Congenital Blindness and Eye Diseases
Inherited eye diseases are attractive targets for gene and cell therapy for three reasons: (1) feasibility (2) accessibility and (3) opportunity. Feasibility: Inherited eyes retinal degenerations have been studied intensively, and the mutations leading to death of photoreceptor cells have been described in over 200 genes (https://sph.uth.edu/retnet/home.htm). Accessibility: Unlike other components of the central nervous system, the retina is accessible from outside the body, and injections may be viewed through the lens. The clarity of the cornea and lens also permits physicians to measure the thickness of retinal layers and the integrity of photoreceptors in the living eye. Such high-resolution methods may become approved as outcome measures in clinical trials. Opportunity: Prevalent age-related diseases beset vision: diabetic retinopathy, glaucoma and age related macular degeneration. These diseases require long-term treatment that may be approachable by gene therapy or cell therapy. Successful gene therapy of a well-defined inherited retinal degeneration would provide the proof-of-concept required to develop a similar treatment for more complex but more frequent target disease.
Monogenic gene therapies in clinical trials:
Gene therapy trials for the treatment of one blinding disease of childhood, Leber Congenital Amaurosis type 2 (LCA2) has attracted the most attention in this field. At least eight phase I or phaseI/II clinical trials are at various stages of completion for this disease. They all use adeno-associated virus serotype 2 (AAV2) to deliver the gene that is defective. RPE65 is expressed in retinal pigment epithelial (RPE) cells, and is required to isomerize a vitamin A derivative to its active form as a component of opsins, the light harvesting components of rod and cone photoreceptor cells. Children with LCA2 have limited light sensitivity, but, importantly, they retain all of the cellular components of the retina, so that delivery of the RPE65 gene to the correct cells was expected to restore vision.
The human clinical trials for LCA2 have proven encouraging. Since 2008, more than 40 patients have been treated with AAV expressing RPE65, and there have been no vector-associated adverse events reported in any of the trials. In the four studies which reported results, improvement in retinal function or visual function was reported in all patients undergoing treatment. In one study, there were increases in visual sensitivity increases of 10- to 10,000 fold above baseline. The areas of visual field improvement corresponded well with the location of the subretinal injection of vector. In all initial studies, patients were treated in only one eye, but a more recent study has tested treatment of the second eye, months or years after subretinal injection of the first eye. Three patients showed only modest inflammation and improved vision after re-administration of the same vector to the second eye.
The LCA2 clinical trials provided causes for caution however. First it became apparent that subretinal injection may present unacceptable risks for the foveal region of a diseased retina. This is important since the fovea is responsible for sharp central vision. Second, despite functional improvement in patients, the rate of retinal degeneration was the same in treated and untreated areas of the LCA2 retinas. While the biological explanation and clinical significance of this result are unclear, it is apparent that quantitative assessment of visual function and retinal structure are required are necessary outcome measures of clinical trials.
Choroideremia is an X-linked disease responsible for degeneration of the retina and choroid, which supplies blood to the outer retina, in approximately 1:50,000 male children. The disease is associated with the deletion of the gene encoding Rab escort protein 1 (REP-1), which is required for post-translational modification and membrane trafficking of Rab proteins.
Unlike the LCA2 trials, patients in this retinal gene therapy trial had good or relatively good central vision at the time of treatment, so that this study is designed to preserve vision rather than to restore it. Also in contrast to LCA, choroideremia affects the photoreceptors, the RPE and the choroid, suggesting that multiple cell layers must be effectively treated.
MERTK associated retinitis pigmentosa: MERTK (mer receptor tyrosine kinase) is required for phagocytosis of photoreceptor outer segments by the RPE, and when absent leads to autosomal recessive retinitis pigmentosa. The disease was first described in the Royal College of Surgeons rat, but occurs in isolated populations in the Middle East and the Faroe Islands. In preclinical studies, a lentivirus expressing MERTK and was able to preserve some retinal function out to 7 months after subretinal injection. More recently, MERTK expressed from an AAV8 vector with a single tyrosine to phenylalanine capsid mutation was shown to preserve retinal structure and function in the Royal College of Surgeons for at least one year after treatment. A phase 1 clinical trial using AAV2 expressing MERTK from an RPE specific promoter is being conducted at King Faisal Specialist Hospital & Research Center in Saudi Arabia. Several patients have been treated by subretinal injection, and no adverse events have been reported.
Leber Hereditary Optic Neuropathy (LHON), the most prevalent of the mitochondrial inherited diseases, leads to degeneration of the optic nerve, beginning usually in the second or third decade of life. LHON is associated with mutations in three different subunits (ND1, ND4 and ND6) in Complex I of the mitochondrial respiratory chain. Since viral vectors normally deliver DNA to the nucleus, the approach to gene therapy for this disease is described as allotropic: mitochondrial reading frames are converted to the nuclear genetic code and an amino-terminal mitochondrial targeting sequence is added. The gene is delivered to the nucleus, but the protein is transported into mitochondrial following its synthesis in the cytoplasm. Gene therapy has been tested in cells containing mutant mitochondria derived from patients, and in a mouse model created by allotopic expression of a mutant ND4 gene.
Two groups, one in Paris and one in Miami are recruiting patients with for gene therapy trials for the most prevalent form of LHON using AAV2 to treat patients with the most common ND4 mutation. While there are some differences in the vectors, both use allotopic delivery of the wild-type gene. Two features of LHON make it at once an attractive and a difficult target for gene therapy. First, there are hundreds of copies of mitochondrial DNA per retinal ganglion cell, but optic neuropathy does not occur in carriers with less than 95% mutant mitochondrial DNA, suggesting that delivery of even a modest amount of normal ND4 protein should be therapeutic. Nevertheless, even in people with uniformly mutated mitochondrial DNA, the disease may not ever manifest itself. This is frequently true of women. For this reason, selecting patients to treat is problematic. Therefore, subjects in the Miami trial will not only have a genetic diagnosis of LHON caused by G11778A, but also have either chronic disease or an acute episode of optic neuropathy affecting one eye.
Autosomal Recessive Stargardt macular dystrophy: Recessive Stargardt disease causes a progressive loss of central vision, associated with the accumulation of lipid-rich deposits in the macula. It is the most prevalent, monogenic retinal degeneration in the United States and Europe. The disease is caused by a mutation in a single gene ABCA4, the product of which helps transport retinyl esters out of the lumen of photoreceptor disc membranes. The size of the ABCA4 cDNA (6.8 kb) is in excess of the packaging limit for AAV, but a lentivirus vector based on equine infectious anemia virus (EIAV) was used to deliver this gene to the abca4 knockout mouse at postnatal day 4 or 5, before tight junctions form between photoreceptors. The authors observed a substantial reduction in A2E accumulation following treatment and a statistically significant improvement in the electroretinogram response. Oxford Biomedica also tested this vector, now dubbed StarGen in healthy macaques and rabbits. They reported transduction of rod and cone photoreceptors and the retinal pigment epithelium in the vicinity of the subretinal injection site. StarGen, now licensed by Sanofi, is under clinical trials in France and the United States.
Usher Syndrome, type 1b: Patients with this form of Usher Syndrome have mutations in the MYO7A gene, are born deaf and develop retinal degeneration in childhood. The shaker 1 mouse carries a mutation in the equivalent mouse gene (Myo7a), and while photoreceptors do not degenerate in these mice, they exhibit defects melanosome localization in the RPE and in opsin transport in photoreceptors. Like patients, the mice are deaf and exhibit vestibular dysfunction. Delivery of MYO7A with an EIAV vector led to restoration of opsin transport and melanosome localization following subretinal injection. The MYO7A vector, UshStat, is in clinical trials in Portland and Paris. The mRNA for this gene is 9 kb in length, and, while this is too long for standard AAV, several strategies have been designed to package the cDNA in pieces for recombination in the cell and functional recovery in the retina.
Other monogeneic diseases in the pipeline for clinical trials: Proof-of-principal of gene therapy has been produced for other inherited retina degenerations, including X-linked retinitis pigmentosa caused by mutations in the RPGR-ORF15 gene, achromatopsia associated with mutations in CNGB3, CNGA3, and GNAT2, LCA1 caused by mutations in the GUCY2D gene, X-linked juvenile retinoschisis, and autosomal dominant retinitis pigmentosa associated with rhodopsin mutations. Clinical trials of gene therapy for some or all of these diseases may be announced in the near future.
Gene Therapy for Age Related Macular Degeneration (AMD): This disease degrades central vision in approximately 10% of those over 70 and 30% of those over 80 years of age. With the aging of the population, the prevalence of this disease is expected to increase. Most of the vision loss in AMD is caused by the abnormal growth of leaking blood vessels from the choroid that supplies blood to the photoreceptor layer of the retina. While the injection of protein inhibitors of the VEGF-A signaling, such as ranibizumab, are effective in preventing choroidal neovascularization in most patients, this treatment involves frequent ocular injections which is literally a pain and increases the risk of infection.
Preclinical data supporting gene therapy to treat wet AMD has made use of a secreted and soluble form a VEGF receptor. Two versions of soluble FLT1 have been tested in AAV: FLT-1 domains 1-7, designated sFLT1, or a chimera, designated sFLT01, comprised of Flt-1 domain 2 linked to a human immunoglobulin Fc fragment. Both have been shown to inhibit choroidal neovascularization in mice and in non-human primates, following laser disruption of the RPE. An important difference between the approaches is the route of injection: AAV2-sFLT1 requires subretinal injection, whereas AAV2-sFLT01 could be delivered by intravitreal injection. Clinical trials are underway in Perth, Australia using subretinal injection of AAV2-sFLT1 and in several sites in the United States using intravitreal injection of AAV2-sFLT01. As an alternative approach, OxfordBiomedica is testing an EIAV vector expressing the angiogenesis inhibitors angiostatin and endostatin in several sites in the United States, this approach also requires subretinal injection of virus.
As the field of gene therapy for eye diseases has matured, a greater share of the research is being conducted in small (e.g., OxfordBiomedica, AGTC, Avalanche) or large companies (e.g., Genzyme/Sanofi). This progression is inevitable, because of the size of clinical trials for prevalent diseases such as AMD requires a large numbers of subjects, in order to compare with the efficacy current therapies. Where does this leave families affected by orphan diseases, such as a particular form of retinitis pigmentosa? Once proof of concept has been established for ocular gene therapy, these rare diseases may be of less interest to large corporations, and government sources are unlikely to continue to fund pre-clinical or clinical gene therapy trials of limited “novelty”. A process must be established to fund clinical trials, and, if successful, to therapies make these therapies available to patients. Certainly, severity and prevalence of disease will influence priority. This is an issue facing all rare diseases, but the success of gene therapy brings it front-and-center for orphan diseases of the eye.