Looking into the future: Gene and cell therapies for glaucoma

Abstract

Glaucoma is a complex group of optic neuropathies that affects both humans and animals. Intraocular pressure (IOP) elevation is a major risk factor that results in the loss of retinal ganglion cells (RGCs) and their axons. Currently, lowering IOP by medical and surgical methods is the only approved treatment for primary glaucoma, but there is no cure, and vision loss often progresses despite therapy. Recent technologic advances provide us with a better understanding of disease mechanisms and risk factors; this will permit earlier diagnosis of glaucoma and initiation of therapy sooner and more effectively. Gene and cell therapies are well suited to target these mechanisms specifically with the potential to achieve a lasting therapeutic effect. Much progress has been made in laboratory settings to develop these novel therapies for the eye. Gene and cell therapies have already been translated into clinical application for some inherited retinal dystrophies and age-related macular degeneration (AMD). Except for the intravitreal application of ciliary neurotrophic factor (CNTF) by encapsulated cell technology for RGC neuroprotection, there has been no other clinical translation of gene and cell therapies for glaucoma so far. Possible application of gene and cell therapies consist of long-term IOP control via increased aqueous humor drainage, including inhibition of fibrosis following filtration surgery, RGC neuroprotection and neuroregeneration, modification of ocular biomechanics for improved IOP tolerance, and inhibition of inflammation and neovascularization to prevent the development of some forms of secondary glaucoma.

1. GLAUCOMA – DEFINITION AND CURRENT THERAPIES

Glaucoma is a complex group of optic neuropathies defined by loss of retinal ganglion cells (RGCs) and their axons within the optic nerve.^1,2 Intraocular pressure (IOP) is a major risk factor that is commonly elevated. Glaucoma is a leading cause of vision loss in both humans and animals, especially dogs.^1–6 Depending on absence or presence of a detectable underlying disease process resulting in impaired aqueous humor drainage through the iridocorneal angle (ICA) and subsequent IOP increase, glaucoma is defined as either primary or secondary. Individuals affected with primary glaucoma have a genetic predisposition. Primary glaucoma can be further categorized as either open-angle or angle-closure glaucoma based on ICA morphology.^1,2 Possible causes of secondary glaucoma include the impairment of aqueous humor outflow through the ICA by invading neoplasms or neovascularization and formation of a preiridal fibrovascular membrane (PIFVM) as induced by chronic anterior uveitis.^4,5,7 In dogs, lens-induced uveitis associated with cataract formation and surgery are common causes of secondary, neovascular glaucoma.^8–14 Regardless of the underlying etiology, the final common pathway for all forms of glaucoma consists of loss of RGCs and their axons with characteristic atrophy and cupping of the optic nerve head (ONH).

Except for addressing known underlying causes in secondary glaucoma, currently accepted treatments to slow disease progression are limited to lowering IOP by medical and surgical methods. These occur by lowering production and increasing drainage of aqueous humor. The most used drug classes by veterinary ophthalmologists, under consideration of species-specific efficacy differences, include prostaglandin-analogues, carbonic anhydrase inhibitors, and beta-adrenergic antagonists.^15–25 Unfortunately, the extent and duration of drug efficacy tends to be limited, resulting in disease progression despite therapy.^1,2 The development of new IOP-lowering glaucoma drugs is slow, and they are generally optimized for human rather than veterinary application.^26–28

A major limitation of medical glaucoma therapy, which has been investigated more thoroughly in human than veterinary patients, is poor adherence. Glaucomatous optic neuropathy may progress because eye drops are not administered as frequently as recommended.^29–31 Poor adherence, especially in elderly patients, can be based on several factors, including forgetfulness or technical difficulties administering eye drops. As an alternative, a number of pharmaceutical companies are working on biodegradable intracameral drug implants that continuously release prostaglandin-analogues inside the eye.³² In March 2020, the U.S. Food and Drug Administration (FDA) approved the bimatoprost implant Durysta™ by Allergan plc to lower IOP in human open-angle glaucoma (OAG) and ocular hypertension patients. Other implants for sustained intraocular drug release are in preclinical and clinical trials.^32,33

Surgical methods to reduce aqueous humor production consists of transscleral or endoscopic laser cyclophotocoagulation, and aqueous humor outflow can be increased via filtration surgery with or without the use of gonioimplants.^1,2,34,35 While technical advances have been made, degree and duration of efficacy still need to be improved considering significant failure rates in some forms of glaucoma as well as animal species and breeds.^2,35

Unless the underlying disease mechanisms can be identified and eliminated, such as in some forms of secondary glaucoma, a cure is not available even with apparently well-controlled IOP. There is a great need for more effective therapies that target specific disease mechanisms and prevent the progressive loss of RGCs and their axons. Gene and cell therapies have the potential to provide more effective and lasting treatments or even cures by addressing specific disease mechanisms. While not yet approved for clinical application, these treatments showed great potential in laboratory settings and several glaucoma animal models and will be reviewed here.

2. UNDERSTANDING DISEASE MECHANISMS AND EARLY DIAGNOSIS

Gene and cell therapies are well suited to be tailored for specific genetic defects or disease pathways and have the potential to be adjustable for individual patients; a strategy defined as personalized or precision medicine. Initial progress in the investigation of canine glaucoma genetics included the identification of mutations in the ADAMTS10, ADAMTS17, and OLFML3 genes causing primary glaucoma in a number of breeds.^36–43 There is also ongoing progress to predict more accurately glaucoma progression based on ICA morphology and aqueous humor outflow pathways; these efforts are facilitated by improvements in high-resolution imaging techniques.^44–49

One of the major challenges in the successful translation of novel therapies from the laboratory bench into the clinic is the limited ability to diagnose glaucoma early so that treatment can be initiated more effectively before too much damage has occurred. Many of the novel gene and cell therapies discussed here were tested under experimental conditions with treatment beginning simultaneously or even before the onset of optic nerve damage.^50–58 This experimental setup does not represent clinical reality. In human patients with primary open-angle glaucoma (POAG), visual field loss is not diagnosed until 50% of RGC axons have already been lost.^59,60 A recent study of canine eyes at risk of developing primary angle-closure glaucoma (PACG) showed significant loss of total retinal and nerve fiber layer thickness measured by high-resolution imaging with optical coherence tomography (OCT) before the occurrence of IOP increase and other clinical signs of glaucoma.⁶¹ Moving forward, our ability to diagnose glaucoma early will be facilitated by a better understanding of disease mechanisms and improved diagnostic technologies, most importantly high-resolution imaging of the anterior and posterior ocular segments, continuous tonometry, and genetic testing.²

3. BASIC PRINCIPLES OF GENE AND CELL THERAPIES

3.1. Gene therapy

Ocular gene therapy is defined as the introduction of genetic material, DNA or RNA, into cells of the eye to modify gene expression and achieve a therapeutic effect. A common, effective method to administer the genetic material into the eye is using a viral vector, such as adeno-associated virus (AAV) or lentivirus.^62,63 Generally, the gene therapy vector is administered in close proximity to the cells to be treated – the specific injection techniques, most importantly intracameral and intravitreal, will be discussed throughout this article. In addition to the injection technique, the ability to transduce and target transgene expression to a certain cell type is determined by the cell tropism of the viral vector, which is influenced by the viral capsid and serotype, and the promoter regulatory sequence that is placed in front of the transgene.^62,63 A genetic switch, for example glucocorticoid response elements (GRE), can be included so that the transgene can be turned on and off by initiating or discontinuing the use of particular drugs.⁶⁴

The advantages of gene therapy include the ability to specifically target known disease mechanisms and cells. Furthermore, if target cells are post-mitotic, i.e. not replaced, a one-time treatment will result potentially in a long-term therapeutic effect and possibly even a cure. Gene therapy cannot replace cells that have already been lost by disease. Overall, the eye is well suited for gene and cell therapies because it forms a separate compartment that is immune privileged.⁶⁵

Most ocular gene therapies have been developed for the treatment of inherited diseases. Gene therapy can be used to address genetic risk factors by replacing, silencing, or editing defective genes that contribute to the development and progression of disease.^62,66 Most ocular gene therapies have been developed for autosomal recessive traits by replacing the mutated, non-functional gene with a normal or wild type copy.^67,68 The treatment of autosomal dominant traits is more challenging because the mutated gene producing a potentially toxic protein may have to be silenced or knocked down first and then replaced with a normal or wild type copy of the gene.⁶⁹ The treatment of non-Mendelian, complex traits or acquired, non-genetic disorders may be more challenging, but possible if specific molecular disease pathways are identified that can be targeted. Gene therapy can also be applied to enable cells within the eye to produce proteins with a therapeutic effect.⁶²

Several ocular gene therapies have already moved into clinical trials for human patients, mostly for the treatment of blinding retinal diseases.⁶⁸ Generally, the eye injections to deliver gene therapies are well established and tolerated. In December 2017, the U.S. FDA approved the first viral ocular gene therapy, Luxturna™ by Spark Therapeutics, Inc., for the treatment of one form of retinal childhood blindness.

3.2. Cell therapy

Cell therapies are defined as the administration of cells with the purpose of achieving a therapeutic effect. Two main strategies are being investigated: (1) The use of stem cells to replace cells that have been lost in a degenerative disease process – this is referred to as regenerative medicine; and (2) the use of genetically modified cells or stem cells as a mean to continuously synthesize and release proteins with a therapeutic effect on adjacent cells and tissues.⁷⁰

Stem cells are undifferentiated or partially differentiated cells with the ability to proliferate and differentiate into many different types of specialized cells. The ability to induce pluripotent stem cells (iPSCs) from differentiated adult cells, such as skin or blood cells, has boosted stem cell research for clinical application since it replaced the need of using human embryologic stem cells (hESCs), which have been associated with some ethical concerns.⁷⁰

Despite much progress in laboratory settings and some ongoing human clinical trials, there is currently no approved ocular stem cell therapy for clinical application in humans or animals. Nevertheless, autologous adult stem cells, for instance hematopoietic stem cells harvested from bone marrow and mesenchymal stem cells harvested from fat tissue, are being used commercially for clinical application in humans and animals. Intraocular administration of these cells is not approved by the U.S. FDA and have not been scientifically proven to provide a safe, therapeutic effect (https://www.fda.gov/consumers/consumer-updates/fda-warns-about-stem-cell-therapies). Recent human case reports have demonstrated the potential adverse effects of such non-approved intraocular stem cell therapies, including intraocular hemorrhage, retinal detachment, and vision loss.^71–73

3.3. Potential applications of gene and cell therapies for glaucoma

Both gene and cell therapies are rapidly evolving in laboratory settings and offer great potential advantages over many conventional drugs, including longer duration with continuous drug release and a therapeutic effect that can be tailored to specific molecular disease pathways within an individual human or animal patient. The following potential applications could be considered for improved glaucoma therapy and will be discussed here (Figure 1): improved IOP control by increased aqueous humor drainage (1), decreased aqueous humor production (2), and/or prevention of gonioimplant bleb fibrosis (3), neuroprotection and neuroregeneration of RGCs and their axons (4), modification of the ocular biomechanical properties for improved IOP tolerance (5), and inhibition of inflammation and neovascularization to prevent the formation of PIFVM and secondary, neovascular glaucoma (6).

Figure 1.

Schematic cross-section of an eye showing potential targets for gene and cell glaucoma therapies: (1) Increased aqueous humor outflow, (2) decreased aqueous humor production, (3) preventing gonioimplant bleb fibrosis, (4) RGC neuroprotection and neuroregeneration, (5) modification of biomechanical properties of sclera, lamina cribrosa, and cornea, and (6) inhibition of inflammation and PIFVM formation.

4. INTRAOCULAR PRESSURE (IOP) CONTROL

4.1. Increasing aqueous humor outflow by treatment of outflow pathways

Lowering IOP is the only approved treatment strategy and will remain the main focus of glaucoma therapy in the foreseeable future. Current treatments consist of medical and surgical techniques to lower production and increase drainage of aqueous humor; however, there are limitations in the extent and duration of IOP-lowering treatments.² As we gain a better understanding of the physiology and pathophysiology of aqueous humor outflow resistance in different species, breeds, and forms of glaucoma, including genetic risk factors, more effective, lasting therapies can be developed. Gene and cell therapies have great potential to be tailored to these specific molecular disease pathways.

Major advances have been made in the robust and safe targeting of transgene expression to the trabecular meshwork and the aqueous humor outflow pathways within the ICA of different animal species and human cell and organ cultures.⁶² While transgene expression can be targeted to the trabecular meshwork with naked nucleic acid,⁷⁴ the use of viral vectors generally provides more robust transduction. Both intracameral and intravitreal administration of adenovirus, lentivirus, adeno-associated virus (AAV), and herpes simplex virus (HSV) gene therapy vectors have been performed successfully to target aqueous humor outflow pathways, with the intracameral route being preferred in larger animal species (Figure 2).^62,75–80 Traditionally, intracamerally injected adenovirus was most effective as shown in several species, including mouse, rat, rabbit, dog, cynomolgus monkey, and human cell and organ cultures, and it appears to be the preferred viral vector to target the murine trabecular meshwork via intracameral or intravitreal injection.^81–86 Because adenovirus is more immunogenic and tends to cause more uveitis compared to AAV and lentivirus, it is rarely used in larger animal species.^62,82 Lentivirus vectors provide robust targeting of the aqueous humor outflow pathways in cats, cynomolgus monkeys, and cultured human donor eyes, as shown most commonly with modified feline and human immunodeficiency virus.^87–89

Figure 2.

Schematic cross-section of an eye showing possible methods for intraocular administration of gene and cell therapies for glaucoma. (1) Gene therapy vectors or cells for the treatment of the aqueous humor outflow pathways within the ICA are administered by intracameral injection. (2) Injections through the pars plana of the ciliary body into the anterior vitreous are used to target the ciliary body epithelium; reagents will also reach the anterior chamber and aqueous humor outflow pathways. (3) Gene therapy vectors or cells for neuroprotective therapy are injected through the pars plana into the posterior vitreous, close to the surface of the retina and ONH. (4) Capsules with genetically modified cells, that release neuroprotective reagents, are attached to the sclera at the pars plana and reach into the anterior vitreous. (5) Transplantation of RGCs to restore vision are injected into the posterior vitreous or (not shown) underneath the retina’s inner limiting membrane. If successful, the cells embed in the retina and form axons that extend through the optic nerve and connect with targets in the brain.

While AAV tends to be the preferred vector for ocular gene therapy with great safety and efficacy in preclinical and clinical trials, including FDA approved Luxturna™, the first commercially available viral ocular gene therapy, it has not shown great results in initial attempts to target the trabecular meshwork.^62,90,91 The use of the self-complementary AAV vectors and the recent development of novel capsid mutated AAV particles resulted in an expanded toolkit for targeting of the trabecular meshwork and other tissues within the anterior segment of the eye (Figure 3).^{62,77,78,80,92–94,95} The self-complementary AAV genome facilitates the generation of double-stranded DNA for more efficient and long-term transduction of trabecular meshwork cells in several species, with long-lasting effects for at least 2 years in monkeys.^62,77,91 Capsid mutant AAV2 (AAV of serotype 2) showed the highest transduction of the trabecular meshwork, followed by AAV5, AAV6 and AAV8 serotypes as shown in the mouse, rat, and cultured human trabecular meshwork cells and perfused anterior chambers.^78,80,93 AAV capsid mutations even resulted in improved trabecular meshwork targeting without the use of self-complementary genome, thereby permitting the insertion of bigger transgenes.^80,93

Figure 3.

Targeting of GFP transgene expression to the canine anterior ocular segment by intravitreal injection of capsid mutated AAV2 (triple Y-F+T-V). (A) Fluorescent gonioscopic imaging showing fluorescence within the ciliary cleft 6 weeks post intravitreal AAV injection. Note the pectinate ligament fibrils demonstrated by the dark bands crossing the region of fluorescence. (B) Immunohistochemical labeling of cryosections of the canine anterior segment shows GFP within the trabecular meshwork (arrow). (C) GFP expression is also seen within the ciliary body epithelium (arrow). GFP, green fluorescent protein; triple Y-F + T-V, AAV2 based capsid was mutated by substitution of three surface-exposed capsid tyrosine (Y) residues with phenylalanine (F) and one threonine (T) residue with valine (V). Scale bar, 100 μm. Boyd et al. 2016, with permission.⁹⁴

Numerous studies in different animal species have provided proof of concept that gene therapy-mediated modification of gene expression within the trabecular meshwork can result in an IOP change. Two types of studies were performed in animal models and perfused human anterior segment cultures: (1) development of potential treatments by lowering IOP, ^{62,64,96–105} and (2) development of glaucoma animal models by increasing IOP.^{62,97,106–113} Sustained IOP reduction, for example, was achieved for at least 5 months in normal cats and monkeys by lentivirus-mediated gene therapy of the trabecular meshwork with either prostaglandin biosynthesis enzyme cyclooxygenase-2 (COX-2) and prostaglandin F2alpha receptor (FPR), or prostaglandin F synthase (PGFS).^104,105

An alternative approach to trabecular meshwork-targeted gene therapy is cell therapy. Trabecular meshwork-like cells can be created from iPSC and injected into the anterior chamber in order to regenerate trabecular meshwork cells that are lost with chronic glaucoma (Figure 2). Proof of concept for this treatment method was provided in transgenic, myocilin (MYOC)-mutant mice with POAG, where the cell therapy resulted in improved outflow facility, IOP control, and halted RGC loss, even in animals with advanced stages of glaucoma.^114,115 Interestingly, the main therapeutic effect of the transplantation occurred via induction of proliferation of the remaining endogenous trabecular meshwork cells. These results were verified in human eye perfusion culture.¹¹⁶ Potential future clinical application may involve the collection of skin cells from glaucoma patients, followed by creation of iPSCs and induction of trabecular meshwork cells. These cells can then be administered intracamerally back into the same patients to achieve a therapeutic effect.

4.2. Decreasing aqueous humor production by treating the ciliary body epithelium

Even though the elevation of IOP in glaucoma patients occurs because of increased aqueous humor outflow resistance, glaucoma treatment strategies include the reduction of aqueous humor production by medical (e.g., carbonic anhydrase inhibitors) and surgical (i.e., cyclophotocoagulation) therapies.^1,2,34 While most glaucoma gene therapy efforts aimed at lowering IOP by treatment of the aqueous humor outflow pathways, there were a few attempts to lower aqueous humor production by the ciliary body epithelium. The advantage of any such gene or cell therapy would be the replacement of daily eye drops for a more effective, sustained control of IOP following a one-time vector injection. Since the underlying disease process is not being addressed with this treatment approach, it is possible that disease progression may only be slowed but not prevented. The ciliary body epithelium can be reached by gene therapy vectors via intravitreal injection (Figure 2 and Figure 3).^94,117 Recently, a team of investigators achieved a reduction of aqueous humor production in experimental mouse models of corticosteroid and microbead-induced ocular hypertension following an intravitreal AAV administration that resulted in disruption of the aquaporin 1 (AQP1) gene within the ciliary epithelium by use of CRISPR-Cas9 gene editing.¹¹⁷

4.3. Increasing aqueous humor outflow by inhibiting fibrosis of conjunctival filtration bleb

Glaucoma filtration surgery is commonly used in humans and dogs when IOP cannot be controlled by maximized medical therapy.^1,2,35 Aqueous humor drainage is increased typically by diversion under the conjunctiva and Tenon’s capsule and the formation of a filtration bleb.¹¹⁸ The most common techniques are trabeculectomy in humans and gonioimplants, for example Ahmed valve or Baerveldt glaucoma drainage device, in humans and dogs.^2,35,119,120 Filtration surgery failures are not uncommon and usually due to excessive wound healing in the subconjunctival space resulting in fibrosis of the filtration bleb, inhibition of aqueous humor drainage, and IOP increase.^{2,35,118,121,122} Susceptibility to bleb failure appears to vary between species, with dogs showing more fibrosis than humans, and even between individual canine breeds indicating potential genetic risk factors that could be addressed by gene therapies.² Antimitotic drugs, most commonly mitomycin C (MMC), are routinely used to soak the conjunctival pocket during surgery and minimize bleb fibrosis.^{121,123–128} Duration of antimitotic drug application and subsequent copious irrigation are critical to minimize adverse effects.¹¹⁸ There is a delicate balance between bleb failure due to fibrosis and toxic effects of antimetabolite overuse, most often conjunctival dehiscence with bleb leakage as well as scleral and conjunctival necrosis.^{118,123,126,129–135}

The search for new, improved and safer antifibrotic treatment strategies (‘scar wars’) has been ongoing for many years and continues to be a high priority.^2,118 Numerous studies in several animal species, including rabbits, rats, dogs, and non-human primates, were conducted to determine the molecular and cellular disease mechanisms of conjunctival bleb fibrosis so that they can be targeted more specifically and effectively.^{118,136–142}

Gene therapy has a great potential to specifically target and modify these pathways for safe and long-term prevention of bleb fibrosis. For example, small interfering RNA (siRNA) was developed to silence transcription factors involved in conjunctival tissue fibrosis: the myocardin-related transcription factor/serum response factor (MRTF/SRF) pathway or secreted protein acidic and rich in cysteine (SPARC).^143–145 Intravitreal antisense oligonucleotides against transforming growth factor-beta2 (TGF-beta2) showed promising results in human subjects with OAG undergoing trabeculectomy.¹⁴⁶ Gene therapy was also successfully performed in rabbits and ocular hypertensive monkeys by recombinant adenovirus with the human p21 transgene (encoding the CDKN1A protein), either by presurgical subconjunctival injection or by topical administration onto the surgical field, in order to modulate wound healing after glaucoma trabeculectomy surgery.¹⁴⁷ The intention was to cause cell cycle arrest of surrounding cells rather than their destruction as with MMC. Another adenovirus-mediated gene therapy method tested in mice consisted of blocking p38 mitogen-activated protein kinase (MAPK) on post-injury conjunctival scarring; it resulted in inhibition of the fibrogenic reaction induced by the subconjunctival fibroblasts.¹⁴⁸

5. NEUROPROTECTION AND NEUROREGENERATION

5.1. Neuroprotection

In many human and animal glaucoma patients the optic nerve continues to degenerate despite apparently effective IOP control within its physiologic range.^1,2 The reasons and mechanisms for the continuing RGC death by apoptosis are not fully understood and may not all be associated with IOP-related biomechanical axon damage at the lamina cribrosa within the ONH.^149–152 Some mechanisms that have been identified and studied include excitotoxicity caused by excessive excitatory amino acid release, most importantly glutamate and aspartate,^153,154 neurotrophin deprivation from blockage of retrograde axonal transport,^155–158 excessive intracellular calcium,¹⁵⁹ compromised blood flow to the ONH and retina,^160–166 oxidative stress,^167–171 inflammation and autoimmunity against retinal and optic nerve antigens,^172–174 and reactive gliosis.^175–178 There is a general understanding among glaucoma clinicians and vision scientists that a more effective glaucoma therapy must include neuroprotective strategies to protect the RGCs and their axons.^1,2,179

Several compounds were identified with neuroprotective properties in laboratory settings, mostly in rodent models of optic nerve injury, acute ocular hypertension, and experimental hypertensive glaucoma.^180,181 So far, translation of these treatments into the clinical application with proven efficacy failed. Most prominently, the N-methyl-D-aspartate (NMDA) receptor antagonist memantine, which counters the toxic effect of excessive glutamate in the extracellular space, reduced RGC death and functional loss in experimental glaucoma in rats and primates.^182–184 Unfortunately, two phase 3 clinical trials failed to demonstrate the protection of visual function by memantine in human glaucoma patients.¹⁸⁵

Gene therapy is well suited for the development of neuroprotective treatments of RGCs because (1) many transgenes can be easily incorporated into viral vectors for specific modification of molecular disease pathways, and (2) the RGCs located in the inner retina can be directly targeted via intravitreal vector administration (Figure 2). The latter has been demonstrated mostly with AAV in several animal species, including mice, rats, dogs, and monkeys with differences in transduction efficiency partially due to species-specific variations in retinal inner limiting membrane thickness (Figure 4).^94,186–192 Techniques for sub-inner limiting membrane injection of AAV have been described to improve transgene delivery to the retina.^193,194 While other intraocular injection methods could also be considered, for example subretinal and suprachoroidal techniques, the intravitreal administration of viral vector appears to be the most effective to target RGCs.^191,195

Figure 4.

Transduction of canine RGC following intravitreal injection of capsid mutated AAV2 (triple Y-F + T-V) with GFP transgene. (A) Representative confocal scanning laser ophthalmoscopy image obtained at 5 weeks post injection demonstrated widespread GFP fluorescence. (B) Immunohistochemical labeling of retinal cryosection with neuronal nuclei (NeuN) antibody (red) to label RGCs demonstrates a high number of cells colabeling with GFP (green). Cell nuclei are shown in blue with DAPI.

DAPI, 4’,6-diamidino-2-phenylindole; GCL, ganglion cell layer; GFP, green fluorescent protein; INL, inner nuclear layer; ONL, outer nuclear layer; triple Y-F+T-V, AAV2 based capsid was mutated by substitution of three surface-exposed capsid tyrosine (Y) residues with phenylalanine (F) and one threonine (T) residue with valine (V). Scale bar, 50 μm. Boyd et al. 2016, with permission.⁹⁴

Tropism for RGCs varies between different viral vectors with AAV2 being superior compared to adenovirus.⁶² The specific targeting of RGCs by gene therapy vectors may not be critical for all neuroprotective strategies; some transgenes encode for secreted proteins that have a regional protective effect following transduction of Müller glia.^196,197 An alternative, less practical approach to target RGCs is by administering the gene therapy vector to specific areas of the brain with the transgene being transported to the RGCs via retrograde axonal transport.¹⁹⁸

Several neuroprotective gene therapy strategies to support RGC survival were successfully tested in rodent models with spontaneous glaucoma (DBA/2J mice), experimental hypertensive glaucoma, acute IOP elevation, or optic nerve injury (optic nerve crush and transection); this was achieved mostly by intravitreal administration of AAV.⁶²

Neurotrophins have been investigated extensively as a possible neuroprotective treatment to promote RGC survival in glaucoma and traumatic optic neuropathies.^62,199 Neurotrophins are essential proteins for differentiation and maintenance of RGCs, but some neurotrophins and their receptors are downregulated in glaucoma and traumatic optic neuropathies.^200,201 Since neurotrophins are proteins, the effect of their direct intravitreal injection is short-lived with a need for sustained production and release that could be achieved by gene or cell therapies.²⁰² Protection of RGCs was demonstrated by viral vector-mediated overexpression of brain-derived neurotrophic factor (BDNF) and its TrkB receptor, ciliary neurotrophic factor (CNTF), glial cell-line derived neurotrophic factor (GDNF), and their downstream pathways.^{50–52,57,196,197,201,203–205}

In addition to neurotrophins, other successfully tested neuroprotective gene therapy strategies for RGCs included increased expression of antiapoptotic,^53,206–208 antioxidant,^54–56, and anti-inflammatory genes,^58,209,210 as well as inhibition of proapoptotic genes.²¹¹ Intramuscular injection of AAV5 with the mutant erythropoietin (EPO) transgene EpoR76E provided RGC neuroprotection in glaucomatous DBA/2J mice and mice with experimental hypertensive glaucoma via the systemic release of EPO from the skeletal muscle.^212–214

A less developed strategy for delivery of non-selective neuroprotective and/or immune-modulatory factors to the RGCs is the intravitreal administration of multipotent stem cells, ocular progenitor cells, or other, genetically modified cells; this technique is time-limited by the viability and secretory capacity of the transplanted cells (Figure 2).⁷⁰ The cells are injected as a single cell suspension but subsequently aggregate in vivo to form small clusters that are free-floating in the vitreous and provide neurotrophic support to the retina without the need for integration into the host tissue.⁷⁰

A well-tested alternative to the intravitreal injection of cells for neuroprotection is the use of encapsulated cell technology (Figure 2). This approach was previously tested in animals and human patients with inherited retinal dystrophies, including Irish Setter dogs with rod-cone dystrophy 1 (rcd1).^215,216 The NT-501 capsule was developed by Neurotech Pharmaceuticals, Inc., and contains genetically modified human cells that synthesize and continuously release the neurotrophic factor CNTF through a semipermeable membrane into the vitreous. The implant is anchored to the sclera at the pars plana of the ciliary body. The continuous release of CNTF by intravitreal encapsulated cell therapy was tested recently in a phase 1 clinical trial on human patients with POAG (ClinicalTrials.gov NCT01408472) and ischemic optic neuropathy (NCT01411657), but results have not yet been published. Two phase 2 clinical trials for the treatment of glaucoma are currently ongoing (NCT02862938 and NCT04577300).

5.2. Neuroregeneration

Like other neurons in the fully developed mammalian central nervous system (CNS), RGCs and their axons within the optic nerve are unable to regenerate after injury, such as glaucoma.²¹⁷ This is different from peripheral neurons and CNS neurons of other animal species, for instance fish and amphibians, which have the ability to regenerate throughout their lifespan.²¹⁸ The failure of optic nerve axons to regenerate is based on a combination of environmental factors within the optic nerve, including inhibition by CNS-specific myelin proteins, as well as reactive scarring and inflammation, and other cellular and molecular responses mediated by glial cells.^218,219 If a RGC is still alive, gene therapy strategies could promote regrowth of its axon within the optic nerve, but a major challenge is the ability of these axons to reach and properly connect with targets in the brain in order to restore functional visual capacity.^51,218,219 While the clinical application is not available yet, major advances have been made in understanding the mechanisms regulating RGC survival and optic nerve axon regeneration following injury.^217,219 Supportive gene therapies and other strategies are being investigated. Promising work was done in adult mice with optic nerve crush injury in which a combination of gene therapy-induced modification of gene expression resulted in RGC axon regeneration and partial restoration of visual function.²²⁰

The replacement of dead and lost RGCs is associated with a number of additional major challenges: lost cells have to be replaced, followed by extension and growth of dendrites and axons, and functional connections with the proper targets in the brain.²²¹ Three main strategies are being pursued for RGC replacement therapy: (1) syngeneic transplantation of adult iPSCs programmed to assume the RGC phenotype, (2) allogeneic transplantation of RGCs from healthy donor eyes, and (3) reprogramming endogenous Müller glial cells into RGCs by gene therapy approach.²¹⁸ The transplantation of cells could potentially occur via sub-inner limiting membrane or intravitreal injection, as long as the cells are able to penetrate the retinal inner limiting membrane and become integrated into the retina (Figure 2).^192–194 Regeneration of the optic nerve is a high priority of several funding agencies, including the U.S. National Eye Institute (NEI)/National Institutes of Health (NIH) and U.S. Department of Defense.²²² Improvements in optic nerve regeneration may even permit whole eye transplantations in the future.²¹⁹

While RGC replacement therapies have not yet reached human patients, clinical trials were conducted or are currently underway using hESCs and iPSCs for dry age-related macular degeneration (AMD) and retinitis pigmentosa (NCT04339764).²²³

6. INFLUENCING THE BIOMECHANICAL PROPERTIES OF THE EYE

There are strong indications that an individual’s susceptibility to IOP increase and variation is determined by the biomechanical properties of the eye, most importantly its fibrous layer consisting of cornea, sclera, and lamina cribrosa.^224,225 In some eyes, the RGC axons are not sufficiently supported by the lamina cribrosa and the peripapillary sclera, so that even physiologic IOP can be harmful.^1,2 This may be the case in human patients with normotensive OAG and progressive optic nerve damage despite consistent IOP measurements below 21 mmHg.^226,227 Disease progression can be slowed or prevented in these patients by a 30% reduction in IOP.²²⁸ Veterinary ophthalmologists also observe individual variations in IOP susceptibility, which tends to be species- and breed-related.² While proof still needs to be provided in dogs that tissue biomechanical properties are linked to IOP susceptibility, initial studies in ADAMTS10-mutant Beagles with OAG showed that their posterior sclera is weaker with reduced fibrous collagen density, possibly explaining the slower optic nerve degeneration with chronic, severe IOP increases compared to dogs of other breeds.^45,229–231 Similar findings can be expected in other ADAMTS10- and ADAMTS17-mutant dogs with OAG, for example Norwegian Elkhound (ADAMTS10),³⁶ Petit Basset Griffon Vendéen (ADAMTS17),^38,232 Basset Fauve de Bretagne (ADAMTS17),⁴⁰ and Chinese Shar-Pei (ADAMTS17).^41,232 Some transgenic mice are also resistant to glaucomatous damage, while hardening of the sclera with cross-linking agents worsens IOP-related damage to the RGC axons.²²⁴

Once we gain a better understanding of which biomechanical properties of the fibrous layer are advantageous, therapeutic tools can be developed to modify these properties to achieve a protective effect from IOP increases. Technologies for specific targeting of transgene expression to the sclera and lamina cribrosa still need to be developed. The peripapillary sclera could potentially be reached by gene therapy vector via suprachoroidal or retrobulbar injections.^{195,233–236} Targeting the lamina cribrosa by gene therapy without damaging the RGC axons will be a more challenging undertaking.

Corneal biomechanical properties not only affect tonometry measurements but also appear to influence susceptibility to glaucomatous damage.^225,237 Corneal hysteresis, a clinical parameter of the cornea’s viscoelastic properties, has emerged as an important risk factor for glaucoma progression in humans.²³⁸ In contrast to sclera and lamina cribrosa, where gene therapy has not yet been performed, corneal AAV-mediated targeting of transgene expression and gene therapy was recently performed successfully in several animal species and disease models, such as dogs with Mucopolysaccharidoses I (MPS I).²³⁹ Transgene expression can be targeted to the cornea by intrastromal, intracameral, or subconjunctival injection, as well as topical eye drop application following removal of the corneal epithelium.²⁴⁰ Based on the available technologies, corneal gene therapy could be considered an additional possible tool to decrease an individual’s susceptibility to glaucomatous optic nerve damage.²⁴¹

7. INHIBITION OF ANTERIOR UVEAL NEOVASCULARIZATION AND INFLAMMATION

While some forms of secondary glaucoma can be successfully treated by identifying and addressing the underlying cause, other forms continue to progress despite symptomatic therapy with IOP-lowering and anti-inflammatory medications. A major concern in veterinary ophthalmology is the development of secondary glaucoma in dogs following phacoemulsification surgery.² While the underlying molecular disease mechanisms are still largely unknown, and may include genetic predisposition of certain breeds, the IOP increase in many affected dogs is based on the formation of PIFVM that covers the aqueous humor drainage pathways, resulting in IOP increase.¹⁰ An upregulation of vascular endothelial growth factor (VEGF) expression associated with canine lens-induced uveitis likely contributes to PIFVM formation post cataract surgery.²⁴² Anti-neovascular gene therapies of the anterior chamber may be an effective treatment. Several such gene therapy strategies have already been developed for the posterior ocular segment in human patients with neovascular disease, most importantly neovascular AMD and diabetic retinopathy, and could be considered for use in dogs.^243,244 These include the gene transfer of endostatin/angiostatin by lentivirus (RetinoStat® by Oxford BioMedica plc; clinicaltrials.gov: NCT01301443) or anti-VEGF antibodies by AAV8 (RGX-314 by RegenxBio®; clinicaltrials.gov: NCT03999801, NCT04567550, NCT04514653, and NCT03066258).²⁴³

More specific and effective gene therapies for anterior uveitis of unknown origin, including non-infectious, autoimmune uveitis, would also help to decrease the incidence of secondary glaucoma. A recently published study showed that intravitreal administration of AAV encoding human leukocyte antigen G1/5 (HLA-G1/5) reduced the severity of experimental autoimmune uveitis in rats.²⁴⁵

7. CONCLUSIONS

Much progress has been made in recent years in the development and clinical testing of ocular gene and cell therapies. Promising novel methods have been developed in the laboratory for the comprehensive treatment of glaucoma by reducing IOP and providing neuroprotection of the retina and optic nerve. As we gain a better understanding of glaucoma disease mechanisms, we will be able to target them specifically using gene and cell therapies in order to achieve safe, effective, and long-term prevention of vision loss. Early diagnosis of glaucoma will be critical for better treatment outcomes; this will be facilitated by improvements in diagnostic tools, for example high-resolution imaging and continuous tonometry. A major obstacle is the translation of new therapies from the laboratory into the clinical setting, especially for veterinary ophthalmic application, due to limited resources for developing animal-specific treatments.