Age-related macular degeneration (AMD) is the primary cause of blindness in adults over 60 years of age, and clinical trials are currently assessing the therapeutic potential of retinal pigmented epithelial (RPE) cell monolayers on implantable scaffolds to treat this disease. However, challenges related to the culture, long-term storage, and long-distance transport of such implants currently limit the widespread use of adherent RPE cells as therapeutics. Here we report a xeno-free protocol to cryopreserve a confluent monolayer of clinical-grade, human embryonic stem cell-derived RPE cells on a parylene scaffold (REPS) that yields viable, polarized, and functional RPE cells post-thaw. Thawed cells exhibit ≥ 95% viability, have morphology, pigmentation, and gene expression characteristic of mature RPE cells, and secrete the neuroprotective protein, pigment epithelium-derived factor (PEDF). Stability under liquid nitrogen (LN2) storage has been confirmed through one year. REPS were administered immediately post-thaw into the subretinal space of a mammalian model, the Royal College of Surgeons (RCS)/nude rat. Implanted REPS were assessed at 30, 60, and 90 days post-implantation, and thawed cells demonstrate survival as an intact monolayer on the parylene scaffold. Furthermore, immunoreactivity for the maturation marker, RPE65, significantly increased over the post-implantation period in vivo, and cells demonstrated functional attributes similar to non-cryopreserved controls. The capacity to cryopreserve adherent cellular therapeutics permits extended storage and stable transport to surgical sites, enabling broad distribution for the treatment of prevalent diseases such as AMD.

Cell-based therapies face challenges, including poor cell survival, immune rejection, and integration into pathologic tissue. We conducted an open-label phase 1/2a clinical trial to assess the safety and preliminary efficacy of a subretinal implant consisting of a polarized monolayer of allogeneic human embryonic stem cell-derived retinal pigmented epithelium (RPE) cells in subjects with geographic atrophy (GA) secondary to dry age-related macular degeneration. Postmortem histology from one subject with very advanced disease shows the presence of donor RPE cells 2 years after implantation by immunoreactivity for RPE65 and donor-specific human leukocyte antigen (HLA) class I molecules. Markers of RPE cell polarity and phagocytosis suggest donor RPE function. Further histologic examination demonstrated CD34+ structures beneath the implant and CD4+, CD68+, and FoxP3+ cells in the tissue. Despite significant donor-host HLA mismatch, no clinical signs of retinitis, vitreitis, vasculitis, choroiditis, or serologic immune response were detected in the deceased subject or any other subject in the study. Subretinally implanted, HLA-mismatched donor RPE cells survive, express functional markers, and do not elicit clinically detectable intraocular inflammation or serologic immune responses even without long-term immunosuppression.

PURPOSE: To report 1-year follow-up of a phase 1/2a clinical trial testing a composite subretinal implant having polarized human embryonic stem cell (hESC)-derived retinal pigment epithelium (RPE) cells on an ultrathin parylene substrate in subjects with advanced non-neovascular age-related macular degeneration (NNAMD). METHODS: The phase 1/2a clinical trial included 16 subjects in two cohorts. The main endpoint was safety assessed at 365 days using ophthalmic and systemic exams. Pseudophakic subjects with geographic atrophy (GA) and severe vision loss were eligible. Low-dose tacrolimus immunosuppression was utilized for 68 days in the peri-implantation period. The implant was delivered to the worst seeing eye with a custom subretinal insertion device in an outpatient setting. A data safety monitoring committee reviewed all results. RESULTS: The treated eyes of all subjects were legally blind with a baseline best-corrected visual acuity (BCVA) of ≤ 20/200. There were no unexpected serious adverse events. Four subjects in cohort 1 had serious ocular adverse events, including retinal hemorrhage, edema, focal retinal detachment, or RPE detachment, which was mitigated in cohort 2 using improved hemostasis during surgery. Although this study was not powered to assess efficacy, treated eyes from four subjects showed an increased BCVA of >5 letters (6-13 letters). A larger proportion of treated eyes experienced a >5-letter gain when compared with the untreated eye (27% vs. 7%; P = not significant) and a larger proportion of nonimplanted eyes demonstrated a >5-letter loss (47% vs. 33%; P = not significant). CONCLUSIONS: Outpatient delivery of the implant can be performed routinely. At 1 year, the implant is safe and well tolerated in subjects with advanced dry AMD. TRANSLATIONAL RELEVANCE: This work describes the first clinical trial, to our knowledge, of a novel implant for advanced dry AMD.