Development and characterization of an on-chip glioma outgrowth model

Development and characterization of an on-chip glioma outgrowth model | 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 Research Article Development and characterization of an on-chip glioma outgrowth model Joshua P. A. Daoud, Alexandra E. Collisson, Daniya Asaid, Ester Clarisse do Couto Lopes, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7785293/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 9 You are reading this latest preprint version Abstract Tumour recurrence in high-grade gliomas, such as glioblastoma, is extremely common due to the challenges associated with complete surgical removal. Approximately 95% of glioblastoma tumours recur within 2 cm of the tumour resection margin; however, unfortunately, the current clinical measurement techniques do not accurately capture cell state during critical stages recurrence. The development of an on-chip model for investigating glioblastoma recurrence outgrowth is necessary to further understanding of the mechanistic understanding of this process at a cellular level and holds potential for translation into patient-specific predictive models. This work presents the development of an on-chip platform which can be used to assess glioblastoma interactions with healthy neural cells. Using a custom single-channel multi-well microfluidic system, 9L/3cmv-GFP (GFP transfected) rat glioma cells were co-cultured with a CTX-TNA2 rat astrocyte cell line to investigate differences in glioblastoma outgrowth between co-culture and monoculture (glioma only) configurations. Quantitative image analysis demonstrated significantly increased outgrowth of glioma cells from a tumour spheroid in co-culture compared to glioma only controls. Additionally, comparison of upregulation of glioma linked CD44 and glial fibrillary acidic protein (GFAP) expression showed significant differences in cellular expression between the two configurations. These findings align with existing literature suggesting that astrocytes facilitate a supportive environment for glioblastoma cells to proliferate and invade healthy tissue by changing to a reactive phenotype. Overall, this work presents a promising on-chip platform that can aid in the quantification of glioblastoma invasion through directional control of outgrowth and offers the potential for further modalities to be added by combining real-time data acquisition elements to the platform. glioma microfluidics lab-on-a-chip co-culture cancer models bioengineering Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 1 – Introduction Glioblastoma multiforme, a high-grade glioma, is the most common intrinsic malignant brain tumour diagnosis in adults [ 1 ]. Patient outcomes are notoriously poor with a 5-year survival rate of less than 7% [ 2 ]. These tumours are characterized as being substantially invasive, with an extremely high recurrence rate, as total surgical resection is challenging due to limitations on total resection volume. Due to the propensity for this cancer to recur, patient deterioration is inevitable, even after several repetitions of chemotherapeutic and radiotherapy treatments [ 3 ]. Current practice relies on outpatient magnetic resonance imaging (MRI) or computed tomography (CT) scans to detect local cancer recurrences after surgical resection and inform neurooncological treatment [ 4 ]. Unfortunately, the time in between these scans is often too long to identify small changes in the tumour margin and rapidly inform neurooncologist leading to worsened patient outcomes [ 5 , 6 ]. Developing predictive in vitro models of glioblastoma recurrence has the potential to provide key feedback on treatment efficacy thus informing decisions about alternative therapies [ 7 ]. Current knowledge of the cellular mechanisms underpinning glioblastoma recurrence is limited by challenges in recording from patients and incomplete recapitulation of the healthy neural environment in glioblastoma organoid models [ 8 ]. To address this, predictive in vitro models and on-chip platforms have gained attention for their ability to provide alternative models to provide a tuneable microenvironment in which to investigate cellular mechanisms. Recently, three-dimensional spheroid models of gliomas have seen an increase in popularity in their use to simulate tumour behaviour [ 9 ]. These models have been shown to retain near-identical genetic expression to the original tumour [ 10 ] which enables researchers to investigate cellular processes as they would naturally occur in the brain [ 11 , 12 ]. However, most of the developed spheroid models are focused on enabling more realistic testing of chemotherapeutic efficacy for drug discovery. Therefore, only a limited amount of research has been focused on developing models that fully recapitulate the neural environment in which glioblastoma occurs. Without these interactions it is challenging to accurately investigate the complex interactions between glioma cells and healthy neural cells like astrocytes, oligodendrocytes, and neurons [ 12 ]. There is clear evidence that the interactions between glioma and healthy neural tissue is a critically important component to the efficacious treatment of gliomas, it is essential to examine their hallmarks using in vitro models to simulate in vivo growth conditions [ 13 ] The development of a co-culture setup to culture healthy neural cells alongside glioma cells to investigate infiltration mechanisms is necessary to facilitate further understanding of the methods behind tumour recurrence [ 14 ]. By simulating a physiologically relevant microenvironment, microfluidic platforms offer solutions to the methodological limitations which are inherent in animal models and traditional culture systems by the ability to engineer the cell culture geometry [ 15 ]. The rationale of this work was to produce a customizable on-chip neural system to enable the combination of a glioma spheroid model with healthy neural tissue. From this model, the aim was to enable the investigation of how glioma cells grow into the surrounding healthy neural tissue in this on-chip system by analysing tumour outgrowth, cell density, and key marker protein expression between healthy and cancerous cells during different stages of outgrowth. This work highlights how the system developed enables an advantageous platform over traditional cell models by providing a easily adapted system to enable clear path for glioma outgrowth. 2 – Materials and Methods 2.1 - Micropatterning Photolithography masks for the microfluidic channels were designed using computer-aided design software AutoCAD (Fig. 1 A) and purchased from Microlitho (UK). Channel moulds were created by spin coating SU-8 photoresist (Kayaku) at 500 rpm for 5 seconds, then 1000 rpm for 30 seconds onto blank silicon wafers and exposing for 8 seconds at 100% power (360 mW/cm 2 ) with a 365 nm UV source (KLOÉ). After developing, masks were rinsed with Propylene Glycol Monomethyl Ether Acetate (PGMEA) and baked at 150°C for 3 minutes. After mould creation, microfluidic channels were fabricated from cast Sylgard-184 Silicone Elastomer (Farnell UK) at 80°C for 1.5 hours. Channel dimensions were analyzed using an optical profilometer (Omniscan), then visualised in Profilm3D. Figure 1 B and 1 C illustrate the workflow of microfluidic channel fabrication. 2.2 - Cell culture Cell preparation The 9L/3cmv-GFP glioma cell line, were previously transfected with green fluorescent protein (GFP) and donated from the laboratory of Prof. Rylie Green at Imperial College London. 9L/3cmv-GFP cells were established and cultured in Dulbecco's Modified Eagle Medium (DMEM), with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin (Pen/Strep) all purchased from ThermoFisher Scientific. Cells were incubated at 37°C and 5% CO 2 , with passaging performed every 4–5 days. For astrocytes, the CTX-TNA2 cell line (UK Health Security Agency) were cultured using the same media formulation. 5 days before co-culture experiments, 9L/3cmv-GFP cells were grown into glioma spheroids using a Corning 96-well spheroid-forming non-adherent microplate (Sigma-Aldrich) and DMEM media. Media was replenished on day 2 to optimise viability and minimise media loss through evaporation. After optimisation of spheroid size, as presented in Results 3.2, the glioma spheroids were grown from a cell concentration of 150,000 cells per mL, which grew spheroids measuring 500–600 mm in diameter within 5 days of growth, which were appropriately sized for visualisation inside microfluidic channels. Microfluidic preparation for cell culture First, microfluidic channels were sterilized using autoclave sterilization at 121°C and flushed with sterile phosphate buffer solution (D-PBS). Microfluidic channels were also coated using a solution which consisted of 25 µg/ml Poly-D-Lysine and 10 µg/ml Laminin, which provided improved adhesion allowing astrocyte cells to settle and adhere onto the glass slide. No further growth factors were added to prevent any interference with cell-cell interactions [ 16 ]. Full coating protocol is provided in supplementary information. Figure 2 shows additional features of the microfluidic chip design, including rings which were cut from a flat disc of cured PDMS and adhered onto the upper surface of the chip using KORASILON silicone paste (Sigma-Aldrich), as this created a reservoir of DMEM media. Small reservoirs of sterile D-PBS were also kept inside of the petri dish throughout the duration of the experiment. These steps prevented excessive media evaporation, enabling longer experimental running times. Outgrowth experiments Co-culture : Astrocyte cell suspension was added into each microfluidic chip at a seeding density of 50,000 cells per mL. This was achieved by adding 2 x 200 µl of astrocyte cell suspension to each end of the microfluidic channel to give a total volume of 800 µl added to each chip. This ensured even distribution of cells throughout the channel. After the addition of astrocytes, one glioma spheroid was added into the centre well of the microfluidic channel. Co-culture samples were incubated at 37°C and 5% CO 2 and allowed to grow without interruption for 48 hours to allow cells to adhere. After the initial 48-hour period, the first media change was performed using fresh media, then after 96 hours, a second media change was performed. After the 96-hour period, subsequent media changes were performed daily to prevent the buildup of cell debris. Co-culture experiments were grown until two timepoints: 7 days and 14 days. Control experiments consisted of glioma spheroids grown in isolation and CTX-TNA2 astrocyte cells grown in isolation inside of the microfluidic platform. Control samples were also grown for 7 and 14 days. Control samples received the same media replenishment schedule as co-culture samples (Supplementary Information Fig. 0 for more detail). Total outgrowth was calculated from fluorescent images in the ImageJ program by measuring the furthest detectable outgrowth of glioma cells from the 9L/3cmv-GFP spheroid. Figure S1 (supplementary information) shows how outgrowth calculations were obtained. CD44 / GFAP Expression Experiments These experiments were carried out by seeding cells into microfluidic chips so that all samples had the same number of cells inside of the chip. Astrocyte control samples and glioma control samples (9L/3cmv-GFP) were both seeded with cell suspensions at concentrations of 100,000 cells per mL, and co-culture samples were seeded with a 50:50 mix of astrocyte and glioma cells with a final concentration of 100,000 cells per mL. Antibody experiments were carried out for two timepoints: 4 and 7 days, to ensure microfluidic chips did not become too confluent for single-cell analysis. 2.3 - Cell staining and imaging Two staining protocols were carried out, one for outgrowth experiments and another for antibody experiments. Full staining protocols are provided in the supplementary information. In all experiments, 9L/3cmv-GFP and CTX-TNA2 cells were fixed inside of microfluidic chips using 4% paraformaldehyde (PFA) and then permeabilised with 0.1% Triton-X solution. Before staining protocol was performed, microfluidic channels were washed several times with phosphate-buffered saline (PBS). Outgrowth staining For the outgrowth experiments, the glioma cell line already expressed GFP, so no fluorescent green stain was needed, but Texas-Red Phalloidin (ThermoFisher Scientific) and 4′,6-diamidino-2-phenylindole (DAPI) (ThermoFisher Scientific) were used to selectively stain F-actin filaments and nuclear DNA in red and blue respectively. Samples were used from day 7 and 14 of the outgrowth experiments. Antibody staining Further staining using primary antibodies CD44 Recombinant Rabbit Monoclonal Antibody (Invitrogen) and GFAP Mouse Monoclonal Antibody (GA5) (Invitrogen), with secondary antibodies Cy™5 AffiniPure Goat-Anti-Rabbit IgG (Jackson ImmunoResearch) and Cy™3 AffiniPure Goat-Anti-Mouse IgG (Jackson ImmunoResearch) were carried out to compare control samples with co-culture between glioma and astrocyte cells. Antibody staining samples were grown for two timepoints: day 4 and day 7. Primary antibodies for the CD44 antigen and Glial Fibrillary Acid Protein (GFAP) were used on all samples. Full antibody staining protocol can be found in supplementary information. Supplementary information Figure S3 depicts how relative intensities were calculated from fluorescent antibody staining images using several regions-of-interest (ROIs) inside of ImageJ. CD44 antigen was stained with Cy5 secondary antibody (far-red), and GFAP was stained with Cy3 secondary antibody (red). GBM cells can be visualised by GFP fluorescence (green). DAPI (dark blue) nuclear DNA stain was used to locate and select cells for further analysis. 2.4 - Analysis Imaging was performed using fluorescent confocal microscopy with the Nikon CSU-W1 SoRa Spinning Disc Confocal microscope. Images were then viewed and processed using the Nikon Elements Viewer software. After processing, fluorescent images were analysed using ImageJ software. Measurements of spheroid outgrowth distance were taken from fluorescent images of the co-culture and spheroid control samples. Measurements of cell density were also calculated for all samples. Supplementary information Fig S2 illustrates how the cell density was calculated from fluorescent images. Antibody staining images were also analysed using ImageJ to determine the relative intensity of antibodies in each sample. 2.5 - Statistics Outgrowth measurements, cell densities, and relative intensities of antibodies were all calculated from fluorescent images in ImageJ (Fiji) software. GraphPad Prism software was used to carry out inferential statistical tests including unpaired and paired sample t-tests, and Mann-Whitney U tests if equal variances could not be assumed. Unless otherwise stated each experiment had a sample size of n = 5. 3 – Results 3.1 - Microfluidic platform dimensions Media loss by evaporation is a common issue in cell culture, which can lead to drastic differences in data as a result, particularly with the periphery of cell culture vessels, known by researchers as the ‘edge effect’ [ 17 ]. To counter this issue and improve data reliability, channel dimensions of 15,600 µm (length) x 300 µm (depth) x 500 µm (width) were used to support larger media volumes (Fig. 3 A). Final dimensions for the microfluidic channel (Fig. 3 B and 2 C) were confirmed by profilometer to be 15,600 µm (length) x 300 µm (depth) x 500 µm (width). 3.2 – Optimisation of Spheroid Size Spheroids were grown from the 9L/3cmv-GFP glioma cell line and compared with previous experimental data obtained from spheroids grown from a non-transfected 9L/lacZ glioma cell line. This was to ensure that transfection did not alter the spheroid growth characteristics, so that transfected cells could be used within co-culture experiments. A full spheroid preparation protocol can be found in supplementary information. A proportional relationship between cell seeding density and spheroid diameter was seen. The desired spheroid diameter was 500–600 mm as this diameter would enable spheroid outgrowth to occur in a linear fashion inside of the microfluidic channel. A cell seeding density of 150,000 cells per mL was selected due to this producing an appropriate size spheroid after 5 days of growth with minimal variation between spheroids. Images taken of the spheroids clearly show an increasingly dense and dark formation of cells known as the necrotic core, which is a physiologically relevant hallmark of a tumour, suggesting that the spheroids closely simulated tumour physiology [ 18 ]. 3.3 – Glioma outgrowth in co-culture and monoculture Fig 5A highlights the differences in spheroid outgrowth distance between co-culture samples (containing glioma and astrocytes), and the glioma spheroid control samples (‘9L only’ on graph). In co-culture samples, the mean outgrowth was measured at 1428 μm (SD = 556) on day 7, which increased to 1969 μm (SD = 796) by day 14. Fig 5B demonstrates how in glioma only controls, the mean outgrowth was measured at 823 μm (SD = 243) on day 7 which increased to 1433 μm (SD = 342) by day 14. Fig 5C reveals the highly significant differences in outgrowth between co-culture and glioma control samples across both day 7 and day 14 timepoints (p < 0.0001). Mean outgrowth data for co-culture samples showed a percentage difference of +73.5% compared to glioma controls on day 7, and +37.4% on day 14. 3.4 - Cell density Figure 6 B highlights significant changes in cell density between co-culture and monoculture samples with. The mean cell density was calculated to be 2781/mm 2 on day 7, which increased to 6259/mm 2 by day 14, representing a significant increase in cell density between days 7 and 14 of growth ( p = 0.0005). For the astrocyte control sample, the mean cell density, was calculated to be 2145/mm 2 on day 7, which increased to 3015/mm 2 by day 14 ( p = 0.033). For the glioma control sample (shown as 9L only), the mean cell density was calculated to be at 1261/mm 2 on day 7 which increased minimally to 1300/mm 2 by day 14, no significant difference was observed. 3.5 – Antibody staining: CD44 and GFAP Figure 7 A shows a statistically significant increase in relative intensity of CD44 (far-red) in co-culture samples, as the relative intensity (a.u) increased from 40 on day 4 to 64 by day 7 ( p = 0.0303). Relative intensity of CD44 for the glioma control samples (shown as 9L only) were non-significant with an increase in relative intensity (a.u) from 14 on day 4 to 19 by day 7. Astrocyte control samples were also non-significant, as the relative intensity (a.u) remained at 34 on both day 4 and day 7. Figure 7 B provides a visual representation of CD44 intensities for all samples. Figure 7C shows a statistically significant increase in relative intensity of GFAP (red) between co-culture timepoints with the relative intensity (a.u) increasing from 10 on day 4 to 13 on day 7 (p = 0.0288). Relative intensities of GFAP between the astrocyte control timepoints were non-significant, as the relative intensity (a.u) remained at 8 for both day 4 and day 7. We observed a significant decrease in GFAP relative intensity (a.u) for glioma control samples from 8 on day 4 to 5 on day 7 (p = 0.0459). 4 - Discussion Mechanistic insights of glioma outgrowth The substantial outgrowth and increased cell density seen in glioma spheroids co-cultured with astrocytes compared to spheroids grown alone is supported by existing research which suggests that nearby astrocytes undergo malignant transformation into a more tumour-permissive phenotype, which may facilitate a supportive environment which promotes glioma invasion [ 19 ]. The results indicated that after prolonged contact with astrocytes in a co-culture environment, that glioma cells showed a relative increase in intensity of CD44, which could suggest that increased contact with astrocytes may contribute to the invasive properties of a glioma by increasing CD44 expression. Heightened CD44 expression has been shown to play a key role in glioma recurrence, as it is associated with invasion and proliferation of tumour cells by initiating cellular signalling pathways that give rise to the highly invasive glioma stem cell phenotype [ 20 ]. High expression of CD44 is also associated with poor prognosis and early tumour recurrence in glioma patients, compared to those with lower CD44 expression [ 21 ]. Furthermore, a substantial increase in cell density for glioma cells co-cultured with astrocytes was observed in comparison to controls. These observations are supported by literature which describes how astrocytes actively facilitate glioma proliferation and invasion by activating CD44 signalling pathways, which induces tumour cell transformation to the highly invasive and proliferative glioma stem cell phenotype [ 22 ]. Additionally, the relative intensity of GFAP increased significantly in when astrocytes were co-cultured with glioma cells, compared to astrocyte control samples. GFAP is a marker for reactive astrocytes whereby astrocytes transform in response to cellular injury, stress or disease [ 23 ]. Supporting evidence suggests that the presence of reactive astrocytes is involved in regulation of glioma progression through many mechanisms, but primarily through direct cell-to-cell interaction with tumour cells which enhances their survival, proliferation and invasion into healthy tissue [ 24 ]. Advantages of Microfluidic platform A notable advantage of this platform is that it allows for precise control over directional spheroid growth by forcing growth into the directional microfluidic channels. This presents a significant advantage over co-cultures grown inside of circular multi-well plates. Recent work in this area has performed spheroid co-cultures in multi-well plates, but have illustrated how spheroid growth is multi-directional, meaning outgrowth measurements cannot be accurately quantified by measurements taken from a set boundary in a single direction [ 25 ]. As a comparison, the same co-culture used here was performed in a stand-alone 96-well plate and imaged. The spheroids exhibited outgrowth, however it was sporadic and inconsistent around the periphery, highlighting the need for culture methods which can improve the precision and accuracy of outgrowth measurements (Figure S4). Critical Evaluations Overall, this research has produced an engineered in vitro model for visualisation and quantification of glioma invasion into healthy tissue using microfluidic “on-chip” technology. One major limitation of the approach taken in this work is that, although the in vitro model included three-dimensional elements such as spheroids, outgrowth was confined to a two-dimensional space along the bottom of the channel. This could have significantly altered the results in comparison to a fully three-dimensional model. Incorporation of a hydrogel into the model could facilitate a fully three-dimensional culture environment by mimicking the extracellular matrix [ 26 ]. Furthermore, the inclusion of other neural cell types such as neurons and microglia could have highlighted additional cellular interactions with tumour cells [ 27 ]. Future Directions This on-chip platform presents a highly feasible model which can be used to culture patient-derived tumour spheroids with healthy neural tissue to produce more physiologically relevant models of unique tumour pathologies [ 28 , 29 ]. Microfluidic models demonstrate a viable method for accurately modelling patient-specific glioma outcomes and provide an indication of treatment efficacy by using heterogenous spheroid cultures from patients to simulate treatment responses [ 30 ]. Next-generation predictive in vitro models are addressing the need for improved preclinical tools which can recapitulate the tumour microenvironment and simulate underlying biological processes that drive treatment response and clinical outcomes [ 31 ] This research adds to an existing evidence base [ 32 , 33 , 34 , 35 ] which strongly supports the practicality of microfluidic platforms for oncological modelling, as they show great potential to advance the future of cancer research. Additionally, the development of this customized on-chip neural systems is further strengthened by the potential for addition of real-time data recording capabilities to provide additional feedback on cell state [ 36 ]. Biosensor integration into microfluidic systems has been widely explored in recent years, with several studies demonstrating their potential for continuous, real-time monitoring of cellular and molecular events [ 37 ][ 38 ][ 39 ]. Building on these advances, future work will focus on incorporating such sensing technologies into this glioma on-chip platform to enhance its analytical capabilities. Electrochemical biosensors, including impedance-based devices will be embedded within the microfluidic model to monitor parameters such as tumour microenvironment changes, metabolic activity, and drug response providing deeper insight into disease progression, recurrence and treatment response [ 40 ]. 5 – Conclusion This work has outlined the developmental stages and application of an on-chip neural model of glioma recurrence. Using microfluidic technologies to guide glioma outgrowth it was possible to observe glioma growth in co-culture and monoculture control samples, as well as measure outgrowth from the tumour spheroid, and obtain data about the expression of relevant cellular markers which influence tumour invasion. Use of a microfluidic platform enabled creation of customisable channel dimensions to improve culture conditions and obtain reproducible data. The initial data presented in this study demonstrated that our microfluidic platform can be used as a practical and viable method for modelling glioma invasion into healthy tissue and provided further evidence toward visualizing cellular crosstalk and its important role in cancer cell migration and metastasis. Additionally, the ability to model visualize these changes in co-culture could have promising implications for further research to determine cancer metastasis and invasion mechanisms so that a deeper understanding can be gained about facilitatory and inhibitory factors which drive growth and survival of residual cancer cells. Declarations Ethics Approval Ethics approval was not required for this work. Author Contribution J.D. performed all experiments including data analysis and authored the manuscript. A.C. assisted with statistical analysis and final edits to the manuscript. D.A. contributed experimental data for supplementary information. E.L. provided written sections of the manuscript. D.P. provided financial support. C.C. led project development, provided financial support, and provided final edits to the manuscript Acknowledgement We would like to thank Dr. Liisa Marketa Blowes for access and support in the use of the CREATE Lab at Queen Mary University of London, as well as key guidance on cell staining, and the School of Engineering and Materials Science for providing access to their state-of-the-art organ-on-chip facilities. We would also like to also thank the Iskratsch Lab at Queen Mary University of London for their assistance with immunofluorescent staining, and the Green Lab at Imperial College London for donation of the 9L/3cmv-GFP cell line. Data Availability All analysis methods are provided in the manuscript and supplementary information. Raw data including images and analyzed numberical data is available upon request. References Louis, D. N., Perry, A., Reifenberger, G., Von Deimling, A., Figarella-Branger, D., Cavenee, W. K., ... & Ellison, D. W. (2016). The 2016 World Health Organization classification of tumors of the central nervous system: a summary. Acta neuropathologica , 131 , 803-820. Woodworth, Davis C., Whitney B. Pope, Linda M. Liau, Hyun J. Kim, Albert Lai, Phioanh L. Nghiemphu, Timothy F. Cloughesy, and Benjamin M. Ellingson. "Nonlinear distortion correction of diffusion MR images improves quantitative DTI measurements in glioblastoma." Journal of Neuro-oncology 116 (2014): 551-558. Lenting, Krissie, Roel Verhaak, Mark Ter Laan, Pieter Wesseling, and William Leenders. "Glioma: experimental models and reality." Acta neuropathologica 133 (2017): 263-282. National Institute for Health and Care Excellence (NICE). Brain tumours (primary) and brain metastases in over 16s: NICE guidelines. NG99. (2019) Available at: https://www.nice.org.uk/guidance/ng99/resources. Mahvi, David A., Rong Liu, Mark W. Grinstaff, Yolonda L. Colson, and Chandrajit P. Raut. "Local cancer recurrence: the realities, challenges, and opportunities for new therapies." CA: a cancer journal for clinicians 68, no. 6 (2018): 488-505. Palmer, G., A. Martling, B. Cedermark, and T. Holm. "A population-based study on the management and outcome in patients with locally recurrent rectal cancer." Annals of surgical oncology 14 (2007): 447-454. Warren, Joan L., and K. Robin Yabroff. "Challenges and opportunities in measuring cancer recurrence in the United States." Journal of the National Cancer Institute 107, no. 8 (2015): djv134. – Atat, Oula El, Zahra Farzaneh, Mahsa Pourhamzeh, Fatima Taki, Ralph Abi-Habib, Massoud Vosough, and Mirvat El-Sibai. "3D modeling in cancer studies." Human Cell (2022): 1-14. Paolillo, Mayra, Sergio Comincini, and Sergio Schinelli. "In vitro glioblastoma models: A journey into the third dimension." Cancers 13, no. 10 (2021): 2449. Edmondson, Rasheena, Jessica Jenkins Broglie, Audrey F. Adcock, and Liju Yang. "Three-dimensional cell culture systems and their applications in drug discovery and cell-based biosensors." Assay and drug development technologies 12, no. 4 (2014): 207-218. Suva, Mario L., and Itay Tirosh. "The glioma stem cell model in the era of single-cell genomics." Cancer cell 37, no. 5 (2020): 630-636. Diao, Wenwen, Xuezhi Tong, Cheng Yang, Fengrong Zhang, Chun Bao, Hao Chen, Liyu Liu et al. "Behaviors of glioblastoma cells in in vitro microenvironments." Scientific reports 9, no. 1 (2019): 85. DePalma, Thomas J., Hemamylammal Sivakumar, and Aleksander Skardal. "Strategies for developing complex multi-component in vitro tumor models: Highlights in glioblastoma." Advanced drug delivery reviews 180 (2022): 114067. Joseph, Justin V., Mathilde S. Blaavand, Thomas Daubon, Frank AE Kruyt, and Martin K. Thomsen. "Three-dimensional culture models to study glioblastoma—Current trends and future perspectives." Current Opinion in Pharmacology 61 (2021): 91-97. Sung, Kyung Eun, and David J. Beebe. "Microfluidic 3D models of cancer." Advanced drug delivery reviews 79 (2014): 68-78. Chauhdari, T., Su, J., Lv, J. and Ding Y. ECM-free lung tumoroids generated by an air-stretchable microfluidic chip. In vitro models. 2025 Sep 4:1-3. Mansoury, Morva, Maya Hamed, Rashid Karmustaji, Fatima Al Hannan, and Stephen T. Safrany. "The edge effect: A global problem. The trouble with culturing cells in 96-well plates." Biochemistry and biophysics reports 26 (2021): 100987. Mirab, Fereshtehsadat, You Jung Kang, and Sheereen Majd. "Preparation and characterization of size-controlled glioma spheroids using agarose hydrogel microwells." PloS one 14, no. 1 (2019): e0211078. Gagliano, Nicoletta, Francesco Costa, Chiara Cossetti, Letizia Pettinari, Rosaria Bassi, Maurizio Chiriva-Internati, Everardo Cobos, Magda Gioia, and Stefano Pluchino. "Glioma-astrocyte interaction modifies the astrocyte phenotype in a co-culture experimental model." Oncology reports 22, no. 6 (2009): 1349-1356. – Inoue A, Ohnishi T, Nishikawa M, et al. A Narrative Review on CD44's Role in Glioblastoma Invasion, Proliferation, and Tumor Recurrence. Cancers (Basel) . 2023;15(19):4898. Published 2023 Oct 9. doi:10.3390/cancers15194898 – Nishikawa, Masahiro, Akihiro Inoue, Takanori Ohnishi, Shohei Kohno, Shiro Ohue, Shirabe Matsumoto, Satoshi Suehiro 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 international 2018, no. 1 (2018): 5387041. – Du Z, Wang Y, Liang J, et al. Association of glioma CD44 expression with glial dynamics in the tumour microenvironment and patient prognosis. Comput Struct Biotechnol J . 2022;20:5203-5217. Published 2022 Sep 9. doi:10.1016/j.csbj.2022.09.003 – Pekny, Milos, Ulrika Wilhelmsson, and Marcela Pekna. "The dual role of astrocyte activation and reactive gliosis." Neuroscience letters 565 (2014): 30-38. – Wu J, Li R, Wang J, et al. Reactive Astrocytes in Glioma: Emerging Opportunities and Challenges. Int J Mol Sci . 2025;26(7):2907. Published 2025 Mar 23. doi:10.3390/ijms26072907 Förster, Nanna, Lotta Isosaari, Oskari Kulta, Oona Junnila, Valtteri Vuolanto, Marjukka Pollari, Kirsi J. Rautajoki, and Susanna Narkilahti. "Functional 3D Human Neuron–Glioblastoma Model Reveals Cellular Interactions Enabling Drug Safety Assessments." The FASEB Journal 39, no. 8 (2025): e70567. Matthiesen, I., Jury, M., Rasti Boroojeni, F., Ludwig, S. L., Holzreuter, M., Buchmann, S., … Herland, A. (2023). Astrocyte 3D culture and bioprinting using peptide functionalized hyaluronan hydrogels. Science and Technology of Advanced Materials , 24 (1). https://doi.org/10.1080/14686996.2023.2165871 Goshi, N., Morgan, R.K., Lein, P.J. et al. A primary neural cell culture model to study neuron, astrocyte, and microglia interactions in neuroinflammation. J Neuroinflammation 17 , 155 (2020). https://doi.org/10.1186/s12974-020-01819-z Song, Hanbing, Simon Bucher, Katherine Rosenberg, Margaret Tsui, Deviana Burhan, Daniel Hoffman, Soo-Jin Cho et al. "Single-cell analysis of hepatoblastoma identifies tumor signatures that predict chemotherapy susceptibility using patient-specific tumor spheroids." Nature communications 13, no. 1 (2022): 4878. Manini, Ivana, Federica Caponnetto, Anna Bartolini, Tamara Ius, Laura Mariuzzi, Carla Di Loreto, Antonio Paolo Beltrami, and Daniela Cesselli. 2018. "Role of Microenvironment in Glioma Invasion: What We Learned from In Vitro Models" International Journal of Molecular Sciences 19, no. 1: 147. https://doi.org/10.3390/ijms19010147 Zhang, Zengli, Huiqian Wang, Qifeng Ding, Yufei Xing, Zhonghua Xu, Chun Lu, Dongdong Luo et al. "Establishment of patient-derived tumor spheroids for non-small cell lung cancer." PloS one 13, no. 3 (2018): e0194016. Thomas, Grace and Ruman Rahman. Evolution of Preclinical Models for Glioblastoma modelling and drug Screening. Current Oncology Reports (2025): pp. 1-24. Khot, M. Ibrahim, Mark A. Levenstein, Greg N. de Boer, Gemma Armstrong, Thomas Maisey, Hafdis S. Svavarsdottir, Helen Andrew, Sarah L. Perry, Nikil Kapur, and David G. Jayne. "Characterising a PDMS based 3D cell culturing microfluidic platform for screening chemotherapeutic drug cytotoxic activity." Scientific Reports 10, no. 1 (2020): 15915. Toh, Yi-Chin, Teck Chuan Lim, Dean Tai, Guangfa Xiao, Danny van Noort, and Hanry Yu. "A microfluidic 3D hepatocyte chip for drug toxicity testing." Lab on a Chip 9, no. 14 (2009): 2026-2035. Kwapiszewska, K., A. Michalczuk, M. Rybka, Radosław Kwapiszewski, and Zbigniew Brzózka. "A microfluidic-based platform for tumour spheroid culture, monitoring and drug screening." Lab on a Chip 14, no. 12 (2014): 2096-2104. Ryu, Gihan, Jingsong Huang, Oliver Hofmann, Claire A. Walshe, Jasmine YY Sze, Gareth D. McClean, Alan Mosley et al. "Highly sensitive fluorescence detection system for microfluidic lab-on-a-chip." Lab on a Chip 11, no. 9 (2011): 1664-1670. Zhu, Yangzhi, Kalpana Mandal, Ana Lopez Hernandez, Satoru Kawakita, Wei Huang, Praveen Bandaru, Samad Ahadian et al. "State of the art in integrated biosensors for organ-on-a-chip applications." Current Opinion in Biomedical Engineering 19 (2021): 100309. Liu, Z., Zhou, Y., Lu, J. et al. Microfluidic biosensors for biomarker detection in body fluids: a key approach for early cancer diagnosis. Biomark Res 12 , 153 (2024). https://doi.org/10.1186/s40364-024-00697-4 Bargahi, N., Ghasemali, S., Jahandar-Lashaki, S. et al. Recent advances for cancer detection and treatment by microfluidic technology, review and update. Biol Proced Online 24 , 5 (2022). https://doi.org/10.1186/s12575-022-00166-y Akgönüllü S, Bakhshpour M, Pişkin AK, Denizli A. Microfluidic Systems for Cancer Diagnosis and Applications. Micromachines (Basel). 2021 Oct 31;12(11):1349. doi: 10.3390/mi12111349. PMID: 34832761; PMCID: PMC8619454. Thenuwara, G.; Javed, B.; Singh, B.; Tian, F. Biosensor-Enhanced Organ-on-a-Chip Models for Investigating Glioblastoma Tumor Microenvironment Dynamics. Sensors 2024 , 24 , 2865. https://doi.org/10.3390/s24092865 Additional Declarations No competing interests reported. Supplementary Files SupplementaryInformationfinal.docx Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 20 Oct, 2025 Reviews received at journal 19 Oct, 2025 Reviews received at journal 15 Oct, 2025 Reviewers agreed at journal 07 Oct, 2025 Reviewers agreed at journal 06 Oct, 2025 Reviewers invited by journal 06 Oct, 2025 Editor assigned by journal 05 Oct, 2025 Submission checks completed at journal 05 Oct, 2025 First submitted to journal 05 Oct, 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-7785293","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":530577329,"identity":"2c808862-bd59-4428-ab14-340c8ac33b16","order_by":0,"name":"Joshua P. A. Daoud","email":"","orcid":"","institution":"Queen Mary University of London","correspondingAuthor":false,"prefix":"","firstName":"Joshua","middleName":"P. A.","lastName":"Daoud","suffix":""},{"id":530577330,"identity":"456abd2d-c156-4e48-b002-380f26aa4f37","order_by":1,"name":"Alexandra E. Collisson","email":"","orcid":"","institution":"Queen Mary University of London","correspondingAuthor":false,"prefix":"","firstName":"Alexandra","middleName":"E.","lastName":"Collisson","suffix":""},{"id":530577331,"identity":"0e48c0f2-0f4a-4d5f-9a77-b6015297e1c9","order_by":2,"name":"Daniya Asaid","email":"","orcid":"","institution":"Queen Mary University of London","correspondingAuthor":false,"prefix":"","firstName":"Daniya","middleName":"","lastName":"Asaid","suffix":""},{"id":530577332,"identity":"c20d6cd2-432a-48ae-89dd-4745860b9e11","order_by":3,"name":"Ester Clarisse do Couto Lopes","email":"","orcid":"","institution":"Queen Mary University of London","correspondingAuthor":false,"prefix":"","firstName":"Ester","middleName":"Clarisse do Couto","lastName":"Lopes","suffix":""},{"id":530577333,"identity":"f8789a4a-35ef-45f0-bb83-4ec678bb5cac","order_by":4,"name":"Dimitrios Paraskevopoulos","email":"","orcid":"","institution":"Barts Health NHS Trust","correspondingAuthor":false,"prefix":"","firstName":"Dimitrios","middleName":"","lastName":"Paraskevopoulos","suffix":""},{"id":530577334,"identity":"cc12f6f5-aa7a-478d-81b4-90ea4756951d","order_by":5,"name":"Christopher A. R. Chapman","email":"data:image/png;base64,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","orcid":"","institution":"Queen Mary University of London","correspondingAuthor":true,"prefix":"","firstName":"Christopher","middleName":"A. R.","lastName":"Chapman","suffix":""}],"badges":[],"createdAt":"2025-10-05 14:38:25","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7785293/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7785293/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":93774899,"identity":"61eda06f-c370-4f84-a09e-b61ce7899411","added_by":"auto","created_at":"2025-10-17 12:26:22","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5174970,"visible":true,"origin":"","legend":"","description":"","filename":"FinalManuscriptfinal.docx","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/7c8571e84c8a04d533151ccd.docx"},{"id":93774884,"identity":"6f2c01d9-b594-4110-881e-f0c4a9ef2f25","added_by":"auto","created_at":"2025-10-17 12:26:21","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8797,"visible":true,"origin":"","legend":"","description":"","filename":"37997ddc908c41d5b75d41bc1dec8389.json","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/1eae5f37f681d59925269180.json"},{"id":93776008,"identity":"290842c9-fe28-426c-a53b-9c685b809290","added_by":"auto","created_at":"2025-10-17 12:34:22","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5483679,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryInformationfinal.docx","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/4eebb338c1dc71e463473ce2.docx"},{"id":93774886,"identity":"8a36aa48-b55a-4139-9acc-cd65a8fd6085","added_by":"auto","created_at":"2025-10-17 12:26:22","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":101450,"visible":true,"origin":"","legend":"","description":"","filename":"37997ddc908c41d5b75d41bc1dec83891enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/154835b046cdee965db6ba38.xml"},{"id":93778913,"identity":"fd418442-406f-4149-9ea4-43b050793561","added_by":"auto","created_at":"2025-10-17 12:50:22","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":354608,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/4fd4ee2d06c3a44f77f52c37.png"},{"id":93777621,"identity":"854a0eb3-4f6c-4734-9ffc-d33fc2a2f8bb","added_by":"auto","created_at":"2025-10-17 12:42:22","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1272144,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/230620b51838eab32ba2a0ac.png"},{"id":93776001,"identity":"6c1b44ec-7c77-4624-82ee-508df9be7ac9","added_by":"auto","created_at":"2025-10-17 12:34:22","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":283944,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/e4fa7d84103baa1cef5377ae.png"},{"id":93774895,"identity":"481ab811-8294-4650-a1fa-86af89ca66fc","added_by":"auto","created_at":"2025-10-17 12:26:22","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":669659,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/d6b9abbc85ece5de7e657d0a.jpeg"},{"id":93776005,"identity":"e0ac0ff1-8a04-4ee0-ae57-37016a65bdef","added_by":"auto","created_at":"2025-10-17 12:34:22","extension":"jpeg","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":796032,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/a38d6edd0fca4c23484072e5.jpeg"},{"id":93776004,"identity":"77437028-13fe-4af6-8a22-7b2a4e124fbe","added_by":"auto","created_at":"2025-10-17 12:34:22","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":468358,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/fbfe6e3e6820e91d8576c059.png"},{"id":93774897,"identity":"c9519e60-5dfd-47bc-a1fc-4dae4b2262f1","added_by":"auto","created_at":"2025-10-17 12:26:22","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":919967,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/3244cf38e791cca6169a4ca8.png"},{"id":93774900,"identity":"ea21633f-a41e-4572-9543-835583c819f6","added_by":"auto","created_at":"2025-10-17 12:26:22","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":48036,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/536436cf45b3c153b00fe398.png"},{"id":93774910,"identity":"d6d78906-f41d-40dd-8ec8-bd329d097711","added_by":"auto","created_at":"2025-10-17 12:26:22","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":169190,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/b12df97239ddd3f7ca69b2a1.png"},{"id":93774906,"identity":"97d6bb7c-5557-48e3-961a-e56aef5e6457","added_by":"auto","created_at":"2025-10-17 12:26:22","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":49605,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/24b2e96eee22e55a1dd7fc16.png"},{"id":93774907,"identity":"5aea079f-7660-42cc-babf-ec8b973a60cf","added_by":"auto","created_at":"2025-10-17 12:26:22","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":125997,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/2821d490585008687d0b2810.png"},{"id":93774908,"identity":"9225a90d-4a5d-4687-8a07-41c212bed883","added_by":"auto","created_at":"2025-10-17 12:26:22","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":229688,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/e42b7325a659f69f8c082af9.png"},{"id":93778914,"identity":"0f25ca63-797d-42ed-812c-08e2ee827ddb","added_by":"auto","created_at":"2025-10-17 12:50:22","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":89885,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/6fda804ca9612be31ed6e190.png"},{"id":93774902,"identity":"0759694c-904f-4db4-a304-9bc3cd54fbd1","added_by":"auto","created_at":"2025-10-17 12:26:22","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":111939,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/f044cb248d7325e3a2667911.png"},{"id":93774901,"identity":"e436e603-dc1e-47a8-b361-323181a17054","added_by":"auto","created_at":"2025-10-17 12:26:22","extension":"xml","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":101374,"visible":true,"origin":"","legend":"","description":"","filename":"37997ddc908c41d5b75d41bc1dec83891structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/a16c2808900cf8efb5bb3be3.xml"},{"id":93777624,"identity":"0fa87f7a-fff3-4d52-8cf0-2454bde989ab","added_by":"auto","created_at":"2025-10-17 12:42:22","extension":"html","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":110605,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/48e9ce45ec91bc8b9ef56faf.html"},{"id":93778912,"identity":"0e286b33-26b0-4fe7-a404-e6892a0e62b2","added_by":"auto","created_at":"2025-10-17 12:50:22","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":354608,"visible":true,"origin":"","legend":"\u003cp\u003e(A) Photomask design showing 4 microfluidic channels (B) Mould construction stages showing how microfluidic channel moulds were fabricated using photolithography (C) Stages of microfluidic chip production using the channel mould and Polydimethylsiloxane (PDMS) which was then plasma bonded onto a glass microscope slide.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/175f877742b00fd372f79f11.png"},{"id":93774882,"identity":"3b298924-6eae-4b8a-8cd3-eca3e16dfcb9","added_by":"auto","created_at":"2025-10-17 12:26:21","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":588232,"visible":true,"origin":"","legend":"\u003cp\u003eImage displaying the set-up of microfluidic chips inside of a glass petri dish including the PDMS rings added to channel surface, and reservoirs of D-PBS.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/7409b1d484a7a50dd7f86c83.png"},{"id":93774885,"identity":"1ed10789-0fef-4c46-a828-2dce993cb08c","added_by":"auto","created_at":"2025-10-17 12:26:22","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":282324,"visible":true,"origin":"","legend":"\u003cp\u003e(A) Image of microfluidic chip displaying the channel dimensions, (B) and (C) Profilometer measurements of microfluidic channel dimensions in Profilm3D.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/d707df56329ac3cec7ad0305.png"},{"id":93774887,"identity":"8ecfba96-01af-460e-96f5-be14fe21ae71","added_by":"auto","created_at":"2025-10-17 12:26:22","extension":"jpeg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":669659,"visible":true,"origin":"","legend":"\u003cp\u003e(A) i-iv = brightfield images of 9L/3cmv-GFP spheroid growth \u003cem\u003ein vitro\u003c/em\u003e; v-viii = brightfield images of 9L/lacZ spheroid growth \u003cem\u003ein vitro \u003c/em\u003e(B) - Average spheroid diameter (µm) after 5 days of growth for 9L/3cmv-GFP and 9L/lacZ cell lines; (C) – 9L/3cmv-GFP spheroid diameters measured over time when grown at different cell densities. Error bars were included for all standard deviation values.\u003c/p\u003e","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/d4e2b3777bd0c91bca8669e6.jpeg"},{"id":93774889,"identity":"32275d88-7585-4812-ab52-ad63e949e104","added_by":"auto","created_at":"2025-10-17 12:26:22","extension":"jpeg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":796032,"visible":true,"origin":"","legend":"\u003cp\u003e(A) Comparison images for day 7 and 14 of co-culture samples with images showing the GFP channel (green - glioma cells), and mixed channels including GFP, DAPI (dark blue - nuclear DNA stain), Phalloidin (red – F-actin stain), (B) Comparison images for day 7 and 14 of glioma control samples (9l only) with images showing the GFP channel (green - glioma cells), and mixed channels including GFP, DAPI (dark blue - nuclear DNA stain), Phalloidin (red – F-actin stain), (C) Mean spheroid outgrowth (mm) for co-culture and 9L/3cmv-GFP only control. Error bars were included for all standard deviation values; ****\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.0001. For each sample, (\u003cem\u003en\u003c/em\u003e= 5), outgrowth measurementswere taken from the channel entrance to the furthest observable regions of outgrowth. For each sample, measurements were taken of the 20 furthest visible points of outgrowth for both sides of the channel, which were subsequently averaged to obtain the mean outgrowth.\u003c/p\u003e","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/d1f1ac089c4da9308b660aa7.jpeg"},{"id":93775999,"identity":"5e07fb72-8cf7-45f5-b2db-c0aa478fda3c","added_by":"auto","created_at":"2025-10-17 12:34:22","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":249942,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003e(\u003c/strong\u003eA) Images showing DAPI staining for each culture type (B) Graph showing calculated mean cell density (per mm\u003csup\u003e2\u003c/sup\u003e), which was calculated by analysing DAPI staining images to quantify individual cells within each culture; *\u003cem\u003ep\u003c/em\u003e=0.033, ****\u003cem\u003ep\u003c/em\u003e=0.0005. DAPI staining was used to identify the cell population in a designated area and repeated 10 times per sample (\u003cem\u003en\u003c/em\u003e = 5) to provide an average cell count for the culture type. Once the average cell counts were calculated, the image area was scaled up to obtain cell density values for number of cells/mm\u003csup\u003e²\u003c/sup\u003e.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/3dd532b596e6d7d97948a700.png"},{"id":93774890,"identity":"14c51de8-0912-4f77-9829-b58569351b96","added_by":"auto","created_at":"2025-10-17 12:26:22","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":598007,"visible":true,"origin":"","legend":"\u003cp\u003e(A) CD44 relative intensity graph showing data from glioma (‘9L only’) and astrocyte control samples and co-culture samples; *\u003cem\u003ep\u003c/em\u003e = 0.0303. (B) Fluorescent images representing CD44 intensity calculated from ROI measurements taken from each sample type (\u003cem\u003en\u003c/em\u003e = 5), Green = GFP, CD44 = far-red, DAPI = blue. (C) GFAP relative intensity graph showing data from astrocyte control samples and co-culture samples; 9L only: *\u003cem\u003ep\u003c/em\u003e = 0.0459, co-culture: *\u003cem\u003ep \u003c/em\u003e= 0.0288. (D) Fluorescent images representing GFAP intensity calculated from ROI measurements taken from each sample type (\u003cem\u003en\u003c/em\u003e = 5), Green = GFP, GFAP = red, DAPI = blue.\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/f6e3d31f5cf39286438b1558.png"},{"id":93780071,"identity":"e69858f6-1893-4114-bea3-a62a701d6447","added_by":"auto","created_at":"2025-10-17 12:58:23","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4237542,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/b9b8aa9d-c6a4-4f99-9b19-64b0eca38a18.pdf"},{"id":93774893,"identity":"34825639-e153-4db0-95b8-e25aa5238121","added_by":"auto","created_at":"2025-10-17 12:26:22","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":5483679,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryInformationfinal.docx","url":"https://assets-eu.researchsquare.com/files/rs-7785293/v1/ffe242eda2193d6030f00c8f.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Development and characterization of an on-chip glioma outgrowth model","fulltext":[{"header":"1 – Introduction","content":"\u003cp\u003eGlioblastoma multiforme, a high-grade glioma, is the most common intrinsic malignant brain tumour diagnosis in adults [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Patient outcomes are notoriously poor with a 5-year survival rate of less than 7% [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. These tumours are characterized as being substantially invasive, with an extremely high recurrence rate, as total surgical resection is challenging due to limitations on total resection volume. Due to the propensity for this cancer to recur, patient deterioration is inevitable, even after several repetitions of chemotherapeutic and radiotherapy treatments [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eCurrent practice relies on outpatient magnetic resonance imaging (MRI) or computed tomography (CT) scans to detect local cancer recurrences after surgical resection and inform neurooncological treatment [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Unfortunately, the time in between these scans is often too long to identify small changes in the tumour margin and rapidly inform neurooncologist leading to worsened patient outcomes [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Developing predictive in vitro models of glioblastoma recurrence has the potential to provide key feedback on treatment efficacy thus informing decisions about alternative therapies [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Current knowledge of the cellular mechanisms underpinning glioblastoma recurrence is limited by challenges in recording from patients and incomplete recapitulation of the healthy neural environment in glioblastoma organoid models [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. To address this, predictive in vitro models and on-chip platforms have gained attention for their ability to provide alternative models to provide a tuneable microenvironment in which to investigate cellular mechanisms.\u003c/p\u003e\u003cp\u003eRecently, three-dimensional spheroid models of gliomas have seen an increase in popularity in their use to simulate tumour behaviour [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. These models have been shown to retain near-identical genetic expression to the original tumour [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e] which enables researchers to investigate cellular processes as they would naturally occur in the brain [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. However, most of the developed spheroid models are focused on enabling more realistic testing of chemotherapeutic efficacy for drug discovery. Therefore, only a limited amount of research has been focused on developing models that fully recapitulate the neural environment in which glioblastoma occurs. Without these interactions it is challenging to accurately investigate the complex interactions between glioma cells and healthy neural cells like astrocytes, oligodendrocytes, and neurons [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. There is clear evidence that the interactions between glioma and healthy neural tissue is a critically important component to the efficacious treatment of gliomas, it is essential to examine their hallmarks using \u003cem\u003ein vitro\u003c/em\u003e models to simulate \u003cem\u003ein vivo\u003c/em\u003e growth conditions [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e] The development of a co-culture setup to culture healthy neural cells alongside glioma cells to investigate infiltration mechanisms is necessary to facilitate further understanding of the methods behind tumour recurrence [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eBy simulating a physiologically relevant microenvironment, microfluidic platforms offer solutions to the methodological limitations which are inherent in animal models and traditional culture systems by the ability to engineer the cell culture geometry [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. The rationale of this work was to produce a customizable on-chip neural system to enable the combination of a glioma spheroid model with healthy neural tissue. From this model, the aim was to enable the investigation of how glioma cells grow into the surrounding healthy neural tissue in this on-chip system by analysing tumour outgrowth, cell density, and key marker protein expression between healthy and cancerous cells during different stages of outgrowth. This work highlights how the system developed enables an advantageous platform over traditional cell models by providing a easily adapted system to enable clear path for glioma outgrowth.\u003c/p\u003e"},{"header":"2 – Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1 - Micropatterning\u003c/h2\u003e\u003cp\u003ePhotolithography masks for the microfluidic channels were designed using computer-aided design software AutoCAD (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA) and purchased from Microlitho (UK). Channel moulds were created by spin coating SU-8 photoresist (Kayaku) at 500 rpm for 5 seconds, then 1000 rpm for 30 seconds onto blank silicon wafers and exposing for 8 seconds at 100% power (360 mW/cm\u003csup\u003e2\u003c/sup\u003e) with a 365 nm UV source (KLO\u0026Eacute;). After developing, masks were rinsed with Propylene Glycol Monomethyl Ether Acetate (PGMEA) and baked at 150\u0026deg;C for 3 minutes. After mould creation, microfluidic channels were fabricated from cast Sylgard-184 Silicone Elastomer (Farnell UK) at 80\u0026deg;C for 1.5 hours.\u003c/p\u003e\u003cp\u003eChannel dimensions were analyzed using an optical profilometer (Omniscan), then visualised in Profilm3D. Figure\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB and \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC illustrate the workflow of microfluidic channel fabrication.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2 - Cell culture\u003c/h2\u003e\u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eCell preparation\u003c/span\u003e\u003c/p\u003e\u003cp\u003eThe 9L/3cmv-GFP glioma cell line, were previously transfected with green fluorescent protein (GFP) and donated from the laboratory of Prof. Rylie Green at Imperial College London. 9L/3cmv-GFP cells were established and cultured in Dulbecco's Modified Eagle Medium (DMEM), with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin (Pen/Strep) all purchased from ThermoFisher Scientific. Cells were incubated at 37\u0026deg;C and 5% CO\u003csub\u003e2\u003c/sub\u003e, with passaging performed every 4\u0026ndash;5 days. For astrocytes, the CTX-TNA2 cell line (UK Health Security Agency) were cultured using the same media formulation.\u003c/p\u003e\u003cp\u003e5 days before co-culture experiments, 9L/3cmv-GFP cells were grown into glioma spheroids using a Corning 96-well spheroid-forming non-adherent microplate (Sigma-Aldrich) and DMEM media. Media was replenished on day 2 to optimise viability and minimise media loss through evaporation. After optimisation of spheroid size, as presented in Results 3.2, the glioma spheroids were grown from a cell concentration of 150,000 cells per mL, which grew spheroids measuring 500\u0026ndash;600 mm in diameter within 5 days of growth, which were appropriately sized for visualisation inside microfluidic channels.\u003c/p\u003e\u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eMicrofluidic preparation for cell culture\u003c/span\u003e\u003c/p\u003e\u003cp\u003eFirst, microfluidic channels were sterilized using autoclave sterilization at 121\u0026deg;C and flushed with sterile phosphate buffer solution (D-PBS). Microfluidic channels were also coated using a solution which consisted of 25 \u0026micro;g/ml Poly-D-Lysine and 10 \u0026micro;g/ml Laminin, which provided improved adhesion allowing astrocyte cells to settle and adhere onto the glass slide. No further growth factors were added to prevent any interference with cell-cell interactions [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Full coating protocol is provided in supplementary information. Figure\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e shows additional features of the microfluidic chip design, including rings which were cut from a flat disc of cured PDMS and adhered onto the upper surface of the chip using KORASILON silicone paste (Sigma-Aldrich), as this created a reservoir of DMEM media. Small reservoirs of sterile D-PBS were also kept inside of the petri dish throughout the duration of the experiment. These steps prevented excessive media evaporation, enabling longer experimental running times.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eOutgrowth experiments\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003cem\u003eCo-culture\u003c/em\u003e:\u003c/p\u003e\u003cp\u003eAstrocyte cell suspension was added into each microfluidic chip at a seeding density of 50,000 cells per mL. This was achieved by adding 2 x 200 \u0026micro;l of astrocyte cell suspension to each end of the microfluidic channel to give a total volume of 800 \u0026micro;l added to each chip. This ensured even distribution of cells throughout the channel.\u003c/p\u003e\u003cp\u003eAfter the addition of astrocytes, one glioma spheroid was added into the centre well of the microfluidic channel. Co-culture samples were incubated at 37\u0026deg;C and 5% CO\u003csub\u003e2\u003c/sub\u003e and allowed to grow without interruption for 48 hours to allow cells to adhere. After the initial 48-hour period, the first media change was performed using fresh media, then after 96 hours, a second media change was performed. After the 96-hour period, subsequent media changes were performed daily to prevent the buildup of cell debris. Co-culture experiments were grown until two timepoints: 7 days and 14 days.\u003c/p\u003e\u003cp\u003eControl experiments consisted of glioma spheroids grown in isolation and CTX-TNA2 astrocyte cells grown in isolation inside of the microfluidic platform. Control samples were also grown for 7 and 14 days. Control samples received the same media replenishment schedule as co-culture samples (Supplementary Information Fig.\u0026nbsp;0 for more detail). Total outgrowth was calculated from fluorescent images in the ImageJ program by measuring the furthest detectable outgrowth of glioma cells from the 9L/3cmv-GFP spheroid. Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e (supplementary information) shows how outgrowth calculations were obtained.\u003c/p\u003e\u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eCD44 / GFAP Expression Experiments\u003c/span\u003e\u003c/p\u003e\u003cp\u003eThese experiments were carried out by seeding cells into microfluidic chips so that all samples had the same number of cells inside of the chip. Astrocyte control samples and glioma control samples (9L/3cmv-GFP) were both seeded with cell suspensions at concentrations of 100,000 cells per mL, and co-culture samples were seeded with a 50:50 mix of astrocyte and glioma cells with a final concentration of 100,000 cells per mL. Antibody experiments were carried out for two timepoints: 4 and 7 days, to ensure microfluidic chips did not become too confluent for single-cell analysis.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e2.3 - Cell staining and imaging\u003c/h2\u003e\u003cp\u003eTwo staining protocols were carried out, one for outgrowth experiments and another for antibody experiments. Full staining protocols are provided in the supplementary information.\u003c/p\u003e\u003cp\u003eIn all experiments, 9L/3cmv-GFP and CTX-TNA2 cells were fixed inside of microfluidic chips using 4% paraformaldehyde (PFA) and then permeabilised with 0.1% Triton-X solution. Before staining protocol was performed, microfluidic channels were washed several times with phosphate-buffered saline (PBS).\u003c/p\u003e\u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eOutgrowth staining\u003c/span\u003e\u003c/p\u003e\u003cp\u003eFor the outgrowth experiments, the glioma cell line already expressed GFP, so no fluorescent green stain was needed, but Texas-Red Phalloidin (ThermoFisher Scientific) and 4\u0026prime;,6-diamidino-2-phenylindole (DAPI) (ThermoFisher Scientific) were used to selectively stain F-actin filaments and nuclear DNA in red and blue respectively. Samples were used from day 7 and 14 of the outgrowth experiments.\u003c/p\u003e\u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eAntibody staining\u003c/span\u003e\u003c/p\u003e\u003cp\u003eFurther staining using primary antibodies CD44 Recombinant Rabbit Monoclonal Antibody (Invitrogen) and\u003c/p\u003e\u003cp\u003eGFAP Mouse Monoclonal Antibody (GA5) (Invitrogen), with secondary antibodies Cy\u0026trade;5 AffiniPure Goat-Anti-Rabbit IgG (Jackson ImmunoResearch) and Cy\u0026trade;3 AffiniPure Goat-Anti-Mouse IgG (Jackson ImmunoResearch) were carried out to compare control samples with co-culture between glioma and astrocyte cells. Antibody staining samples were grown for two timepoints: day 4 and day 7.\u003c/p\u003e\u003cp\u003ePrimary antibodies for the CD44 antigen and Glial Fibrillary Acid Protein (GFAP) were used on all samples. Full antibody staining protocol can be found in supplementary information. Supplementary information Figure S3 depicts how relative intensities were calculated from fluorescent antibody staining images using several regions-of-interest (ROIs) inside of ImageJ. CD44 antigen was stained with Cy5 secondary antibody (far-red), and GFAP was stained with Cy3 secondary antibody (red). GBM cells can be visualised by GFP fluorescence (green). DAPI (dark blue) nuclear DNA stain was used to locate and select cells for further analysis.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003e2.4 - Analysis\u003c/h2\u003e\u003cp\u003eImaging was performed using fluorescent confocal microscopy with the Nikon CSU-W1 SoRa Spinning Disc Confocal microscope. Images were then viewed and processed using the Nikon Elements Viewer software.\u003c/p\u003e\u003cp\u003eAfter processing, fluorescent images were analysed using ImageJ software. Measurements of spheroid outgrowth distance were taken from fluorescent images of the co-culture and spheroid control samples. Measurements of cell density were also calculated for all samples. Supplementary information Fig S2 illustrates how the cell density was calculated from fluorescent images. Antibody staining images were also analysed using ImageJ to determine the relative intensity of antibodies in each sample.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003e2.5 - Statistics\u003c/h2\u003e\u003cp\u003eOutgrowth measurements, cell densities, and relative intensities of antibodies were all calculated from fluorescent images in ImageJ (Fiji) software. GraphPad Prism software was used to carry out inferential statistical tests including unpaired and paired sample t-tests, and Mann-Whitney U tests if equal variances could not be assumed. Unless otherwise stated each experiment had a sample size of \u003cem\u003en\u003c/em\u003e\u0026thinsp;=\u0026thinsp;5.\u003c/p\u003e\u003c/div\u003e"},{"header":"3 – Results","content":"\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003e3.1 - Microfluidic platform dimensions\u003c/h2\u003e\u003cp\u003eMedia loss by evaporation is a common issue in cell culture, which can lead to drastic differences in data as a result, particularly with the periphery of cell culture vessels, known by researchers as the \u0026lsquo;edge effect\u0026rsquo; [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. To counter this issue and improve data reliability, channel dimensions of 15,600 \u0026micro;m (length) x 300 \u0026micro;m (depth) x 500 \u0026micro;m (width) were used to support larger media volumes (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). Final dimensions for the microfluidic channel (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB and \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC) were confirmed by profilometer to be 15,600 \u0026micro;m (length) x 300 \u0026micro;m (depth) x 500 \u0026micro;m (width).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003e3.2 \u0026ndash; Optimisation of Spheroid Size\u003c/h2\u003e\u003cp\u003eSpheroids were grown from the 9L/3cmv-GFP glioma cell line and compared with previous experimental data obtained from spheroids grown from a non-transfected 9L/lacZ glioma cell line. This was to ensure that transfection did not alter the spheroid growth characteristics, so that transfected cells could be used within co-culture experiments. A full spheroid preparation protocol can be found in supplementary information.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eA proportional relationship between cell seeding density and spheroid diameter was seen. The desired spheroid diameter was 500\u0026ndash;600 mm as this diameter would enable spheroid outgrowth to occur in a linear fashion inside of the microfluidic channel. A cell seeding density of 150,000 cells per mL was selected due to this producing an appropriate size spheroid after 5 days of growth with minimal variation between spheroids. Images taken of the spheroids clearly show an increasingly dense and dark formation of cells known as the necrotic core, which is a physiologically relevant hallmark of a tumour, suggesting that the spheroids closely simulated tumour physiology [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e].\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003e3.3 \u0026ndash; Glioma outgrowth in co-culture and monoculture\u003c/h2\u003e\u003cp\u003eFig 5A highlights the differences in spheroid outgrowth distance between co-culture samples (containing glioma and astrocytes), and the glioma spheroid control samples (‘9L only’ on graph). In co-culture samples, the mean outgrowth was measured at 1428 μm (SD = 556) on day 7, which increased to 1969 μm (SD = 796) by day 14. Fig 5B demonstrates how in glioma only controls, the mean outgrowth was measured at 823 μm (SD = 243) on day 7 which increased to 1433 μm (SD = 342) by day 14. Fig 5C reveals the highly significant differences in outgrowth between co-culture and glioma control samples across both day 7 and day 14 timepoints (p \u003c 0.0001). Mean outgrowth data for co-culture samples showed a percentage difference of +73.5% compared to glioma controls on day 7, and +37.4% on day 14. \u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003e3.4 - Cell density\u003c/h2\u003e\u003cp\u003eFigure\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB highlights significant changes in cell density between co-culture and monoculture samples with. The mean cell density was calculated to be 2781/mm\u003csup\u003e2\u003c/sup\u003e on day 7, which increased to 6259/mm\u003csup\u003e2\u003c/sup\u003e by day 14, representing a significant increase in cell density between days 7 and 14 of growth (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0005). For the astrocyte control sample, the mean cell density, was calculated to be 2145/mm\u003csup\u003e2\u003c/sup\u003e on day 7, which increased to 3015/mm\u003csup\u003e2\u003c/sup\u003e by day 14 (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.033). For the glioma control sample (shown as 9L only), the mean cell density was calculated to be at 1261/mm\u003csup\u003e2\u003c/sup\u003e on day 7 which increased minimally to 1300/mm\u003csup\u003e2\u003c/sup\u003e by day 14, no significant difference was observed.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003e3.5 \u0026ndash; Antibody staining: CD44 and GFAP\u003c/h2\u003e\u003cp\u003eFigure\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eA shows a statistically significant increase in relative intensity of CD44 (far-red) in co-culture samples, as the relative intensity (a.u) increased from 40 on day 4 to 64 by day 7 (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0303). Relative intensity of CD44 for the glioma control samples (shown as 9L only) were non-significant with an increase in relative intensity (a.u) from 14 on day 4 to 19 by day 7. Astrocyte control samples were also non-significant, as the relative intensity (a.u) remained at 34 on both day 4 and day 7. Figure\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eB provides a visual representation of CD44 intensities for all samples.\u003c/p\u003e\u003cp\u003eFigure 7C shows a statistically significant increase in relative intensity of GFAP (red) between co-culture timepoints with the relative intensity (a.u) increasing from 10 on day 4 to 13 on day 7 (p = 0.0288). Relative intensities of GFAP between the astrocyte control timepoints were non-significant, as the relative intensity (a.u) remained at 8 for both day 4 and day 7. We observed a significant decrease in GFAP relative intensity (a.u) for glioma control samples from 8 on day 4 to 5 on day 7 (p = 0.0459). \u003c/p\u003e\u003c/div\u003e"},{"header":"4 - Discussion","content":"\u003cp\u003e\u003cb\u003eMechanistic insights of glioma outgrowth\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe substantial outgrowth and increased cell density seen in glioma spheroids co-cultured with astrocytes compared to spheroids grown alone is supported by existing research which suggests that nearby astrocytes undergo malignant transformation into a more tumour-permissive phenotype, which may facilitate a supportive environment which promotes glioma invasion [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eThe results indicated that after prolonged contact with astrocytes in a co-culture environment, that glioma cells showed a relative increase in intensity of CD44, which could suggest that increased contact with astrocytes may contribute to the invasive properties of a glioma by increasing CD44 expression. Heightened CD44 expression has been shown to play a key role in glioma recurrence, as it is associated with invasion and proliferation of tumour cells by initiating cellular signalling pathways that give rise to the highly invasive glioma stem cell phenotype [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. High expression of CD44 is also associated with poor prognosis and early tumour recurrence in glioma patients, compared to those with lower CD44 expression [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. Furthermore, a substantial increase in cell density for glioma cells co-cultured with astrocytes was observed in comparison to controls. These observations are supported by literature which describes how astrocytes actively facilitate glioma proliferation and invasion by activating CD44 signalling pathways, which induces tumour cell transformation to the highly invasive and proliferative glioma stem cell phenotype [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eAdditionally, the relative intensity of GFAP increased significantly in when astrocytes were co-cultured with glioma cells, compared to astrocyte control samples. GFAP is a marker for reactive astrocytes whereby astrocytes transform in response to cellular injury, stress or disease [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. Supporting evidence suggests that the presence of reactive astrocytes is involved in regulation of glioma progression through many mechanisms, but primarily through direct cell-to-cell interaction with tumour cells which enhances their survival, proliferation and invasion into healthy tissue [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e].\u003c/p\u003e\u003cp\u003e\u003cb\u003eAdvantages of Microfluidic platform\u003c/b\u003e\u003c/p\u003e\u003cp\u003eA notable advantage of this platform is that it allows for precise control over directional spheroid growth by forcing growth into the directional microfluidic channels. This presents a significant advantage over co-cultures grown inside of circular multi-well plates. Recent work in this area has performed spheroid co-cultures in multi-well plates, but have illustrated how spheroid growth is multi-directional, meaning outgrowth measurements cannot be accurately quantified by measurements taken from a set boundary in a single direction [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. As a comparison, the same co-culture used here was performed in a stand-alone 96-well plate and imaged. The spheroids exhibited outgrowth, however it was sporadic and inconsistent around the periphery, highlighting the need for culture methods which can improve the precision and accuracy of outgrowth measurements (Figure S4).\u003c/p\u003e\u003cp\u003e\u003cb\u003eCritical Evaluations\u003c/b\u003e\u003c/p\u003e\u003cp\u003eOverall, this research has produced an engineered \u003cem\u003ein vitro\u003c/em\u003e model for visualisation and quantification of glioma invasion into healthy tissue using microfluidic \u0026ldquo;on-chip\u0026rdquo; technology. One major limitation of the approach taken in this work is that, although the \u003cem\u003ein vitro\u003c/em\u003e model included three-dimensional elements such as spheroids, outgrowth was confined to a two-dimensional space along the bottom of the channel. This could have significantly altered the results in comparison to a fully three-dimensional model. Incorporation of a hydrogel into the model could facilitate a fully three-dimensional culture environment by mimicking the extracellular matrix [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. Furthermore, the inclusion of other neural cell types such as neurons and microglia could have highlighted additional cellular interactions with tumour cells [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e].\u003c/p\u003e\u003cp\u003e\u003cb\u003eFuture Directions\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThis on-chip platform presents a highly feasible model which can be used to culture patient-derived tumour spheroids with healthy neural tissue to produce more physiologically relevant models of unique tumour pathologies [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. Microfluidic models demonstrate a viable method for accurately modelling patient-specific glioma outcomes and provide an indication of treatment efficacy by using heterogenous spheroid cultures from patients to simulate treatment responses [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eNext-generation predictive in vitro models are addressing the need for improved preclinical tools which can recapitulate the tumour microenvironment and simulate underlying biological processes that drive treatment response and clinical outcomes [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e] This research adds to an existing evidence base [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e] which strongly supports the practicality of microfluidic platforms for oncological modelling, as they show great potential to advance the future of cancer research. Additionally, the development of this customized on-chip neural systems is further strengthened by the potential for addition of real-time data recording capabilities to provide additional feedback on cell state [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eBiosensor integration into microfluidic systems has been widely explored in recent years, with several studies demonstrating their potential for continuous, real-time monitoring of cellular and molecular events [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e][\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e][\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]. Building on these advances, future work will focus on incorporating such sensing technologies into this glioma on-chip platform to enhance its analytical capabilities. Electrochemical biosensors, including impedance-based devices will be embedded within the microfluidic model to monitor parameters such as tumour microenvironment changes, metabolic activity, and drug response providing deeper insight into disease progression, recurrence and treatment response [\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e].\u003c/p\u003e"},{"header":"5 – Conclusion","content":"\u003cp\u003eThis work has outlined the developmental stages and application of an on-chip neural model of glioma recurrence. Using microfluidic technologies to guide glioma outgrowth it was possible to observe glioma growth in co-culture and monoculture control samples, as well as measure outgrowth from the tumour spheroid, and obtain data about the expression of relevant cellular markers which influence tumour invasion. Use of a microfluidic platform enabled creation of customisable channel dimensions to improve culture conditions and obtain reproducible data. The initial data presented in this study demonstrated that our microfluidic platform can be used as a practical and viable method for modelling glioma invasion into healthy tissue and provided further evidence toward visualizing cellular crosstalk and its important role in cancer cell migration and metastasis. Additionally, the ability to model visualize these changes in co-culture could have promising implications for further research to determine cancer metastasis and invasion mechanisms so that a deeper understanding can be gained about facilitatory and inhibitory factors which drive growth and survival of residual cancer cells.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics Approval\u003c/strong\u003e\u003cp\u003eEthics approval was not required for this work.\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eJ.D. performed all experiments including data analysis and authored the manuscript. A.C. assisted with statistical analysis and final edits to the manuscript. D.A. contributed experimental data for supplementary information. E.L. provided written sections of the manuscript. D.P. provided financial support. C.C. led project development, provided financial support, and provided final edits to the manuscript\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eWe would like to thank Dr. Liisa Marketa Blowes for access and support in the use of the CREATE Lab at Queen Mary University of London, as well as key guidance on cell staining, and the School of Engineering and Materials Science for providing access to their state-of-the-art organ-on-chip facilities. We would also like to also thank the Iskratsch Lab at Queen Mary University of London for their assistance with immunofluorescent staining, and the Green Lab at Imperial College London for donation of the 9L/3cmv-GFP cell line.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eAll analysis methods are provided in the manuscript and supplementary information. Raw data including images and analyzed numberical data is available upon request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eLouis, D. N., Perry, A., Reifenberger, G., Von Deimling, A., Figarella-Branger, D., Cavenee, W. K., ... \u0026amp; Ellison, D. W. (2016). The 2016 World Health Organization classification of tumors of the central nervous system: a summary. \u003cem\u003eActa neuropathologica\u003c/em\u003e, \u003cem\u003e131\u003c/em\u003e, 803-820.\u003c/li\u003e\n\u003cli\u003eWoodworth, Davis C., Whitney B. Pope, Linda M. Liau, Hyun J. Kim, Albert Lai, Phioanh L. Nghiemphu, Timothy F. Cloughesy, and Benjamin M. Ellingson. \u0026quot;Nonlinear distortion correction of diffusion MR images improves quantitative DTI measurements in glioblastoma.\u0026quot; \u003cem\u003eJournal of Neuro-oncology\u003c/em\u003e116 (2014): 551-558.\u003c/li\u003e\n\u003cli\u003eLenting, Krissie, Roel Verhaak, Mark Ter Laan, Pieter Wesseling, and William Leenders. \u0026quot;Glioma: experimental models and reality.\u0026quot; \u003cem\u003eActa neuropathologica\u003c/em\u003e 133 (2017): 263-282.\u003c/li\u003e\n\u003cli\u003eNational Institute for Health and Care Excellence (NICE). Brain tumours (primary) and brain metastases in over 16s: NICE guidelines. NG99. (2019) Available at: https://www.nice.org.uk/guidance/ng99/resources.\u003c/li\u003e\n\u003cli\u003eMahvi, David A., Rong Liu, Mark W. Grinstaff, Yolonda L. Colson, and Chandrajit P. Raut. \u0026quot;Local cancer recurrence: the realities, challenges, and opportunities for new therapies.\u0026quot; \u003cem\u003eCA: a cancer journal for clinicians\u003c/em\u003e 68, no. 6 (2018): 488-505.\u003c/li\u003e\n\u003cli\u003ePalmer, G., A. Martling, B. Cedermark, and T. Holm. \u0026quot;A population-based study on the management and outcome in patients with locally recurrent rectal cancer.\u0026quot; \u003cem\u003eAnnals of surgical oncology\u003c/em\u003e 14 (2007): 447-454.\u003c/li\u003e\n\u003cli\u003eWarren, Joan L., and K. Robin Yabroff. \u0026quot;Challenges and opportunities in measuring cancer recurrence in the United States.\u0026quot; \u003cem\u003eJournal of the National Cancer Institute\u003c/em\u003e 107, no. 8 (2015): djv134.\u003c/li\u003e\n\u003cli\u003e\u0026ndash; Atat, Oula El, Zahra Farzaneh, Mahsa Pourhamzeh, Fatima Taki, Ralph Abi-Habib, Massoud Vosough, and Mirvat El-Sibai. \u0026quot;3D modeling in cancer studies.\u0026quot; \u003cem\u003eHuman Cell\u003c/em\u003e (2022): 1-14.\u003c/li\u003e\n\u003cli\u003ePaolillo, Mayra, Sergio Comincini, and Sergio Schinelli. \u0026quot;In vitro glioblastoma models: A journey into the third dimension.\u0026quot; \u003cem\u003eCancers\u003c/em\u003e 13, no. 10 (2021): 2449.\u003c/li\u003e\n\u003cli\u003eEdmondson, Rasheena, Jessica Jenkins Broglie, Audrey F. Adcock, and Liju Yang. \u0026quot;Three-dimensional cell culture systems and their applications in drug discovery and cell-based biosensors.\u0026quot; \u003cem\u003eAssay and drug development technologies\u003c/em\u003e12, no. 4 (2014): 207-218.\u003c/li\u003e\n\u003cli\u003eSuva, Mario L., and Itay Tirosh. \u0026quot;The glioma stem cell model in the era of single-cell genomics.\u0026quot; \u003cem\u003eCancer cell\u003c/em\u003e 37, no. 5 (2020): 630-636.\u003c/li\u003e\n\u003cli\u003eDiao, Wenwen, Xuezhi Tong, Cheng Yang, Fengrong Zhang, Chun Bao, Hao Chen, Liyu Liu et al. \u0026quot;Behaviors of glioblastoma cells in in vitro microenvironments.\u0026quot; \u003cem\u003eScientific reports\u003c/em\u003e 9, no. 1 (2019): 85.\u003c/li\u003e\n\u003cli\u003eDePalma, Thomas J., Hemamylammal Sivakumar, and Aleksander Skardal. \u0026quot;Strategies for developing complex multi-component in vitro tumor models: Highlights in glioblastoma.\u0026quot; \u003cem\u003eAdvanced drug delivery reviews\u003c/em\u003e 180 (2022): 114067.\u003c/li\u003e\n\u003cli\u003eJoseph, Justin V., Mathilde S. Blaavand, Thomas Daubon, Frank AE Kruyt, and Martin K. Thomsen. \u0026quot;Three-dimensional culture models to study glioblastoma\u0026mdash;Current trends and future perspectives.\u0026quot; \u003cem\u003eCurrent Opinion in Pharmacology\u003c/em\u003e 61 (2021): 91-97.\u003c/li\u003e\n\u003cli\u003eSung, Kyung Eun, and David J. Beebe. \u0026quot;Microfluidic 3D models of cancer.\u0026quot; \u003cem\u003eAdvanced drug delivery reviews\u003c/em\u003e 79 (2014): 68-78.\u003c/li\u003e\n\u003cli\u003eChauhdari, T., Su, J., Lv, J. and Ding Y. ECM-free lung tumoroids generated by an air-stretchable microfluidic chip. In vitro models. 2025 Sep 4:1-3.\u003c/li\u003e\n\u003cli\u003eMansoury, Morva, Maya Hamed, Rashid Karmustaji, Fatima Al Hannan, and Stephen T. Safrany. \u0026quot;The edge effect: A global problem. The trouble with culturing cells in 96-well plates.\u0026quot; \u003cem\u003eBiochemistry and biophysics reports\u003c/em\u003e 26 (2021): 100987.\u003c/li\u003e\n\u003cli\u003eMirab, Fereshtehsadat, You Jung Kang, and Sheereen Majd. \u0026quot;Preparation and characterization of size-controlled glioma spheroids using agarose hydrogel microwells.\u0026quot; \u003cem\u003ePloS one\u003c/em\u003e 14, no. 1 (2019): e0211078.\u003c/li\u003e\n\u003cli\u003eGagliano, Nicoletta, Francesco Costa, Chiara Cossetti, Letizia Pettinari, Rosaria Bassi, Maurizio Chiriva-Internati, Everardo Cobos, Magda Gioia, and Stefano Pluchino. \u0026quot;Glioma-astrocyte interaction modifies the astrocyte phenotype in a co-culture experimental model.\u0026quot; \u003cem\u003eOncology reports\u003c/em\u003e 22, no. 6 (2009): 1349-1356.\u003c/li\u003e\n\u003cli\u003e\u0026ndash; Inoue A, Ohnishi T, Nishikawa M, et al. A Narrative Review on CD44\u0026apos;s Role in Glioblastoma Invasion, Proliferation, and Tumor Recurrence. \u003cem\u003eCancers (Basel)\u003c/em\u003e. 2023;15(19):4898. Published 2023 Oct 9. doi:10.3390/cancers15194898\u003c/li\u003e\n\u003cli\u003e\u0026ndash; Nishikawa, Masahiro, Akihiro Inoue, Takanori Ohnishi, Shohei Kohno, Shiro Ohue, Shirabe Matsumoto, Satoshi Suehiro et al. \u0026quot;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.\u0026quot; \u003cem\u003eStem cells international\u003c/em\u003e 2018, no. 1 (2018): 5387041.\u003c/li\u003e\n\u003cli\u003e\u0026ndash; Du Z, Wang Y, Liang J, et al. Association of glioma CD44 expression with glial dynamics in the tumour microenvironment and patient prognosis. \u003cem\u003eComput Struct Biotechnol J\u003c/em\u003e. 2022;20:5203-5217. Published 2022 Sep 9. doi:10.1016/j.csbj.2022.09.003\u003c/li\u003e\n\u003cli\u003e\u0026ndash; Pekny, Milos, Ulrika Wilhelmsson, and Marcela Pekna. \u0026quot;The dual role of astrocyte activation and reactive gliosis.\u0026quot; \u003cem\u003eNeuroscience letters\u003c/em\u003e 565 (2014): 30-38.\u003c/li\u003e\n\u003cli\u003e\u0026ndash; Wu J, Li R, Wang J, et al. Reactive Astrocytes in Glioma: Emerging Opportunities and Challenges. \u003cem\u003eInt J Mol Sci\u003c/em\u003e. 2025;26(7):2907. Published 2025 Mar 23. doi:10.3390/ijms26072907\u003c/li\u003e\n\u003cli\u003eF\u0026ouml;rster, Nanna, Lotta Isosaari, Oskari Kulta, Oona Junnila, Valtteri Vuolanto, Marjukka Pollari, Kirsi J. Rautajoki, and Susanna Narkilahti. \u0026quot;Functional 3D Human Neuron\u0026ndash;Glioblastoma Model Reveals Cellular Interactions Enabling Drug Safety Assessments.\u0026quot; \u003cem\u003eThe FASEB Journal\u003c/em\u003e 39, no. 8 (2025): e70567.\u003c/li\u003e\n\u003cli\u003eMatthiesen, I., Jury, M., Rasti Boroojeni, F., Ludwig, S. L., Holzreuter, M., Buchmann, S., \u0026hellip; Herland, A. (2023). Astrocyte 3D culture and bioprinting using peptide functionalized hyaluronan hydrogels. \u003cem\u003eScience and Technology of Advanced Materials\u003c/em\u003e, \u003cem\u003e24\u003c/em\u003e(1). https://doi.org/10.1080/14686996.2023.2165871\u003c/li\u003e\n\u003cli\u003eGoshi, N., Morgan, R.K., Lein, P.J. \u003cem\u003eet al.\u003c/em\u003e A primary neural cell culture model to study neuron, astrocyte, and microglia interactions in neuroinflammation. \u003cem\u003eJ Neuroinflammation\u003c/em\u003e\u003cstrong\u003e17\u003c/strong\u003e, 155 (2020). https://doi.org/10.1186/s12974-020-01819-z\u003c/li\u003e\n\u003cli\u003eSong, Hanbing, Simon Bucher, Katherine Rosenberg, Margaret Tsui, Deviana Burhan, Daniel Hoffman, Soo-Jin Cho et al. \u0026quot;Single-cell analysis of hepatoblastoma identifies tumor signatures that predict chemotherapy susceptibility using patient-specific tumor spheroids.\u0026quot; \u003cem\u003eNature communications\u003c/em\u003e 13, no. 1 (2022): 4878.\u003c/li\u003e\n\u003cli\u003eManini, Ivana, Federica Caponnetto, Anna Bartolini, Tamara Ius, Laura Mariuzzi, Carla Di Loreto, Antonio Paolo Beltrami, and Daniela Cesselli. 2018. \u0026quot;Role of Microenvironment in Glioma Invasion: What We Learned from In Vitro Models\u0026quot; \u003cem\u003eInternational Journal of Molecular Sciences\u003c/em\u003e 19, no. 1: 147. https://doi.org/10.3390/ijms19010147\u003c/li\u003e\n\u003cli\u003eZhang, Zengli, Huiqian Wang, Qifeng Ding, Yufei Xing, Zhonghua Xu, Chun Lu, Dongdong Luo et al. \u0026quot;Establishment of patient-derived tumor spheroids for non-small cell lung cancer.\u0026quot; \u003cem\u003ePloS one\u003c/em\u003e 13, no. 3 (2018): e0194016.\u003c/li\u003e\n\u003cli\u003eThomas, Grace and Ruman Rahman. Evolution of Preclinical Models for Glioblastoma modelling and drug Screening. \u003cem\u003eCurrent Oncology Reports\u003c/em\u003e (2025): pp. 1-24.\u003c/li\u003e\n\u003cli\u003eKhot, M. Ibrahim, Mark A. Levenstein, Greg N. de Boer, Gemma Armstrong, Thomas Maisey, Hafdis S. Svavarsdottir, Helen Andrew, Sarah L. Perry, Nikil Kapur, and David G. Jayne. \u0026quot;Characterising a PDMS based 3D cell culturing microfluidic platform for screening chemotherapeutic drug cytotoxic activity.\u0026quot; \u003cem\u003eScientific Reports\u003c/em\u003e 10, no. 1 (2020): 15915.\u003c/li\u003e\n\u003cli\u003eToh, Yi-Chin, Teck Chuan Lim, Dean Tai, Guangfa Xiao, Danny van Noort, and Hanry Yu. \u0026quot;A microfluidic 3D hepatocyte chip for drug toxicity testing.\u0026quot; \u003cem\u003eLab on a Chip\u003c/em\u003e 9, no. 14 (2009): 2026-2035.\u003c/li\u003e\n\u003cli\u003eKwapiszewska, K., A. Michalczuk, M. Rybka, Radosław Kwapiszewski, and Zbigniew Brz\u0026oacute;zka. \u0026quot;A microfluidic-based platform for tumour spheroid culture, monitoring and drug screening.\u0026quot; \u003cem\u003eLab on a Chip\u003c/em\u003e 14, no. 12 (2014): 2096-2104.\u003c/li\u003e\n\u003cli\u003eRyu, Gihan, Jingsong Huang, Oliver Hofmann, Claire A. Walshe, Jasmine YY Sze, Gareth D. McClean, Alan Mosley et al. \u0026quot;Highly sensitive fluorescence detection system for microfluidic lab-on-a-chip.\u0026quot; \u003cem\u003eLab on a Chip\u003c/em\u003e 11, no. 9 (2011): 1664-1670.\u003c/li\u003e\n\u003cli\u003eZhu, Yangzhi, Kalpana Mandal, Ana Lopez Hernandez, Satoru Kawakita, Wei Huang, Praveen Bandaru, Samad Ahadian et al. \u0026quot;State of the art in integrated biosensors for organ-on-a-chip applications.\u0026quot; \u003cem\u003eCurrent Opinion in Biomedical Engineering\u003c/em\u003e19 (2021): 100309.\u003c/li\u003e\n\u003cli\u003eLiu, Z., Zhou, Y., Lu, J. \u003cem\u003eet al.\u003c/em\u003e Microfluidic biosensors for biomarker detection in body fluids: a key approach for early cancer diagnosis. \u003cem\u003eBiomark Res\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 153 (2024). https://doi.org/10.1186/s40364-024-00697-4\u003c/li\u003e\n\u003cli\u003eBargahi, N., Ghasemali, S., Jahandar-Lashaki, S. \u003cem\u003eet al.\u003c/em\u003e Recent advances for cancer detection and treatment by microfluidic technology, review and update. \u003cem\u003eBiol Proced Online\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 5 (2022). https://doi.org/10.1186/s12575-022-00166-y\u003c/li\u003e\n\u003cli\u003eAkg\u0026ouml;n\u0026uuml;ll\u0026uuml; S, Bakhshpour M, Pişkin AK, Denizli A. Microfluidic Systems for Cancer Diagnosis and Applications. Micromachines (Basel). 2021 Oct 31;12(11):1349. doi: 10.3390/mi12111349. PMID: 34832761; PMCID: PMC8619454.\u003c/li\u003e\n\u003cli\u003eThenuwara, G.; Javed, B.; Singh, B.; Tian, F. Biosensor-Enhanced Organ-on-a-Chip Models for Investigating Glioblastoma Tumor Microenvironment Dynamics. \u003cem\u003eSensors\u003c/em\u003e \u003cstrong\u003e2024\u003c/strong\u003e, \u003cem\u003e24\u003c/em\u003e, 2865. https://doi.org/10.3390/s24092865\u003c/li\u003e\n\u003c/ol\u003e"}],"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":"in-vitro-models","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [In vitro models](https://link.springer.com/journal/44164)","snPcode":"44164","submissionUrl":"https://submission.springernature.com/new-submission/44164/3","title":"In vitro models","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"glioma, microfluidics, lab-on-a-chip, co-culture, cancer models, bioengineering","lastPublishedDoi":"10.21203/rs.3.rs-7785293/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7785293/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eTumour recurrence in high-grade gliomas, such as glioblastoma, is extremely common due to the challenges associated with complete surgical removal. Approximately 95% of glioblastoma tumours recur within 2 cm of the tumour resection margin; however, unfortunately, the current clinical measurement techniques do not accurately capture cell state during critical stages recurrence. The development of an on-chip model for investigating glioblastoma recurrence outgrowth is necessary to further understanding of the mechanistic understanding of this process at a cellular level and holds potential for translation into patient-specific predictive models. This work presents the development of an on-chip platform which can be used to assess glioblastoma interactions with healthy neural cells. Using a custom single-channel multi-well microfluidic system, 9L/3cmv-GFP (GFP transfected) rat glioma cells were co-cultured with a CTX-TNA2 rat astrocyte cell line to investigate differences in glioblastoma outgrowth between co-culture and monoculture (glioma only) configurations. Quantitative image analysis demonstrated significantly increased outgrowth of glioma cells from a tumour spheroid in co-culture compared to glioma only controls. Additionally, comparison of upregulation of glioma linked CD44 and glial fibrillary acidic protein (GFAP) expression showed significant differences in cellular expression between the two configurations. These findings align with existing literature suggesting that astrocytes facilitate a supportive environment for glioblastoma cells to proliferate and invade healthy tissue by changing to a reactive phenotype. Overall, this work presents a promising on-chip platform that can aid in the quantification of glioblastoma invasion through directional control of outgrowth and offers the potential for further modalities to be added by combining real-time data acquisition elements to the platform.\u003c/p\u003e","manuscriptTitle":"Development and characterization of an on-chip glioma outgrowth model","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-17 12:26:17","doi":"10.21203/rs.3.rs-7785293/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-10-20T09:56:12+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-19T14:56:01+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-15T15:01:38+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"85119911232887109927550919899648117026","date":"2025-10-07T13:35:10+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"9930033728760309190836383439975416083","date":"2025-10-07T02:33:35+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-10-06T10:35:03+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-06T03:07:42+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-10-06T03:07:14+00:00","index":"","fulltext":""},{"type":"submitted","content":"In vitro models","date":"2025-10-05T14:34:14+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"in-vitro-models","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [In vitro models](https://link.springer.com/journal/44164)","snPcode":"44164","submissionUrl":"https://submission.springernature.com/new-submission/44164/3","title":"In vitro models","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"5ad8a7ba-327c-45ff-9075-426ac7ac0d7c","owner":[],"postedDate":"October 17th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2025-11-04T08:39:01+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-17 12:26:17","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7785293","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7785293","identity":"rs-7785293","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}