Automated High-Content, High-Throughput Spatial Analysis Pipeline for Drug Screening in 3D Tumor Spheroid Inverted Colloidal Crystal Arrays

• Published in: ACS applied materials & interfaces Vol. 17; no. 35; pp. 49210 - 49226
• Main Authors: Jeon, Hyunsu, Kim, Gaeun, Carpenter, James, Colón, Yamil J., Wang, Yichun
• Format: Journal Article
• Language: English
• Published: United States American Chemical Society 03.09.2025
• Subjects: Antineoplastic Agents - chemistry; Antineoplastic Agents - pharmacology; Biological and Medical Applications of Materials and Interfaces; Cell Line, Tumor; Colloids - chemistry; Doxorubicin - chemistry; Doxorubicin - pharmacology; Drug Screening Assays, Antitumor - methods; High-Throughput Screening Assays - methods; Humans; Hydrogels - chemistry; Spheroids, Cellular - drug effects; spatial signal profiling; high yield; automated processing; drug discovery; cancer model; hydrogel
• ISSN: 1944-8244; 1944-8252; 1944-8252
• DOI: 10.1021/acsami.5c10049

High-content, high-throughput (HCHT) screening platforms are essential for drug discovery, yet conventional 2D assays lack physiological relevance, and current 3D spheroid systems often face challenges to scalability, uniformity, and the analytical efficiency required for statistically robust screening. Here, we present a fully integrated 3D HCHT platform that synergizes tumor spheroid arrays generated from a bioinert inverted colloidal crystal (iCC) hydrogel framework with an automated, high-speed image analysis pipeline for rapid and spatially resolved therapeutic profiling. The iCC framework enables spontaneous self-assembly of highly ordered tumor spheroid array at high spheroid density (∼79.8 spheroids·mm–2) with tight size uniformity (<10% standard deviation), supporting reproducible and high-content imaging (∼40 spheroids·image–1). The automated image-processing algorithm achieves robust region-of-interest segmentation, fluorescence-weighted-spheroid centroiding (FwSC), and multiparametric spatial analysis in <5 s per imagemarkedly faster than well-plate analysis (e.g., ∼5 min per 96-well plate) or sequential single-spheroid analysis methods (∼1 s per spheroid). Using this platform, we capture dose-dependent diffusion of doxorubicin, tumor penetration profile of small extracellular vesicles, and cell-type-specific infiltration behaviors of monocytes versus macrophages. Comparative viability profiling across chemotherapeutics further reveals distinct spatial toxicity signatures, highlighting the importance of spatial context in drug response assessment. Collectively, this platform enables rapid, reproducible, and spatially informative screening in 3D, offering a powerful tool for drug discovery, tumor modeling, and immunotherapeutic development.