Retinal degeneration is an irreversible phenomenon caused by various disease conditions including age-related macular degeneration (AMD) and retinitis pigmentosa (RP). During the course of these diseases, photoreceptors (PRs) are susceptible to degeneration due to their malfunctions or to a primary dysfunction of the retinal pigment epithelium (RPE). Once lost, these cells could not be endogenously regenerated in humans, and cell therapy to replace the lost cells is one of the promising strategies to recover vision. Depending on the nature of the primary defect and the stage of the disease, RPE cells, PRs, or both might be transplanted to achieve therapeutic effects. We describe in this review the current knowledge and recent progress to develop such approaches. The different cell sources proposed for cell therapy including human pluripotent stem cells are presented with their advantages and limits. Another critical aspect described herein is the pharmaceutical formulation of the end product to be delivered into the eye of patients. Finally, we also outline the future research directions in order to develop a complex multilayered retinal tissue for end-stage patients.
Lining the back of the eye, the retina is a light-sensitive tissue composed of several neuronal layers that convert light stimuli into electrical impulses that are further processed and integrated. The resulting signal is then transmitted to the brain through the optical nerve. Photoreceptors (PRs), which convert these light inputs, are in contact with a specific epithelial layer, the retinal pigment epithelium or RPE, which provides a trophic support and maintains PR homeostasis. Among other functions, the RPE is involved in the elimination of photoreceptor debris, the secretion of growth factors, the transport of nutrients, and the recycling of proteins involved in the visual cycle [
RP is a heterogeneous group of inherited disorders that could affect either the RPE or the PRs or both [
AMD is the other condition in which PRs degenerate. It represents the leading cause of blindness in Western countries. The elderly population is at risk with 12% of people older than 80 years being affected. As the life expectancy increases worldwide, AMD is becoming a global burden [
For the abovementioned diseases, a specific cell therapy strategy could be applied depending on the disease stage. For early cases of AMD, where RPE cells are lost and Bruch’s membrane is damaged but PRs are still preserved, an RPE cell therapy could be applied. When PRs are also lost, a combination of PR and RPE cell therapies corresponds to the more appropriate treatment. For RP associated with RPE defects, RPE cell therapy should be applied when PRs are still preserved. Otherwise, a complete therapy with RPE and PRs has to be applied, as for other forms of RP.
The first demonstration of an RPE cell therapy was performed in monkeys in the early 80s [
Scheme recapitulating the different cell sources that could be used for the cell therapy of the eye.
Human fetal RPE cells could be harvested from fetuses (10- to 21-week gestational age). These cells could be then passaged up to 10 times [
Fetal allogeneic sheets containing both RPE and the neural retina were grafted in patients with RP and AMD [
Procurement of RPE cells or PRs obtained from human fetuses is regulated and needs ethical agreement from the civil society. This agreement is extremely variable from one country to another making its use, as a cell source, difficult. Moreover, the Declaration of Helsinki has set the requirements for use of human materials: informed consent; no incentive for abortion; procurement of a human material that, if not used for research, would be discarded; review of the protocol from ethical committees; a detailed medical history of the donor; and a complete separation of the donor and the recipient [
Human adult RPE cells could be obtained directly from cadavers (allogeneic RPE) or from patients (autologous RPE). The ARPE-19 cell line, commonly used in research laboratories, has been obtained from a 19-year-old man who died following a road accident (in 1986). Following transplantation into the RCS rat, PRs were maintained compared to the nontreated control and the electrical response to light of the retina remained preserved [
Autologous RPE cells are harvested from a nasal location and grafted in another site with depleted RPE. The surgical approach was first experimented in rabbits [
PR sheets obtained from cadavers were transplanted into 8 patients with advanced RP without any signs of visual improvements (a rescue of central vision or a delay in visual loss) [
These different studies demonstrated the feasibility for the use of human adult RPE cells or PRs harvested from cadavers as a source for cell therapy. The major limitation for the use of allogeneic cells from cadavers and autologous RPE is the quantity of cells obtained, thus limiting the large-scale use due to a low amplification potential. As for fetal cell sources, another concern is related to the variations from one batch to another due to the heterogeneity of donors. Finally, autologous cell harvesting can generate complications due to the surgery. A robust industrialized manufacturing process based on another cell source is required to treat the millions of potential patients.
In amphibians, endogenous RPE cells are able to proliferate and reconstitute retinal cells after injuries [
Ideally, the cell source should allow obtaining directly or indirectly PRs or RPE cells. However, many publications described a trophic effect engendered by the grafted cells that were not directly related to RPE cell functions. In addition, some RPE typical functions like phagocytosis, while specifically regulated, are not restricted to this cell type. That is the reason why alternative cell sources, which are not typical retinal cells, were proposed for cell therapy of the eye. It includes iris pigment epithelium (IPE), Schwann cells, fetal brain-derived neural progenitors, bone marrow mesenchymal stem cells (MSCs), retinal neurospheres, and umbilical cord stem cells.
IPE cells possess phagocytic properties and form a monolayer with tight junctions suitable to form a de novo blood-retinal barrier [
As Schwann cells secrete a variety of trophic factors necessary for PR survival, they were grafted in the RCS rat model [
Like Schwann cells, human cortical progenitors are able to secrete trophic factors that could sustain the survival of PRs [
Bone marrow MSCs are precursors of the bone, cartilage, and adipocytes
Some rare and quiescent cells located in humans in the pars plicata and pars plana of the retinal ciliary margin can be stimulated with bFGF to proliferate
Human umbilical tissue-derived cells (hUTCs) are another cell source that has high amplification potentials: at least
Human pluripotent stem cells (hPSCs) comprise human embryonic stem cells (hESCs) and human-induced pluripotent stem cells (hiPSCs). hESCs were first derived in 1998 from the inner cell mass of a human embryo at the blastocyst stage [
hiPSCs are obtained from any differentiated cell of the human adult body after forced expression of a cocktail of pluripotency factors [
All the previously cited cell sources for retinal cell therapy have a lot of limitations in terms of supply chain, donor to donor variability, and scale up (limited cell amplification). Moreover, all strategies based on nonretinal cells may only have a transient effect and will not replace lost PRs or RPE cells. With the emergence of hPSCs, the potential to generate every cell type of the human retina
The differentiation of hPSCs into the various retinal cell types follows the sequential steps of the normal development occurring in the human embryo as identified by time-dependent expression of marker characteristics of these steps [
hPSCs grow as small colonies in a medium containing the fibroblast growth factor (FGF2) that maintains their undifferentiated state. The withdrawal of FGF2 from the culture medium is sufficient to initiate the differentiation. After few weeks of culture under these conditions, pigmented patches start to appear in the culture dish. These patches could be isolated and further amplified to obtain RPE cells [
The efficacy of RPE differentiation could be improved by the addition of growth factors or small molecules in the culture medium [
To further progress in the large-scale production of RPE cells, automated systems of manufacturing could help increase the reproducibility and consistency from batch to batch of the therapeutic cells [
The differentiation of hPSCs into the neural retina follows the same embryonic developmental pathway [
Scheme describing the sequential developmental steps to generate RPE cells and PRs from hPSCs and the different markers that could be used to discriminate between them.
A variety of protocols have been developed to generate PRs, mainly through the formation of organoids that resemble the optic vesicle or optic cup stage
Due to the heterogeneity of the cell types obtained, different strategies were developed to obtain transplantable PRs that are purified and depleted from mitotic cells. The selection and isolation, using a panel of cell surface markers like CD73 specific to PRs in the retina or through a negative selection with markers not expressed in PRs, were successfully achieved [
To be successful, the cell therapy should lead to the replacement of the dead or defective cell type. The way to formulate it, for delivery into the recipient eye, has major consequences on therapeutic outcomes in patients. The scheme below recapitulates the different modes of delivery for a potential cell or tissue replacement therapy (Figure
Scheme recapitulating the strategy to formulate a cell or a tissue therapy to treat patients with various stages of retinal degeneration.
When RPE cells derived from hPSCs are manufactured, they could be banked and cryopreserved in liquid nitrogen at the end of the process. It allows determining the quality of the production in terms of purity, safety, potency, and stability of cells over time of cryoconservation. From such banks, different formulation strategies were proposed for the delivery to patients: either as cell suspension or after a prior step of epithelial reconstitution
However, such cell suspension formulation raises some concerns about efficient integration, functionality, and survival of the grafted material. RPE cell functions rely on its epithelial organization with the secretion of cytokines, the resistance to a stressed environment, or the apicobasal polarization of key proteins [
Several strategies were developed to generate an RPE epithelium ready for transplantation including the use of a supporting scaffold. Such a scaffold would also replace Bruch’s membrane. The selection of a candidate scaffold should meet the following properties: mechanical resistance and flexibility for easy handling, permeability to allow exchange of nutrients and waste materials, thickness that is compatible with a subretinal implantation, and, if biodegradable, nontoxic byproducts [
Nonbiodegradable synthetic polymers like the ultrathin parylene or the porous polyester are currently tested as scaffolds for the RPE tissue preparation in ongoing clinical trials [
Despite being physiologically relevant, the tissue formulation complicates drastically the logistical procedures. First, it requires a prior culture step of RPE cells to reconstruct an epithelium which leads to mobilization of a manufacturing suite for each patient. The scheduling of the surgery should be anticipated to generate the grafting material on time. Such tissue therapy requires the development of specific injection systems in order to correctly graft the polarized epithelial tissue [
The stage of development of grafted PRs is crucial for their correct integration [
Different attempts were described in the literature where PRs were dissociated and injected as a cell suspension [
Functional outcomes of PR transplantation remained limited due to the few numbers of PRs integrated or which were involved in the cytoplasmic exchange [
Few studies were published using PR sheets. Retinal organoids derived from pluripotent stem cells were cut into 0.5 mm sections prior transplantation [
Our understanding of the mechanisms of the hPSC differentiation into RPE cells and PRs has considerably progressed. However, we still need to optimize the way to formulate the tissue to be grafted in order to achieve the best functional recovery. This is particularly true to produce a complex retinal tissue containing layers of PRs and RPE cells suitable for grafting. The proof of concept of visual recovery following transplantation in models of retinal degeneration is globally still missing, as only some forms of rescue were observed using tests that require low vision performances. This will be achieved only by a future optimization of the grafted material in order to be able to graft PRs and PR/RPE sheets that would not fold or form rosettes in the recipient eye. A microstructured 3D scaffold that would guide the correct orientation of PRs is a promising strategy to generate organized layers of PRs [
The field is progressing carefully, and first clinical trial results are starting to be published [
Current or planned clinical trials based on hPSCs to treat RP, Stargardt’s macular dystrophy (SMD), or AMD.
Stage of development | Targeted disease | Sponsor/company | Therapy |
---|---|---|---|
Phases I and II |
AMD, SMD | Astellas Institute for Regenerative Medicine | Allogenic hESC-RPE, cell suspension injection |
Phase I/II |
Dry AMD | CHA Biotech Co. Ltd. | Allogeneic hESC-RPE, cell suspension injection |
Phase I/II |
Dry AMD, GA | Cell Cure Neurosciences Ltd. | Allogeneic hESC-RPE, cell suspension injection |
Phase I/II |
Dry AMD, GA | Regenerative Patch Technologies LLC | Allogeneic hESC-RPE, epithelium on a parylene membrane |
Phase I/II | AMD | RIKEN Centre for Developmental Biology | Autologous and allogenic hiPSC-RPE, epithelium without substrate |
Phase I |
Wet AMD | Pfizer/London Project to Cure Blindness | Allogenic hESC-RPE, epithelium on polyester membrane |
cGMP optimization | AMD, SMD, RP | Cellular Dynamics International/NEI | Autologous hiPSC-RPE, epithelium on biodegradable scaffold |
Phase I/II | RP with genetic defects in LRAT, MERTK, RPE65 | I-Stem/Institut de la Vision | Allogeneic hESC-RPE, epithelium on amniotic membrane |
Preclinical | AMD, best disease, LCA | HMC, Israel | Allogeneic hESC-RPE, cell suspension injection |
Phase I/II |
Dry AMD, SMD | Chinese Academy of Sciences, Southwest Hospital, China | Allogeneic hESC-RPE |
Phase I/II |
Dry AMD, wet AMD, SMD | Federal University of São Paulo | Allogeneic hESC-RPE as cell suspension versus as a sheet on a polymeric substrate |
Most of ongoing and planned clinical trials use allogeneic material and therefore are using an immunosuppression therapy. The immunosuppression strategy is intensely debated despite the fact that the eye is generally considered as an immune-privileged site [
Cell therapy for retinal dystrophies is at the forefront of the clinical investigation compared to other diseases in which such treatment approach could be applied. Indeed, with the ease to access and the available imaging techniques as well as the sophisticated functional assays, retinal diseases serve as a proof of concept of regenerative medicine applications. However, we should remain cautious with the completion of all milestones before treating patients. In particular, the quality, safety, and expected functionality of the cell therapy product to be delivered to patients should be precisely evaluated. In the coming years, results from many clinical trials will help determine the best formulation strategy to be applied for each category of patients as well as the level of immune suppression required.
KB and CM are inventors of a pending patent related to medical devices for the preparation of retinal tissues for regenerative medicine.
This work was supported by grants from the ANR (SightREPAIR: ANR-16-CE17-008-02), the Fondation pour la Recherche Médicale (Bioengineering program: DBS20140930777), and the LabEx Revive (ANR-10-LABX-73) to Christelle Monville. It was supported by NeurATRIS, a translational research infrastructure (Investissements d’Avenir) for biotherapies in neurosciences (ANR-11-INBS-0011), and INGESTEM, the national infrastructure (Investissements d’Avenir) engineering for pluripotent and differentiated stem cells (ANR-11-INBS-000), to Christelle Monville. I-Stem is part of the Biotherapies Institute for Rare Diseases supported by the Association Française contre les Myopathies- (AFM-) Téléthon.