Machine learning-based estimation of spatial gene expression pattern during ESC-derived retinal organoid development

Pluripotent stem cell-derived models of retinal disease: Elucidating pathogenesis, evaluating novel treatments, and estimating toxicity

Blindness poses a growing global challenge, with approximately 26% of cases attributed to degenerative retinal diseases. While gene therapy, optogenetic tools, photosensitive switches, and retinal prostheses offer hope for vision restoration, these high-cost therapies will benefit few patients. Understanding retinal diseases is therefore key to advance effective treatments, requiring in vitro models replicating pathology and allowing quantitative assessments for drug discovery. Pluripotent stem cells (PSCs) provide a unique solution given their limitless supply and ability to differentiate into light-responsive retinal tissues encompassing all cell types. This review focuses on the history and current state of photoreceptor and retinal pigment epithelium (RPE) cell generation from PSCs. We explore the applications of this technology in disease modelling, experimental therapy testing, biomarker identification, and toxicity studies. We consider challenges in scalability, standardisation, and reproducibility, and stress the importance of incorporating vasculature and immune cells into retinal organoids. We advocate for high-throughput automation in data acquisition and analyses and underscore the value of advanced micro-physiological systems that fully capture the interactions between the neural retina, RPE, and choriocapillaris.