Cancer metastasis remains the dominant cause of cancer-related death, accounting for roughly 90 percent of mortality among cancer patients, and colorectal cancer is among the most aggressive in this respect because of its high incidence and its strong tendency to spread to the liver, lungs and peritoneum. A central obstacle to studying how tumours begin this journey has always been that tumour cells do not behave in isolation. Their migration and invasion are shaped by the surrounding microenvironment, including the composition and stiffness of the extracellular matrix, the presence of stromal and endothelial cells, soluble signalling gradients and the physical confinement of three-dimensional tissue architecture. Traditional two-dimensional cultures on plastic fail to reproduce these constraints, while animal models are expensive, ethically constrained and often poor predictors of clinical outcomes because of interspecies differences. A team of researchers led by Adrian García and Daniel Nieto at the University of A Coruña has now developed a platform that bridges this gap, combining a microfluidic chip, digital light processing bioprinting and a hybrid extracellular-matrix bioink into a single reconfigurable tumour-on-chip system described in Materials Today Bio.
The new device addresses a long-standing limitation of existing tumour-on-chip models, which typically rely on manually loaded hydrogels and fixed architectures that make it difficult to separate the contributions of matrix properties, cellular organisation and soluble signalling to tumour cell behaviour. The researchers designed a resin-printed microfluidic chip with a central culture chamber measuring 12 by 8 millimetres and one millimetre in height, connected to inlet and outlet microchannels, with an optimised non-rectangular geometry to improve filling and reduce bubble trapping. After screening three commercial photopolymer resins for geometric fidelity and cytotoxicity, the team selected BioMed Clear, which supported 84.74 percent cell viability in direct contact assays compared with only 12.15 percent for the worst-performing resin. Imaging is performed through a bonded glass coverslip at the bottom of the device rather than through the resin itself, which minimises autofluorescence and optical artefacts. In flow tolerance testing, the assembled chips withstood perfusion rates from 0.05 to 80 millilitres per minute without leaks or delamination, demonstrating mechanical robustness far beyond routine culture requirements.
The heart of the platform is in situ digital light processing bioprinting, in which a top-down projector illuminates the hydrogel precursor inside the chip through a transparent PDMS lid, photopolymerising defined patterns layer by spatial definition rather than by layer accumulation. Sequential loading, exposure and washing steps allow different bioinks to be printed one after another into the same chamber, generating multicompartment structures in which each region contains a distinct cell population or matrix condition. Because the projected image determines the position, geometry and light dose of every region, the researchers can change the entire experimental design without redesigning the chip, fabricating new moulds or transferring preformed gels. A proof-of-concept print using three visually distinguishable materials confirmed that the workflow produces cleanly separated central cores, intermediate rings and outer rings within the device, establishing that the chip supports controlled multi-material patterning with spatial precision at the sub-millimetre scale.
Formulating the bioink required balancing biological complexity against printability. Gelatin methacryloyl, or GelMA, was chosen as the structural backbone because it crosslinks under visible light, carries cell-adhesive motifs derived from its gelatin origin and allows mechanical tuning. Because GelMA alone does not capture the biochemical richness of the tumour extracellular matrix, the researchers incorporated Matrigel, a basement membrane extract, and type I collagen, the principal fibrillar protein of stromal tissue. Screening metabolic activity of encapsulated HCT116 colorectal cancer cells revealed a clear composition-dependent response: higher GelMA concentrations reduced activity while increasing Matrigel content enhanced it, and the addition of a low concentration of collagen further improved the cellular response. The final formulation of 5 percent GelMA, 15 percent Matrigel and 0.2 percent type I collagen produced a soft hydrogel with an apparent compressive modulus of approximately 2.11 kilopascals, a value squarely within the range reported for soft tumour and stromal microenvironments in human colon tissue.
Perhaps the most elegant technical feature of the platform is its use of greyscale-controlled exposure to tune matrix mechanics without changing the bioink’s chemical composition. Because the photocrosslinking density of a hydrogel depends on the delivered light dose, the researchers can adjust stiffness simply by dimming the projected image. Calibrated measurements on a 4K digital light processing projector showed a nearly linear relationship between greyscale level and irradiance at the printing plane, and four conditions at 100, 75, 50 and 25 percent projected intensity delivered irradiances from 6.776 down to 1.898 milliwatts per square centimetre over 45-second exposures. This optical dial tuned the apparent compressive modulus of the identical bioink from 2.11 kilopascals at full intensity down to 0.43 kilopascals at quarter intensity, while rhodamine B diffusion measurements confirmed that apparent diffusivity increased correspondingly from 4.95 to 6.21 times ten to the minus six square centimetres per second. In other words, a single bioink can now be printed across a continuum of mechanically distinct microenvironments within the same experiment.
To verify that the transport properties of these soft matrices were compatible with static culture, the team performed finite element simulations in COMSOL Multiphysics using the experimentally measured diffusion coefficients. One model tracked how concentration gradients relax within the printed hydrogel domains, confirming that the Matrigel- and collagen-containing formulations reach equilibrium faster than GelMA alone. A second model coupled glucose diffusion with cellular consumption described by Michaelis-Menten kinetics, simulating nutrient depletion over four days at different cell densities. These calculations established that a working density of one million cells per millilitre was conservative for the short-term experiments, ensuring that the observed cellular behaviour would reflect matrix and co-culture effects rather than metabolic starvation.
The biological experiments exploited both variables of the platform simultaneously. Using fluorescently labelled cells, the researchers printed concentric structures comprising a central HCT116 tumour core 0.5 millimetres in diameter, a surrounding ring of human umbilical vein endothelial cells 100 micrometres thick and an outer acellular ring completing a one-millimetre construct. Tumour-only controls replaced the endothelial ring with cell-free bioink. One construct was printed at each of the four greyscale conditions within every chip, allowing matrix stiffness and cellular organisation to be compared in a single device. Over four days of static culture, fluorescence imaging revealed that HCT116 cells progressively redistributed outward from the printed core, and quantitative image analysis showed that this movement depended strongly on both variables. Two-way statistical analysis found a significant interaction between light dose and culture configuration, with the co-cultured constructs showing markedly greater redistributed areas at 50, 75 and 100 percent intensity, and the strongest response occurring at 75 percent projected intensity.
The magnitude of the endothelial effect was striking: the mean redistribution distance of HCT116 cells increased from 48 micrometres in tumour-only constructs to 101 micrometres when the HUVEC compartment was present, an overall increase driven by paracrine chemotactic signalling that is well documented in co-culture studies of tumour-endothelial interaction. Crucially, live/dead staining on day four showed high viability of both cell types across all greyscale conditions with no significant intensity-dependent differences, confirming that the redistribution patterns arose from genuine matrix- and co-culture-dependent behaviour rather than printing-induced cytotoxicity. The authors note that the aim was not to present a fully established model of invasion but to demonstrate that the platform can monitor early tumour cell redistribution from a defined initial geometry under controlled, experimentally variable conditions.
The significance of this work lies in its integration. Direct in-chip bioprinting, sequential multi-material patterning, light-dose modulation and fluorescence-based monitoring have previously existed as separate capabilities, but combining them within one microfluidic workflow creates a genuinely reconfigurable system in which geometry, stiffness, composition and cellular architecture can each be varied independently. Because greyscale exposure alters crosslinking density while keeping ligand density, viscosity and biochemical composition constant, matrix mechanics can now be interrogated as an isolated variable in tumour cell behaviour, something composition-based approaches cannot achieve cleanly. The platform also opens a path toward future versions incorporating perfusion, endothelial barrier measurements, drug-loaded hydrogel compartments and nanocarrier delivery, consistent with emerging hydrogel-based approaches for colorectal cancer therapeutics.
For the broader field of cancer research, the study offers a versatile, controlled tool for dissecting how microenvironmental architecture shapes the earliest steps of metastatic behaviour, the stage at which therapeutic intervention remains most effective. As the authors conclude, the goal is to enable mechanistic studies of migration, invasion, endothelial interaction and therapeutic response under better-defined tumour-microenvironmental conditions, moving the field beyond the limitations of flat plastic and toward models that honour the true three-dimensional complexity in which cancer cells live, move and invade.
Subject of Research: Development of a 3D-bioprinted tumour-on-chip platform with tunable hydrogel properties for monitoring colorectal cancer cell migration and invasion
Subject of Research: Technology and Engineering
Article Title: Development of a 3D-bioprinted tumour-on-chip with tunable hydrogel properties for monitoring colorectal cancer cell migration and invasion
Article References: García, A., Jove, L., Pereira, M., Figueroa, A., & Nieto, D. (2026). Development of a 3D-bioprinted tumour-on-chip with tunable hydrogel properties for monitoring colorectal cancer cell migration and invasion. Materials Today Bio, 40, Article 103646. https://doi.org/10.1016/j.mtbio.2026.103646
Image Credits: AI Generated
DOI: 10.1016/j.mtbio.2026.103646
Keywords: tumour-on-chip, 3D bioprinting, digital light processing, colorectal cancer, GelMA hydrogel, cell migration, greyscale crosslinking, extracellular matrix, HUVEC co-culture, microfluidics, HCT116, matrix stiffness
Cite Scienmag News
APA
MLA
Chicago
Nathaniel Bowman. (September 7, 2026). 3D-bioprinted colorectal tumor-on-chip with tunable hydrogels tracks cancer invasion. Scienmag. https://scienmag.com/3d-bioprinted-colorectal-tumor-on-chip-with-tunable-hydrogels-tracks-cancer-invasion/
Nathaniel Bowman. “3D-bioprinted colorectal tumor-on-chip with tunable hydrogels tracks cancer invasion.” Scienmag, 7 September 2026, https://scienmag.com/3d-bioprinted-colorectal-tumor-on-chip-with-tunable-hydrogels-tracks-cancer-invasion/. Accessed 7 September 2026.
Nathaniel Bowman. “3D-bioprinted colorectal tumor-on-chip with tunable hydrogels tracks cancer invasion.” Scienmag. September 7, 2026. https://scienmag.com/3d-bioprinted-colorectal-tumor-on-chip-with-tunable-hydrogels-tracks-cancer-invasion/
Copy citation
Download RIS
Tags: 3D bioprinted tumor-on-chipadvanced biofabrication for cancer researchbioprinting of cancer tissue constructscancer metastasis to liver and lungscolorectal cancer metastasis modelcolorectal cancer metastasis modelingcolorectal tumor microenvironment modelingdigital light processing bioprintingextracellular matrix stiffness in cancer progressionextracellular matrix stiffness in cancer researchhybrid bioink for cancer researchhybrid bioink for tissue engineeringlimitations of traditional 2D culturesmicrofluidic tumor microenvironmentorgan-on-chip for metastasis studyreconfigurable tumor-on-chip platformtumor cell invasion in 3D tissue modelstumor microenvironment simulationtumor-cell migration and invasiontunable hydrogels for cancer invasion
