Background: Glioblastoma (GBM) remains one of the most challenging cancers to model in vitro because of its marked heterogeneity and the critical role of the tumor microenvironment (TME) in disease progression and therapy resistance. Advanced bioengineered platforms integrating biomimetic matrices, microfluidics, and computational modeling offer new opportunities to reproduce key physiological features of GBM for mechanistic studies and drug testing.
Methods: We developed a biomimetic GBM-on-Chip platform combining patient-derived glioblastoma stem cells (GSCs) with a GelMA/HAMA hydrogel engineered to mimic the brain extracellular matrix. Finite element modeling guided the design of the microfluidic platform by predicting oxygen transport and defining culture conditions capable of generating physiologically relevant hypoxic niches under dynamic perfusion. Computational predictions of oxygen gradients were subsequently validated experimentally. Proneural (L0512) and mesenchymal (C1312) GSCs were characterized in 2D and cultured within the device to investigate the impact of the engineered TME on tumor organization, phenotype, and therapeutic response.
Results: The optimized hydrogel supported subtype-specific spheroid formation and promoted transcriptional changes associated with lipid metabolism, extracellular matrix remodeling, and cellular plasticity, consistent with the establishment of GBM-like tumor niches. The predicted hypoxic regions were experimentally confirmed by increased HIF-1α expression, demonstrating the ability of the platform to reproducibly generate physiologically relevant oxygen gradients. Drug screening with Temozolomide, alone or combined with Pitavastatin, revealed subtype-dependent responses, with mesenchymal GSCs exhibiting enhanced resistance under hypoxic conditions, highlighting the central role of hypoxia in therapy resistance.
Conclusions: By integrating biomaterials engineering, microfluidics, and finite element modeling, this GBM-on-Chip enables the rational design and validation of hypoxic tumor microenvironments while recapitulating key features of glioblastoma heterogeneity. The platform provides a robust and predictive tool for investigating hypoxia-driven tumor biology and accelerating the development of patient-specific therapeutic strategies.