Retinal Pigment Epithelium Transplantation: Past, Present, and Future

Abstract Retinal pigment epithelium (RPE) is a monolayer of cells situated between photoreceptors and the underlying choroid. It is essential for normal retinal function. Damaged RPE is associated with diseases such as age-related macular degeneration, Stargardt's macular dystrophy, and retinitis pigmentosa. RPE cells can easily be visualized in vivo, sustainable in vitro, and differentiated from stem cells with a relatively straightforward protocol. Due to these properties and the clinical significance of this epithelium in various retinal diseases, RPE transplantation as a treatment modality has gained considerable interest in the last decade. This paper presents the main techniques for RPE transplantation and discusses recent clinically relevant publications.


INTRODUCTION
Retinal pigment epithelium (RPE) is a cell layer sandwiched between photoreceptors apically and Bruch's membrane basally. RPE cells are cuboidal with apical microvilli gaping photoreceptor outer segments. [1] Their cytoplasms contain melanin pigments, lipofuscin granules, melanolipofuscins, and phagosomes. [1] These cells adhere to each other via zonula occludentes, creating a mosaic-patterned layer of tightly adjoined hexagonally shaped cells. RPE performs several functions essential for vision, such as adsorption of excessive light, transport of nutrients to and from the neuroretina, protection against photooxidation, regeneration of 11 cis-retinal for the visual cycle, and phagocytosis of shed photoreceptor outer segments. [1,2] It also constitutes the outer part of the blood-retinal barrier. [1] RPE dysfunction is seen in several retinal disorders, such as age-related macular degeneration (AMD), [3] proliferative vitreoretinopathy (PVR), and retinitis pigmentosa (RP). Regarding AMD, it is still unclear whether drusen formation is a consequence or a cause of RPE dysfunction. [4] However, both are strongly associated with each other. [5] In PVR, RPE cells contribute to the folding of the retina by detaching and translocating from the underlying basement membrane. [6][7][8] Similarly, RPE disintegration and migration are responsible for the bone-spicule pigmentation that is pathognomonic for RP. [9] Thus, when RPE fails to function correctly, the retina suffers.
Replacing RPE cells is an idea that arose shortly after the characterization of these cells about 40 years ago. [10] Their anatomical accessibility, in vitro resilience, and clinical importance made them attractive for transplantation scientists. The first proof-of-concept study of RPE transplantation was performed in 1984 by Gouras et al in monkeys. [11] Four years later, Li and Turner published a report demonstrating that subretinal injection of RPE cells prevented photoreceptor degeneration in rats. [12] The field has subsequently matured at a rapid pace with not only animal studies, [13][14][15] but also human trials, using RPE of fetal, [16,17] post-mortem adult, [18][19][20] autologous, [21][22][23] induced pluripotent stem cell (iPSC)-derived [24] and embryonic stem cell (ESC)-derived origin. [25][26][27] The introduction of IPSC-and ESC-derived RPE have been critical for the advancement of clinical RPE research during the recent years, and has, to a large extent, solved the issue of sourcing RPE donor tissue. While iPSCs offer an unlimited supply of autologous cells and (to some degree) do not require the use of immunosuppressants, they may carry patients' own genetic vulnerabilities contributing to disease processes. [28] This can be avoided by using ESCs. However, ESCs raise ethical concerns [29][30][31] and are, in contrast to iPSCs, neither autologous nor unlimited in supply. The methods for generating RPE cells from iPSCs [32,33] and ESCs [34,35] have been described in detail. As the succeeding paragraphs will demonstrate, ongoing clinical trials employ these two stem cell sources to generate mature, transplantation-ready RPE.

Transplantation Techniques
There are currently three techniques for subretinal RPE transplantation: (1) surgical placement of RPE as an intact cell sheet (with or without scaffold), (2) injection of RPE as a cell suspension, and (3) macular translocation.
Transplantation of RPE as an intact cell sheet was first described in 1991 by Peyman et al, who treated one patient with an autologous pedicle flap and another with an allogeneic graft consisting of RPE with an underlying choroid. [19] Similar procedures have later been performed by others. [24,26,36,37] Delivering RPE as a patch increases the likelihood of correct anatomical placement and the structural integrity of the graft. Major complications associated with this technique are subretinal hemorrhage and proliferative vitreoretinopathy. Delivery by injection is technically easier and probably less traumatizing to the adjacent tissue. However, cell clumping, poor attachment, and disorganization of RPE upon injection [38] are important drawbacks. The third approach, macular translocation, involves rotating the retina away from a subretinal pathology to an area of healthy RPE. This technique is surgically less straightforward and complicated by cataract, retinal detachment, and diplopia. [39]

Current State of the RPE Transplantation
To give the reader an idea of the current state of RPE transplantation, selected recent studies are briefly discussed below.
In a phase 1/2 clinical trial, Schwartz and associates injected human ESC-derived RPE subretinally via a small 38-gauge retinotomy in 18 patients; nine with AMD and nine with Stargardt's Macular Dystrophy (SMD). [27] Visual acuity improved in the majority of patients. No ocular or systemic safety issues were registered, apart from surgery-associated complications, such as vitreous inflammation, cataract, and endophthalmitis. This study suggested that ESCderived RPE is a potential and safe source of cells for the treatment of retinal disorders.
While Schwartz et al delivered the cells via injection, Kashani and colleagues described a patch of ESC-derived RPE monolayer attached to a synthetic parylene substrate. [40] The initiative is called California Project to Cure Blindness. They implanted this engineered patch in 16 patients with advanced non-neovascular AMD with a median age of 78 years in a singlearm, open-label, prospective, non-randomized, phase 1/2 study. [41] Critical inclusion criteria were advanced non-neovascular AMD, pseudophakia, and severe vision loss. Mild to moderate subretinal hemorrhages and macular holes were reported as adverse events. One patient developed ischemic colitis, possibly related to immunosuppression. Employing a similar approach, an initiative called "The London Project to Cure Blindness" published data on a phase 1 trial encompassing two patients with severe wet AMD. [26] Their patch was also made of differentiated ESC-derived RPE, but the scaffold was made of polyester membrane coated with human vitronectin. Using purposebuilt microsurgical tools, the RPE patches were delivered subretinally to one eye in each of the two patients with severe exudative AMD. They reported successful delivery and survival of the RPE patch and a visual acuity gain of 29 and 21 letters in the two patients, respectively, over 12 months. Importantly, preclinical safety studies did not reveal tumorigenicity or notable proliferative capacity of the ESC-derived RPE cells. Undifferentiated ESCs were not detected in the final differentiated RPE product. Furthermore, investigation of systemic biodistribution 26 weeks after implantation of the RPE grafts in pigs did not provide any evidence of migration of cells distant to the administration site. Although the number of patients included in the study is too low to conclude on the clinical efficiency of the RPE patch, the report provides valuable information about the surgical technique, stability of the transplant, and the safety of the ESCderived cells.

CONCLUSION
As the aforementioned reports indicate, human RPE transplantation is still in its early days. Significant challenges related to graft composition, graft vehicle, and surgical technique need to be addressed. However, it is encouraging that, despite these challenges, the mentioned studies report positive outcomes following transplantation. Future clinical trials are therefore eagerly awaited. As of this writing, 12 RPE transplantation projects are listed on the clinical trials database ClinicalTrials.gov [ Table 1], all being phase 1/2 trials assessing safety, side effects, and dosing. The most frequently occurring indication remains AMD. However, SMD, RP, and Leber congenital amaurosis are other diseases listed as indications for some ongoing trials. The majority of the projects seem to employ RPE cells derived from ESC. Based on this and the aforementioned recent studies, it is the authors' impression that the current front-line of RPE transplantation is based on cell delivery as a patch using ESC-derived RPE.

Financial Support and Sponsorship
None.

Conflicts of Interest
There are no conflicts of interest.
TPU and JRE hold a patent on the culture of retinal pigment epithelial cells: https://patents. google.com/patent/US20180119097A1/en.