Symmetric cancer spheroid-fibroblast organization revealed in 3D by high-throughput microscopy

High-throughput microscopy enables large-scale analysis of cancer spheroid interactions. Integrating this technique with point-spread function (PSF) engineering, a customized modification of the optical system, provides 3D imaging capabilities that enable volumetric analysis of the samples from a minimal number of images, thereby reducing phototoxicity and improving temporal resolution. However, existing methods for extracting volumetric information using PSF engineering have been demonstrated primarily on point sources or rely on supervised training data, limiting their general applicability to complex biological samples. Here, we reveal a unique symmetric spheroid-fibroblast interaction pattern using high-throughput imaging. We further characterize these interactions using a PSF-engineered high-throughput microscope, introducing unsupervised methods for 3D localization from single or dual image acquisitions. Our analysis uncovers surprising spatial cellular trajectories of fibroblasts as they approach and interact with spheroids. It further identifies a characteristic 3D distance between adjacent fibroblast clusters and enables live detection of drug-induced perturbations to these interaction patterns.