**Background:** The outer blood-retina barrier (oBRB), composed of the retinal pigment epithelium (RPE), Bruch's membrane (BM), and the choroid, is essential for retinal homeostasis and immune privilege. Disruption of the oBRB occurs in numerous retinal diseases, including age-related macular degeneration (AMD), diabetic macular edema (DME), retinitis pigmentosa (RP), choroideremia, gyrate atrophy, oculocutaneous albinism (OCA), Stargardt disease, Doyne honeycomb retinal dystrophy (DHRD), and Sorsby's fundus dystrophy (SFD). These conditions lead to photoreceptor loss and vision impairment, with most lacking effective treatments once degeneration is established. To study oBRB pathophysiology and develop therapies, in vitro models have evolved over three decades, and cellular therapy using RPE sheets on biomaterial scaffolds has progressed to clinical trials.
**Methods:** This narrative review systematically examines the literature on oBRB engineering, covering in vitro models and cellular therapy implants. It describes the biology of the oBRB and associated diseases, then categorizes models by complexity: RPE-only models (using Transwell inserts, custom synthetic or natural membranes) and co-culture models incorporating endothelial cells (ECs) and support cells (fibroblasts, pericytes, mesenchymal stem cells). Cell types discussed include immortalized lines (ARPE-19, D407), primary human fetal and adult RPE cells, and stem cell-derived RPE (from adult RPE stem cells, human embryonic stem cells [hESC], and induced pluripotent stem cells [hiPSC]). For in vivo implantation, the review details material properties (thickness, topography, wettability, mechanical properties, permeability, degradation) and summarizes preclinical animal studies and clinical trials.
**Key Results:** The review reports that ARPE-19 cells, while widely used, have limitations including low transepithelial electrical resistance (TER) values (typically 30–100 Ω·cm² vs. native ~150 Ω·cm²), lack of pigmentation, and gene expression profiles closer to fibroblasts than native RPE. In contrast, hiPSC-derived RPE cells express typical markers (RPE65, BEST1, CRALBP, ZO-1), achieve TER values up to 1000 Ω·cm² after extended culture, and enable patient-specific disease modeling. Co-culture of RPE with ECs improves RPE maturation, increasing tight junction formation, pigmentation, and basement membrane protein deposition (laminin, collagen IV). For example, Paek et al. showed increased RPE65 expression and melanosome number in co-culture, while Song et al. demonstrated formation of a 2–4 μm BM-like structure after 6 weeks. Inclusion of support cells (fibroblasts, pericytes) prolonged culture from 2 weeks to 2 months (Manian et al.) and promoted capillary formation with lumens of 20–50 μm (Song et al.). For cellular therapy, five clinical trials have been registered: three have reported phase I/IIa results. The Regenerative Patch Technologies trial (NCT02590692) using a parylene-C membrane with hESC-RPE in 15 patients (average age 78) showed 86.9% defect coverage, with ETDRS letter score improvements of 6–13 letters at 1 year. The Moorfields Eye Hospital trial (NCT01691261) using a PET membrane with hESC-RPE in 2 patients reported ETDRS improvements of 29 and 21 letters, and reading speed increased from 1.7 to 82.8 words/min (patient 1) and 0 to 47.8 words/min (patient 2). The RIKEN Center trial (UMIN000011929) using a cell sheet without substrate in 1 patient (77 years) showed VFQ-25 score improvement from 48.8 to 58.3 at 1 year. Common complications included intra-retinal hemorrhage (systematic in Kashani et al.), subconjunctival hemorrhage (n=12), and sub-retinal hemorrhage (n=9).
**Clinical Implications:** The review underscores that hiPSC-derived RPE cells are the most versatile cell source for both disease modeling and therapy, enabling personalized medicine and reducing ethical concerns. Co-culture models incorporating ECs and support cells are essential for recapitulating native oBRB physiology and should be adopted in future studies. For cellular therapy, the five ongoing clinical trials demonstrate surgical feasibility, graft survival, and visual improvement in patients with advanced retinal degeneration. However, complications such as hemorrhage and the need for immunosuppression remain challenges. Future directions include developing biodegradable scaffolds that mimic BM topography (e.g., electrospun nanofibers with diameters ~200 nm), incorporating choroidal cells into implants, and extending trials to patients with earlier-stage disease to evaluate therapeutic efficacy.