Human eye formation involves a complex process requiring elaborate epithelial movements and cellular choreography. The retinal pigmented epithelium (RPE), Bruch’s membrane/choroid and retinal photoreceptor cells are the dominant cell types participating in light-perception. Any progressive degeneration of these cells can lead to retinal degeneration (RD). RD includes diverse ocular diseases, namely, age-related macular degeneration (AMD), retinitis pigmentosa (RP), diabetic retinopathy and glaucoma. In particular, the main characteristic of AMD (mainly affecting elderly individuals worldwide) is an abnormal decrease of RPE leading to secondary photoreceptor dysfunction, while RP (the leading cause of irreversible blindness in paediatric and young populations) is a hereditary retinal degenerative disease characterized by the progressive death of photoreceptors [
1]. Although AMD and RP differ in pathological progress, they impinge upon a common final pathway of photoreceptor loss. It is evident that the prevalence of blindness due to RD is increasing according to the latest systematic analysis of causes of blindness around the world from 1990 and 2010, and the prevalence linked to macular degeneration was 5 and 7%, respectively [
2]. Blindness is associated with devastating impacts on functional abilities and quality of life, leading to increased health care resource utilization and higher patient support cost [
3]. In addition, the degree of vision impairment for different individuals varies depending on age, disease stage, and occurrence time emphasizing the need for new proposals to help prevent or reverse RD. A number of related strategies have been explored, such as neurotrophic factor supports, electronic retinal prostheses and pharmacological treatments (e.g., anti-vascular endothelial growth factor therapy used in wet AMD treatment [
4]). However, the current strategies only retard the progression of RD, which is not yet curable. The regrowth of retinal cells is still limited although Joel Schuman et al. have shown neurogenesis in the adult mice retina [
5]. Stem/progenitor cell-based therapy could play a critical role in sight restoration by replacing missing retinal cells and/or rescuing remaining cells. Here, different stem/progenitor cells can be obtained from two broad lines. (1) Ocular-derived progenitor cells, e.g., retinal progenitor cells (RPCs), are located in the inner layer of the optic cup where nearly all retinal cell types initially differentiate from [
6]. It has been reported that foetal and postnatal-derived RPCs could express immature markers, indicative of a retinal stem-cell state [
7,
8]. (2) Non-ocular-derived stem cells (with the potential to self-renew and produce different cells including RPE, photoreceptors, etc.), include embryonic stem cells (ESCs) [
9], induced pluripotent stem cells (iPSCs) [
10], mesenchymal stromal cells (MSCs) (particularly bone marrow mesenchymal stromal cells (BM-MSCs) [
11] and adipose-derived stromal cells (ADSCs) [
12]).
In this review, an in-depth analysis of RPCs, ESCs, iPSCs and MSCs was conducted. Identifying the main advantages and disadvantages of these cells is the key to selecting the most promising candidates that could be applied in RD treatment (Table
1). Among these four cell types, RPC populations used as allografts have shown immune privilege and a relatively simple manufacturing process [
13], and they are one of the best options. However, the challenge of acquiring adequate progeny still remains. ESCs and iPSCs have the potential to replace retinal cells. However, the vital issues in the use of ESCs and iPSCs are ethical and biosafety concerns (like genetic abnormalities [
14]). In contrast, MSCs mainly provide trophic support to slow down retinal cell degeneration instead of replacing the missing retinal cells. At present, clinical trials are underway to evaluate three major issues (Table
2): (1) safety, (2) efficacy, and (3) efficiency. Based on the ability of transplanted cells to differentiate and replace the missing photoreceptors or simply protect the remaining photoreceptors during degenerative process, cell-based therapy appears to be valid so far [
15‐
17]. Stem/progenitor cells present challenges related to their proliferation and/or differentiation into target cells in vitro, but that does not apply to RPE [
18]. Other factors to consider are limited likelihood of long-term graft survival and host functional restoration in vivo. Even then, it is anticipated that this will progressively become a promising method for visual restoration in the near future because of the concerted research efforts worldwide.
Table 1
Comparison of four types of stem/progenitor cells for RD clinical application
Derivation/generation sources | Foetal and postnatal retina | Developing embryos | Terminally differentiated tissues | Developmentally mature organs |
Advantages | Simplicity, accessibility and safety (minimal trauma); immune privilege; ready neuroprotection; no tumourigenicity; no requirement of immunosuppressive drugs | Differentiation into various retinal cell types; providing abundant donor cells | Without ethical concerns; low risk of immune rejection (autologous hiPSC derivatives); gene therapy | Trophic support; immunosuppression |
Disadvantages | Low rate of cell proliferation | Ethical concerns; tumourigenicity; requirement of immunosuppressive treatment throughout life | Low differentiation efficiency; biosafety concerns (e.g., genetic abnormalities) | Low rate of cell migration and differentiation |
Table 2
Clinical trials using stem/progenitor cell-based therapeutics in RD
hRPCs | NCT02320812 | I/IIa | jCyte | California, US | Henry Klassen | June 2015 | Ongoing but not enrolling patients | 28 RP | 500,000 –3,000,000 cells | 12 | Intravitreal | No data | No data | No data | No data | ReNeuron (Boston: Phase I/II, 15 RP) |
ESCs → RPE | NCT01345006 | I/II | Advanced Cell Technologies | California, US | Steven Schwartz | April 2011 | Completed | 9 SMD | 50,000–150,000 cells | 22 | Subretinal | None | 10 | 7 | 1 | Pfizer [London: Phase I, 10 AMD (wet)] |
NCT01344993 | 9 AMD (dry) |
NCT01625559 | I | CHABiotech | Seoul, South Korea | Won Kyung Song | September 2012 | Unknown | 2 SMD | 50,000 cells | 12 | Subretinal | None | 3 | 1 | None |
NCT01674829 | I/IIa | 2 AMD (dry) |
NCT02286089 | I/II | Cell Cure Neuroscience | Jerusalem, Israel | Ytzhak Hemo | April 2015 | Enrolling patients | 15 AMD (dry) | 50,000–500,000 cells | 12 | Subretinal | No data | No data | No data | No data |
iPSCs → RPE | UMIN000011929 | I | RIKEN | Kobe, Japan | Masayo Takahashi | September 2014 | Suspended | 1 AMD (wet) | 1.3 mm × 3 mm RPE sheet | 12 | Subretinal | None | None | 1 | None | National Eye Institute (Preclinical, AMD) |
BM-MSCs | NCT01068561 | I | University of Sao Paulo | São Pauloin, Brazil | Rubens C Siqueira | May 2009 | Completed | 3 RP and 2 cone-rod dystrophy | 10,000,000 cells | 10 | Intravitreal | None | 4 | 1 | None | Red de Terapia Celular (Spain: Phase I, 10 RP); Al-Azhar University [Egypt: Phase I/II, 1 AMD (dry)] |
NCT01560715 | II | January 2011 | Completed | 20 RP | 12 | 20 (transitorily) | None | None |
BM CD34+ cells | NCT01736059 | I | University of California, Davis | California, US | Susanna S Park | July 2012 | Enrolling patients | 6 RD or ischaemic disorders | 3,400,000 cells | 6 | Intravitreal | None | 6 | None | None |
Progress in the study of stem/progenitor cells in RD
We focused on advances in two broad categories of stem/progenitor cells, i.e., ocular-derived progenitor cells and non-ocular-derived stem cells, which were studied broadly in various animal models of RD (mouse, rat, rabbit, pig, monkey, etc.) and applied in clinical trials.
Difficulties and prospects
RD commonly results from RPE and photoreceptor apoptosis. Much progress in stem/progenitor cell therapy for RD have been made through a succession of studies on RPCs, ESCs, iPSCs, and MSCs. They are currently regarded as promising therapeutic approaches for RD. However, one critical point is to choose the best stem/progenitor cell source for successful clinical application. Here, RPC populations are one of the most promising candidates because the manufacturing process is relatively simple, safe and straightforward. More importantly, RPCs originating from the developing retina have exhibited immune privilege as allografts so that better neuroprotection can be attained relative to other pluripotent cells [
13]. Although the proliferation of donor RPCs is limited utilizing primary culture, RPCs can be viable beyond passage 20 by making full use of novel culture techniques. Compared with RPCs, the other three stem cell types have their own characteristics (Table
1). The ethical concerns are particular to clinical applications of hESCs involving the use of early human embryos. Regardless of the low risk of graft-host immune rejection, iPSCs can lead to tumourigenicity, mutations and epigenetic changes. Whether the extraneous four transcription factors induce reprogrammed iPSC abnormalities is still unclear [
101]. MSCs predominantly protect retinal neurons from further dysfunction at early stages of RD rather than replace the lost and dead retinal cells at late stages. In addition, clinical trials are underway. The primary objectives of Phase I and II clinical trials are safety and efficacy, respectively. Both types of clinical trials require enough time and patient samples, although no major adverse event has been reported so far. At the same time, these successes of RD treatment will further represent a solid milestone for the treatment of other degenerative diseases in the brain and spinal cord in the near future because they all belong to central nervous system and share most common characteristics of the regenerative response.