Autologous Adipose Tissue Stem Cell-Derived Terminally Differentiated Retinal Pigment Epithelium Monolayer on a Biocompatible Scaffold, to Restore Vision in Age-Related Macular Degeneration Patients

Abstract

In Western countries, age-related macular degeneration (AMD) is the common cause of central visual loss in the elderly leading to gradual blindness. Studies implicate the retinal pigment epithelium (RPE) as an important player in AMD pathogenesis, as progressive loss of RPE cells and photoreceptors lead to poor visual acuity. Several strategies aim to prevent the RPE degeneration by saving the damaged retinal cells or replacing them. Cell rescue provides retinal cells with trophic or immunomodulatory factors, whereas cell replacement aims to repair and regenerate neuroretina providing new cells. Currently, a major limitation is cell loss from subretinal injections of bolus cell suspensions. The most promising studies focus on tissue transplantation or RPE cell patches hosted on implantable scaffolds made of advanced biomaterials. The latter support the development of artificial 3D retinal constructs improving functional integration capacity and increasing the survival of implanted cells into the damaged retina. However, there is no consensus on the optimal RPE source, delivery strategy, cell dose and animal model for testing. This chapter aims to improve the efficacy of RPE grafting suggesting an optimal cell source, an efficient delivery method, and the topography of retina damage as a determining factor to calculate the effective therapeutic dose.

1. Introduction

In current industrialized countries, approximately 1.75 million individuals are affected by loss of ocular light-sensing retinal tissue. However, the disease does not always lead to rapid visual loss; depending on the type and severity of AMD, blindness may occur for months to many years [1].

The human RPE epithelium forms early in development and subsequently remains dormant, undergoing minimal proliferation throughout normal life [2]. Many blinding diseases are characterized by the progressive loss of photoreceptor cells (cones and rods) and lack the ability to regenerate in mammals, including humans. Very few existing conventional therapies for retinal diseases slow the progression of the diseases. Basically, there are two potential options to deal with this progressive condition: First, to rescue the damaged tissue by immunomodulation or with retinal support cells that secrete trophic cytokines [3]; second, to repair and regenerate the impaired neuroretinal tissue with cell replacement therapy [4].

RPE cell loss occurs early in age-related macular degeneration (AMD), and cell replacement is the aim of several stem cell therapy programs. Cell replacement therapy is an attractive option for retinal diseases as the eye is an easily accessible and readily monitored human organ; additionally, the risk of cell transplant rejection is limited due to the autonomous and partially privileged immune structure [4]. Therefore, the development of advanced therapy medicinal products (ATMPs) has become a fast-growing field and presents an opportunity for preserving or restoring vision even in advanced stages of retinal degeneration, when photoreceptor degeneration has already occurred [5]. As there are currently almost no elective therapies available for AMD and other RPE disorders, RPE replacement, especially in later disease stages, has significantly accelerated progress in the field of RPE biology [2]. The effective integration of the transplanted RPE cells into the damaged area to achieve the reconstruction of the neural circuitry and the long-term survival and functionality of the transplanted cells are currently the major targets for successful transplantation [6, 7]. However, a major limitation is donor cell loss associated with common delivery methods such as subretinal injections of bolus cell suspensions. Advanced biocompatible materials to develop implantable artificial 3D retinal constructs and support their functional integration in the damaged retinal microenvironment may be a potential solution to the main problems of cell transplantation, improving the integration capacity and increasing the survival of implanted cells [8].

Currently, there is no consensus regarding the optimal strategy for RPE replacement therapy, and multiple RPE sources, delivery methods, cell doses, and recipient animal models have been investigated with variable results [9]. This chapter aims to clarify the importance of these factors in successful RPE transplantation and focuses on selecting the optimal RPE source, the most efficient method of cell delivery, and the calculation of therapeutic dose. We suggest that the degree of retina damage should be a determining factor in assessing the therapeutic cell dose. Moreover, this proposal could provide a way out of the existing problem of determining the effective therapeutic dose of cell preparations in the field of regenerative medicine to achieve the best results for the benefit of AMD patients and other people, who resort to regenerative medicine to restore their damaged health.

2. Architecture and function of normal retina

The human retina is the light-sensitive tissue lining the inner surface of the eye. It is a complex neural network composed of multiple cell layers with different features and functions (Figure 1). The main objective of this network is light transduction through electrical pulse, encoding, and packing visual input for visual pathways and superior visual elaboration [2].

Figure 1.

Anatomy of the human eye. The light path is represented as a rainbow. Inset sections represent the retinal cell types present in the macula (top) and the fovea (bottom). Key describes retinal cell types. Credit: Ref. [10] (with permission).

Retina is a 2-layered cup formed by the retinal pigment epithelium (RPE) monolayer (outer layer) and multilayer neurosensory retina (inner layer), separated by the subretinal space (Figure 2). The RPE is a cuboidal, pigmented cobblestone monolayer (Figure 3) of polarized columnar epithelium of post-mitotic epithelial cells. These cells are located in Bruch’s basement membrane, while their apical surface contacts the outer segment of retinal photoreceptors. (Figure 4) [14]. The hexagonal network formed by RPE cells does not allow cell overlapping or for the formation of voids, thus permitting higher area coverage with the lower cell surface tension. The innermost layer or neuroretina is made up by a complex circuitry of six main neuronal cell types, including (from inner to outer): photoreceptors (PRCs) – rods and cones – horizontal, bipolar, amacrine and retinal ganglion cells (RGCs) (Figures 1 and 4). PRCs are located in the most external layer of neuroretina, and they have a close anatomical relation with the RPE monolayer. Bruch’s membrane (BrM, Figure 5) is an organized 2–5 mm-thick acellular sheet divided into five layers and mainly composed of elastin, collagen I-V, laminin and fibronectin. Structurally, this multi-laminar and highly specialized structure separates the RPE layer from the choroid; functionally, it supports the exchange of nutrients and oxygen supply between the neural retina and the capillaries of the choroid layer. The choroid is a vascular layer of the eye and lies between the sclera and the BrM (Figure 2). Its capillaries supply oxygen and nourishment to the outer layer of RPE. Sclera is a tough white sheath around the outside of the eye ball (Figure 2). The fenestrated choriocapillaris, BrM, and RPE form the outer blood-retinal barrier [7, 8, 15, 16]. Fovea is a small depression in the middle of the retina and is responsible for high-resolution vision [7, 17]. This retinal region is a relatively small area of 1.8 mm² and represents a cone-rich area extremely important for human vision, as it facilitates important tasks such as facial recognition, driving, reading, writing, etc. It is estimated that there are approximately 200,000 cones, a cell number easily produced by organoid technology [8]. Region surrounding the fovea is called the macula (Figures 1 and 6), the 6-mm centermost area of the neurosensory retina rich in cones.

Figure 2.

Intravitreal versus subretinal delivery. The first is less technical with fewer risks associated with structural damage to ocular tissues; however, PRE cell delivery to the disease tissue can be problematic because of the diffusional volume of the vitreous. Subretinal delivery is more technically challenging, however, the cell concentration is much higher. Credit: Ref. [11] (permission copyright clearance Center, Inc).

Figure 3.

The RPE layer is a sheet of polygonal mostly hexagonal cells. Cells differentiated from the iPSC of a healthy donor and cultured for 3 months. Credit: Ref. [12] (with permission).

Figure 4.

(a) Retina structure and its main neural cell types. (b) Schematic of a healthy versus diseased retina. Retinal degenerative diseases damage the physiology of the retina firstly affecting the RPE cells and the rod and cone photoreceptors, causing irreversible vision disfunctions in the patient. – Credit: Ref. [13] (with permission).

Figure 5.

Main functions of RPE. Implementation of the visual cycle - barrier function - secretion of hormones and growth factors - phagocytosis of used photoreceptor discs - absorption of light. Credit: Ref. [12] (with permission).

Figure 6.

Loss of central vision in AMD patients. (A) Visual field seen by a person with normal vision and by an AMD patient (loss of central color vision). (B) A normal human eye compared with an AMD eye depicting deposits of drusen (yellow spots) and choroidal neovascularization with subretinal hemorrhage in the macula (red). (C) Cross-section of a normal eye through the macula shows retinal neuronal layers, RPE, Bruch’s membrane, and choroid vessels. Credit: Ref. [18] Oct (with permission).

RPE represents a pillar to maintain the structural and functional integrity of the entire retina. It supports the PRCs, providing metabolic and transport functions essential for homeostasis of the neural retina (Figure 5). The hallmark of the RPE layer is the presence of melanosomes. Stray light passing through the PRC layer is absorbed by the contained melanin granules and is also prevented from reflecting on the PRCs themselves. In the absence of melanin, the visual image would be degraded [7]. RPE also phagocytoses the shed photoreceptor outer segments; when it degenerates, as happens in AMD, eventually, PRCs degenerate with it [7, 16].

Light enters the eye through the cornea, passes through the pupil, lens and strikes the retina (Figure 1). Visual information from the retina transmits to the brain through optic nerve fiber. In this visual signal processing, a sequence of events takes place, and every cell plays a definite role. PRCs, the first-order specialized neurons, have the structure and metabolic privilege to convert light into electricity through the visual cycle, a molecularly-driven complex procedure. The PRC-generated electrical impulse is then transmitted to bipolar cells (second-order neurons) and finally to the optic nerve core, which is composed of ganglion cells (third-order neurons) [19].

3. Age-related macular degeneration (AMD)

RPE layer is a fragile photosensitive part of the central nervous system prone to degenerative diseases, including AMD. With the adjacent light-sensitive photoreceptors form a functional unit, lining the back of the eye. Its dysfunction plays an important role in degenerative retinal diseases, characterized by non-functional or loss of PRCs leading to deterioration or permanent vision loss [9].

AMD is the common cause of progressive central visual loss in the elderly >60 years old, leading to blindness and affecting 1.75 million individuals in Western countries [14]. Several studies point to the RPE as an important player in AMD pathogenesis. Progressive dysfunction and loss of central – not peripheral retinal pigment epithelium cells and photoreceptors lead to poor visual acuity. Central vision is required for detailed work and for tasks like reading and driving. Early disease may be without appreciable vision loss. Complete blindness is not always the worst outcome for AMD patients. It may occur from months to many years, depending on the type and severity of the disease. Nonetheless, central vision disorder is a debilitating event that affects patients’ quality of life, e.g., loss of ability to drive, read, facial recognition. (Figure 6). Multiple risk factors drive the pathogenesis of AMD, most prominently age, smoking, dietary modifications, and genetic variation in multiple genes [16]. In the retinal microenvironment, loss of the balance between oxidants and anti-oxidants can cause nitrogen species and reactive oxygen to accumulate, triggering chronic inflammation. RPE becomes dysfunctional and, in severe cases, necrotic enhancing drusen deposits and complement system activation [20]. Visual loss caused by AMD cannot normally be reversed; currently, there is no effective therapy for this highly prevalent disorder to reverse the degenerative conditions.

There are two main types of AMD (Figure 7) [21]: the initial non-exudative or dry AMD characterized by pigment alterations in the macular area and accumulation of extracellular whitish-yellow deposits of protein, lipid and carbohydrate termed drusen (the pathological hallmark of AMD). Drusens are deposited subretinally at the RPE-Bruch’s membrane interface and are accompanied by RPE cell atrophy. Dry AMD constitutes about 90% of AMD cases in the United States. The later exudative stage, or wet AMD, occurs after the ingrowth of the underlying choroidal vasculature into the retina. Drusen formation is accompanied by progressive atrophy of the RPE layer and dysfunction of the overlying PRCs (Figure 8). Coalescence of large drusen droplets may cause separation of the RPE layer from Bruch’s membrane with subsequent apoptosis of RPE cells; geographic atrophy represents an advanced AMD type with extensive apoptosis leading to gradual loss of central vision. A small ratio of all dry AMD cases (about 10%) progress to the wet type of disease, characterized by choroid neocapillaries penetrating the damaged Bruch’s membrane, overgrowing into the RPE layer and overlying the neurosensory retina causing higher damage to central vision [2]. Wet AMD is the most severe but more treatable, and it can be delayed by repeated anti-VEGF medication.

Figure 7.

What is age-related macular degeneration. The intimate relation among photoreceptors, RPE, and choroid is disrupted by drusen (lipid and cellular debris). This separates RPE from Bruch’s membrane and the underlying choroidal vessels in association with RPE atrophy and photoreceptor degeneration (hallmarks of dry AMD). In wet AMD, abnormal leaky choroidal neovessels proliferate and penetrate the altered Bruch’s membrane protruding into the subretinal space, causing hemorrhage and rapid loss of vision. Credit: https://learn.eyecheck.com/what-is-macular-degeneration (with permission - Mr. J. Anthony Moses CEO Eyecheck (https://eyecheck.com).

Figure 8.

Healthy macula vs. macula in AMD. The “dead zone” in the macula has minimal viable PRs and RPE cells. The healthy zone is usually toward the periphery of the macula and still has functional PRs with healthy RPEs. In the “transition zone” RPE cells are dead, but the PRs are intact. This is the potential target zone for RPE transplantation. Credit: Ref. [22] (with permission).

4. Cell replacement therapy

In mammalian organs of the central nervous system, including the ocular PRE, the lack of regenerative capacity represents a significant reason for using stem cells to replace damaged tissues, e.g., in dry AMD, an atrophic condition of PRE. Importantly, RPE cells normally divide and eventually differentiate during early embryogenesis; the resulting post-mitotic RPE cells generally remain dormant but can be activated to proliferate in various circumstances of adult life [2].

RPE loss of function, which occurs early in the course of AMD and precedes photoreceptor death, has been a key therapeutic target for the last three decades [7]. RPE cell transplantation holds promise by replacing dysfunctional RPE cells and lost PRs to slow disease progression and/or restore vision. In clinical trials, RPE cells were injected in suspension and biomaterials [23, 24]. A key factor in increasing a retinal implant’s effectiveness is the accomplishment of a substantial integration of the implanted cells [8]. Limitations such as the delivery method, low integration rate, and short-term survival of grafted cells hinder a successful cell-based therapy; moreover, it is unknown if disease pathogenesis due to a genetic mutation may affect a grafted RPE; this would downgrade RPE cell intervention as a temporary treatment and not a cure [7].

Clinical trials have demonstrated safety and modest vision improvements, but their number has been too small. They also do not provide specific data to elucidate if implants act via cell replacement, neuro-protection, immune-modulation or other mechanisms [15]. Due to the absence of predictive or diagnostic markers and the lack of long-term follow-ups to ensure safety, the in vivo functionality of RPE grafts has not been fully studied, and their survival remains uncertain even if anatomical integration is achieved [7].

4.1 Cell sources

Cell source is a key component to the success of any cell therapy. Identifying the appropriate cell source and suitable type, as well, represents a major challenge for the successful treatment of human retinal diseases. Current therapy focuses on the direct differentiation of four kinds of somatic stem cell therapy medicinal products (sCTMPs) into RPE and PRC phenotypes: human embryonic stem cells (hESCs), human inducible pluripotent stem cells (hiPSCs), retinal progenitor cells (RPCs) and mesenchymal stem/stromal cells (MSCs) [4, 8, 25]. All of the above stem cell types have been extensively investigated in several preclinical and clinical trials to confirm their effectiveness in repairing damaged retinal tissues. In addition, MSC-derived paracrine factors and exosomes have been tested to understand their efficiency in reversing retinal degeneration.

Embryonic stem cells (ESCs) have unlimited self-renewal capability. These totipotent stem cells, which originate from the inner cell mass of the embryonic blastocyst, can differentiate into cell types from all three embryonic germ layers, namely, ectoderm, mesoderm and endoderm [26]. The generation and establishment of murine and human ESCs and iPSC was a decisive step in the advancement of clinical translation, as their capacity for unlimited expansion provided an inexhaustible source of viable, easy-to-use donor cells. Numerous in vitro studies have demonstrated the ability of both human and mouse ESCs to differentiate into different functional retinal types, such as photoreceptors and RPE cells [27]. Treatment of retinal degenerative diseases by hESC-derived photoreceptors with or without RPE cells offers a vast source of transplanted cells for retinal replacement therapy. Using a specific combination of factors, hESCs can be directed to a retinal fate to generate retinal progenitors [28]. The Food and Drug Administration approved the first clinical trial using hESC-derived RPE cells for the treatment of dry AMD and Stargardt disease [29]. Despite encouraging results, the potential of using ESCs in cell replacement therapy for the treatment of retinal diseases is still limited. Ethical issues associated with obtaining hESCs and serious complications, including xeno-transplant immune rejection, have limited their clinical application [30]. Additionally, after several months of naïve ESCs transplantation, teratoma formation containing cells of all three embryonic germ layers from residual undifferentiated stem cells may have been observed [19]. Therefore, stringent release testing strategies are designed to detect cells with tumorigenic potential prior to transplantation.

iPSCs have several advantages as a source of retinal cells for transplantation. They possess unlimited self-renewal capacity and can be obtained from the patients themselves by directly reprogramming adult somatic cells to transit to a pluripotent state [31, 32]. Depending on the combination of soluble agents used, these cells can be induced to proliferate and differentiate into various retinal cell types, including rods, cones, retinal ganglion cells, etc. They can be collected easily and produced at a relatively reasonable price, present no ethical problems, and can be autologous, thus preventing rejection issues. However, they could conserve the epigenetic characteristics of the original cells, harbor the disease genes from the donor (the mutation load of iPS cells passes through several stresses and long culture periods), and lead to tumor formation [33]. A major drawback of iPSCs is their unstable ability to differentiate into the specifically programmed cell type. Additionally, in the case of autologous therapies, it would be advisable for patients with genetic mutations to have their genome edited. When therapies are allogenic, potential rejection may spoil long-term results; therefore, human leukocyte antigens (HLA) matching and possibly immunosuppression would be needed [7].

MSCs are a population of multipotent stromal cells that can be found abundantly in a variety of adult tissues including bone marrow, adipose tissue and dental pulp [34]. They can also be isolated from neonatal tissues and fluids such as umbilical cord, Wharton’s jelly, amniotic membrane, amniotic fluid, and placenta. Despite the different tissues from which they are generated, MSCs have some distinct common characteristics [5]. Below we list the specific minimal criteria established in 2006 by the International Society for Cellular Therapy (ΙSCT), to define a cell as an MSC [35]:

• MSCs are able to adhere to plastic support in standard culture conditions, and this ability is exploited for their isolation.

• Have the potential to differentiate into an array of cell types including osteoblasts, chondrocytes, myocytes, adipocytes, neural cells, etc. (Figure 9).

• Do not express hematopoietic and endothelial cell markers such as CD31, CD34 CD45, CD11b, CD11c, CD14, CD19, CD79a, CD86, Stro-1, SSEA-4 and HLA class II molecules.

• Defined by the expression of specific cell surface markers as CD105, CD90, CD44, CD9, CD13, CD29, CD271, CD146 and CD73.

• Capable of avoiding immune cell recognition due to these properties, low levels of expression of class I HLA, zero expression of class II HLA, and low immunogenic capacity; all these properties render MSCs an appropriate source for allogeneic and autologous transplants.

• Wide distribution and ease to harvest.

• Possess minimal susceptibility to malignant transformation.

• Therapeutic efficacy to promote regeneration of multiple cells including central nervous system neurons.

Figure 9.

Mesenchymal stem cell differentiation potential. Credit: Ref. [36] (reproduced under the terms of the CC attribution 4.0 International License).

Recent clinical trials to repair retinal degeneration with MSCs have demonstrated their safety profiles and promising potential to delay further retinal dysfunction. Administration of MSCs has revealed significant restoration of the visual system. MSCs have the potential to survive for a long time after injection into the vitreous body and have the capacity to regenerate a crushed optic nerve or protect retinal ganglion cell functions. Neuroprotective effects on degenerated retinal cells could be associated with delaying or even stopping uncontrolled cell death [5]. Due to their neurotrophic properties, pluripotency, hypo-immunogenicity, modulation of immune response to tissue injury, limited ethical concerns, and low risk of tumor formation, MSCs have garnered significant interest in regenerative medicine and emerged as a prospective leading cell source for the treatment of retinal diseases [37]. However, in some clinical trials for the repair of retinal disorders with MSCs, various problems regarding the efficacy and safety profiles were reported, e.g., severe side effects or failure ti achieve the primary end-points of stem cell efficacy. Several clinical trials have shown that some AMD patients – after receiving intravitreal transplantation of autologous bone marrow/adipose tissue-derived MSCs experienced devastating outcomes (dense vitreous hemorrhage and rhegmatogenous retinal detachments following severe proliferative vitreoretinopathy and finally lost their vision) [1, 38, 39, 40]. Similarly, a high rate of retinal detachment and perforation was observed in AMD patients with geographic atrophy after subretinal injection of MSCs derived from human umbilical cord blood [41]. A probable cause is the trans-differentiation of injected MSCs into myofibroblast-like cells, which can induce fibrosis or proliferative vitreoretinopathy [25]. Complications, which are a significant issue of constant research, have additionally been attributed to the surgical procedure applied, number of administered MSCs, transplantation timing and the intrinsic mechanism of their in vivo action [42]. Therefore, for ocular or intraocular delivery of MSCs, it is imperative to follow more careful surgical procedures and more stricter control regarding the quality of MSC preparations [43].

Table 1 contains the most important advantages and disadvantages of the three main stem cell types used for the treatment of human retinal diseases.

	ESCs		iPSC		MSCs
A/A	Advantages	Disadvantages	Advantages	Disadvantages	Advantages	Disadvantages
1.	Stable genome	Forbidden in several countries due to ethical reasons.	Differentiation capacity in diverse retinal populations	Genomic instability and tendency to carcinogenesis	Desirable differentiation potential	Limited differentiation potential (osteoblasts, chondrocytes, myocytes, adipocytes, etc.) compared to ESCs/iPSCs
2.	Endless self-renewal ability	High costs for production and lower availability	Affordable collection and easy production scaling	Changing ability to differentiate into the desired cell linage	No ethical bans – Low immunogenicity	Susceptibility (minimal) to malignant transformation
3.	Reduced rejection in allogenic transplants	Risk of teratoma formation	Can perform either autologous or allogeneic transplantation	Rejection of allogeneic grafts	Ability to regenerate multiple human-damaged tissues, including CNS neurons
4.	Totipotent cells, able to differentiate into any type of human somatic cells	Ethical issues arising from collection sources and their use	Without any ethical dilemmas		Wide distribution in human body – easy to isolate, expand and be maintained

Table 1.

Advantages and disadvantages of ESCs – iPSCs – MSCs [7, 25, 44].

4.2 Cell delivery formulations

In simple cell replacement therapy techniques, such as subretinal delivery of cell suspension boluses, a major challenge is the loss of donor cells in the intraocular microenvironment (Figure 10). Currently, cell transplantation with supportive biomaterials is considered as an effective alternative to avoid cell loss, improve survival rate, and increase cell integration with the host retina microenvironment [45].

Figure 10.

Scheme of possible RPE transplant delivery system. Credit: Ref. [7] (with permission).

4.2.1 Cell suspensions

RPE cell suspensions do not appear to be the ideal solution for restoring defective vision in AMD patients. When administered in a suspension formulation, there is an increased risk of cultured or freshly harvested RPE cells not surviving or integrating into the retinal network of aged or damaged Bruch membrane, causing serious complications such as proliferative vitreo-retinopathy [16]. Cell suspension injections are a suboptimal procedure with three major drawbacks: First, loss of retinal cells can be observed due to their dispersion into the host’s intraocular tissue (Figure 11) [8]. Second, areas of basal lamina can form, alternating with multilayers formed by the uneven distribution of the transplanted RPE cells. Third, cells cannot integrate and function normally for a prolonged period in a defective or diseased basal lamina; additionally, the hostile sub-foveal microenvironment is not conductive to successful retinal engraftment due to glia and host immune insults [13].

Figure 11.

Surgical approaches for subretinal delivery. (A) Intravitreal injection: Fills the vitreous body with the therapeutic suspension, (B) subretinal injection: Delivers the therapeutic suspension under the sensory retina into the subretinal space (between photoreceptors-RPE), (C) suprachoroidal injection: Delivers cells into to the space suprachoroidal space (between choroid-sclera). Credit: Ref. [46] (with permission).

4.2.2 Scaffold implants

A transplant-ready RPE monolayer can be generated on certain biomaterials to facilitate cell delivery and in vivo engraftment. RPE cells can be grown on bioengineered, nanofibrous scaffolds, and the RPE cell-enriched scaffolds can be transplanted into patients’ eyes [13]. An ocular scaffold is a biocompatible construct fabricated to mimic the native retinal microenvironment. It aims to support essential properties of the grafted cells, e.g., adhesion, proliferation, differentiation and long-term survival. In different experimental models, implanted retinal cells on supportive biomaterials slightly improved vision capabilities, achieving a better but limited integration, than suspension cells in the same area [8].

Recent advances in material science helped to develop strategies to grow cells as intact monolayers or as sheets on biodegradable or non-degradable polymer scaffolds for transplantation into the eyes. Biocompatible polymers are either synthetic substances, e.g., polycaprolactone (PCL), polylactic acid (PLA) and poly lactic-co-glycolic acid (PLGA), or materials of natural origin, e.g., celluloses, zein, alginates, etc. [47]. Such implants are found to be more promising than the bolus injection approach. Obviously, the supportive biomaterial construct along with the extracellular matrix formed can help maintain cellular integrity, hold the cells in place to increase cell survival, encourage proper cellular alignment, and improve integration with the host retina and visual function [16, 45]. A crucial factor to the successful long-term function of a synthetic biomaterial construct is its ability to replace native damaged BrM, in which the grafted RPE cells can grow and integrate into the native retinal circuit. The surface properties of the underlying substrate scaffold directly influence the cells’ ability to form a differentiated monolayer. The physical properties of synthetic membranes include biostability, porosity and suitable mechanical strength for surgical handling [16].

Biomaterial scaffolds provide a platform for reconstituting RPE sheets with the neighboring choroid and neurosensory retina. They:

• must eliminate the possibility of injected RPE cell suspensions refluxing into the vitreous cavity, which can cause fibrosis, leading to retinal detachment and blindness [16];

• may ensure the correct orientation of rods and cones and their functional vicinity with underneath RPE cells [48];

• should create a functional microenvironment to favor adhesion, survival and differentiation of transplanted retinal cells;

• must be nontoxic when implanted in the subretinal area;

• must be below 10 μm in thickness, in order not to affect the focal length of the eye, provide mechanical stability and flexibility to resist implantation strain and facilitate retina curvature;

• are positioned between the RPE and choriocapillaris, so they should ideally degrade and be replaced by the natural extracellular matrix that separates these tissues;

• should be biocompatible, biologically inert (not induce local tissue damage) and sterilizable, neither triggering cytotoxic processes nor inflammatory responses. A very promising biomaterial is silkworm silk fibroin because it fulfills all these requirements [8];

• could be derived from natural sources (chitosan, gelatin, collagen, cellulose, alginates, etc., are more preferred as they have higher biocompatibility, excellent biodegradability, and minimal toxicity) or made synthetically (polylactide-co-glycolide (PLGA), polycaprolactone (PCL), polylactic acid (PLA), fibronectin, polyurethane, etc.) [49].

Researchers have used biomaterials based on 3D printed Bruch’s membranes, alginate, silk, cellulose, and hyaluronic acid [13] to create scaffolds to:

• support the development of transplanted retinal cells

• act as substitutes for Bruch’s membrane or

• act as retinal cell delivery systems.

Comparing biodegradable scaffolds with non-biodegradable counterparts seem to perform better: they keep the focal length of the eye unchanged, they do not leave foreign materials in the subretinal space, they allow better vascularization, the new RPE sheet is smoothly incorporated into the host’s RPE network [50]. In addition, transplanted polymers that undergo rapid bulk degradation such as poly lactic-co-glycolic acid (PLGA) significantly lower local pH and cause widespread retinal injury when delivered in sufficient quantities.

5. Terms and conditions for a long-term functional retinal cell implant

Precise integration of the implanted cells into the existing host cellular network in the damaged macula area is a prerequisite to increasing the effectiveness of retinal implant therapies. Approaches using cell suspensions involve easer technical maneuvers; however, their functional outcomes seem to be inferior. Indeed, the success of retinal transplantation based on cell suspension injection is hindered by limited cell survival and lack of cellular integration. Currently, the ultimate balance between practicability and clinical outcomes remains unclear. Among the different experimental approaches tested, patch implants showed satisfactory outcomes with a reasonable incidence of adverse events. For the treatment of AMD, we propose to administrate – on appropriate scaffolds – in the form of monolayer as many mature RPE cells as are required to functionally cover a damaged area [7, 16].

5.1 Factors to consider in establishing a long-term functional RPE implant

RPE cell implants are able to survive and make connections to the host retina with visual improvement, if the graft integrates and/or communicates with the host retina [17]. The limitations of suspension cell injections can be overcome by an efficient strategy using 2D or 3D scaffolds. The current major obstacles to successful retinal cell transplantation are the integration of the scaffold into the impaired area of the host retina and the functional reconstruction of the neural circuitry [6]. In contrast to cell suspensions, ex vivo growth of an RPE monolayer has the advantage of performing experimental readouts of its functionality before surgical implantation [23]. Importantly, most biomaterials used for retinal regeneration have not been tested under in vivo conditions [8]. For AMD patients, because implanted RPE cells will need to be functional, possibly for decades, properties such as cell attachment to an aged Bruch’s membrane and survival of transplanted cells in a degenerative environment present a challenge [17]. Although safety has been demonstrated, the cell replacement mechanism and efficacy remain difficult to validate [15].

5.1.1 The basic principle of cellular implant generation: Replicating the in vivo healthy RPE layer

Aged human Bruch’s membrane does not support attachment and survival of transplanted RPE. Compared to an RPE monolayer patch, injectable cell suspensions do not reliably form a functional RPE monolayer and exhibit poor long-term survival. Transplantation of an intact RPE sheet onto a suitable substrate mimicking normal Bruch’s membrane represents a feasible approach to overcome this hurdle [16]. Scaffolds must be engineered in such a way that they can mimic the conditions of the host retinal microenvironment and promote the structural and functional integration of cells into the existing native neuronal network. A monolayer RPE patch aims to recapitulate the anatomy of a normal RPE layer. Also, it promotes the formation of tight junctions between neighboring retinal cells (Figure 12) to establish cell polarity, which is necessary for the RPE cells to support the photoreceptor functions.

Figure 12.

RPE replacement therapy for wet AMD. (a) Choroidal blood vessels penetrate through the RPE, leaking fluid in the subretinal space. (b) An hESC-RPE monolayer on a plastic substrate (green) is delivered to the areas of vessel leakage and RPE - retina detachment. (c) Patient photoreceptors (orange) recover through interaction with transplanted RPE cells; cytokines from RPE cells cause choroidal vessels to stop proliferating and leaking. Credit: Ref. [50] (with permission).

5.1.2 Disadvantages of suspension-injected RPE grafts

There have been several investigations into the inoculation of RPE cell suspensions. There is also evidence that suspension-injected RPEs do not always survive long -term and frequently fail to form a functional monolayer. This is a significant challenge for injections of suspended cells because the therapeutic effects of grafted RPE cells depend on their polarization and spatial arrangement in a subretinal monolayer. Unfortunately, the ability of cells to migrate from the transplant site to the damaged host retina is strongly prevented by the hostile environment of the degenerating retina [19]. Both preclinical and clinical trials have shown that suspension-injected RPE eventually loses most of its morphological and functional properties, resulting in low efficacy of this type of cell replacement therapy [43]. RPE cells have a limited migratory ability, and it is unknown if they can migrate away from the injection site, how far they will travel in a diseased eye, and if the benefit is similar to animal studies. Additional limitations include low survival, low cell integration at the transplant site, and difficulty in maintaining the injected cells in the target areas. Conclusively, the current golden standard technique for cell delivery in retinal therapies, that is, subretinal bolus injection of RPE cells, is governed by limitations inherent to the transplantation process. It frequently results in disorganized and poorly localized grafts with low rates of cell survival due to poor incorporation of donor cells and cell regression into the vitreous cavity [28].

5.1.3 Cell delivery system: Retinal damage should determine the properties of engineered scaffolds

It has been shown that aged or damaged human Bruch’s membrane does not support a functional suspension-injected RPE implant in the long-term. Due to the disappointing outcomes, infusion of cell suspension is currently considered not to be a suitable cell replacement approach for retinal regeneration [16]. Presently, cell attachment on a scaffold represents the preferred therapeutic approach. Critical to successful subretinal transplantation is the ability to grow a functioning RPE monolayer on a substrate, having the ability to replace a damaged Bruch’s membrane. Such a biocompatible construct may have the potential to preserve or restore vision in patients affected by blinding diseases such as AMD (Figure 13). The choice of substrates with proper physical and chemical properties should assist the engineered cell retinal construct be functional even for decades [7]. The type of support should be more biologically relevant and mimic, at some points, the endogenous Bruch’s membrane [8]. Scaffolds for retinal cell transplants can be fibrous or cylindrical or made of hydrogels to closely imitate the microstructure of the extracellular matrix or the vertical cell arrangement in the retina layers or the mechanical properties of retina, respectively. The candidate scaffolds on which RPE cells will be placed may be designed to act either as supporting devices or as artificial BrMs, and their selection should preferably have the following characteristics [16]:

• for easy handling, to have proper flexibility and mechanical resistance

• for the exchange of nutrients, to be porous

• for subretinal implantation, to have compatible thickness

• for biodegradable substances, not to produce toxic by-products.

Figure 13.

Wet- and dry- type AMD cell therapy: A conceptual mode of action using pluripotent stem cell-derived RPE cell products. Credit: Ref. [51] (with permission).

Various scaffolds with well-defined characteristics such as physico-mechanical properties, ultrafine structure and biocompatibility have been investigated for retinal tissue engineering and repair. Many synthetic polymers such as poly (e-caprolactone) and poly (ethylene glycol) diacrylate, polylactic acid, poly-glycerol sebacate, polylactic-co-glycolic acid, etc., have also been tested for their ability to deliver RPE cells or photoreceptors to the subretinal area, or as a Bruch’s membrane substitute. Studies have shown that polymers present different advantages [13]:

• increased rate of proliferation and differentiation,

• higher cell survival and number of cells delivered to the subretinal area,

• induction of retinal differentiation to generate functional neurons.

The needs of both RPE and photoreceptor replacement therapy are not likely to be met by a single polymer substrate alone. It is known that RPE cells have a limited ability to adhere and function on a damaged or aged Bruch’s membrane; therefore, the support provided by a polymeric material must be essential for RPE cell localization, survival, or directed functional differentiation. However, a permanent barrier in the subretinal space would compromise photoreceptor transplant functionality [13]. To overcome this obstacle, we would have to make a porous scaffold with natural materials, e.g., collagen, elastin, etc., in order to mimic healthy Bruch’s membrane as closely as possible. Biodegradable polymer scaffolds may increase survival cell rate, and it also appears that the adhesion and orientation of transplanted retinal cells are promoted by the scaffold porosity (dimensional conformation). In addition, cell orientation can be affected by scaffold topography because of its positive influence on cell morphology, proliferation, migration and adhesion [52]. Good results on biomaterials and RPE cells have been obtained from both, in vitro and in vivo studies; however, how long the transplanted cells could survive has still to be determined [53]. Unfortunately, research in the field is still in the early stages, clinical trials are very few and in very early stages of development, and the lifetime of retinal grafts has not been correlated with any prognostic indices [8].

5.1.4 Interventional timing

In RPE cell suspension replacement therapy, the time point at which retinal cells should be injected is a critical factor in successful transplantation. The remaining dysfunctional RPE cells do not allow the attachment of new cells, as they still occupy Bruch’s membrane. An ideal invasive timing would be when the endogenous host-damaged RPE cells have been removed, but photoreceptors are still present and functional. For photoreceptor preservation strategies, early intervention leads to better vision preservation [17].

5.1.5 Site of graft placement

However, in some cases, transplantation may cause a retinal detachment (Figure 14) [17]. The point of injecting an RPE implant is also a significant factor. It should be near areas of the macula or geographic atrophy as the AMD macula is affected. However, in some cases, the transplant can cause a retinal detachment, which represents a high risk for macular vision disorder. In patients with geographic atrophy, there are probably several foci of degenerative cells, and inoculation of multiple retinal implants in an equal number of injection sites is recommended [17].

Figure 14.

Depiction of the subretinal injection site. Credit: https://jirehdesign.com/stock-eye-illustrations/eye-surgery/retina-surgery-illustrations/subretinal-injection-suvr0049/ (with permission - Mr. Mark Erickson CEO JirehDesign www.JirehDesign.com).

5.1.6 Cell density of RPE layer: An important factor to calculate the effective injected cell number

The RPE sheet is a monolayer in the outer retina composed of polygonal, mostly hexagonal cells and is highly adapted to local physiological requirements (Figure 3). Increasing RPE cell numbers by transplantation can effectively treat atrophic RPE diseases, including AMD. Currently, there is no standardized guidance for the injection volume and cell concentration, and the exact number of cells required for efficient transplantation remains unclear [17].

Patient age, location of retinal damage, and disease status are three factors that affect RPE cell density and shape. Across the posterior pole, significant spatial variations can be seen in terms of RPE cell morphology, photoreceptor density and distribution, as well. The healthy foveola is composed of strictly mononuclear cells, mostly hexagonal and of uniform size. It exhibits the highest RPE cell density, and only cones are found. In the human eye, there are approximately 137 million photoreceptors, 7 million cones and 130 million rods. The rods are located at a distance of approximately 6 mm from the foveal center and about 200 from it. Their number decreases toward the periphery but more slowly than the cone numbers. Instead, the cones can be seen in the center of the fovea. Their density lowers from the center to the periphery in a radial manner. Regarding apoptotic RPE cells, age significantly increases their proportion in erderly patients. The majority of them are mainly confined to the macula area and their number is more decreased in peripheral regions of Bruch’s membrane [54]. In aging, although RPE cell apoptosis occurs in the macula, the overall number remains constant and decreases in the periphery. This observation suggests that peripheral RPE cells migrate from the periphery to the center of macula where they can replace the dead cells of the RPE layer. Indeed, although during aging, the density of RPE cells in the macula region remains unchanged, a more variable number of neighboring cells can be detected in this RPE area of the macula.

Various structural changes can be observed in an aged RPE layer, such as basal deposits, accumulation of drusen, calcification of the elastin layer, diffuse layer thickening, laminar deposits, collagenic cross-linking, etc. Age-dependent extracellular matrix changes within Bruch’s membrane may induce apoptosis in cells in the overlying RPE because disruption of epithelium-basement membrane interactions leads to apoptosis. These changes in the extracellular matrix of Bruch’s membrane are the hallmark of elderly AMD patients at autopsy and may be related to apoptosis in the overlying RPE [54].

The number of RPE cells in normal human eyes decreases by approximately 0.3% per year (in younger eyes it is estimated to be ~0.07 per 100,000 cells) [55]. In older eyes loss increases up to 1.96% of macular RPE cells per year. Τhis can translate into an annual loss of 1960 cells per 100,000 of RPE cells in the human macular; that is, a maximum of nearly 20% of RPE cells in the macula would be lost per decade, which represents a significant rate of loss. The local variations in normal cell density, combined with variations in disease presentation, make systematic assessment of the clinical status of AMD patients a prerequisite prior to the production of any RPE implant. Accurate topography of the lesion is required as a guide – by making further anatomical and integrity studies, such as optical coherence tomography (OCT) analysis – to calculate the extent of the damaged area and the approximately (exact) number of dead RPE/photoreceptor cells to engineer a potential fully functional implant.

5.1.7 Autologous adipose tissue mesenchymal stem cell-derived terminally differentiated RPE (ADMSC-RPE): a promising cell population for retinal regeneration

ADMSCs are a promising cell therapy with regenerative properties. These multipotent stem cell-like reside in the stromal vascular fraction (SVF) of adipose tissue, and their properties can be compared to those of bone marrow stromal cells (BMSCs) properties comparable to those of bone marrow stromal cells (BMSCs). A major advantage over BMSCs is that much larger amounts of regenerative ADMSCs can be harvested by applying minimally invasive procedures during liposuction. ADMSCs can perform their paracrine action by secreting a plethora of pro-regenerative cytokines and growth factors to enforce the regeneration of damaged tissues [16]. MSCs meet all the specifications recommended for human ADMSCs by the relevant department of the International Society for Cellular Therapy (ISCT) [35]. Hunan ADMSCs under standard culture conditions exhibit adherent properties, proliferate, differentiate, express the surface markers CD13, CD29, CD44, CD73, CD90, CD105, CD146 and CD271 and are negative for the antigens CD31, CD34, CD45, Stro-1, SSEA-4 and HLA-DR [44]. Similar to many other types of stem cells, they express a high proliferative rate and have the ability to differentiate into cells originated from other embryonic germ layers. Under in vitro conditions, they have the potency to differentiate into osteoblasts, adipocytes and chondroblasts; they can also be induced to differentiate into myogenic, neurogenic, angiogenic, hepatic and retinal cell lines [56]. By choosing autologous fatty tissue to isolate ADMSCs, we avoid the destructive side effects of a potential RPE transplant rejection. These properties make ADMSCs attractive candidates for cell replacement therapies and a proper cell source for the treatment of several retinopathies [48, 57].

In humans, RPE cells undergo early terminal differentiation starting at 4–6 weeks of gestation. These cells normally remain dormant throughout life, dividing rarely or not at all [2]. They can be activated upon release from the niche that normally maintains a quiescent state. Escape from the niche causes massive RPE proliferation, leading to devastating outcomes, e.g., destructive proliferative vitreoretinopathy (PVR) or disc-shaped scar formation. Although MSCs have emerged as a prospective leading cell source for the treatment of retinal diseases, such destructive clinical results have reduced their credibility as a safe source. The proliferative response of RPE can be reparative in dry AMD to partially counter RPE cell atrophy [2], or after receiving intraocular injection of autologous BM- or AT-derived MSCs [40]. Mature RPE cells derived preferentially from the most readily available ΑΤ-ΜSCs appear to be a safe and functional cell source for retinal replacement therapy. This technique allows for the easy production of a considerable number of MSC-derived RPE cells seeded on an appropriate scaffold and maintains the physiological and functionally crucial RPE polarization [7]. Due to their terminal differentiation state, the scaffold-implanted RPE cells lack proliferative capacity and remain dormant within the retinal microenvironment, mimicking the in vivo physiological state. Therefore, the repaired RPE monolayer is not compromised by cell overgrowth which may cause uneven distribution of RPE cells, multilayer formation or even retinal detachment.

5.1.8 Photoreceptor replacement therapy

Photoreceptor damage is one of the main characteristics found in retinal degeneration diseases, such as Retinitis Pigmentosa or AMD. Such degenerative diseases are currently turning into irreversible conditions attributed – on the one hand, to the insufficient intrinsic regeneration rate in the human retina and – on the other hand, to the lack of effective treatments to arrest the progressive loss of photoreceptors. Cell and tissue engineering approaches focus on the preservation of photoreceptors by replacing damaged RPE; this technique has been shown to slow the degeneration of photoreceptors and even restore partial visual functions [21, 58]. To tackle their loss and restore vision in affected AMD patients, photoreceptor cell replacement therapy has been proposed as a potential treatment modality [17, 45]. Although it seems that photoreceptor degeneration may be retarded by RPE replacement, a ‘retinal sheet’, that is, an organized construct that can be easely incorporated into native tissue will likely be needed to replace impaired photoreceptors to restore vision [6].

5.1.8.1 Photoreceptor transplantation in advanced AMD stages

In advanced stages of AMD, photoreceptor replacement can be an effective therapeutic strategy to substitute this type of retinal cell, which eventually dies due to support inability by RPE cells. This therapeutic option either replaces damaged photoreceptors with new ones or induces the secretion of neuro-protective or immune-modulatory factors that promote the functional recovery of damaged cells [43]. Τhere is a wide time-frame to apply a therapeutic photoreceptor intervention, which increases the probability of successful cell transplantation because of the prolonged periods of survival time of neurons of the inner retinal layers after photoreceptor degeneration. However, there is a very difficult condition, namely, the implantation of functional photoreceptors that will make new synapses with the inner neighboring bipolar cells and the reversal of the synaptic remodeling that occurs after the degeneration of photoreceptors [59]. Although post-mitotic transplanted photoreceptors have been proved to be able to regain relative vision functionality [60], to date, no one has achieved a significant restoration of visual function.

5.1.8.2 Photoreceptor preservation

The development of physical substrates for artificial RPE sheets is the main objective for biomaterial engineers. Preserving photoreceptors is not an easy task. Difficulties include keeping them alive until transplantation time and replacing those already been lost, by successful integration of new into neural retina. Depending on the donor age, cultured photoreceptors can be maintained alive for 24–48 h and after transplantation the survival rate is less than 20% [61]. Importantly, the ratio of functional photoreceptors among alive ones is unclear. It is confirmed that a polarized photoreceptor monolayer construct may successfully get integrated into the neural retina by using high-tech 3D scaffolds [62].

5.1.8.3 Photoreceptor transplantation with biomaterials

Both natural biological and synthetic polymeric materials are used to construct scaffolds for photoreceptor implants. Such biological materials that mimic natural architectures include alginate, collagen, etc. Synthetic polymers, such as poly-caprolactone (PCL), poly (lactic-co-glycolic) acid (PLGA), etc., express higher mechanical strength and controlled degradation rates [52]. We would not recommend the use of scaffolds with the capacity to facilitate the differentiation of photoreceptor progenitors (PRPs). Similar to the destructive behavior of MSCs in the repaired RPE monolayer, uncontrolled overgrowth of PRPs could compromise the functional integrity and survival of the repaired photoreceptor network.

5.1.8.4 Retinal cell transplantation must be done in a timely manner before loss of photoreceptors

Among patients with progressing AMD there is a fear that a risky surgical procedure will further worsen their already impaired vision, so they consider their remaining reduced vision. For this reason, in clinical trials, only severely disabled patients are treated with few healthy native photoreceptors capable of interacting with the implanted patch. However, one should take into consideration the adverse molecular microenvironment that prevails throughout the central nervous system and makes the regeneration of photoreceptors (and all cell repair processes) hardly reversible [8]. The greater the loss of photoreceptors, the more hostile the retinal microenvironment.

5.1.9 About animal models: Rat models are a favorable choice for the study of retinal degeneration

Pig eyes are considered ideal for improving translational surgical techniques due to size similarity to human eyes [63], however, large numbers of pigs are costly to house and difficult to immunosuppress for a long time. These inhibitory factors make it difficult to use pig eyes to assess the tolerability of a human retinal cell implant. There are many mouse models for studying hereditary retinal degeneration in humans. However, the mouse eye is too small to reliably inject into the subretinal space and functionally evaluate a photoreceptor graft on a polymer scaffold. Current research is using rat models to study retinal degeneration diseases and even to implant polymer-supported RPE grafts and full-thickness retinal sheets [64]. Their wide experimental use is attributed to (i) the cheap price that allows housing in large numbers and testing many variables simultaneously; (ii) the size of eyes, which is large enough for subretinal transplantation of a biodegradable cell implant.

5.1.10 Clinical validation is essential for post-transplantation

The retinal transplant can be safely and inexpensively observed with bio-microscopy at frequent intervals following transplantation [45]. In addition, advanced optical systems have been invented for the in in vivo imaging of retinal grafts, including spectral domain-optical coherence tomography (SD-OCT) and adaptive optics-scanning laser ophthalmoscopy (AO-SLO). These imaging systems are valuable for studying cellular changes of retinal transplants monitoring treatment responses and disease progression [17].

6. Conclusion: key points for successful retinal transplantation

1. A cell product (suspension or implant) is a drug, “living drug,” and as with conventional drugs, we must know its therapeutic dose.

2. MSCs (and stem cells in general) have unlimited proliferative capacity. Once infused, they expand beyond the initial administered dose:

   • making the determination of the therapeutic cell dose a serious problem

   • causing RPE detachment due to overproduction of MSC-derived RPE cells.

3. MSCs are considered the safest stem cells (Table 1), and their selection represents the best cell source for RPE transplantation.

4. Adipose tissue-derived mesenchymal stem cells are increasingly used in retinopathies due to their ease of obtaining and preparing. Implantation of autologous ADMSC-derived RPE cells protects from the destructive side effects of possible transplant rejection.

5. There is no risk of rejection after administration of live drugs consisting only of autologous ADMSC-derived terminally differentiated RPE cells.

6. Due to their terminal differentiation state, the in vivo proliferative capacity of scaffold-implanted RPE cells is negligible or zero; they will remain dormant within the retinal microenvironment, mimicking the in vivo physiological state. The repaired RPE monolayer will not be compromised by cell overgrowth which may cause uneven distribution of RPE cells, multilayer formation or even retinal detachment.

7. In AMD replacement therapy, in vivo simulation of a healthy RPE layer or photoreceptor network should be the target of subretinal delivery of adipose-derived and scaffold-engineered human retinal cells.

8. Cell suspension preparations proved to be ineffective in achieving reconstruction of the neural circuitry and long-term survival and functionality of transplanted cells.

9. Cell implants on biocompatible scaffolds are the current trend in AMD treatment because they have been shown to integrate efficiently and functionally into the damaged retina.

10. The selection of a substrate with appropriate physical and chemical properties should be a determining factor for the long-term functional performance of the engineered retinal tissue and mimics at some points the endogenous Bruch’s membrane. Depending on the clinical status of patients, the candidate scaffold on which RPE cells will be placed may be designed to act either as a supporting device or as an artificial BrM.

11. Clinical parameters should determine the design of AMD cell therapy, which will be different for each patient, not only for patients belonging to different types (dry or wet) but also for patients of the same type. Damage extent and location will determine the:

    • concentration of implanted cells

    • composition and physicochemical properties of scaffold (biodegradable, or non-biodegradable)

    • interventional timing.

12. Retinal cell transplantation must be done in a timely manner before loss of photoreceptors.

13. Long-term post-transplantation monitoring with advanced in vivo imaging systems (AO-SLO, SD-OCT) will be a prerequisite.

14. The proposed cell replacement procedure for AMD patients will be:

    • an indicative personalized therapeutic regimen, where dose of cell preparation and scaffold properties will be determined by scientific criteria and not empirically

    • a possible solution to the existing problem of determining the therapeutic dose of cell preparations in the field of regenerative medicine.

Glossary

Adaptive optics-scanning laser ophthalmoscopy (AO-SLO): It is an imaging method for monitor living retinal cells; it makes use of adaptive optics to provide sharper images of the retina or cornea obtained with scanning laser ophthalmoscopy (SLO), removing any optical aberration. SLO utilizes confocal laser microscopy to examine the human retina or cornea for diagnostic imaging purposes.

Adipose tissue-derived mesenchymal/stromal stem cells (AD-MSCs): A highly promising class of multipotent mesenchymal stem cell-like cells for cellular regenerative therapies. They reside in the stromal vascular fraction (SVF) of adipose tissue and have the ability to secrete a multitude of pro-regenerative growth factors. In general, their properties are comparable to those of stromal cells in the bone marrow (BMSCs). All criteria for MSCs recommended by the relative Committee of the International Society for Cellular Therapy (ISCT) are met by human/mouse ADMSCs. They have the potency to differentiate into osteoblasts, adipocytes, and chondroblasts, as well as neural and retinal cells in vitro.

Age-related macular degeneration (AMD): The common cause of severe central visual loss in the elderly in industrialized Western countries, associated with loss of face recognition and ability to read or drive a vehicle. Gradual blindness is attributed to damage to Bruch’s membrane or choriocapillaris or photoreceptors or RPE layer. There are two distinct AMD types, neovascular (NV) or dry and non-neovascular (NN) or wet. About 10% of dry AMD patients progress to the wet form of AMD. NV AMD is more severe but it can be delayed with repeated anti-VEGF medication, in contrast, no effective treatment exists for NN-AMD.

Amacrine cells (ACs): Retinal inter-neuronal cells with inhibitory properties. In the inner plexiform layer (IPL), they make synaptic connections with bipolar and ganglion cells. They reach across several bipolar cells to regulate signals directed at retinal ganglion cells. So far, around 30 subtypes have been identified.

Bipolar cells (BCs): A type of retinal interneurons which transmits signals to retinal ganglion cells that have received from photoreceptors.

Bruch’s membrane (BrM): It is described as a highly specialized and pentalaminar collagenous structure that separates the RPE cells from the choroid, the underlying vascular layer. It is an organized acellular sheet 2–5 mm-thick, and its five layers are mainly composed of elastin, collagen I-V, laminin and fibronectin. Functionally, it supports the exchange of nutrients and oxygen supply between the neural retina and the capillaries of the choroid layer.

Choriocapillaris: The innermost layer of the choroid (vascular layer of the eye that lies between sclera and BrM) underlying Bruch’s membrane. It consists of compacted anastomotic capillaries, which supply oxygen and nourishment to the outer layer of RPE and photoreceptors and removes their metabolic wastes. The fenestrated choriocapillaris, BrM, and RPE form the outer blood-retinal barrier.

Cone photoreceptors: Retinal photoreceptors that mediate color vision and function optimally in bright light. There are three types of cones that mediate trichromatic vision in humans: S-M-L-cones sensitive to short-medium-long wavelength light, respectively. The human retina contains 6 million cones residing in the foveola in a densely packed form. Cones mediate the high-acuity vision necessary for face recognition, writing, reading, driving, and fine tasks such as sewing.

Drusen: They are extracellular whitish-yellow deposits of protein, lipid, and carbohydrate deposits located under the retina. The exact relationship between degenerative macular disease and drusen is not clear. Drusen formation, the pathological hallmark of AMD is accompanied by progressive atrophy of the RPE layer and resulting dysfunction of the overlying photoreceptor cells. Merge of large drusen can cause separation of RPE from Bruch’s membrane leading to RPE cell apoptosis. If the latter is extensive, it leads to an an advanced form of AMD called geographic atrophy.

Dry or non-neovascular AMD: The initial non-exudative step or dry AMD, which is characterized by pigment alterations in the macular area and accumulation of drusen. They are deposited subretinally at the RPE-Bruch’s membrane interface and accompanied by RPE cell atrophy. Dry AMD constitutes about 90% of AMD cases in the United States, and visual loss caused by AMD cannot normally be reversed, as there is currently no effective therapy for this highly prevalent disorder to reverse the degenerative conditions.

Fovea: In the middle of the retina, there is a small depression called fovea (which constitutes an area of 1.8 mm²), which is responsible for high-resolution vision. Fovea is rich in cone photoreceptors; within the fovea there are approximately 200,000 cones, a number of cells that can easily be produced by the current organoid technology. This relatively small region is extremely important for human vision; it makes tasks such as reading, facial recognition and driving.

Ganglion cell layer: The innermost retina layer is composed of ganglion cells and displaced amacrine cells. Retina ganglion cells transmit the synaptic input – received by bipolar cells and amacrine cells – to the brain thalamus, hypothalamus, and midbrain via long projections that form the optic nerve.

Geographic atrophy (GA): An advanced form of AMD characterized by well-defined areas of dead photoreceptors, dead underlying RPE cells and atrophic subjacent choriocapillaris. If GA occurs in the fovea, the patient loses high-acuity vision.

Human embryonic stem cells (hESCs): RPE cells and/or photoreceptors have been demonstrated by numerous protocols, and ESCs are omnipotent cells with unlimited self-renewal capability. They originate from the inner cell mass of the embryonic blastocyst and can differentiate into cell types from all three embryonic germ layers, namely, ectoderm–mesoderm–endoderm. The generation of RPE and/or photoreceptor cells from in vitro differentiation of mouse and human ESCs has been demonstrated by numerous protocols and offers huge potential for cell replacement therapy in treating retinal degenerative diseases. Clinical application has been limited by ethical issues related to obtaining hESCs. In addition, serious complications, including xeno-transplant immune rejection, as well as teratoma formation after several months of transplantation (containing cells of all three embryonic germ layers) have prevented their widespread application in cell replacement therapies.

Human inducible pluripotent stem cells (hiPSCs): Τhey can be obtained from the patients’ themselves by directly reprogramming adult somatic cells to transit to a pluripotent state. Several types of retinal cells including rods, cones and ganglion cells have been derived from in vitro differentiation of iPSCs using a combination of soluble factors. They can be collected easily and produced at a relatively reasonable price, possess unlimited self-renewal capacity, present no ethical problems, and can be autologous, thus preventing rejection issues. However, they could conserve the epigenetic characteristics of the original cells, harbor the disease genes from the donor (the mutation load of iPS cells passes through several stresses and long culture periods), and lead to tumor formation. When cells are autologous, genome editing would be required in patients with genetic mutations.

Horizontal cells (HCs): Regulate the signal that emerges from several rods and cones.

Inner nuclear layer (INL – also inner synaptic layer): It consists of the cellular bodies of the following retinal neurons: amacrine, horizontal, bipolar; it also contains the bodies of Müller cells (the major type of glial cells in the retina).

Inner plexiform layer: An area of the retina that is made up of a dense reticulum of synaptic connections (fibrils) formed by interlaced dendrites of ganglion cells, bipolar cells, and amacrine cells (cells of the inner nuclear layer). The IPL contains the synapse between the second-order and third-order neurons in the visual pathway.

Inner segment: It is a part of the photoreceptor cell where the Golgi apparatus, endoplasmic reticulum, ribosomes and abundant mitochondria are contained; there is also a cilium connecting it with the corresponding outer segment and a synaptic connection with horizontal cells.

Intravitreal injection: It is used to administer medications or other substance to treat a variety of retinal disorders. It is performed to place medicines inside the eye, near the retina. The drug is delivered into the vitreous humor, the jelly-like substance inside the eye. Intravitreal drug delivery has become the gold standard for the treatment of many retinal diseases, including neovascular AMD, diabetic retinopathy and retinal vein occlusion, the most common conditions treated with intravitreal anti-VEGF drugs.

Macula: The macula is an oval-shaped small area in the central retina of humans and other animals. It is the pigmented part of the retina located in the very center of the neurosensory retina (in the center of the macula is the fovea). The diameter of the human macula is approximately 5.5 mm and is subdivided into six smaller areas: fovea, parafovea, perifovea, umbo, foveola, and foveal avascular zone. The human macula contains 6 million cones in total, residing in the foveola in a densely packed form. The macula is responsible for the central, high-resolution color vision mediated by cones. If the macula is damaged, e.g., in macular degeneration disorders, this type of vision is impaired. Clinical monitoring of the macula is done from the pupil, as in ophthalmoscopy or retinal photography.

Mesenchymal stem/stromal cells (MSCs): They are a population of multipotent stromal cells that can be found abundantly in a variety of adult tissues such as adult bone marrow, adipose tissue, and dental pulp. They can also be isolated from neonatal tissues and fluids such as the umbilical cord, Wharton’s jelly, amniotic membrane, amniotic fluid, and placenta. Although MSCs are derived from different tissues, they share some common features. In recent clinical trials for the treatment of retinal disorders, MSC has shown encouraging safety profiles. In addition, they have demonstrated promising efficacy in having the ability to delay retinal degeneration.

Müller cells (MCs): The most common and abundant type of glial cells in the vertebrate retina (followed by astroglia and microglia). The MC is the only retinal glial cell that shares a common cell lineage with retinal neurons. Müller cells provide trophic and anti-oxidative support to photoreceptors and retinal neurons and regulate the tightness of the blood-retinal barrier.

Nerve fiber layer (NFL): The retinal nerve fiber layer (RNFL) or nerve fiber layer, comprises the axons of retinal ganglion cells. These ganglion cell axons collect the visual impulses produced by the rods and cones and travel through the bodies of ganglion cells.

Outer segment (OS): A subcellular photoreceptor organelle filled with proteins essential for the photoconversion cascade, including light-sensitive pigments. They undergo renewal on a daily basis by discarding their discs (tips), which are phagocytosed by the RPE cells. These physiological mechanisms enable the OSs to maintain constant length while shedding accumulated toxic photo-oxidative compounds.

Photoreceptors (PRCs): Retinal neurons specialized in photo-transduction. They convert light, electromagnetic radiation into cell membrane potential, whose changes result in altered neurotransmitter release from the synaptic terminals of photoreceptors. The light is converted into electrical signals that travel through other retinal neurons to reach the optic nerve. There are two main types of light-sensitive cell photoreceptors in the human eye: rods and cones. Rods are responsible for vision in poor light and at night (scotopic vision); they do not mediate color vision and have a low spatial acuity. Cones enable vision at higher light levels (photopic vision), are capable of color vision, and have a high spatial acuity.

Retina: A thin light-sensitive layer located at the back of the eye. It is a two-layered cup formed by the multilayered neurosensory retina and RPE monolayer. The subretinal space lies between them. This intricate structure is essential for vision. In the center of the retina is the macula, the part responsible for central and fine-detail vision needed for tasks such as reading.

Retinal pigment epithelium (RPE): A retinal cell monolayer attached to Bruch’s membrane and closely associated with the outer segments of the overlying photoreceptors. These segments are phagocytosed by the apical villous processes of RPE cells on a daily basis. RPE cells have a hexagonal morphology and are usually pigmented containing melanin granules. The RPE layer forms a natural barrier to the choroidal capillaries and removes harmful metabolic waste produced by photoreceptors in response to light.

Retinal progenitor cells (RPCs): A population of human RPE stem cells.

Rod photoreceptor: The second type of retinal photoreceptor cells that mediates vision under dim light conditions. One hundred twenty million rods are contained in the human retina, and unlike cones, rods have only one type of photosensitive pigment called rhodopsin. Rods do not mediate color vision, and they have poor visual acuity because many rod cells share one connection to the optic nerve.

Scaffolds: They are natural or synthetic biomaterials that mimic the microenvironment of native retinal tissue. They can increase the survival of transplanted cells, support their integrity, encourage proper cell integration with the host retinal cellular network and organize the orientation of photoreceptor cells. The scaffolds significantly decrease the likelihood that injected suspensions of RPE cells regress into the vitreous cavity, thereby helping to avoid potential fibrosis and retinal detachment.

Spectral domain-optical coherence tomography (SD-OCT): OCT angiography is one of the breakthroughs in the advancing field of ophthalmic imaging, a non-invasive in vivo imaging modality combining both quantitative and qualitative assessment of vasculature seen on standard angiography. This imaging modality offers volumetric resolution of the head of the optic nerve, retina, and choroid. Early hemodynamic alterations in the vasculature of choroid and retina layers can be readily detected. Because of its non-invasive nature, standard angiographic techniques that monitor changes in the superficial layers of choroid and retina, e.g., indocyanine green angiography (ICGA) and fundus fluorescein angiography (FFA), respectively, utilize OCT.

Subretinal space: A potential space that lies between RPE monolayer and the neurosensory retina. Subretinal infusion is considered a specialized cell delivery method for retinal implants.

Vitreous cavity: An intraocular cavity that occupies the area between the retina and crystalline lens. It is filled with a clear gel, the vitreous humor, which may regulate the concentration of ions in the retina.

Wet or neovascular AMD: The advanced exudative stage or wet AMD occurs after ingrowth of the underlying choroidal neovascular vessels (CNV) through the damaged Bruch’s membrane into the RPE layer and neurosensory retina. An estimated proportion of 10% of dry AMD patients progress to the wet form of AMD, leading to gradual loss of central vision.

“Wet”” AMD is the most severe but more treatable, and it can be delayed by repeated anti-VEGF medication.

Notes

The reason for writing this chapter was a published paper reporting three cases of vision loss after bilateral intravitreal injections of autologous adipose tissue-derived stem cells in patients with non-neovascular AMD, obtained in 2015 at a private clinic in Miami Florida – United States [1]. Unfortunately, stem cell injection caused rhegmatogenous retinal detachment combined with vitreous hemorrhage, ocular hypertension and hemorrhagic retinopathy. One year after cell therapy both patients’ clinical conditions worsened finally without any light perception.

The next generation of therapeutics approved for the clinic given to regeneration or repair of damaged tissues or organs is expected to be cell preparations termed “living drugs”; this name is due to their ability to dynamically and temporally respond to changes during their ex vivo manipulation and after in vivo administration. A cell product (suspension or implant) is a living drug containing living cells that retain the in vivo ability to proliferate. Once injected, the cells expand beyond the initial administered dose, making determination of the therapeutic cell dose a serious problem in the modern field of regenerative medicine. In contrast, it is unthinkable to administer a conventional drug whose dose of active ingredients is unknown. Stem cells in general have unlimited proliferative capacity and could be likened to a new unruly horse without a bridle in a meadow (obviously, its rider is in danger at any moment from its unpredictable behavior). Although stem cell biology is very developed, there are still many gaps in basic properties such as nature, structure, proliferation, differentiation, mutagenesis, oncogenesis, mechanism of action, etc. Therefore, the enthusiasm for their therapeutic administration in various diseases is still premature and dangerous.

It should be preferable to avoid the administration of stem cells and their undifferentiated progeny for regenerative purposes; instead, we prefer live drugs containing only stem cell-derived terminally differentiated cells, whose proliferative capacity is negligible or zero. AD-MSCs are increasingly used due to the ease of their isolation and manipulation; subretinal delivery of human RPE cells, photoreceptors, or both – differentiated from AD-MSCs to replace the damaged cells in AMD patients – is currently being investigated in several approved clinical trials worldwide. However, one must take into account the individual clinical parameters that are different for each patient, which inevitably lead to personalized therapeutic regimens. Cell replacement therapy should be considered as an individualized therapy and be planned as such. In the present chapter, the proposed treatment process for AMD is illustrated in the individualized treatment method of each patient who is given cell replacement therapy. It could also serve as a solution to the existing problem of determining the therapeutic dose of cell preparations in regenerative medicine to avoid devastating outcomes as in the aforementioned cases of the three women.