A Spatially Controlled Glioblastoma-on-a-Chip for Dissecting Tumor-Immune-Vascular Crosstalk and Bevacizumab Resistance.

Wang, Zixuan; Gao, Jie; Chan, Junned; Fang, Yongcong; Zhang, Yanmei; Liu, Xin; Xia, Jingjing; Mao, Heyun et al. · Adv Healthc Mater · 2026

basic_science · Level V

Where this comes from

Abstract

Glioblastoma (GBM) is an aggressive brain malignancy with a median survival of only 15 months despite current therapeutic interventions. A major obstacle to effective treatment is the intricate tumor microenvironment (TME), which drives tumor progression and therapy resistance. However, existing in vitro models lack micrometer-scale spatial control over multiple cellular components and fail to preserve their dynamic interactions, limiting our understanding of tumor biology and treatment responses. Here, we present a spatially controlled GBM-on-a-chip (scGoC) platform that integrates a novel organoid positioning strategy, enabling the reconstruction of the TME within a 3D matrix at micron-level precision. This platform allows real-time monitoring of dynamic tumor-immune-vascular interactions, providing refined insights into tumor progression and drug responses. Using the scGoC, we demonstrate that endothelial cells not only promote microglial migration toward tumor organoids but also drive their polarization toward an immunosuppressive phenotype, faithfully recapitulating hallmark features of GBM progression. Importantly, the platform reveals a pro-tumorigenic adaptive response to bevacizumab monotherapy, driven by the upregulation of HIF-1 signaling and metabolic reprogramming pathways, providing mechanistic insights into GBM treatment. Collectively, the scGoC establishes a versatile and standardized tool for decoding complex cellular communications and investigating mechanisms of treatment response with broad applicability across cancer research.