’s College London","correspondingAuthor":false,"prefix":"","firstName":"Davide","middleName":"","lastName":"Danovi","suffix":""}],"badges":[],"createdAt":"2025-09-04 13:23:22","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7536545/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7536545/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-025-30914-5","type":"published","date":"2025-12-13T15:59:14+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":92876155,"identity":"7e18f379-fbc5-43ad-bd78-b324a0d567af","added_by":"auto","created_at":"2025-10-06 14:42:40","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":148764,"visible":true,"origin":"","legend":"","description":"","filename":"TsangScientificreportsManuscript.docx","url":"https://assets-eu.researchsquare.com/files/rs-7536545/v1/dfcf6c20660e33b7048e20ef.docx"},{"id":92875146,"identity":"33f8793d-4df5-4089-8b2d-45a65065a2d1","added_by":"auto","created_at":"2025-10-06 14:34:40","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":19638,"visible":true,"origin":"","legend":"","description":"","filename":"TsangScientificReportsTables.docx","url":"https://assets-eu.researchsquare.com/files/rs-7536545/v1/b869763f5e08193e01c0a0f7.docx"},{"id":92875145,"identity":"b81ee140-873d-4f6f-a98c-d2459c4fb4d9","added_by":"auto","created_at":"2025-10-06 14:34:40","extension":"json","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":11278,"visible":true,"origin":"","legend":"","description":"","filename":"3ea06b32593e43d0af7d9b8f3eec26c9.json","url":"https://assets-eu.researchsquare.com/files/rs-7536545/v1/fb915734b460929ac853f1f0.json"},{"id":92874357,"identity":"55d9d6d1-7eb4-41b5-a3a1-9939e3fd47c8","added_by":"auto","created_at":"2025-10-06 14:26:40","extension":"docx","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":69429,"visible":true,"origin":"","legend":"","description":"","filename":"TsangScientificReportsSupplementaryinformation.docx","url":"https://assets-eu.researchsquare.com/files/rs-7536545/v1/813c384f16f6d15afc6f774c.docx"},{"id":92874361,"identity":"045761d1-f917-444b-a548-5a8f821805ef","added_by":"auto","created_at":"2025-10-06 14:26:40","extension":"xml","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":144892,"visible":true,"origin":"","legend":"","description":"","filename":"3ea06b32593e43d0af7d9b8f3eec26c91enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7536545/v1/9a5a18ffe8fd67970d6aa153.xml"},{"id":92875150,"identity":"a2d0e59f-9e74-4081-a074-debd28d99666","added_by":"auto","created_at":"2025-10-06 14:34:41","extension":"pdf","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6338821,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1Neuralspheroid.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7536545/v1/ea04fa36fbcb7d6f5a249885.pdf"},{"id":92876156,"identity":"6c105914-4da6-4cc6-abe0-89222718378e","added_by":"auto","created_at":"2025-10-06 14:42:41","extension":"pdf","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3325364,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2Liveassay.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7536545/v1/25e86069cead032e5f774a04.pdf"},{"id":92874363,"identity":"47a31d98-9a49-450a-be17-5939131923c8","added_by":"auto","created_at":"2025-10-06 14:26:41","extension":"pdf","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":988811,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3Liveassay.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7536545/v1/7d56957997793fd7fbbd58e5.pdf"},{"id":92874367,"identity":"277d0450-3ced-43b4-be2e-51aa98ee827d","added_by":"auto","created_at":"2025-10-06 14:26:42","extension":"pdf","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":46872872,"visible":true,"origin":"","legend":"","description":"","filename":"Figure4Infiltration.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7536545/v1/5759d180bbf5a4c14140fb9c.pdf"},{"id":92874366,"identity":"2832dccc-b5db-42cf-801b-9ad34d1aab9c","added_by":"auto","created_at":"2025-10-06 14:26:41","extension":"pdf","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":19164044,"visible":true,"origin":"","legend":"","description":"","filename":"Figure5Inhibitors.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7536545/v1/f3375882180af3be78004f30.pdf"},{"id":92875149,"identity":"2f3c2d48-9f88-4e37-9a9e-3ce629f8fd58","added_by":"auto","created_at":"2025-10-06 14:34:40","extension":"pdf","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":93218,"visible":true,"origin":"","legend":"","description":"","filename":"Figure6RNASeq.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7536545/v1/ebfcc91e2b6560f840c04be8.pdf"},{"id":92874359,"identity":"c1e0d6e4-18b2-43da-b560-8815f4be8cbd","added_by":"auto","created_at":"2025-10-06 14:26:40","extension":"xml","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":144565,"visible":true,"origin":"","legend":"","description":"","filename":"3ea06b32593e43d0af7d9b8f3eec26c91structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7536545/v1/422a1d52137e25cf1c394578.xml"},{"id":92874362,"identity":"836f350c-c2c1-4ac1-ba30-2424fa554eb1","added_by":"auto","created_at":"2025-10-06 14:26:41","extension":"html","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":157051,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7536545/v1/3f9fcba075e1532659337b4d.html"},{"id":92874351,"identity":"e88ee891-777a-459e-9057-5b8ddb8a2527","added_by":"auto","created_at":"2025-10-06 14:26:40","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":460671,"visible":true,"origin":"","legend":"\u003cp\u003eCo-culture of human iPSC-derived neural spheroids and patient-derived GB cells\u003c/p\u003e\n\u003cp\u003eA) Timeline and schematic diagram of the transcription-factor mediated neural induction (Fernandopulle et al., 2018), neural spheroid generation using Pluronic acid-coated V-bottom plates, maturation, axon outgrowth, and co-culture with patient-derived cells.\u003c/p\u003e\n\u003cp\u003eB) Representative brightfield image captured on the x10 microscope of a neural spheroid generated through forward reprogramming. Scale bar = 200 μm.\u003c/p\u003e\n\u003cp\u003eC) Representative confocal images of the neural spheroid and axons (Hoechst in blue and Tubb3 in green). Scale bar = 200 μm.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7536545/v1/004fa85819d7fe679f4ba95c.png"},{"id":92874347,"identity":"5801b554-12b0-4312-819d-754f0e1fa49f","added_by":"auto","created_at":"2025-10-06 14:26:40","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":309166,"visible":true,"origin":"","legend":"\u003cp\u003eLive imaging assays of neural spheroids co-cultured with NS17, GBM20, and GBM1 cells\u003c/p\u003e\n\u003cp\u003eSchematic diagram of the layout of the cells cultured with or without axons and phase contrast images of the time-lapse assay of NS17, GBM20, and GBM1 cells cultured with or without axons at time point 0 (00:00) with the cell mask overlay. Scale bar = 500 μm.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7536545/v1/a3def9174d32951912128cbf.png"},{"id":92876153,"identity":"9fb3ed78-47be-449b-b9f3-78b24700abd0","added_by":"auto","created_at":"2025-10-06 14:42:40","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":164743,"visible":true,"origin":"","legend":"\u003cp\u003eMotility analysis of live imaging assays of neural spheroids co-cultured with NS17, GBM20, and GBM1 cells\u003c/p\u003e\n\u003cp\u003eA) Polar histograms summarize the percentage of cells migrating by angle in degrees. Number of cells analyzed = 1000 cells.\u003c/p\u003e\n\u003cp\u003eB) Quantification of the instantaneous velocity, speed, and confinement ratio of the NS17, GBM20, and GBM1 cells with or without axons. Number of cells analyzed = 1000 cells. ****, \u003cem\u003ep\u003c/em\u003e\u0026lt; 0.0001 calculated from Welch’s t-test.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7536545/v1/331bb66d6204ae64a0034dad.png"},{"id":92876154,"identity":"cf76a58d-7c53-480f-868a-b6beda11ba9a","added_by":"auto","created_at":"2025-10-06 14:42:40","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1351251,"visible":true,"origin":"","legend":"\u003cp\u003eEndpoint assays of neural spheroids co-cultured with NS17, GBM20, and GBM1 cells\u003c/p\u003e\n\u003cp\u003eA) Representative immunofluorescence images of the co-culture of the neural spheroids with NS17, GBM20, and GBM1 cells at Day 0-5. Scale bars = 100 μm.\u003c/p\u003e\n\u003cp\u003eB) Quantification of NS17, GBM20, and GBM1 infiltration of the neural spheroid at Day 0-5, normalized to Day 0. Error bars represent mean ± SD. number of independent experiments (n) = 3. ****, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.0001 and **, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001 calculated from 2way ANOVA of baseline-corrected data.\u003c/p\u003e\n\u003cp\u003eC) Normal distribution curves for NS17, GBM20, and GBM1 infiltration calculated from the means and standard deviation at Day 3. Z-factor = 0.701246093.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7536545/v1/05aeafabce8ad375be546e8d.png"},{"id":92874348,"identity":"4bcea81d-36fe-4751-8725-d7364db06051","added_by":"auto","created_at":"2025-10-06 14:26:40","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":672009,"visible":true,"origin":"","legend":"\u003cp\u003ePilot drug screen on the co-culture of neural spheroids with GBM1 or GBM20 cells\u003c/p\u003e\n\u003cp\u003eA) Representative immunofluorescence images of the co-culture of the neural spheroids with GBM1 cells from Day 1-3 in the untreated control and with the inhibitor Motixafortide at 1 μM. Scale bar = 100 μm.\u003c/p\u003e\n\u003cp\u003eB) Quantification of GBM20 and GBM1 cellular infiltration at Day 3, normalized to Day 0. Error bars represent mean ± SD. number of independent experiments (n) = 3. **, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.005 and *, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05 calculated using the Brown-Forsythe and Welch ANOVA tests.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-7536545/v1/8ca2d847caa10b45765dcd27.png"},{"id":92875143,"identity":"bf046cc8-5372-4a80-a2f6-ba41de543c6e","added_by":"auto","created_at":"2025-10-06 14:34:40","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":123795,"visible":true,"origin":"","legend":"\u003cp\u003eDifferential gene expressions between GBM1 and GBM20 cell lines.\u003c/p\u003e\n\u003cp\u003eA) Volcano plot showing the differentially expressed genes between GBM1 (orange) and GBM20 (blue). Genes with an adjusted p-value \u0026lt; 0.05 and |log₂ fold change| \u0026gt; 1 were considered significant.\u003c/p\u003e\n\u003cp\u003eB) Bar plots of GO biological process enrichment for significantly upregulated genes in GBM1 and GBM20. The q-score represents enrichment strength (−log₁₀ adjusted p-value). Color intensity corresponds to the adjusted p-value (FDR-corrected), with red indicating higher significance and blue lower significance.\u003c/p\u003e\n\u003cp\u003eC) GSEA enrichment scores of relevant pathways in GBM1 (orange) and GBM20 (blue).\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-7536545/v1/1bbe741545c4c67ca3031e75.png"},{"id":98245831,"identity":"acd2e77e-69f3-4fda-a559-4ebcf524e72b","added_by":"auto","created_at":"2025-12-15 16:18:30","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3471307,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7536545/v1/e1c78486-183d-41f3-a335-523e0df4c2c3.pdf"},{"id":92874354,"identity":"c45715d3-bec0-45cb-8595-133a0fde1bed","added_by":"auto","created_at":"2025-10-06 14:26:40","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":69429,"visible":true,"origin":"","legend":"","description":"","filename":"TsangScientificReportsSupplementaryinformation.docx","url":"https://assets-eu.researchsquare.com/files/rs-7536545/v1/e7693673b42cf9e1454f9930.docx"}],"financialInterests":"Competing interest reported. D.D. is co-founder of Migration Biotherapeutics, and the rest of the authors declare no competing interests.","formattedTitle":"A human iPSC-based neural spheroid platform for modeling glioblastoma infiltration using high-content imaging","fulltext":[{"header":"Introduction","content":"\u003cp\u003eGlioblastoma (GB) is the most common and aggressive form of primary brain tumor in adults, with a median overall survival of just 12 to 15 months following diagnosis and standard-of-care treatment with temozolomide \u003csup\u003e\u003cspan additionalcitationids=\"CR2 CR3 CR4 CR5 CR6 CR7\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. GB is defined by its IDH1/IDH2-wildtype status and classified as a World Health Organization (WHO) grade IV tumor, reflecting its high malignancy \u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. A major barrier to effective treatment is the pronounced inter- and intra-tumoral heterogeneity in genetic mutations and cellular phenotypes, which complicates therapeutic targeting and drives drug resistance \u003csup\u003e\u003cspan additionalcitationids=\"CR10\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. Moreover, the highly infiltrative nature of GB precludes complete surgical resection and temozolomide treatment, promoting tumor recurrence \u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eA key contributor to the aggressive invasiveness of GB is a subpopulation of neoplastic cells called the GB stem-like cells (GSCs) \u003csup\u003e\u003cspan additionalcitationids=\"CR15 CR16\" citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. GSCs exhibit hallmark features of stemness, including self-renewal and multipotency, and are found at the tumor core and in infiltrative margins, where they play a key role in tumor recurrence and therapy resistance \u003csup\u003e\u003cspan additionalcitationids=\"CR18 CR19\" citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. Most importantly, GSCs are highly invasive and utilize conserved mechanisms of invasion, emphasizing their critical role in disease progression. Their invasive capacity is mediated by intrinsic processes, including cytoskeletal remodeling and upregulation of surface receptors that regulate adhesion and motility \u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. These mechanisms facilitate GSC migration along two primary routes, the vascular endothelium and axonal tracts within the white matter of the brain, although the mechanisms underlying the route selection remain unclear \u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eAn increased understanding of the prognostic significance of tumor invasion has driven the development of advanced \u003cem\u003ein vitro\u003c/em\u003e models that more accurately recapitulate GB complexity for drug screening purposes \u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. Among these, patient-derived GSC lines have emerged as a valuable research tool. These lines faithfully preserve the stem cell-like properties, genetic landscape, and transcriptional profile of the parental tumor, even after extended passaging, thereby maintaining key features of the \u003cem\u003ein vivo\u003c/em\u003e tumor phenotype \u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e,\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. Particularly, patient-specific GSCs have been used to generate physiologically relevant model systems, including glioma spheroids and GB cerebral organoids that better represent the complexity of GB and the tumor microenvironment \u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eHigh-content imaging combines automated microscopy with quantitative single-cell analysis, allowing detailed, large-scale assessment of cellular phenotypes \u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e,\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. In cancer research, high-content imaging enables dynamic studies of cell behaviors such as migration, especially when using label-free modalities like phase contrast to avoid phototoxicity during long-term live-cell assays \u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e,\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e. These advancements, alongside computational and multi-omics approaches, have significantly enhanced drug screening and target discovery in complex tumor models like GB.\u003c/p\u003e\u003cp\u003eIn this study, we present a novel \u003cem\u003ein vitro\u003c/em\u003e co-culture system that combines patient-derived GB cells and axons extending from neural spheroids derived from human iPSCs. This model allows for the interaction between GB cells and neuronal axons, providing a more physiologically relevant microenvironment for the study of GB cells migration and infiltration, which are critical processes driving GB recurrence. Using high-content imaging, we reveal significant differences in the migration patterns of two different GB cell lines and provide insights into the mechanisms underlying GB cell line-specific infiltration strategies, with potential implications for anti-infiltrative therapies. Additionally, we correlate migration phenotypes with drug responses identified through a screening of relevant inhibitors. This drug screen provides proof-of-principle that migration is a measurable phenotype with potential for guiding precision therapies against GB infiltration.\u003c/p\u003e"},{"header":"Results","content":"\u003ch2\u003ePatient-derived GB cells migrate along axons toward the neural spheroid in vitro\u003c/h2\u003e\n\u003cp\u003eTo generate an \u003cem\u003ein vitro\u003c/em\u003e model suitable for the characterization of infiltration of GB cells and amenable to drug-screening studies, we combined the culture of patient-derived GB cell lines with human iPSC-derived neural spheroids with radially extending axons. Wildtype human iPSCs are differentiated into excitatory cortical-like neurons using NGN2-mediated forward programming differentiation\u0026nbsp;\u003csup\u003e30\u003c/sup\u003e. Following the induction phase, early neuronal progenitors are placed into non-adherent Pluronic acid-coated V-bottom plates where they self-aggregate into individual spheroids of reproducible size and shape (Figure 1A). Then, spheroids are transferred onto laminin-coated plates where they adhere to the culture substrate and extend long, radially organized neurites (Figure 1B). Immunofluorescence staining performed on the spheroids revealed positive expression of the axonal marker \u0026beta;III-tubulin (Tubb3) in the projecting neurites (Figure 1B), validating the generation of a neuronal substrate suitable for GB co-culture assays.\u003c/p\u003e\n\u003cp\u003eHaving established a co-culture of neural spheroids with emerging axons, we next examined how two different patient-derived GB cell lines, GBM1 and GBM20, behave in this axon-rich microenvironment. GBM1, first described by Wurdak et al., was derived from a primary, treatment-na\u0026iuml;ve tumor\u0026nbsp;\u003csup\u003e9,31\u003c/sup\u003e, while GBM20, reported by Polson et al., was obtained from a recurrent tumor previously treated with chemoradiotherapy and the IMA950 peptide vaccine\u0026nbsp;\u003csup\u003e31\u003c/sup\u003e. A human neural stem cell line called NS17 was examined in parallel as a non-tumorigenic control with features of\u0026nbsp;self-renewal and multipotency\u0026nbsp;\u003csup\u003e32\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eLive-cell imaging of the GB and NS17 lines revealed that all three cell lines aligned with and migrated along neuronal axons, indicating a direct interaction between the cells and the axons (Figure 2 and supplementary videos). To determine whether the direction of cell migration along axons was random or biased toward the neural spheroid, we quantified the directionality of individual cell tracks (Figure 3A). In the absence of axons, migration was isotropic, as shown by the uniform distribution of the polar histograms. In contrast, cells plated on axons exhibited directional migration along axons, with a clear enrichment of movement toward the neural spheroid positioned on the right side of the imaging field. This directional bias was evident across all cell lines, including the non-cancerous NS17 control (10% of cells moving at an angle 0\u003csup\u003eo\u003c/sup\u003e), and was most pronounced in GBM20 (12% of cells moving at an angle 0\u003csup\u003eo\u003c/sup\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eA significant increase in instantaneous velocity was observed when all three cell lines were cultured on axons compared with axon-free conditions (****, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.0001) (Figure 3B). Interestingly, a change in overall mean speed reached significance only for GBM1 cells on axons, with no such effect on the other cell lines. Although migration directionality and velocity were strongly influenced by the presence of neuronal axons, the confinement ratio remained unchanged between axon-present and axon-absent conditions for all cell lines (Figure 3B). These findings suggest that axons function as a permissive, orienting substrate that enhances guidance cues, particularly for GBM1 cells. Moreover, the increased variability in directionality and velocity on axons highlights the heterogeneity of cell responses and supports the need for further investigation into temporal dynamics and GB subtype-specific migration mechanisms.\u003c/p\u003e\n\u003ch2\u003eGB cells exhibit cell line-specific infiltration in neural spheroid co-culture\u003c/h2\u003e\n\u003cp\u003eWe hypothesized that cells migrating toward the neural spheroid would subsequently infiltrate it. To assess the infiltration potential of patient-derived GB cells in comparison to non-tumorigenic neural stem cells, GBM1, GBM20, and NS17 cells were labelled with a permeable cytoplasmic dye, added to neural spheroids in culture, and imaged every day for 5 days. A customized image analysis pipeline was applied to endpoint images acquired from Day 1 to 5 to quantify the number of infiltrating cells across conditions (Figure 4A).\u003c/p\u003e\n\u003cp\u003eAll three cell lines showed a progressive increase in the number of infiltrated cells over time (Figure 4B). On Day 5, NS17 cells reached an average infiltration of 6.5 \u0026plusmn; 0.9 cells per spheroid, whereas GBM1 and GBM20 demonstrated significantly greater infiltration, with 17.2 \u0026plusmn; 0.6 and 11.9 \u0026plusmn; 1.6 cells, respectively. Statistically significant differences emerged by Day 2, at which point GBM1 cells had infiltrated the spheroids at a higher rate (5.5 \u0026plusmn; 1.0 cells) compared to NS17 (1.6 \u0026plusmn; 0.4 cells). Normal distribution analysis of cell infiltration values revealed progressive divergence between patient-derived GB lines and the non-tumorigenic control (Figure 4C). On Day 3, data distribution showed a clear separation between NS17 and GBM1/GBM20, with a calculated Z-factor of 0.701, indicating excellent assay robustness and strong discriminatory power suitable for screening applications.\u0026nbsp;While the Z-factor decreased to 0.218 by Day 4, reflecting increased variability or overlap in infiltration measurements, it improved again on Day 5, reaching 0.575, consistent with a reliable and moderately robust assay window (See also Figure S1A).\u0026nbsp;These data suggest that Day 3 represents the optimal time point for distinguishing differential infiltration phenotypes in this model, with potential applicability for high-throughput anti-invasion compound screening.\u003c/p\u003e\n\u003ch2 id=\"_Toc128035599\"\u003eGB and neural spheroid co-cultures enable identification of tumor-specific infiltration inhibitors\u003c/h2\u003e\n\u003cp\u003eHaving developed a quantitative pipeline to measure GB cell migration along axons and infiltration into the neural spheroid core, we next used this co-culture platform for a pilot drug screen. To investigate mechanisms regulating GB infiltration, a panel of 18 inhibitors (Table 1) targeting cytoskeletal dynamics, kinase signaling, and axonal guidance pathways that mediate glioma invasion was selected based on literature and clinical relevance \u0026nbsp;\u003csup\u003e15,17,33\u003c/sup\u003e. This included established cytoskeletal disruptors such as inhibitors to actin polymerization (Latrunculin B), myosin II ATPase (Blebbistatin), and ROCK1/2 (Y-27632), commonly used to reduce cell motility, along with targeted canonical GB signaling pathways involved in cell migration and invasion, including inhibitors to TGF-\u0026beta; receptor (SB 431542), PI3K/Akt (API-2), Protein Kinase C (GF 109203X), and integrins (Cilengitide), pathways often targeted in clinical evaluation\u0026nbsp;\u003csup\u003e34\u0026ndash;38\u003c/sup\u003e. Additional targets included non-canonical regulators such as hedgehog (HPI-1) and Wnt (IWP-2) pathways, and receptor tyrosine kinases, including FAK (PF 573228), FGFR (PD 173074), PDGFR (AC 710), and EGFR (Iressa)\u0026nbsp;\u003csup\u003e39\u0026ndash;41\u003c/sup\u003e. Given the importance of axon-guided migration in GBM, we included inhibitors of Ephrin receptors (ALW-II-41-27 for EphA2, NVP-BHG712 for EphB4, and Ehp-inhibitor-1), and the CXCL12/CXCR4 axis (Motixafortide), known to regulate directed migration in the hypoxic tumor core in immunotherapy\u0026nbsp;\u003csup\u003e42,43\u003c/sup\u003e. Finally, we tested repurposed drugs such as pranlukast (ONO-1078), a leukotriene receptor antagonist shown to impair GB spread along white matter tracts in preclinical models\u0026nbsp;\u003csup\u003e44\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eOn Day 0, GB and NS17 cells were labelled and seeded in co-culture with the neural spheroids. Inhibitors were applied at 1 \u0026mu;M on Days 1 to 3, and cultures were imaged each day after 6 hours of incubation. Representative images illustrate the increasing accumulation of GBM1 cells within and around the spheroid from Day 1 to Day 3 in the untreated controls and with Motixafortide treatment (Figure 5A). As expected from previous observations, both GBM1 and GBM20 cells showed progressive infiltration into the neural spheroid over three days, and inhibitor-specific effects became apparent by Day 3. For the GBM20 line, significant reduction of infiltration was observed with PF 573228 (**, \u003cem\u003ep\u003c/em\u003e = 0.0043), Y-27632 (*, \u003cem\u003ep\u003c/em\u003e = 0.0244), Motixafortide (*, \u003cem\u003ep\u003c/em\u003e = 0.0232), and GF 109203X (*, \u003cem\u003ep\u003c/em\u003e = 0.0262) (Figure 5B). Conversely, only Latrunculin B (*, \u003cem\u003ep\u003c/em\u003e = 0.0386) and Motixafortide (*, \u003cem\u003ep\u003c/em\u003e = 0.0293) significantly reduced GBM1 infiltration (Figure 5B). Importantly, none of the tested inhibitors significantly affected infiltration of the non-cancerous NS17 neural stem cells (See also Figure S1B). Their infiltration remained consistently low and comparable to vehicle controls, underscoring the specificity of the drug responses observed in the GB models.\u003c/p\u003e\n\u003ch2\u003eGBM1 and GBM20 exhibit distinct transcriptomic signatures that align with differential drug sensitivities\u003c/h2\u003e\n\u003cp\u003eTo investigate the molecular basis underlying the different infiltrative behaviors and drug responses of GBM1 and GBM20 observed in our co-culture system, we conducted bulk RNA-sequencing followed by differential gene expression analysis. The two GB lines exhibited distinct transcriptomic profiles, as visualized in a volcano plot. GBM20 showed elevated expressions of \u003cem\u003ePCOLCE\u003c/em\u003e, \u003cem\u003eGABBR2\u003c/em\u003e, \u003cem\u003eERAP2\u003c/em\u003e, \u003cem\u003ePCDHB5\u003c/em\u003e, \u003cem\u003eERAP2\u003c/em\u003e and \u003cem\u003eCCL2\u003c/em\u003e, while GBM1 was characterized by higher expressions of \u003cem\u003eCOL1A2\u003c/em\u003e, \u003cem\u003eMGST1\u003c/em\u003e, \u003cem\u003eCDKN2A\u003c/em\u003e, \u003cem\u003eMCAM\u003c/em\u003e, and \u003cem\u003eRGMB\u003c/em\u003e, suggesting distinct biological programs (Figure\u0026nbsp;6A).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eGene Ontology (GO) enrichment analysis of significantly dysregulated genes (adjusted p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, |log₂\u0026nbsp;FC|\u0026thinsp;\u0026gt;\u0026thinsp;1) further highlighted these differences (Figure\u0026nbsp;6B). GBM1 was enriched for neurodevelopmental programs such as\u0026nbsp;\u0026ldquo;positive regulation of nervous system development\u0026rdquo;\u0026nbsp;and\u0026nbsp;\u0026ldquo;axonogenesis,\u0026rdquo;\u0026nbsp;as well as cell adhesion and guidance processes (\u0026ldquo;cell\u0026ndash;cell adhesion via plasma membrane adhesion molecules\u0026rdquo;\u0026nbsp;and\u0026nbsp;\u0026ldquo;axon guidance\u0026rdquo;) and cilium-associated programs (\u0026ldquo;cilium assembly\u0026rdquo;\u0026nbsp;and\u0026nbsp;\u0026ldquo;cilium organization\u0026rdquo;). In contrast, GBM20 showed enrichment for extracellular matrix (ECM) organization pathways (\u0026ldquo;ECM structural organization\u0026rdquo; and \u0026ldquo;external encapsulating structure organization\u0026rdquo;) and genes involved in neurite outgrowth and synapse formation (\u0026ldquo;pattern specification process\u0026rdquo; and \u0026ldquo;regulation of neuron projection development\u0026rdquo;).\u003c/p\u003e\n\u003cp\u003eTo explore how these transcriptional differences may relate to drug responses, we conducted pre-ranked Gene Set Enrichment Analysis (GSEA) using MSigDB gene sets (GO, KEGG, Reactome, and Hallmark). In the drug screen, infiltration of GBM1 cells was significantly reduced by Latrunculin B (actin polymerization inhibitor) and Motixafortide (CXCR4 antagonist). Correspondingly, GSEA revealed enrichment of actin remodeling and wound healing pathways in GBM1, suggesting a dependence on actin-driven motility. Enrichment of immune-related pathways, including PD-1 signaling and antigen presentation, also supports involvement of the CXCL12\u0026ndash;CXCR4 axis, targeted by Motixafortide (Figure 6C). GBM20 infiltration was significantly inhibited by GF 109203X (Protein Kinase C inhibitor) and Y-27632 (ROCK inhibitor). GSEA revealed enrichment of presynaptic organization and oxidative stress\u0026ndash;related pathways in GBM20, consistent with ROCK- and PKC-mediated cytoskeletal remodeling and redox signaling dependencies (Figure 6C).\u0026nbsp;\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eWe present a physiologically relevant \u003cem\u003ein vitro\u003c/em\u003e co-culture model combining patient-derived GB cells with human iPSC-derived neural spheroids. These spheroids form a dense core of neuronal bodies with radially extending axons, allowing a reproducible and scalable assessment of GB cell infiltration. Using high-content imaging, we established automated pipelines to quantify axon engagement, directional migration in live assays, and spheroid infiltration in endpoint assays. Both GBM1 and GBM20 actively migrated toward neural spheroids, but GBM1 showed a higher infiltration capacity. A focused drug screen revealed line-specific sensitivities, with four compounds reducing GBM1 infiltration and two affecting GBM20, establishing proof-of-principle for our model as a platform for identifying anti-infiltration compounds. Transcriptomic profiling further highlighted distinct gene expression programs underpinning the observed migratory phenotypes and drug responses. This integrative approach correlates cell behavior with molecular signatures, laying the foundation for functional precision-medicine studies in GB.\u003c/p\u003e\u003cp\u003eWe demonstrated that both GB lines and a non-tumorigenic neural stem cell line (NS17) migrated toward neural spheroids, but only the GB lines infiltrated them. This suggests that while neural-derived cues attract all lines, only GB cells possess the mechanism for infiltration. Consistent with prior work, we observed that GBM1 infiltrated more readily compared to neural progenitors \u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e. However, our previous findings reported greater aggressiveness with patient-derived GBM20 spheroids spontaneously fusing and infiltrating human cerebral organoids at a faster rate than GBM1 spheroids \u003csup\u003e\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e. These differences may reflect shifts in phenotype following treatment, as GBM20 is derived from a recurrent tumor that was treated with radiotherapy and chemotherapy. Notably, recurrent GB cells have been shown to adopt less proliferative, pre-oligodendrocyte-like states in response to damage, potentially explaining the reduced invasiveness \u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eOur model also identified compounds that differentially impair GB cell infiltration. GBM20 was sensitive to inhibition of FAK (PF 573228), PKC (GF 109203X), and ROCK (Y-27632), while GBM1 responded to actin polymerization inhibition (Latrunculin B). Both lines were sensitive to inhibition of CXCR4 (Motixafortide), suggesting reliance on CXCL12/CXCR4 signaling, a known target in GB immunotherapy \u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e. Transcriptomic analyses revealed expression patterns consistent with drug sensitivity. GBM20 showed upregulation of FAK pathway\u0026ndash;related genes (\u003cem\u003eSRC\u003c/em\u003e, \u003cem\u003eITGA5\u003c/em\u003e, \u003cem\u003eITGB3\u003c/em\u003e, \u003cem\u003ePPFIA1\u003c/em\u003e and \u003cem\u003ePRKCB\u003c/em\u003e), aligning with sensitivity to PF 573228 and GF 109203X. Previous studies support these findings, as GF 109203X effectively targeted Protein Kinase C in GB \u003csup\u003e47\u003c/sup\u003e and PF 573228 has also reduced the migration of established GB cell lines, T98G and U-118, in other models \u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e. High \u003cem\u003eCXCL12\u003c/em\u003e expression in GBM1 matched its higher sensitivity to Motixafortide with an increased reduction in the number of infiltrated cells in GBM1 compared to GBM20.\u003c/p\u003e\u003cp\u003eGSEA revealed additional mechanistic differences: GBM1 showed enrichment for actin cytoskeleton remodeling and immune-related pathways, supporting a model of CXCR4- and actin-driven motility. GBM20 exhibited enrichment for synaptic signaling and oxidative stress pathways, consistent with its dependence on ROCK/PKC signaling. These findings reinforce the value of transcriptomics-informed migration assays in uncovering subtype-specific vulnerabilities. Differential drug responses may reflect distinct pathway dependencies or compensatory mechanisms, underscoring the need for tailored therapeutic strategies.\u003c/p\u003e\u003cp\u003eThe alignment between transcriptional programs and drug responses supports behavior-based functional stratification as a complementary strategy to conventional molecular profiling. This approach offers a powerful tool for evaluating patient-specific vulnerabilities and complements recent advances using machine learning and high-throughput drug screening \u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e. Our results not only highlight divergent invasive programs between GBM1 and GBM20 but also demonstrate the utility of patient-derived co-culture systems in identifying functionally relevant pathways. This work provides proof-of-principle that migration can serve as a robust, measurable, and functionally relevant phenotype to screen therapeutic vulnerabilities in GB.\u003c/p\u003e\u003cp\u003eLimitations and future direction\u003c/p\u003e\u003cp\u003eThis pilot study involved only two GB lines and a single control, limiting broader applicability. Expanding the model to include a diverse panel of patient-derived lines will enhance its predictive utility for therapeutic development. Future improvements include incorporating iPSC-derived oligodendrocytes to mimic myelinated tracts, as glioma cells preferentially migrate on myelinated axons \u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e,\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e. Moreover, while phase-contrast imaging enabled live tracking of GB migration and pseudopodia formation \u003csup\u003e\u003cspan additionalcitationids=\"CR51\" citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e, it was not optimal for imaging infiltration into dense neural spheroids. Implementing high-content Z-stack imaging platforms such as the Operetta CLS permitted higher-throughput and higher-resolution quantification of infiltration dynamics.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eAlthough glioma growth has been widely studied, its invasion and migration remain poorly understood. Our co-culture model, pairing patient-derived GB cells with axon-extending human neural spheroids, provides a physiologically relevant platform for migration-based drug screening. By integrating high-content imaging, drug screening, and transcriptomic data, we captured cell line–specific behaviors and subtype-dependent migratory mechanisms. These findings underscore the heterogeneity of GB invasion and demonstrate the potential of our platform to guide personalized therapeutic strategies.\u003c/p\u003e"},{"header":"Methods ","content":"\u003ch2\u003eCell culture\u003c/h2\u003e\n\u003ch3\u003eGB cells culture\u003c/h3\u003e\n\u003cp\u003eThe patient-derived GBM1 and GBM20 cell lines were derived under informed consent according to ethical approvals (LREC 115/ES/0094; courtesy of Dr Heiko Wurdak, University of Leeds\u0026nbsp;\u003csup\u003e9,31\u003c/sup\u003e. GBM1 and GBM20 cell lines were derived respectively from a primary tumor of a 58-year-old female patient and a 50-year-old male patient with a recurrent GB, previously treated with radiotherapy, temozolomide, and IMA950. The cell models maintain the stem cell-like characteristics of GB under adherent culture conditions\u0026nbsp;\u003csup\u003e9,31,45\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eCells were cultured 5 μg/mL poly-L-ornithine (PLO) (Sigma) and 2 μg/mL laminin (Thermofisher) coated plates in Neurobasal medium (Invitrogen)\u0026nbsp;supplemented with 0.5X B27 (Invitrogen), 0.5X N2 (Invitrogen), 40 ng/mL epidermal growth factor (EGF) (R\u0026amp;D Systems), and 40 ng/mL basic-fibroblast growth factor (b-FGF) (Invitrogen).\u0026nbsp;Cells were passaged using 1X TrypLE (Gibco) when approximately 80% confluency was reached, and passages remained between passage numbers 14 to 20.\u0026nbsp;Cell cultures were routinely tested for mycoplasma, and all were negative for contamination.\u003c/p\u003e\n\u003ch3 id=\"_Toc138265418\"\u003eNeural stem cells\u003c/h3\u003e\n\u003cp\u003eThe cells GCGR.NS17ST_A (shortened to NS17) from the Glioma Cellular Genetics Resource (GCGR) was acquired under informed consent according to local ethical approvals (08/S1101/1)\u0026nbsp;\u003csup\u003e32\u003c/sup\u003e and kindly provided by Dr Steven Pollard (University of Edinburgh). The cells were cultured according to previously published methods DMEM/HAMS-F12 medium (Sigma) supplemented with 1.5 mg/mL of glucose (Sigma), 100 μM MEM NEAA (Gibco), 50 U/mL and 50 mg/mL Pen Strep (Gibco), 0.01% BSA (Gibco), 100 μM beta-mercaptoethanol (Gibco),\u0026nbsp;0.5X B27, 0.5X N2,\u0026nbsp;EGF to 10 ng/mL, and b-FGF to 10 ng/mL. The cells were passaged using StemPro Accutase Cell Dissociation Reagent (Gibco) when they reached approximately 80% confluency at a split no lower than 1:6.\u0026nbsp;\u003c/p\u003e\n\u003ch3 id=\"_Toc138265419\"\u003eInduced glutamatergic neuron cell culture\u003c/h3\u003e\n\u003cp\u003eHPSI1013I-PAMV_1 human iPSCs obtained from the HipSci biobank, where\u0026nbsp;pluripotency characterization, such as Pluritest, was performed as part of their standard quality control pipelines (\u003ca href=\"http://www.hipsci.org\"\u003ewww.hipsci.org\u003c/a\u003e). The line was derived from skin fibroblasts from a healthy male donor aged 65-69 from a White British ethnicity. HPSI1013I-PAMV_1 cells were transduced with a doxycycline-inducible Neurogenin-2 (iNGN2) transgene following the protocol from the Ward laboratory\u0026nbsp;\u003csup\u003e30\u003c/sup\u003e. The integration of the transgene was achieved by electroporation and TALEN-mediated integration into the AAVS1 safe-harbor locus.\u0026nbsp;iNGN2\u0026nbsp;PAMV_1 cells were cultured in feeder-free conditions on 4% Vitronectin XF (Stem Cell Technologies) coated plates in Essential 8 medium (Gibco) supplemented with 50 U/mL and 50 mg/mL Penicillin/Streptomycin respectively (Life Technologies). The medium was changed daily, and cells were passaged using TryplE and resuspended in Essential 8 medium with 10 μM Y-27632 Rho-kinase (ROCK) inhibitor (ENZO Life Sciences) at approximately 1.5 million cells per well.\u0026nbsp;All cell cultures were routinely screened for mycoplasma and confirmed to be negative for contamination.\u003c/p\u003e\n\u003ch3 id=\"_Toc138265420\"\u003eTranscription factor-mediated differentiated CNS-like glutamatergic neurons\u003c/h3\u003e\n\u003cp\u003eThe transcription factor-mediated differentiation protocol was adapted from the Ward laboratory\u0026nbsp;\u003csup\u003e30\u003c/sup\u003e. On Day 0, iNGN2 PAMV_1 cells were dissociated using Accutase at 37°C for 5 minutes. The cells were resuspended with the Induction Medium - DMEM/F12 Hepes medium (Life Technologies), supplemented with 0.5X N2, 100 μM MEM NEAA, 2 mM L-Glutamine (Gibco), Doxycycline (2 μg/ml) (Sigma), and 10 μM Y-27632 ROCKi (1:100), on GFR Matrigel (BD Biosciences) (1:50) at a density of 1.5 million cells per well. Doxycycline will induce the differentiation into Glutamatergic Excitatory Neurons. On Days 1 and 2, nascent neurites began to be evident, and the media was replaced with fresh induction media with Doxycycline. Cells were dissociated with Accutase for freezing in 10% dimethyl sulfoxide (Sigma) in medium or for neural spheroid generation.\u0026nbsp;\u003c/p\u003e\n\u003ch3 id=\"_Toc138265421\"\u003eGeneration of neural spheroids\u003c/h3\u003e\n\u003cp\u003eFollowing dissociation, differentiated cells were resuspended in the induction media and were seeded at a density of 10,000 cells/well in the 5% Pluronic F127 (Sigma-Aldrich) coated V-bottom plate and centrifuged for 2 min at 200 x g to aggregate the cells at the bottom of the well according to the in-house protocol\u0026nbsp;\u003csup\u003e53\u003c/sup\u003e. The cells were left for 48 h with daily half-medium changes.\u003c/p\u003e\n\u003ch3 id=\"_Toc138265422\"\u003eMaturation and generation of axon bundles\u003c/h3\u003e\n\u003cp\u003eOn Day 6, the neural spheroids were transferred to laminin-coated 24-well or 96-well glass-bottom plates (Cellvis) in Cortical Neuron Culture Medium - BrainPhys neuronal medium (Stemcell Technologies) supplemented with 0.5X B27, 10 ng/mL Brain-Derived Neurotrophic Factor (BDNF) (PeproTech), 10 ng/mL Neurotrophin-3 (NT-3) (PeproTech), and 1 μg/mL Laminin. The cells were checked daily for the presence of cell debris and morphological changes with bi-weekly half-medium changes.\u003c/p\u003e\n\u003ch3 id=\"_Toc138265423\"\u003eCo-culture of GB cells with neural spheroids\u003c/h3\u003e\n\u003cp\u003eOn Day 11, 1X BioTracker 488 Green Nuclear Dye (Sigma-Aldrich) was added to the cell culture medium to stain the neural spheroids for 20 min at 37°C and washed twice. GB cells were labelled with 1X BioTracker 655 Red Cytoplasmic Membrane Dye (Sigma-Aldrich) in suspension for 20 min at 37°C and washed three times by centrifugation. The GB cells were seeded on top of the neural spheroids at a density of 5,000 cells/well in a 24-well plate or 840 cells/well in a 96-well plate with a 50/50 medium.\u003c/p\u003e\n\u003ch2 id=\"_Toc138265427\"\u003eStaining and Imaging\u003c/h2\u003e\n\u003ch3\u003eImmunostaining\u003c/h3\u003e\n\u003cp\u003eNeural spheroids were fixed at 4% PFA/PBS at RT for 20 min and permeabilized with 3% BSA/0.1% Triton-X-100/PBS for 1 h at RT. Cells were incubated with the primary mouse axonal antibody βIII-tubulin (Tubb3) (R\u0026amp;D Systems) diluted in 3% BSA/0.1% Triton-X-100/PBS (1/1000) overnight at 4°C in the dark. The next day, primary antibodies were rinsed away with 3% BSA/PBS. The cells were incubated with the secondary anti-mouse antibody conjugated to AF 488 (Invitrogen), diluted in 3% BSA/0.1% Triton-X-100/PBS (1:500) at RT for 45 min and rinsed with PBS. To stain the nucleus, the cells are incubated with 10 μM Hoechst (Invitrogen) for 10 min at RT and rinsed with PBS. Finally, the cells were incubated with 80% glycerol as a clearing agent for 1 h min at RT and rinsed with PBS before being placed in fresh PBS at 4°C until imaging.\u003c/p\u003e\n\u003ch3\u003eConfocal microscopy\u003c/h3\u003e\n\u003cp\u003eConfocal images were acquired using a Leica TCS SP8 Confocal laser scanning microscope, using a 10x dry objective, and viewed with the Leica software. Each fluorochrome was excited with the corresponding laser line (DAPI, UV (355 nm); AF488, Green (530 nm); AF594, Red (639 nm) laser line). Immunofluorescence data were analyzed in Fiji open-source software (2.0.0-rc-64/1.51s version). Images from different fields were tiled and stitched, and the maximum projection was obtained using the Z-stack.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eImaging assay\u003c/h2\u003e\n\u003ch3 id=\"_Toc138265432\"\u003eLive-cell imaging of non-labelled cells\u003c/h3\u003e\n\u003cp\u003eLive-cell imaging of non-labelled cells on laminin-coated glass plates and on axons was performed on the Livecyte quantitative phase imager (Phasefocus), which uses quantitative phase imaging and has an accurate tracking apparatus. ROIs (10 mm x 25 mm) on the right and the top of the neural spheroids were selected, and images were acquired every 15 min for 24 h.\u0026nbsp;\u003c/p\u003e\n\u003ch3 id=\"_Toc138265433\"\u003eEndpoint assay\u003c/h3\u003e\n\u003cp\u003eCo-cultures of patient-derived GB cells and human iPSC-derived cortical-like neural spheroids were imaged every 24 h at Day 0, 1, 2, and 3 using the Operetta CLS High Content Analysis (PerkinElmer) with the brightfield, 488 nm, and 655 nm channels. The PreciScan Intelligent Acquisition plug-in for Harmony software was used to locate the neural spheroid within the well. The plug-in allows the accurate targeting of the region of interest (ROI) whilst reducing acquisition and analysis times. Once the PreciScan was performed and the region of the neural spheroid was located, a Z-stack of 10 images over 20 μm (distance 2 μm) was acquired.\u003c/p\u003e\n\u003ch3 id=\"_Toc138265434\"\u003eInhibitor screen\u003c/h3\u003e\n\u003cp\u003eInhibitors were added to the co-culture at a final concentration of 1 μM. Following a 6 h antagonist treatment, images were acquired using the Operetta CLS Imaging System. This was repeated for the next two days.\u003c/p\u003e\n\u003ch2 id=\"_Toc138265435\"\u003eImage analysis\u003c/h2\u003e\n\u003ch3 id=\"_Toc138265437\"\u003eLive-cell imaging assay analysis using the Phasefocus Livecyte\u003c/h3\u003e\n\u003cp\u003eImages were analyzed using the built-in Analysis Software with the Motility Assay on the Livecyte quantitative phase imager (Phasefocus). Using the built-in applications, the cells were identified and tracked, and the tracking properties were calculated to generate cell migration parameters. Directionality results were analyzed in a polar histogram using R Studio (version 2025.05.1). Analyzes of instantaneous velocity, speed, and confinement ratio were performed using GraphPad Prism software, and the Welch’s t-test was performed between cells plated with or without axons. A p-value below 0.05 was considered significant and was indicated with an asterisk: ****, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.0001.\u003c/p\u003e\n\u003ch3 id=\"_Toc138265439\"\u003eEndpoint assay analysis using Harmony\u003c/h3\u003e\n\u003cp\u003eImages were analyzed using the built-in Harmony HCI and Analysis Software on the Operetta CLS device. The maximum projection of the Z-stack was taken, and the neural spheroid was stained with the BioTracker 488 and identified as an ROI using the green channel. The GB cells stained with the BioTracker 655 in the ROI were identified and segmented using the far-red channel (Segmentation Method C in Harmony Software).\u003c/p\u003e\n\u003cp\u003eAll experiments were performed in technical triplicate and were independently repeated at least three times to perform statistical analysis. All statistical analyses were performed using GraphPad Prism software (version 9.2). Results represented as means with standard deviation (SD), and different ANOVA tests with multiple comparisons between control and other conditions were performed to calculate the statistical significance of multiple experimental conditions. A p-value below 0.05 was considered significant and was indicated with an asterisk: *, \u003cem\u003ep\u003c/em\u003e ≤ 0.05; **,\u003cem\u003e\u0026nbsp;p\u003c/em\u003e ≤0.01; ***, \u003cem\u003ep\u0026nbsp;\u003c/em\u003e≤0.001. The number of independent experiments is indicated in the figure legend as N.\u003c/p\u003e\n\u003ch2\u003eRNA sequencing\u003c/h2\u003e\n\u003cp\u003eThe GBM1 and GBM20 cell lines were prepared for RNA sequencing. They were rinsed once with PBS, detached with TrypLE, and snap-frozen in pellets of 10\u003csup\u003e6\u003c/sup\u003e cells. The cells were submitted to Eurofins Genomics, Germany, under Project ID: NG-29040. Raw sequencing data were pre-processed using the fastp software to generate clean data, termed quality control. This involves checking the quality of the raw sequencing filtering for high-quality reads to remove poor-quality bases (below Phred Quality 20)\u003csup\u003e54\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eHigh-quality sequence reads were aligned using the STAR (Spliced Transcripts Alignment to a Reference) to the reference genome UCSC\u0026nbsp;Homo sapiens\u0026nbsp;version hg38\u003csup\u003e55\u003c/sup\u003e. Gene-wise quantification was achieved to inspect the transcriptome alignments using the RSEM tool\u0026nbsp;\u003csup\u003e56\u003c/sup\u003e. For the differential gene expression between cell lines, genes with fewer than 10 average reads were removed. Using the R/Bioconductor DESeq2 package, the abundance counts of each gene were then used to perform differential gene expression\u0026nbsp;\u003csup\u003e57\u003c/sup\u003e. Eurofins Genomics provided this pre-processing.\u003c/p\u003e\n\u003ch2\u003eRNA-seq data analysis\u003c/h2\u003e\n\u003cp\u003eFurther analysis was performed on R Studio using the ggplot2 package to generate a volcano plot of differential gene expression. Sample-wise comparison values (log₂ fold change and p-value) provided by Eurofins Genomics were used. Gene Ontology (GO) enrichment and Gene Set Enrichment Analysis (GSEA) were performed using the clusterProfiler and fgsea packages to identify biological processes and pathways associated with differential expression. Significantly enriched terms were visualized using bar plots and enrichment plots.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eAcknowledgements \u003c/p\u003e\n\u003cp\u003eWe appreciate the support, advice, and reagents provided by everyone at CSCRM. We are grateful to Dr Bronwyn Irving, Dr James Williams, and Dr L\u0026eacute;a R\u0026rsquo;Bibo for their valuable expertise in GB cell culture, stem cell culture, and neural differentiation, respectively. We thank Thomas Williams, Dr Lazaros Fotopoulos, and Erika Wiseman at the Stem Cell Hotel for their expert assistance and management of the imaging facilities. We would also like to thank Ioanna Kourouzidou, Oluwaseun Adegbite, Layla Kadhim for their involvement in preliminary experiments. Furthermore, we acknowledge Dr Andrea Serio, Dr Ciro Chiappini and Dr Steve Pollard for their valuable support and input in the project initiation. \u003c/p\u003e\n\u003cp\u003eHipSci Lines samples were collected from consented research volunteers recruited from the NIHR Cambridge BioResource through https://www.cambridgebioresource.group.cam.ac.uk/. The HipSci consortium obtained ethics approval for a revised consent (REC ref. 09/H0304/77, V3 15/03/2013), under which all data, except for the Y chromosome from males, can be made openly available (Y chromosome data can be used to de-identify men by surname matching). For open access, the author has applied a CC BY public copyright license to any author-accepted manuscript version arising from this submission.\u003c/p\u003e\n\u003cp\u003eFunding\u003c/p\u003e\n\u003cp\u003eV.T. and F.R. were funded by the Wellcome Trust as part of the \u0026ldquo;Cell Therapies \u0026amp; Regenerative Medicine\u0026rdquo; PhD Programme (108874/Z/15/Z). The authors acknowledge financial support Rosetrees Trust to D.D., from the Medical Research Council (grant MR/N025865/1) to I.L, and from Innovate UK (TSB/89370) from J.D.C and D.D.\u003c/p\u003e\n\u003cp\u003eAuthor contributions\u003c/p\u003e\n\u003cp\u003eV.T., D.D., and I.L., conceptualization; V.T. and F.R., data curation; V.T., F.R., I.L., and D.D., investigation; V.T., F.R., I.L., and D.D., funding acquisition; V.T., formal analysis; V.T., F.R., H.N., and J.C., methodology; D.D. and I.L., supervision; V.T., and D.D., visualization; V.T., and D.D., writing \u0026ndash; original draft; V.T., F.R., I.L., H.N., H.B., and D.D., writing \u0026ndash; review \u0026amp; editing. All authors contributed to the article and approved the submitted version. \u003c/p\u003e\n\u003cp\u003eData availability\u003c/p\u003e\n\u003cp\u003eProcessed data have been deposited in the Gene Expression Omnibus (GEO) under accession number GSE308010 (accessible during review at: https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.ncbi.nlm.nih.gov%2Fgeo%2Fquery%2Facc.cgi%3Facc%3DGSE308010\u0026amp;data=05%7C02%7Cv.tsang%40qmul.ac.uk%7C73893a06442148f1678508ddf46aabc3%7C569df091b01340e386eebd9cb9e25814%7C0%7C0%7C638935461170639673%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C\u0026amp;sdata=%2BW3x03LMJE%2BRKGccA12v15qzHemSMEIWJ4wYTU4kkpE%3D\u0026amp;reserved=0\u003c/p\u003e\n\u003cp\u003eReviewer token: efixgocyffgbpmb).\u003c/p\u003e\n\u003cp\u003eRaw sequencing data (BAM files) are available in the Sequence Read Archive (SRA) under BioProject accession number PRJNA1327912 (accessible view only at:\u003c/p\u003e\n\u003cp\u003ehttps://dataview.ncbi.nlm.nih.gov/object/PRJNA1327912?reviewer=1p614t1hlcnep52di60emi03d3).\u003c/p\u003e\n\u003cp\u003eData will be released publicly upon acceptance and publication.\u003c/p\u003e\n\u003cp\u003eCompeting interests\u003c/p\u003e\n\u003cp\u003eD.D. is co-founder of Migration Biotherapeutics, and the rest of the authors declare no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eCeccarelli, M. \u003cem\u003eet al.\u003c/em\u003e Molecular Profiling Reveals Biologically Discrete Subsets and Pathways of Progression in Diffuse Glioma. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e164\u003c/strong\u003e, 550\u0026ndash;563 (2016).\u003c/li\u003e\n\u003cli\u003eGuo, X. \u003cem\u003eet al.\u003c/em\u003e Clinical updates on gliomas and implications of the 5th edition of the WHO classification of central nervous system tumors. \u003cem\u003eFront. Oncol.\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, (2023).\u003c/li\u003e\n\u003cli\u003eHertler, C. \u003cem\u003eet al.\u003c/em\u003e Long-term survival with IDH wildtype glioblastoma: first results from the ETERNITY Brain Tumor Funders\u0026rsquo; Collaborative Consortium (EORTC 1419). \u003cem\u003eEur. J. Cancer\u003c/em\u003e \u003cstrong\u003e189\u003c/strong\u003e, (2023).\u003c/li\u003e\n\u003cli\u003eLouis, D. N. \u003cem\u003eet al.\u003c/em\u003e The 2021 WHO Classification of Tumors of the Central Nervous System: a summary. \u003cem\u003eNeuro-Oncol.\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, 1231 (2021).\u003c/li\u003e\n\u003cli\u003eMelhem, J. M., Detsky, J., Lim-Fat, M. J. \u0026amp; Perry, J. R. Updates in IDH-Wildtype Glioblastoma. \u003cem\u003eNeurotherapeutics\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 1705\u0026ndash;1723 (2022).\u003c/li\u003e\n\u003cli\u003eOstrom, Q. T. \u003cem\u003eet al.\u003c/em\u003e CBTRUS Statistical Report: Primary Brain and Other Central Nervous System Tumors Diagnosed in the United States in 2011\u0026ndash;2015. \u003cem\u003eNeuro-Oncol.\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, iv1 (2018).\u003c/li\u003e\n\u003cli\u003eOstrom, Q. T. \u003cem\u003eet al.\u003c/em\u003e Editor\u0026rsquo;s choice: The epidemiology of glioma in adults: a \u0026ldquo;state of the science\u0026rdquo; review. \u003cem\u003eNeuro-Oncol.\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 896 (2014).\u003c/li\u003e\n\u003cli\u003eWorld Health Organization Classification of Tumours of the Central Nervous System. \u003cem\u003eInt. Agency Res. Cancer\u003c/em\u003e \u003cstrong\u003e5th editio\u003c/strong\u003e, (2021).\u003c/li\u003e\n\u003cli\u003eWurdak, H. \u003cem\u003eet al.\u003c/em\u003e An RNAi Screen Identifies TRRAP as a Regulator of Brain Tumor-Initiating Cell Differentiation. \u003cem\u003eCell Stem Cell\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 37\u0026ndash;47 (2010).\u003c/li\u003e\n\u003cli\u003eGimple, R. C., Bhargava, S., Dixit, D. \u0026amp; Rich, J. N. Glioblastoma stem cells: lessons from the tumor hierarchy in a lethal cancer. \u003cem\u003eGenes Dev.\u003c/em\u003e \u003cstrong\u003e33\u003c/strong\u003e, 591\u0026ndash;609 (2019).\u003c/li\u003e\n\u003cli\u003eDirks, P. B. Brain tumor stem cells: the cancer stem cell hypothesis writ large. \u003cem\u003eMol. Oncol.\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 420\u0026ndash;430 (2010).\u003c/li\u003e\n\u003cli\u003eStupp, R. \u003cem\u003eet al.\u003c/em\u003e Effects of radiotherapy with concomitant and adjuvant temozolomide versus radiotherapy alone on survival in glioblastoma in a randomised phase III study: 5-year analysis of the EORTC-NCIC trial. \u003cem\u003eLancet Oncol.\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 459\u0026ndash;466 (2009).\u003c/li\u003e\n\u003cli\u003eOzdemir-Kaynak, E., Qutub, A. A. \u0026amp; Yesil-Celiktas, O. Advances in glioblastoma multiforme treatment: New models for nanoparticle therapy. \u003cem\u003eFront. Physiol.\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, (2018).\u003c/li\u003e\n\u003cli\u003ede Gooijer, M. C., Guill\u0026eacute;n Navarro, M., Bernards, R., Wurdinger, T. \u0026amp; van Tellingen, O. An Experimenter\u0026rsquo;s Guide to Glioblastoma Invasion Pathways. \u003cem\u003eTrends Mol. Med.\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 763\u0026ndash;780 (2018).\u003c/li\u003e\n\u003cli\u003eArmento, A. \u003cem\u003eet al.\u003c/em\u003e Molecular mechanisms of cell motility. in \u003cem\u003eCancer and Metastasis Review\u003c/em\u003e vol. 11 73\u0026ndash;93 (Codon Publications, 2017).\u003c/li\u003e\n\u003cli\u003eRobertson, F. L., Marqu\u0026eacute;s-Torrej\u0026oacute;n, M. A., Morrison, G. M. \u0026amp; Pollard, S. M. Experimental models and tools to tackle glioblastoma. \u003cem\u003eDMM Dis. Models Mech.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, (2019).\u003c/li\u003e\n\u003cli\u003eNishikawa, M. \u003cem\u003eet al.\u003c/em\u003e Significance of glioma stem-like cells in the tumor periphery that express high levels of CD44 in tumor invasion, early progression, and poor prognosis in glioblastoma. \u003cem\u003eStem Cells Int.\u003c/em\u003e \u003cstrong\u003e2018\u003c/strong\u003e, (2018).\u003c/li\u003e\n\u003cli\u003eDarmanis, S. \u003cem\u003eet al.\u003c/em\u003e Single-Cell RNA-Seq Analysis of Infiltrating Neoplastic Cells at the Migrating Front of Human Glioblastoma. \u003cem\u003eCell Rep.\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, 1399\u0026ndash;1410 (2017).\u003c/li\u003e\n\u003cli\u003eSingh, S. K. \u003cem\u003eet al.\u003c/em\u003e Identification of human brain tumour initiating cells. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e432\u003c/strong\u003e, 396\u0026ndash;401 (2004).\u003c/li\u003e\n\u003cli\u003eGarnier, D., Renoult, O., Alves-Guerra, M. C., Paris, F. \u0026amp; Pecqueur, C. Glioblastoma stem-like cells, Metabolic strategy to kill a challenging target. \u003cem\u003eFront. Oncol.\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 118 (2019).\u003c/li\u003e\n\u003cli\u003eCha, J. \u0026amp; Kim, P. Biomimetic strategies for the glioblastoma microenvironment. \u003cem\u003eFront. Mater.\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, (2017).\u003c/li\u003e\n\u003cli\u003eSelvi, H., Br\u0026uuml;ning-Richardson, A. \u0026amp; Danovi, D. Systematic Review of Pre-Clinical Systems Using Artificial Microenvironments and Anti-Migratory Drugs to Control Migration of Glioblastoma Cells. \u003cem\u003eExpert Rev. Mol. Med.\u003c/em\u003e \u003cstrong\u003e27\u003c/strong\u003e, e6 (2025).\u003c/li\u003e\n\u003cli\u003eXie, Y. \u003cem\u003eet al.\u003c/em\u003e The Human Glioblastoma Cell Culture Resource: Validated Cell Models Representing All Molecular Subtypes. \u003cem\u003eEBioMedicine\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 1351\u0026ndash;1363 (2015).\u003c/li\u003e\n\u003cli\u003eLee, J. \u003cem\u003eet al.\u003c/em\u003e Tumor stem cells derived from glioblastomas cultured in bFGF and EGF more closely mirror the phenotype and genotype of primary tumors than do serum-cultured cell lines. \u003cem\u003eCancer Cell\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 391\u0026ndash;403 (2006).\u003c/li\u003e\n\u003cli\u003ePine, A. R. \u003cem\u003eet al.\u003c/em\u003e Tumor Microenvironment Is Critical for the Maintenance of Cellular States Found in Primary Glioblastomas. \u003cem\u003eCancer Discov.\u003c/em\u003e CD-20-0057 (2020) doi:10.1158/2159-8290.cd-20-0057.\u003c/li\u003e\n\u003cli\u003eLiu, E. \u0026amp; Nolan, J. P. Surface Enhanced Raman Scattering (SERS) Image Cytometry for High-Content Screening. \u003cem\u003eFluoresc. Microsc. Super-Resolut. Nov. Tech.\u003c/em\u003e 93\u0026ndash;108 (2014) doi:10.1016/B978-0-12-409513-7.00007-5.\u003c/li\u003e\n\u003cli\u003eBon, C., Si, Y. \u0026amp; Arimondo, P. B. Targeting DOT1L for mixed-lineage rearranged leukemia. \u003cem\u003eHistone Modif. Ther.\u003c/em\u003e 81\u0026ndash;99 (2020) doi:10.1016/B978-0-12-816422-8.00005-2.\u003c/li\u003e\n\u003cli\u003eGarrett, A. M., Lastakchi, S. \u0026amp; McConville, C. The Personalisation of Glioblastoma Treatment Using Whole Exome Sequencing: A Pilot Study. \u003cem\u003eGenes\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, (2020).\u003c/li\u003e\n\u003cli\u003eLee, S. \u003cem\u003eet al.\u003c/em\u003e Targeting tumour-intrinsic neural vulnerabilities of glioblastoma. \u003cem\u003ebioRxiv\u003c/em\u003e 2022.10.07.511321 (2022) doi:10.1101/2022.10.07.511321.\u003c/li\u003e\n\u003cli\u003eFernandopulle, M. S. \u003cem\u003eet al.\u003c/em\u003e Transcription Factor\u0026ndash;Mediated Differentiation of Human iPSCs into Neurons. \u003cem\u003eCurr. Protoc. Cell Biol.\u003c/em\u003e \u003cstrong\u003e79\u003c/strong\u003e, e51 (2018).\u003c/li\u003e\n\u003cli\u003ePolson, E. S. \u003cem\u003eet al.\u003c/em\u003e KHS101 disrupts energy metabolism in human glioblastoma cells and reduces tumor growth in mice. \u003cem\u003eSci. Transl. Med.\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 2718 (2018).\u003c/li\u003e\n\u003cli\u003eAl Shboul, S. \u003cem\u003eet al.\u003c/em\u003e Kinomics platform using GBM tissue identifies BTK as being associated with higher patient survival. \u003cem\u003eLife Sci. Alliance\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, (2021).\u003c/li\u003e\n\u003cli\u003eLiu, X. \u003cem\u003eet al.\u003c/em\u003e Overview of the molecular mechanisms of migration and invasion in glioblastoma multiforme. \u003cem\u003eJ. Chin. Med. Assoc.\u003c/em\u003e \u003cstrong\u003e84\u003c/strong\u003e, 669\u0026ndash;677 (2021).\u003c/li\u003e\n\u003cli\u003eStr\u0026ouml;bele, S. \u003cem\u003eet al.\u003c/em\u003e A Potential Role for the Inhibition of PI3K Signaling in Glioblastoma Therapy. \u003cem\u003ePLOS ONE\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, e0131670 (2015).\u003c/li\u003e\n\u003cli\u003eDjuzenova, C. S. \u003cem\u003eet al.\u003c/em\u003e Differential effects of the Akt inhibitor MK-2206 on migration and radiation sensitivity of glioblastoma cells. \u003cem\u003eBMC Cancer 2019 191\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 1\u0026ndash;18 (2019).\u003c/li\u003e\n\u003cli\u003eBogdahn, U. \u003cem\u003eet al.\u003c/em\u003e Targeted therapy for high-grade glioma with the TGF-\u0026beta;2 inhibitor trabedersen: results of a randomized and controlled phase IIb study. \u003cem\u003eNeuro-Oncol.\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 132 (2011).\u003c/li\u003e\n\u003cli\u003eWick, A. \u003cem\u003eet al.\u003c/em\u003e Phase 1b/2a study of galunisertib, a small molecule inhibitor of transforming growth factor-beta receptor I, in combination with standard temozolomide-based radiochemotherapy in patients with newly diagnosed malignant glioma. \u003cem\u003eInvest. New Drugs\u003c/em\u003e \u003cstrong\u003e38\u003c/strong\u003e, 1570 (2020).\u003c/li\u003e\n\u003cli\u003eWen, P. Y. \u003cem\u003eet al.\u003c/em\u003e Buparlisib in Patients With Recurrent Glioblastoma Harboring Phosphatidylinositol 3-Kinase Pathway Activation: An Open-Label, Multicenter, Multi-Arm, Phase II Trial. \u003cem\u003eJ. Clin. Oncol.\u003c/em\u003e \u003cstrong\u003e37\u003c/strong\u003e, 741 (2019).\u003c/li\u003e\n\u003cli\u003eAnderson, H. J. \u0026amp; Galileo, D. S. Small-molecule inhibitors of FGFR, integrins and FAK selectively decrease L1CAM-stimulated glioblastoma cell motility and proliferation. \u003cem\u003eCell. Oncol.\u003c/em\u003e \u003cstrong\u003e39\u003c/strong\u003e, 229\u0026ndash;242 (2016).\u003c/li\u003e\n\u003cli\u003eParker, J. J. \u003cem\u003eet al.\u003c/em\u003e Gefitinib selectively inhibits tumor cell migration in EGFR-amplified human glioblastoma. \u003cem\u003eNeuro-Oncol.\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 1048 (2013).\u003c/li\u003e\n\u003cli\u003eSingh, J., Sharma, K. \u0026amp; Pillai, P. P. PDGFR inhibition mediated intracellular signalling in C6 glioma growth and migration: role of ERK and ROCK pathway. \u003cem\u003eCytotechnology\u003c/em\u003e \u003cstrong\u003e70\u003c/strong\u003e, 465 (2018).\u003c/li\u003e\n\u003cli\u003eAffinito, A. \u003cem\u003eet al.\u003c/em\u003e Targeting Ephrin Receptor Tyrosine Kinase A2 with a Selective Aptamer for Glioblastoma Stem Cells. \u003cem\u003eMol. Ther. Nucleic Acids\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 176\u0026ndash;185 (2020).\u003c/li\u003e\n\u003cli\u003eZhou, W., Guo, S., Liu, M., Burow, M. E. \u0026amp; Wang, G. Targeting CXCL12/CXCR4 Axis in Tumor Immunotherapy. \u003cem\u003eCurr. Med. Chem.\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, 3026 (2019).\u003c/li\u003e\n\u003cli\u003eBrooks, L. J. \u003cem\u003eet al.\u003c/em\u003e The white matter is a pro-differentiative microenvironment for glioblastoma. \u003cem\u003eJ. Chem. Inf. Model.\u003c/em\u003e \u003cstrong\u003e01\u003c/strong\u003e, 1689\u0026ndash;1699 (2020).\u003c/li\u003e\n\u003cli\u003eda Silva, B. B., Mathew, R. K., Polson, E. S., Williams, J. \u0026amp; Wurdak, H. Spontaneous Glioblastoma Spheroid Infiltration of Early-Stage Cerebral Organoids Models Brain Tumor Invasion. \u003cem\u003eSLAS Discov.\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, 862\u0026ndash;868 (2018).\u003c/li\u003e\n\u003cli\u003eIrving, B. K. Modelling invasion, migration and cellular networks in Glioblastoma stem cells. (University of Leeds, 2020).\u003c/li\u003e\n\u003cli\u003eGeribaldi-Dold\u0026aacute;n, N. \u003cem\u003eet al.\u003c/em\u003e Targeting Protein Kinase C in Glioblastoma Treatment. \u003cem\u003eBiomedicines\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, (2021).\u003c/li\u003e\n\u003cli\u003eSmer-Barreto, V. \u003cem\u003eet al.\u003c/em\u003e Identification of drug candidates against glioblastoma with machine learning and high-throughput screening of heterogeneous cellular models. 2025.03.06.641926 Preprint at https://doi.org/10.1101/2025.03.06.641926 (2025).\u003c/li\u003e\n\u003cli\u003eGiese, A. \u003cem\u003eet al.\u003c/em\u003e \u003cem\u003eMigration of Human Glioma Cells on Myelin\u003c/em\u003e\u003cem\u003e : Neurosurgery\u003c/em\u003e. \u003cem\u003eNeurosurgery\u003c/em\u003e vol. 38 755\u0026ndash;764 (1996).\u003c/li\u003e\n\u003cli\u003eZepecki, J. P. \u003cem\u003eet al.\u003c/em\u003e Regulation of human glioma cell migration, tumor growth, and stemness gene expression using a Lck targeted inhibitor. \u003cem\u003eOncogene\u003c/em\u003e \u003cstrong\u003e38\u003c/strong\u003e, 1734\u0026ndash;1750 (2019).\u003c/li\u003e\n\u003cli\u003eBayat, N. \u003cem\u003eet al.\u003c/em\u003e The anti-angiogenic effect of atorvastatin in glioblastoma spheroids tumor cultured in fibrin gel: In 3D in vitro model. \u003cem\u003eAsian Pac. J. Cancer Prev.\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 2553\u0026ndash;2560 (2018).\u003c/li\u003e\n\u003cli\u003eGudbergsson, J. M. \u003cem\u003eet al.\u003c/em\u003e A tumorsphere model of glioblastoma multiforme with intratumoral heterogeneity for quantitative analysis of cellular migration and drug response. \u003cem\u003eExp. Cell Res.\u003c/em\u003e \u003cstrong\u003e379\u003c/strong\u003e, 73\u0026ndash;82 (2019).\u003c/li\u003e\n\u003cli\u003eAlsehli, H. \u003cem\u003eet al.\u003c/em\u003e An integrated pipeline for high-throughput screening and profiling of spheroids using simple live image analysis of frame to frame variations. \u003cem\u003eMethods San Diego Calif\u003c/em\u003e \u003cstrong\u003e190\u003c/strong\u003e, 33 (2021).\u003c/li\u003e\n\u003cli\u003eChen, S., Zhou, Y., Chen, Y. \u0026amp; Gu, J. fastp: an ultra-fast all-in-one FASTQ preprocessor. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cstrong\u003e34\u003c/strong\u003e, i884\u0026ndash;i890 (2018).\u003c/li\u003e\n\u003cli\u003eDobin, A. \u003cem\u003eet al.\u003c/em\u003e STAR: ultrafast universal RNA-seq aligner. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cstrong\u003e29\u003c/strong\u003e, 15\u0026ndash;21 (2013).\u003c/li\u003e\n\u003cli\u003eLi, B. \u0026amp; Dewey, C. N. RSEM: Accurate transcript quantification from RNA-Seq data with or without a reference genome. \u003cem\u003eBMC Bioinformatics\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 1\u0026ndash;16 (2011).\u003c/li\u003e\n\u003cli\u003eLove, M. I., Huber, W. \u0026amp; Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. \u003cem\u003eGenome Biol.\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 1\u0026ndash;21 (2014).\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Table","content":"\u003ch3\u003eTable 1. Table summarizing the inhibitors used in the drug screen, including the compound name, pharmacological target, IC50, provider, and catalogue number\u0026nbsp;\u003c/h3\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 149px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCompound name\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 177px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePharmacological Target\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eIC50 in nM\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCompany\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCatalogue number\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 149px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAPI-2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 177px;\"\u003e\n \u003cp\u003eProtein kinase B (PKB/Akt)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eTocris\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e2151\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 149px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePF 573228\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 177px;\"\u003e\n \u003cp\u003eFocal Adhesion Kinase (FAK)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eTocris\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e3239\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 149px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eONO 1078\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(Pranlukast)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 177px;\"\u003e\n \u003cp\u003eCysteinyl leukotriene receptor 1 (CysLT1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eTocris\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e3026/50\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 149px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBlebbistatin\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 177px;\"\u003e\n \u003cp\u003eMyosin II ATPase\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e2000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eTocris\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e1760\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 149px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLatrunculin B\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 177px;\"\u003e\n \u003cp\u003eactin polymerization\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eTocris\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e3974\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 149px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eY-27632\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 177px;\"\u003e\n \u003cp\u003eRho-associated protein kinase (ROCK)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e140\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eTocris\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e1254\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 149px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePD 173074\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 177px;\"\u003e\n \u003cp\u003eFibroblast Growth Factor Receptor (FGFR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eTocris\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e3044\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 149px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eHPI 1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 177px;\"\u003e\n \u003cp\u003eSonic Hedgehog\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e150\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eTocris\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e3839\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 149px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eIWP 2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 177px;\"\u003e\n \u003cp\u003eWnt\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e27\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eTocris\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e3533\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 149px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eALW II-41-27\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(Compound 7)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 177px;\"\u003e\n \u003cp\u003eEphrin receptor EphA2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eSelleck Chem\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003eS6515\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 149px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNVP-BHG712 S2202\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 177px;\"\u003e\n \u003cp\u003eEphrin receptor EphB4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eSelleck Chem\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003eS2202\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 149px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eEhp-inhibitor-1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 177px;\"\u003e\n \u003cp\u003eEphrin receptors (EphB2, EphB4 and related Eph kinases)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eSelleck Chem\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003eS0256\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 149px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCilengitide\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(EMD121974)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 177px;\"\u003e\n \u003cp\u003eintegrins \u0026alpha;v\u0026beta;3 and \u0026alpha;v\u0026beta;5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e4.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eTocris\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e5870\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 149px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMotixafortide\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(BL-8040)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 177px;\"\u003e\n \u003cp\u003estromal derived factor 1 (SDF-1, CXCL12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eSelleck\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003eS9665\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 149px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eGF 109203X\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 177px;\"\u003e\n \u003cp\u003eProtein Kinase C\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e8.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eTocris\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0741\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 149px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSB 431542\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 177px;\"\u003e\n \u003cp\u003eTGF-\u0026beta; receptors\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e94\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eTocris\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e1614\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 149px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAC 710\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 177px;\"\u003e\n \u003cp\u003ePlatelet-Derived Growth Factor Receptor (PDGFR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e1.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eTocris\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e5013\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 149px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eIressa\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 177px;\"\u003e\n \u003cp\u003eEpidermal growth factor receptor (EGFR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 90px;\"\u003e\n \u003cp\u003eTocris\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e3000\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Glioblastoma, cancer migration, induced-pluripotent stem cells, neural spheroid, high-content imaging, stem cell modelling ","lastPublishedDoi":"10.21203/rs.3.rs-7536545/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7536545/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eGlioblastoma is the most aggressive adult brain tumor, characterized by resistance to therapy and high recurrence due to diffuse infiltration. To mimic glioblastoma migration, we developed a physiologically relevant co-culture model, combining patient-derived glioblastoma cell lines with cortical-like neural spheroids differentiated from human induced pluripotent stem cells. Using high-content imaging, we demonstrate that GBM1 and GBM20 cell lines migrate directionally along axons toward neural spheroids in live imaging assays and infiltrate spheroids extensively in endpoint assays, unlike non-cancerous neural stem cells. A proof-of-principle drug screen identified PF-573228 (FAK inhibitor) and Motixafortide (CXCR4 inhibitor) as potent suppressors of GBM1 and GBM20 infiltration, respectively. Bulk RNA sequencing revealed gene expression profiles correlating with invasive behavior and drug sensitivity. This platform offers a valuable model for studying glioblastoma infiltration along axons and provides proof-of-principle that migration can serve as a measurable and actionable phenotype to screen therapeutic vulnerabilities in glioblastoma.\u003c/p\u003e","manuscriptTitle":"A human iPSC-based neural spheroid platform for modeling glioblastoma infiltration using high-content imaging","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-06 14:26:35","doi":"10.21203/rs.3.rs-7536545/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-10-20T05:31:56+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-17T22:34:50+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-07T00:49:44+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"27592937990417498788865472091261712189","date":"2025-09-29T11:21:23+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"289407950425351225824233778660172904260","date":"2025-09-26T06:22:38+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"210869120100411724386025197911101401575","date":"2025-09-23T20:30:14+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"325477836392101255566600243657156501320","date":"2025-09-23T19:41:13+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-09-23T18:17:07+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-09-23T18:05:26+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-09-23T17:59:34+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-09-21T16:33:19+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-09-21T16:29:13+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"7bd46a37-74cf-4886-b839-669545d60cd3","owner":[],"postedDate":"October 6th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":55511175,"name":"Biological sciences/Cancer"},{"id":55511176,"name":"Biological sciences/Cell biology"},{"id":55511177,"name":"Biological sciences/Neuroscience"}],"tags":[],"updatedAt":"2025-12-15T16:16:59+00:00","versionOfRecord":{"articleIdentity":"rs-7536545","link":"https://doi.org/10.1038/s41598-025-30914-5","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2025-12-13 15:59:14","publishedOnDateReadable":"December 13th, 2025"},"versionCreatedAt":"2025-10-06 14:26:35","video":"","vorDoi":"10.1038/s41598-025-30914-5","vorDoiUrl":"https://doi.org/10.1038/s41598-025-30914-5","workflowStages":[]},"version":"v1","identity":"rs-7536545","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7536545","identity":"rs-7536545","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}