Page 16 - manuscript_ijb05590

achieves tumor models replicating the in vivo environment, enabling real-time high-
96
97
resolution imaging for drug testing . For example, Xiong et al. developed a
customized bladder-shaped microfluidic device with biomimetic guides and triangular
markers. This platform enabled long-term culture (>4 weeks) of patient-derived

spheroids while preserving histopathological and genetic signatures. By optimizing

drug diffusion kinetics (uniform distribution within 6 seconds) and hydrodynamic

conditions, the system achieved high-fidelity drug susceptibility testing across eight

patients. Each microfluidic device was seeded with 5–15 tumor spheroids and served

as a preclinical model to predict drug response by assessing the effects of different

chemotherapeutic agents and their concentrations on the spheroids. Results correlated

closely with PDX models and clinical responses, demonstrating their capacity to

recapitulate the in vivo TME. Crucially, the device overcame limitations of traditional

methods (e.g., spheroid loss in ultra-low attachment (ULA) plates), providing a

clinically translatable tool for rapid personalized chemotherapy screening (Figure 4C,
98
D). Skubal et al. utilized a microfluidic platform to capture the vascularization

process during the development of renal cell carcinoma spheroids in real time, and
successfully obtained high-resolution images of the tumor-on-a-chip model before and

after treatment with bevacizumab to evaluate the efficacy of the vascular-targeting

81
therapy (Figure 4E). Similarly, Oh et al. constructed a TME with spatiotemporal
control by integrating components (ECM, vasculature, stromal cells, interstitial flow)

on-chip, monitoring nanoparticle accumulation/uptake in target cells to address

84
nanomedicine translation challenges. Du et al. developed a microfluidic platform
simulating/controlling multiple TME factors, evaluating 3D tumor invasion into stroma

and investigating paclitaxel (PTX) effects on cancer cell migration, survival, and

85
morphology. Pavesi et al. designed a platform reconstructing HUVEC microvascular
networks to observe cancer cell extravasation and capture T cell movement

spatiotemporal data/cytotoxic efficiency, validating T cell therapy.

15