Summary of the Ninth Annual Retinal Cell & Gene Therapy Innovation Summit 2024

The Retinal Cell and Gene Therapy Innovation Summit is the premier event for researchers and companies developing treatments and cures for retinal degenerative diseases. In its ninth year, the Innovation Summit featured more than 30 presentations from retina experts from around the world with nearly 420 people in attendance.

Hosted by the Foundation Fighting Blindness and the Casey Eye Institute at Oregon Health & Science University, the sold-out event was held on May 3, 2024, two days prior to the 2024 annual meeting of the Association for Research in Vision and Ophthalmology (ARVO) in Seattle. While the ARVO meeting is the world's largest eye research conference, with more than 10,000 attendees, the Innovation Summit provided a special focus on potential retinal therapies, many of which are in clinical trials.

The Summit featured a compelling patient perspective from Martha Steele, a Foundation Board Director, and an insightful keynote address from David Gamm, MD, PhD, University of Wisconsin-Madison, on the benefits and challenges in using retinal organoids for disease modeling and photoreceptor transplantation

Summit co-organizers were Casey's Paul Yang, MD, PhD, and Renee Ryals, PhD, and the Foundation’s Amy Laster, PhD. 

Summit sponsors:

• Platinum: Johnson & Johnson, Spark
• Silver: Atsena, Ocuphire, Restore Vision, Sepul Bio
• Patron: Adverum Biotechnologies, Annexon BioSciences, MeiraGTx, Nacuity Pharmaceuticals, REGENXBIO, SparingVision

Session Summaries

SESSION 1: PRE-CLINICAL MODELS

Metformin Eye Drops: A Potential Treatment for Stargardt Patients
Dr. Ruchi Sharma, National Eye Institute, National Institutes of Health

Stargardt disease leads to progressive vision loss of central and night vision. It is usually caused by a mutation in the gene ABCA4, which is expressed in photoreceptors and, as was recently shown, in the retinal pigment epithelium (RPE). To model Stargardt disease, Dr. Sharma uses  induced pluripotent stem cells (iPSC)–derived RPE, which accurately mimics the Stargardt disease condition, allows researchers in preclinical settings to safely assess potential new therapies for signs of efficacy.

One such potential new therapy for Stargardt disease is metformin. The drug itself is not new—providers use it in the treatment of type 2 diabetes and gestational diabetes—and it is known to enhance cellular lipid metabolism and increase lysosomal activity (which is decreased in tissue affected by Stargardt disease).

Researchers from the National Eye Institute applied metformin to the Stargardt iPSC-RPE models, and found that metformin rescued defective photoreceptor outer segment processing in ABCA4A null iPSC-RPE tissue. Researchers then moved to an ABC4A knockout mouse model in which they administered metformin via drinking water. They found that metformin administered over 32 weeks in ABCA4 knockout mice resulted in reduced lipid and cholesterol accumulation, as well as reduced accumulation of A2E in the RPE.

Researchers have now moved from cell and small animal models into humans. An ongoing Phase 1/2 study is assessing the safety and efficacy of oral metformin in patients with Stargardt disease. All patients are at least 12 years old, have ABCA4 retinopathy and vision problems, and receive a maximum daily dose of 2000 mg metformin via oral administration for 24 months.

Oral metformin may carry risks, including birth defects and gastrointestinal issues. Future studies will include exploration of a metformin eye drop developed by Curative Bio, which may provide a safer and more efficient route for getting metformin to the tissue that needs it most.

In Vitro Modeling and Rescue of Pigmentation Defect Associated with TYR Mutations Using OCA1A Patient-derived Cells
Dr. Rajarshi Pal, Eyestem Research

Oculocutaneous albinism (OCA) is a group of rare inherited autosomal recessive disorders caused by lack of melanin biosynthesis. Approximately 7 to 9 types of OCA have been identified, with the OCA1A subtype representing the most severe and aggressive form of OCA.

OCA1A is associated with mutations in the TYR gene, which results in a lack of pigmentation in skin, hair, and eyes. Foveal development is derailed in patients with OCA1A, resulting in poor functional vision and intense photophobia.

Using peripheral blood collected from a patient with OCA1A, researchers developed an induced pluripotent stem cell (iPSC)-derived retinal pigment epithelium (RPE) line. The RPE provides critical support for photoreceptors. The retinal pigment epithelium (RPE) in this iPSC model failed to show any evidence of pigmentation upon maturity, whereas the RPE in healthy tissue showed pigmentation.  Researchers have shown that the TYR protein is responsible for melanin production—and that TYR is absent in OCA1A RPE. Further, in the RPE of an OCA1A model, TYRP1 and TYRP2 show abnormal expression. A gene expression analysis found that the WNT signaling pathway was dysregulated in patients with OCA1A, further building a model for disease that researchers can use to understand targets for therapy.

These initial findings elucidate the disease pathway for OCA1A, and are undergoing validation using multiple readouts. It appears that the OCA1A-RPE model system could be useful for studying molecular pathogenesis and could facilitate discovery of a novel treatment.

Photoreceptor Transplantation: Update on Experience with Alternate Cone-Enriched Species
Dr. Deepak Lamba, Genentech

Rod photoreceptors dominate the anatomy of many of the rodent species used in research. However, humans rely on cones for visual acuity and daytime and color vision. Therefore, finding cone-rich species of rodents that are amenable to non-invasive retinal imaging (and therefore structural and functional assessments) is an asset. Two particularly cone-rich rodent species have proven to be advantageous for retinal research purposes: the 13-Lined Ground Squirrel and the Northern Tree Shrew.

Each animal comes with benefits and drawbacks. The 13-Lined Ground Squirrel is endemic to the Midwest United States, which makes it easy to access for US-based researchers. However, the animal hibernates, and there is a small window of time in which studies can be performed. The Northern Tree Shrew has a very dense cone anatomy (approximately 95% cone cells) that is similar to primate anatomy, but the animals chiefly live in Southeast Asia. US researchers are limited to a few academic centers with access to Northern Tree Shrews.

After initiating methods of retinal degeneration in both animal models, human induced pluripotent stem cells (hiPSC) were transplanted into retinal tissue. These cells included green fluorescent protein (GFP)–expressing photoreceptors, which enable for researchers to identify these cells within the animal model.

Immune suppression was needed for cone-dominant animal models to ensure the survival and integration of transplanted photoreceptors. Those efforts worked.

Researchers observed proliferation of GFP-expressing photoreceptors in the models. In the Northern Tree Shrew, hiPSC-derived retinal cells survived up to 8 weeks; in the 13-Lined Ground Squirrel, hiPSC-derived retinal cells survived up to 4 months. Researchers have tracked the origin and fate of these cells via ex vivo analysis, providing greater insights into how they integrate into surrounding tissue. These insights may ultimately prove valuable to researchers investigating pathways to cell transplantation for retinal degeneration, and could facilitate research into this therapeutic approach.

SESSION 2: PRE-CLINICAL GENE THERAPY

IND-Enabling Studies to Support the Clinical Development of ATSN-201, a Subretinally Delivered, Laterally Spreading Gene Replacement Therapy for X-Linked Retinoschisis (XLRS)
Dr. Shannon Boye, University of Florida

The RS1 gene is linked to cell-cell adhesion. Mutations along RS1 lead to monogenic X-linked retinoschisis (XLRS), which affects approximately 30,000 patients in the United States and Europe. As an X-linked condition, XLRS generally affects males. Females are usually unaffected carriers. Anatomically, XLRS is characterized by abnormal splitting of central retinal layers and loss of photoreceptors. Patients with XLRS are diagnosed as children and experience progressive vision loss into late adulthood. This means that, should researchers develop a safe and effective treatment, there is a wide treatment window in which patients could undergo therapy.

Finding an effective target and administrative route for therapy are key. Expression of the RS1 protein by photoreceptors has recently been shown to provide an effective and long-lasting rescue of tissue in an XLRS animal model, suggesting that photoreceptors could be a viable target for drug development. Earlier clinical trials investigating potential XLRS treatments found that intravitreal delivery resulted in inefficient photoreceptor transduction and an inflammatory response.

Subretinal injection in the peripheral retina shows promise as a means of optimized delivery. The peripheral administration via a subretinal bleb reduces the likelihood of foveal detachment, allows therapy to cover a larger area, and efficiently transduces foveal photoceptors, allowing expression of the RS1 protein by central photoreceptors.

Many retinal gene therapies in development use human-engineered adeno-associated viruses (AAVs) to deliver therapeutic genes to the retina. The AAVs are comprised of millions/billions of containers (capsids) which penetrate retinal cells with their therapeutic genetic cargo. Researchers are investigating use of a laterally spreading AAV called AAV.SPR, developed by Atsena Therapeutics, in animal models. In 45 non-human primates that were dosed with AAV5 and AAV.SPR, researchers found significant expression of green fluorescent protein beyond the bleb site at 7 weeks in the AAV.SPR group compared with the AAV5 group. The protein was able to reach  the fragile fovea (central retina) which was just outside of the bleb site.

RS1-deficient mice were followed for 6 months after subretinal administration of ATSN-201, a therapy utilizing AAV.SPR. Mice were dosed at low, medium, and high doses, with vehicle serving as a control arm. Researchers found that following administration of ATSN-201, complete resolution of schisis cavities (splitting of retinal layers) was observed in all treatment groups. Next, a hybrid toxicology/pharmacology study of ATSN-201 was performed in RS1 knockout mice. The treatment was well tolerated at all dose levels in this animal model, and adverse events related to the drug were observed. When conducted in non-human primatesl, ATSN-201 treatment resulted in no adverse systemic effects, ophthalmic examination findings, and changes in intraocular pressure.

In all, the findings of these investigational new drug (IND)-enabling studies support the clinical use of ATSN-201 for the treatment of XLRS. Atsena’s LIGHTHOUSE clinical trial, a Phase 1/2 trial exploring ATSN-201 in patients with XLRS, is underway. The RD Fund, the Foundation’s venture philanthropy arm, is an investor in Atsena, a biotech co-founded by Shannon Boye, PhD, University of Florida.

Efficient AAV Capsids for Intravitreal Gene Therapy Created Using the scAAVengr Workflow
Dr. Leah Byrne, University of Pittsburgh

Two dynamics in the delivery of gene therapy—difficulties with AAV transduction and increased inflammation following high doses of therapy—have negatively impacted some efforts to use AAV vectors in safe and effective ways. Generally speaking, there is an inverse relationship between the ease of the route of administration and the efficiency by which drug reaches its target: subretinal injections are difficult to administer but lead to more efficient delivery, intravitreal injections are easy but lead to inflammatory and transduction issues, and suprachoroidal injections split the difference. Add to this that existing AAVs are inefficient in nature (and therefore require high dosages, leading to inflammation), and the challenges to the AAV-based therapeutic landscape become clear.

The company Avista has created AAV vectors with higher efficiencies that spread across the entire retina, enabling increased efficiency without triggering an immune response. The goal is to create a gene therapy that is conducive to intravitreal injection, a less-invasive outpatient procedure with the lowest barrier to administration.

Avista has leveraged their high-throughput computationally guided platform scAAVengr (which stands for single-cell AAV engineering) to build the next generation of AAV vectors. Using scAAVengr, researchers have created a highly diverse library of mutated AAVs, each of which is barcoded so that performance can be tracked. These were pooled and administered via intravitreal injection into non-human primates. A small percentage of the variants were highly successful, leading to transgene expression.

Researchers identified approximately 200 variants that showed potential for infecting retinal cells, and re-ran those variants in clinically meaningful doses in non-human primate models. AAVs with which most clinicians are familiar, such as wildtype AAV9 and wildtype AAV2, were outperformed by vectors such as 7m8 (a vector developed by directed evolution in 2013) and K912 (developed via directed evolution in canine models in 2022). All of these were surpassed by a handful of top-performing AAV variants, the most impressive of which is ATX002.

Further non-human primate models have found that ATX002 outperforms 7m8 (a well-regarded AAV), leading to efficient gene expression in the periphery, the fovea, and through all retinal layers. Researchers found that RS1 transgene expression with the ATX002 vector was anywhere between 6 to 326 times higher than AAV8 vector, underscoring the efficacy of this vector. Further, both photoreceptor and bipolar cells expressed nearly endogenous (normal) levels of retinoschisin after ATX002 use.

The Foundation Fighting Blindness funded research that is informing Avista’s vector development.

ELOVL2 Gene Therapy: Increasing Retinal Concentrations of Long Chain and Very Long Chain Polyunsaturated Fatty Acids to Treat Dry AMD
Dr. Marty Emanuele, Visgenx

Long chain (LC) and very long chain (VLC) polyunsaturated fatty acids (PUFAs) modulate the viscosity and fluidity of biologic membranes, which in turn affect the transduction of visual signals, structural integrity, and bioenergetics of retinal tissue. Animal models without LC and VLC PUFAs have a decline in mitochondrial function (cells’ energy production), which significantly affects visual transduction.

An analysis of human donor retinal tissue found that patients with dry age-related macular degeneration (AMD) had lower volumes of LC and VLC PUFAs compared with age-matched controls. Patients with Stargardt disease type 3 present with a dry AMD–like phenotype (condition) and show lower concentrations of LC and VLC PUFAs.

The elongation of fatty acids is controlled by a family of enzymatic proteins called Elongation of Very Long Chain Fatty Acids-Like (ELOVL).  In the case of Stargardt disease type 3 patients, mutations in ELOVL4 lead to a deficit in VLC PUFAs. Given that retinal degeneration is the only known phenotype of Stargardt type 3, researchers have theorized that reduced VLC PUFAs may be the cause of photoreceptor cell death. Mouse models lacking ELOVL2, which also have lower LC and VLC PUFA concentrations, present with dry AMD-like phenotypes. This has led researchers to hypothesize that declines in LC and VLC PUFAs underpin the pathology of dry AMD, and had led to an effort to restore healthy LC and VLC PUFAs in retinal tissue via ELOVL2 gene therapy.

The company Visgenx is leading the way on this initiative and has developed an ELOVL2 gene therapy called VGX-0111. So far, animal models have shown promise for this approach. Mice with age-related retinal thinning were treated with ELOVL2 gene therapy or AAV8 vehicle (placebo). At 6 months, thicker photoreceptor layers were observed in mice that were treated with gene therapy compared with those treated only with vehicle (who themselves showed age-related photoreceptor loss).

Subretinal administration of VGX-0111in African Green Monkeys resulted in higher levels of ELOVL2 expression within the macula and the retinal pigment epithelium compared with control eyes. Specifically, 76% to 99% of macular cones and RPE expressed hELOVL2 at 56 days after gene therapy was administered. Further, the specific VLC PUFAs associated with AMD in humans were expressed at increased rates in African Green Monkeys. VGX-0111 at the lowest dose (4E9) was well tolerated based on electroretinogram (ERG), optical coherence tomography (OCT), and histopathology findings.

The finding that ELOVL2 mutation is linked with reduced LC and VLC PUFAs, which in turn is linked with dry AMD development, may be a major step forward in the fight against AMD. That VGX-0111 was well tolerated in animals at a low dose and led to increases in the specific VLC PUFAs linked to macular degeneration indicates that ELOVL2 gene therapy could be a viable strategy for addressing dry AMD.

Design of a Phase 1/2 Clinical Trial Using a Dual Vector Strategy for the Treatment of MYO7A-related Usher Syndrome (USH1B)
Dr. Francesca Simonelli, Università degli studi della Campania Luigi Vanvitelli

Usher syndrome type 1 leads to early onset retinitis pigmentosa (RP) and hearing loss. A subtype of this condition, called Usher syndrome type 1B (USH1B) is caused by mutations in the MYO7A gene. There is no approved treatment for USH1B, which accounts for approximately 45% of cases of Usher Syndrome type 1. The UshTher Project, a European research consortium, aims to test the safety and efficacy of dual AAV gene therapy approaches for USH1B patients with MYO7A-mediated disease.

Because the MYO7A gene is too large for a single AAV vector, researchers split a transgene coding sequence into two AAV vectors, which has been shown to be an effective means of delivering the full length MOY7A in mice, pig, and non-human primate models. Production of clinical-grade dual AAV vectors is complete, as is a natural history study of USH1B patients.

During the natural history study, researchers sought to identify patients who would be most suitable for clinical trial enrollment. Qualification for such a study relied on ophthalmic findings and the rate of disease progression over 2 years in 53 patients enrolled at 3 sites. Functional and structural assessments occurred at baseline, year 1, and year 2, and 50 of the 53 patients remained enrolled for the duration of the study.

A plurality of patients (41%) showed a hyperautofluorescent ring on fundus autofluorescence (FAF) imaging at baseline, with patchy patterns (33%) and severely decreased FAF patterns (18%) also among the most common clinical presentations. (FAF is an approach used to image diseased and degenerated areas of the retina.) Best corrected visual acuity (BCVA) at baseline was significantly related to age: Pediatric patients showed good vision until the second decade of life, at which point BCVA declined significantly (2-4 letters over 2 years of follow-up). Patients with a hyperautofluorescent ring had approximately 79 letters at baseline compared with patients who had patchy patterns (62 letters) and severely decreased FAF patterns (40 letters); both differences were statistically significant.

Although visual field findings showed an association with age, no differences were observed during the study (which may be due to learning effects). Visual field was significantly larger in eyes with a hyperautofluorescent ring pattern compared with the other two patterns. Central macular thickness measurements were not correlated with age and were unchanged during the study.

This prospective natural history study of USH1B documented for the first time the slow natural course of the disease, and demonstrated that FAF could serve as an objective readout to accurately monitor and stage progression of RP in patients with USH1B. Researchers concluded that the patchy pattern represents an intermediate disease stage that could be suitable for research, as it would illustrate both the potential benefits and any toxic adverse effects of treatment.

A clinical trial assessing the safety and efficacy of a gene therapy for USH1B has been designed. That study is awaiting approval from various European regulatory authorities.

Intein-based Dual-AAV Gene Therapy for Stargardt Disease
Dr. Aniz Girach, SpliceBio

ABCA4 mutations lead to loss of the identically named protein ABCA4, which is responsible for clearing byproducts of the visual process such as lipofuscin and A2E. Without clearance of those visual byproducts, photoreceptor loss occurs. Gene therapy for ABCA4 faces a challenge: the gene is too large to fit in a single AAV vector. However, splitting the gene and packaging it into a dual AAV system may prove effective.

SpliceBio, using an AAV8 vector, has found a means by which to split ABCA4 into two vectors. Incorporation of the gene into the nucleus via this pathway leads to the transcription and translation of two different intein fragments, which form a full-length protein that becomes functional. This may be an effective model for a gene therapy that treats Stargardt disease, which is usually caused by ABCA4 mutations.

SB007, a treatment developed by SpliceBio, is currently in Investigational New Drug Application (IND)-enabling studies for gaining authorization to launch a clinical trial. In one such study, subretinal administration of SB007 in knockout mice models found that expression of ABCA4 was restored, with clear expression of the protein observed in the outer segment and the outer nuclear layer. This increased expression resulted in a 50% reduction in A2E accumulation.

These findings were also found in other animal models. In ABCA4 knockout rats, both A2E and A2GPE (a precursor to A2E) reductions were observed following SB007 administration. High levels of mRNA and protein expression were achieved in minipig and non-human primate models after SB007 was dosed, showing that hABCA4 expression was possible in larger animals; in the non-human primate model, ABC4A RNA expression was observed along the outer nuclear layer, retinal pigment epithelium, and the cone-rich fovea. Inflammation issues were not observed.

An observational study, called POLARIS, has begun. POLARIS aims to evaluate disease progression via multimodal imaging and various endpoints, refine eligibility for a Phase 1/2 study, and identify patients whose disease progresses quickly so they could be potentially enrolled in future clinical trials assessing SB007. Researchers will also use POLARIS to refine endpoints used for interventional trials. Enrolled adult and adolescent patients having pathogenic ABCA4 mutations with defined macular lesions will participate in POLARIS for up to 24 months.

When cleared for initiation, a Phase 1/2 study evaluating the safety and efficacy of SB007 will begin. This study will be divided into 2 parts. In part A, adults with advanced disease will be enrolled in a dose-escalation study at 3 dose levels. In part B, adults and adolescent patient with early-to-mid-stage disease will receive high-dose or low-dose SB007, and will include a control group. Anatomic and functional endpoints that will satisfy regulatory agencies and payers have yet to be determined.

ACDN-01: First-in-class RNA Exon Editor for the Treatment of Stargardt Disease
Dr. Jay Barth, Ascidian Therapeutics

There are inherit challenges to gene-editing approaches to genetic diseases. In particular, the need to develop distinct therapies for each mutation. Similarly, the need for multi-component editing machinery similarly leads to slow progress. This means that only a narrow portion of a population with a specific disease can be treated even after the painstaking process of engineering, evaluating, and earning regulatory approval for a gene editing treatment

Researchers at Ascidian Therapeutics are taking a different approach to gene editing therapy: RNA exon editing, which has the potential to replace multiple contiguous exons with a single therapeutic agent. (Exons are the regions that provide coding information for producing the proteins crucial to cell health and function. They are also the regions where disease-causing mutations are most likely to occur.) Addressing multiple mutations with a single technology could lead to a platform that treats a range of patients.

RNA exon editing could, like gene therapy approaches, require only a single dose to realize efficacy. This is in part because RNA exon editing requires the delivery of a DNA episome, which continuously expresses RNA exon editors. The resulting exon editing via trans-splicing corrects mRNA mutations, leading to the production of a functional protein. DNA episomes have the capacity to enable multiple RNA exon editing mechanisms, meaning that a therapy customized to a specific genetic mutation is not required.

Ascidian’s focus is on the therapy ACDN-01, an ABCA4 RNA exon editor designed to replace 22 exons of ABCA4 pre-mRNA in patients with ABCA4-mediated retinal degenerations; this accounts for approximately 30,000 to 45,000 patients in the United States. ACDN-01 is delivered via an AAV vector as a one-time treatment.

In human retinal explants, robust ABCA4 RNA exon editing was achieved following administration of ACDN-01. Compared with control tissue, tissue treated with ACDN-01 realized a 30% to 50% increase in RNA replacement at day 21, suggesting a therapeutic benefit could be realized after treatment. In non-human primate models, in vivo RNA exon editing resulted in a 30% protein replacement rate at 6 months, and the rate of RNA replacement was within the range seen in donor retinal tissue. This makes ADCN-01 the first treatment to demonstrate RNA exon editing in a large animal, the first treatment to show durability out to 6 months, and the first treatment to produce of a full-length protein following dosing with a single AAV vector.

The US Food and Drug Administration authorized Ascidian’s IND filing, and the Phase 1/2 STELLAR clinical trial is expected to begin soon. Patients with ABCA4-related retinopathy (including those with Stargardt disease) will be enrolled. There will be 3 ascending doses delivered via subretinal administration. The primary endpoint is safety and tolerability, and early efficacy endpoint will also be explored. A 5-year safety follow-up as required for AAV-based gene therapy trials will be conducted as well.

SESSION 3: CLINICAL GENE THERAPY

PRODYGY: A First-in-Human Trial of Rod-Derived Cone Viability Factor (RdCVF) Gene Therapy in Subjects with Rod-Cone Dystrophy
Dr. Daniel Chung, Sparing Vision

Inherited retinal diseases (IRDs) are associated with over 300 different mutated genes, approximately 70 to 80 of which are linked specifically with rod-cone dystrophies. Although gene therapy approaches have proven to be impressive scientific breakthroughs, developing individual drugs for each genetic mutation would be expensive, time-consuming, and inefficient. That makes gene agnostic therapies, those that are designed to work regardless of the mutated gene causing the IRD, very attractive.

Most rod-cone dystrophies begin with loss of rods. Loss of rods subsequently leads to loss of cones. Foundation-funded researchers from the Insitut de la Vision discovered a protein (rod-derived cone viability factor or RdCVF) produced by rods that is crucial for cone survival.

The discovery of RdCVF and its full-strength isoform RdCVFL may be key to unlocking a gene-agnostic therapy for RP. RdCVF stimulates glucose metabolism in cones, prompting renewal of the outer segment, and RdCVFL mitigates the effects of oxidative stress that increase in cones following rod death. This naturally occurring process is a means by which healthy retinal cells protect and renew themselves. However, in patients with dying rods, rod-derived factors such as RdCVF and RdCVFL are not produced, leading to eventual cone death and then blindness.

If a therapy supplemented the production of RdCVF and RdCVFL, then patients may be able offset the effects of rod death (ie, death of cone cells) long enough that vision is preserved. With this approach, it doesn’t matter which genetic mutation is at the root of a patient’s RP, so long as the retina’s natural activities are boosted.

The scientists at SparingVision tested this approach via the drug SPVN06, the safety of which is under investigation in the Phase 1/2 PRODYGY trial. In the first step of PRODYGY, 9 total patients (3 at each dosing level) will receive a single dose of SPVN06 at a low dose, medium dose, or high dose. To date, all patients in the low-dose and medium-dose groups have received treatment, and patients in the high-dose group are still being dosed and monitored.

So far, no adverse events, study discontinuation, or dose-limited toxicities have been reported. Some mild side effects were observed, most of which resolved on their own. After reviewing these outcomes, a data safety and monitoring board provided a positive recommendation to proceed with the trial as planned. 

Nuclear Hormone Receptor-Based Gene Modifier Therapy: Safety and Efficacy from Phase 1/2 Clinical Trials for Retinitis Pigmentosa
Dr. Benjamin Bakall, Associated Retina Consultants | University of Arizona, College of Medicine, Phoenix

OCU400 from Ocugen is a gene-agnostic approach that would leverage the power of a master gene regulator to modify the expression of many genes, allowing for one or more mutations to be treated with a single gene therapy approach rather than gene-specific technologies. Use of a single-dose master gene regulator would enable providers to treat patients with different genotypes. In particular, researchers investigating OCU400 are exploring treatments for retinitis pigmentosa (RP) and Leber congenital amaurosis (LCA).

After concluding that OCU400 was safe and showed signs of efficacy in mouse models, researchers initiated a Phase 1/2 trial. A total of 18 adults with RP were enrolled in the study; patients had either an autosomal dominant RHO mutation, dominant NR2E3 mutation, or recessive NR2E3 mutation. As with other early phase studies, researchers primarily investigated the safety of this intervention, but also looked for signs of efficacy. Patients were dosed with low-dose, medium-dose, or high-dose OCU400.

The drug appears to be generally safe and well-tolerated. Two safety events worth noting were related to the treatment: one case of panuveitis (inflammation) occurred in the medium-dose group (which was resolved with regular treatment) and one case of decreased vision occurred in the high-dose group (which is ongoing).

Signs of efficacy were observed at 12 months, with 72% of patients showing stability or improved best corrected visual acuity measurements, and 78% of patients showing stability or improvement in low-luminance visual acuity and multi-luminance mobility testing. A Phase 3 study is enrolling 150 RP patients in the United States.

Researchers speculate that, if a master gene regulator such as OCU400 were found to be safe and effective, it could restore retinal cell homeostasis, increase the likelihood of photoreceptor survival, initiate a “molecular reset” of health and survival gene networks, and improve visual outcomes and quality of life.

Preliminary Safety of ATSN-201 Subretinal Gene Therapy in Patients with X-linked Retinoschisis
Dr. Christine Kay, Atsena Therapeutics

X-linked retinoschisis (XLRS), which is caused by mutations to the gene RS1, is one of the most common causes of juvenile macular degeneration. In XLRS, split retinal layers result in vision loss and increased risk of retinal detachment. Patients with healthy RS1 genes produce a protein (also called RS1) that promotes retinal tissue stability by promoting cell-cell adhesion. In patients with XLRS, the RS1 protein in not produced at sufficient levels and retinal instability follows.

Atsena Therapeutics’ gene therapy ATSN-201 is designed to address RS1 genetic mutations by introducing a functional RS1 gene to photoreceptors. ATSN-201 is a single-dose therapy, meaning patients may only need the drug once in their lifetime.

In a Phase 1/2 study, patients with XLRS due to a pathogenic mutation of RS1 and schisis (retinal splitting) involving the fovea as seen on optical coherence tomography are receiving ATSN-201. There are low-dose, medium-dose, and high-dose groups in this part of the study, with 3 patients assigned to each dose level. All patients are at least 18 years old. Safety will be assessed at 1 and 5 years, and signs of efficacy will also be assessed at those timepoints as well. So far, 3 patients have been enrolled in the low-dose group and have been followed for 6 months.

As the most recent assessment, there have been no serious treatment-emergent adverse events or dose-limiting toxicities. There have been 24 adverse events (AEs) observed in the study, with 21 of them considered unrelated to the study drug, and 18 of them considered related to the surgical procedure. All of the AEs have been categorized as either grade 1 or grade 2 AEs using a 5-point severity scale. Any ocular inflammation so far has been minimal and has responded well to steroid treatment. None of the 3 patients in the study have discontinued treatment, and no retinal detachments or macular holes have been observed. Vision has remained stable in patients following an expected postoperative decline in vision, which resolved within 1 to 3 months in the 3 patients dosed so far. Foveal thickness improved in 2 of the 3 patients in this cohort.

Part A of this Phase 1/2 study will continue with enrolling patients in the mid-dose and high-dose cohorts. In part B of this trial, some patients from part A will continue to be followed, and some new patients between the ages of 6 and 18 will be dosed with ATSN-201.

The RD Fund, the Foundation’s venture philanthropy arm, is an investor in Atsena.

Preliminary Safety and Efficacy Results from an Open-label, Dose-escalation Pilot Gene Therapy Study of FT-002 in Subjects with RPGR Gene Mutation Associated X-linked Retinitis Pigmentosa
Dr. Ruifang Sui, Peking Union Medical College Hospital

X-linked retinitis pigmentosa (XLRP) is a severe form of RP, resulting in night blindness and decreased visual acuity. It is often diagnosed in patients under 10 years old. It often leads to legal blindness around the age of 45. Variants of the gene RPGR account for more than 70% of XLRP cases and approximately 15% of all RP cases. Current estimates place the prevalence of XLRP secondary to RPGR variants at 20,000 patients in the United States and Europe and 50,000 patients in China. Finding treatments for XLRP driven by RPGR variants has been a challenge, in part because the ORF15 region (a hot spot for mutations) of RPGR has proven difficult to stabilize when producing gene therapy vectors.

FT-002, a gene therapy under investigation from the team at Frontera, is being assessed in a Phase 1/2 study. FT-002 is a one-time treatment delivered via subretinal injection and uses AAV5 as its gene delivery vector.

Preclinical models found that FT-002 significantly preserved retinal structure and rescued photoreceptor function in RPGR knockout mice. Distribution of vector genomes were seen in the retinal cells of mice and non-human primates (NHP) after administration of FT-002. It was also well tolerated in NHPs.

Based on the safety data from animal models, an in-human clinical trial was initiated. Patients diagnosed with XLRP secondary to mutation of RPGR who were aged 8 to 45 years and had best corrected visual acuity (BCVA) of 35 to 78 letters (20/200 to 20/32 Snellen equivalent) were enrolled. Investigators looked for dose-limiting toxicity, safety endpoints, and signs of efficacy.

There were three dose cohorts (low-dose, medium-dose, and high-dose), with 6 patients enrolled in each cohort. Patients will be followed for 1 year. So far, 4 patients in the low-dose cohort have completed 1 year, and all remaining patients are still within the first year of dosing. (Note: one patient from the low-dose cohort withdrew from the study for personal reasons.)

Thus far, FT-002 has been well-tolerated in humans. No dose-limiting toxicity or ocular serious adverse events (SAEs) have been observed. There have been 7 reported cases of ocular inflammation, all of which resolved with application of topical steroids. Among all 18 patients in the trial, 15 have experienced a total of 51 treatment-emergent adverse events (TEAEs); 26 of these 51 TEAEs were related to the study procedure itself.

Early efficacy has been observed on microperimetry on days 84 and 168 in patients who received the medium dose of FT-002. Of the 6 patients in the low-dose cohort, 2 (33%) achieved an improvement of at least 7 dB in at least 5 microperimetry foci (retinal locations) at day 84. At day 168, 3 patients (50%) had achieved this threshold. In the medium-dose group, mean BCVA change from baseline was +1.8 letters at day 168. Among the high-dose group at day 84, mean BCVA change from baseline was +6.2. Mobility test scores improved from baseline in the medium- and high-dose groups as well.

Gene Therapy for X-linked Retinitis Pigmentosa Caused by Mutations in RPGR
Dr. Robert MacLaren, University of Oxford

Developing gene therapies for X-linked retinitis pigmentosa (XLRP) faces two major hurdles: quirks related to the gene sequence for RPGR and the alternative splicing mechanics of RPGR.

RPGR is a gene implicated in XLRP. Much of RPGR is comprised of G and A nucleotides, resulting in highly repetitive sequences. These sequences are prone to in-frame deletions of the coding sequence during cloning, which serve as a snag during basic research for gene therapies treating XLRP.

Further complicating the creation of gene therapies for XLRP, RPGR in photoreceptors is alternatively spliced. What should be intron 15 is incorporated in a 15 exon RPGR mRNA transcript, resulting in a region called RPGR^ORF15. If this region experiences a mutation, then retinal degeneration occurs. RPGR glutamylation, which must occur for photoreceptors to properly function, is impaired by RPGR^ORF15 variants, which interfere with the enzyme TTLL5. This is particularly true of cone cells, which rely on RPGR glutamylation for survival and function.

Thus far, subretinal delivery has proven to be the most effective means of realizing therapeutic gains; intravitreal delivery, while easier to administer, has yet to prove as effective at delivering an effective payload. A recent report from the XIRIUS study (a clinical trial conducted by Biogen) found that delivery of full-length RPGR via subretinal delivery using an AAV8 vector improves cone function out to 2.5 years, showing some potential for long-term therapeutic effect. Early imaging results have found evidence of anatomic reversal in XLRP patients in this study, with increased cone activity and retinal thickness from baseline out to 3 months.

This represents a major step forward in treatment for XLRP. Rather than merely arresting development of retinal atrophy, gene therapy may reverse some degree of retinal atrophy and restore some vision to patients. These trials are, still years away from completion, and it remains to be seen whether vision restoration will continue, halt, or diminish with time. Still, these recent advancements provide direction for future research and bring into focus the benefits of full-length RPGR delivery.

Note: The XLRP gene therapy did not meet its primary efficacy endpoint in the Phase 2/3 XIRIUS clinical study and therefore did not receive regulatory approval.

Voretigene Neparvovec-rzyl (VN), which is marketed under the name LUXTURNA and manufactured by Spark Therapeutics, was approved by the FDA in 2017. It is the first FDA-approved AAV vector–based gene therapy, and is indicated for the treatment of patients with confirmed biallelic RPE65 mutation associated with retinal dystrophy and viable retinal cells.
Dr. Aaron Nagiel, Children’s Hospital Los Angeles, University of Southern California.

Now that VN has been in use for several years, researchers have sought to collect long-term safety information in real-world patients. This post-authorization safety study (PASS) was a multicenter, longitudinal, observational safety registry study that enrolled 87 patients (169 eyes).

The mean age at VN administration was 17.1 years, and the mean duration of follow-up was 3.8 years. Nearly all patients (89.7%) who received VN were diagnosed with Leber congenital amaurosis type 2, with the remaining patients reporting diagnoses of retinitis pigmentosa (9.2%) and severe early childhood onset retinal dystrophy (1.1%).

Most eyes (60.4%) were injected via a single retinotomy site, with 21.9% requiring 2 retinotomies and the remainder requiring at least 3 retinotomies. Despite these disparities, 92.9% of eyes received the recommended dose of 0.30 mL of VN. Among eyes that received bilateral VN, 79.3% were injected 6 to 18 days apart from each other, which is consistent with the surgical manual’s instructions.

Despite adherence to the surgical manual regarding timing of bilateral injections, most eyes were treated with a modified injection technique that differed from the manual’s instructions. The use of visualization dyes (68.0% of cases) and use of intraoperative OCT imaging (62.7% of cases) were the leading deviations from the surgical manual’s operating instructions.

Treatment-emergent adverse events (TEAEs) in real-world patients were mostly transient, mild, and treatable. Many of the TEAEs were related to the procedure itself or to corticosteroid use, and 9.0% of TEAEs were considered related to VN. Many of the TEAEs observed in the PASS were also found in the clinical trials assessing VN.

Chorioretinal atrophy (CRA) occurred in 27.6% of eyes in the PASS. If eyes with touchdown CRA (which is related to surgery and not to the drug itself) are excluded, the rate falls to 17.8%. No correlation between genotype and incidence of CRA was observed, although it should be noted that 3 of 3 sibling patients experienced non-touchdown CRA following VN administration at the same treatment center. CRA itself was not observed in the clinical trials for VN, although a review of data shows that study findings reminiscent of CRA had been observed in 5 patients.

Regarding vision, no clinically significant change in visual acuity was observed from baseline to follow-up visits in the PASS at year 3. Visual function as measured by full-field scotopic (rod-mediated vision) threshold testing improved at a clinically meaningful rate at year 3, which was similar in amplitude to what was observed in the Phase 3 clinical trials.

The PASS is plans to follow patients to 5 years post-treatment, and more data regarding real-world safety outcomes should be expected in the coming years.

A Patient’s Perspective

Martha Steele, Board Director, Foundation Fighting Blindness

Martha Steele, a Foundation Board Director and President of the Foundation’s Boston Chapter, brings unwavering passion to driving the Foundation’s mission. She delivered an honest and compelling story of her journey with vision and hearing loss from Usher syndrome type 2A (USH2A) and her hope and appreciation for research.

USH2A led to hearing difficulties early in her life. An audiologist’s confirmation of her hearing loss validated her parents’ observations, but also worried them. The decision to send a 6-year-old to a school specializing in hearing-impaired students meant that she must live with a foster family, as the school itself was too far from Ms. Steele’s home. After a year of learning how to adjust to academic settings, Martha returned to her local school. She excelled athletically and academically.

Deterioration of vision followed, but progressed so slowly that Martha was unaware of it. It wasn’t until her mid-20s when, living in Columbia, she fell into an open manhole. Alarmed, her friends conducted an at-home test to assess her field of vision, which they determined was about the size of a basketball. Martha was unfazed. This was something that could be dealt with, surely.

But upon an ophthalmic examination during a return to the US, a doctor diagnosed her with retinitis pigmentosa and offered a prognosis of blindness within 10 years. The diagnosis shattered her. It was not a fatal diagnosis, but to a young, capable woman, it certainly felt like one.

Martha was referred to a retina specialist, who better contextualized her condition and prognosis. Most importantly, she learned that her central vision would remain intact for most of the next decade. Her retina specialist advised living the life she envisioned for herself despite the inevitable, forthcoming vision loss. She took the advice, completing college, earning a graduate degree, moving to Boston, and spending her career in public health. 

Still, the prospect of losing central vision worried Martha. As a hearing-impaired person, central vision was critical to reading lips. Loss of central vision occurred rapidly, especially compared with the slow loss she experienced with her peripheral vision. Ms. Steele experienced significant emotional fluctuations, and the despair associated with the disruption of routine life events—knowing if her husband was next to her on the couch, for example, or realizing that she could not see herself in the mirror—devastated her.

Today, Martha is totally blind. Although she lives a rich life, she reminded the audience that having any level of functional vision would be wonderful. Seeing the white lines on a street, the outline of a faraway house, or the shape of a person standing next to her could be the difference between safety versus danger, confidence versus disorientation, and social poise versus embarrassment.

Martha closed her remarks by celebrating the progress that researchers have made in the past 15 years since she joined Foundation Fighting Blindness. That progress inspires fundraising, generates hope for patients, and empowers engagement with communities who can better craft a world for tomorrow’s patients. The fight is long and filled with challenges, but from that fraught fabric grows a community of patients, researchers, and supporters who reinforce each other’s strengths for the sake of progress.

Keynote Lecture

Dr. David Gamm, University of Wisconsin, Madison

Human pluripotent stem cells (hPSCs) offer a window into human retinal development. They presents a unique opportunity for researchers to track natural tissue development over time. Still, hPSC models are imperfect: hPSC modeling systems take time to develop, and thus any thorough evaluation of models requires constant monitoring. As such, checkpoints must be established to ensure that development occurs as authentically as possible.

If researchers do indeed craft a model sufficiently authentic, the model is conducive to several applications, including disease modeling; better understanding mechanisms of disease; development of gene-, genome-, and RNA-based therapies; and production of cell therapies. Further, researchers have grown retinal cells from patients with suspected mutation-based blindness who nevertheless lack a genetic diagnosis. Further examination of these cells can uncover splicing differences or transcriptome changes that can be used to identify the genetic mutation responsible for the patient’s phenotype, helping to clarify the source of disease and aiding in diagnosis.

One important distinction to remember in working with hPSCs is that hPSCs are cell lines, but that hPSC-derived cells are not. It is for this reason that human photoreceptors cannot be maintained in a dish—although they can be continuously grown from a stock of undifferentiated pluripotent stem cells.

Disease models leveraging retinal organoids developed from hPSCs have several advantages in addition to their theoretically unlimited supply. These include the recapitulation of human photoreceptor structure and function, a conduciveness to editing disease gene mutations into or out of lines, the ability to be derived from any patients with any disease, and the power as a platform for development of potency assays. Retinal organoids from hPSCs do have some drawbacks, however, including that these are not “true” disease models, only exhibit cell or tissue autonomous effects when modified, demonstrate wide variability between and within cultures, and follow a developmental timeline beholden to human retinal development.

Tracking the differentiation of hPSC-derived cells is key to assessing their viability as a model. By day 40, photoreceptors can be seen; by days 60 to 80 migration toward the outer rim is observed and cells start packing together. At day 200, outer segment structures present as hairlike appendages. At this point, something akin to retinal tissue is present, and researchers begin to assess whether the photoreceptors in the dish are authentic enough for clinical research. If they are, then they must meet clinical Good Manufacturing Processes (cGMP) standards for compatibility, scalability, and reproducibility before they can be used in a clinical trial setting.

It should be noted that disease models cannot answer all questions about a particular disease. This means that a model may be able to answer a specific question or two, and that the next model can be built to answer other questions or to respond to questions raised by the findings from the first model. In a sense, then, any understanding of disease from an hPSC model is derived from a patchwork of models rather than from a single model itself. This is, in part, is because an hPSC retinal organoid model does not undergo a growth period of disease before a vision-disrupting phenotype is realized. Thus, human tissue that may undergo decades of normal phenotypical expression before onset of disease cannot be recreated with hPSC retinal organoid models.

One way of measuring whether lab-grown retinal cells could functionally connect with the patient’s retina is to assess the phototransduction qualities of hPSC-derived photoreceptors. The team at the University of Wisconsin, Madison, found that in some cases, hPSC-derived photoreceptor cells are nearly as responsive to different wavelengths of light as human photoreceptors. Another method of assessing function potential is to observe evidence of metabolism. Also, axon extension following removal of cone cells from the organoids suggests that the organoid is functioning authentically, as does the redistribution of synaptic markers once axons stop extending.

From this research, Opsis Therapeutics (in partnership with Bayer, BlueRock Therapeutics, Fujifiilm Cellular Dynamics) was founded to determine if hPSC-derived cell therapies for eye diseases could be manufactured according to cGMP standards. BlueRock, which licensed the cell therapy OpCT-001 for the treatment of primary photoreceptor disease, plans to submit an IND application later this year. At the same time, BlueRock has partnered with Foundation Fighting Blindness to add a new cohort to the Uni-Rare natural history study, which is assessing outcomes in patients with a broad range of inherited retinal diseases, including retinitis pigmentosa, Usher syndrome, and Leber congenital amaurosis.

SESSION 4: PRE-CLINICAL CELL THERAPY

Simulation of Neurogenesis from Muller Glia to Repair Degenerated Retina
Dr. Tom Reh, University of Washington

Researchers interested in organ regeneration (including ocular tissue regeneration) in humans have observed self-regeneration in the animal kingdom. Amphibians such as newts and salamanders can regenerate various ocular tissues, including retinal tissue after damage. Post-hatch chickens and zebrafish also regenerate retinal neurons, which has been traced to the presence of Müller glia cells (sometimes called Müller cells). Exploration of mouse models, however, show that mice do not naturally regenerate Müller glia–derived neurons after retinal injury—and researchers want to know if such regeneration could be induced to address retinal degenerative diseases.

Progenitor cells mature into both Müller glia and neurons in retinal tissue during natural development. In some of the models described above (e.g., amphibians, birds, and fish), glia cells and progenitor cells exist within a mutually reinforcing feedback loop, such that glia cells remake progenitor cells that in turn make new glia cells, some of which go on to generate new neurons.

Researchers theorized that if transcription factors present in normally developing mammal retinal tissue were to be expressed in Müller glia, such expression could induce reversion of Müller glia into progenitor cells, which might lead to neuron regeneration. Two foundational pieces of research helped guide this strategy: Yamanaka found adding transcription factors to most somatic cells transform such cells back to induced pluripotent stem cells (iPSCs), and Thompson showed that some amphibian and fish tissues revert back only to progenitor cells rather than revert fully to iPSCs during the regenerative feedback loop.

After assessing the numerous transcription factors found in healthy retinal cells, it was determined that Ascl1 was the transcription factor that would best induce reprogramming of Müller glia into progenitor cells. Using a mouse model, genetic engineering directed Ascl1 expression in Müller glia, with the aim of assessing outcomes in 2 to 5 weeks. In vivo imaging showed that neural regeneration indeed occurred, and functional assessments showed that the regenerated bipolar cells demonstrated a response to light. Further investigations into this strategy have shown that AAV delivery may be a viable route of administration and that similar outcomes are observed in non-human primates.

Most of the Müller glia–derived cells in this model were bipolar cells, with some amacrine cell generation observed. Researchers also found that the addition of other transcription factors in addition to Ascl1 promotes generation of retinal ganglion cell-like cells (which could have application in various disease states), and have theorized that new cells generated from Ascl1 expression in Müller glia could mature and survive up to 1 year in a mouse model.

Questions remain about the viability of this strategy in humans. Although signs of Müller glia–derived photoreceptor-like cells have been observed in ex vivo human retinal tissue following increased expression of Ascl1, it remains to be seen if these cells would be numerous enough to actualize meaningful change in patients’ functional vision. Research in non-human primates is ongoing, with financial support from Foundation Fighting Blindness, and more data are forthcoming.

Optogenetic Engineering of Stem Cell–derived Photoreceptors to Improve Visual Restoration
Dr. Oliver Goureau, Institut de la Vision, Sorbonne Université, INSERM

The transplantation of photoreceptors into ocular tissue faces several challenges that undermine the photoreceptors’ maturation. Transplanted photoreceptors often have difficulty connecting with the remaining inner nuclear layer (leading to synaptic transmission gaps), and they sometimes struggle to grow a photoreceptor outer segment (which is necessary for functional phototransduction). Further, transplanted tissue may be unable to physically contact retinal pigment epithelium and Müller cells, thereby interfering with chromophore replenishment.

Researchers are investigating the possibility of bypassing these challenges by conferring artificial light sensitivity to such tissue and transplanting that tissue into the retina. If effective, such an approach would represent an end-run around the challenges outlined above.

It is important to understand the role of opsins in bypassing the maturation of photoreceptors. Opsins are light-responsive proteins bound to retinal tissue that act as ion pumps. In the case of this specific research, microbial opsins are used. Microbial opsins are often derived from algae and are activated by specific wavelengths of light.

The normal phototransduction cascade—a highly complex, light-sensitive process—results in the hyperpolarization of photoreceptors. An alternative approach using microbial opsins results in hyperpolarization of cells regardless of whether those cells bind to surrounding tissue. This alternative process is less light sensitive than the normal phototransduction cascade, and starts when orange light (590 nm in wavelength) activates the opsin known as Jaws.

To actualize this concept, researchers leveraged AAV vectors in human induced pluripotent stem cells (iPSCs) to grow Jaws-expressing cones in retinal organoids. These cones were transplanted into an RHO knockout mouse model, tissue from which was harvested 1 month after transplantation. Light responses were observed in photoreceptors, bipolar cells, and ganglion cells.

Improvements upon this strategy have been implemented. CRISPR technology targeting iPSCs before they are developed to form retinal organoids has been used in lieu of AAV vectors targeting retinal organoids. Also, careful selection of pure Jaws-expressing photoreceptors (both rods and cones) has been implemented prior to transplantation into a mouse model, enabling a more potent dose of Jaws-expressing cells to be implanted.

Opsins could be a next major tool in the armamentarium of researchers finding therapies for blinding diseases. The more researchers understand about the role opsins could play in cell therapy, the closer they may get to understanding how best to apply this emerging technology.

A Gene Agnostic Stem Cell Treatment for Mid-stage Retinitis Pigmentosa
Dr. Deepti Singh, InGel Therapeutics

The first (and only) drug, voretigene neparvovec-rzyl, for the treatment of LCA and RP represented a milestone in the battle against genetic retinal diseases. However, voretigene only works for less than 1% of patients with these conditions, and the unmet need for treatments remains high. A gene-agnostic stem cell therapy that treats RP could represent a breakthrough for a significant portion of this untreated population, as it would enable treatment for patients with a general phenotype regardless of a specific genotype.

IGT001 from InGel Therapeutics hopes to be that therapy. In IGT001, fetal-derived rod precursor cells that have been purified via a microfluidic system are embedded in an engineered hydrogel that structurally and chemically mimics human vitreous. This hydrogel protects the rod precursor cells from shear forces (e.g., moving through a syringe needle), is immune evasive, and allows therapy to exist within a controlled microenvironment. InGel is targeting an intravitreal injection route of delivery.

In the early stages of RP, rod cells slowly die, leading to reduced vision and increased oxidative stress. During mid-stage RP, retinal tissue undergoes loss of cone outer segments due to the overall loss of rod cells. Without outer segments, phototransduction cannot occur, and cone cells become dormant and eventually die.

IGT001 aims to intervene at mid-stage RP, when cone outer segments are still viable. IGT001 works in two steps. In the first step, the hydrogel is injected intravitreally and then adheres to the inner limiting membrane. The hydrogel fully degrades during a 6-to-8-week period and the rod cells in the hydrogel are cleared away. The rod-derivedproteins secreted into the surrounding retinal tissue mitigate dysfunctional cell signaling responsible for outer segment cone degeneration, improving phototransduction, cone fueling, synaptic connections and promoting protection from oxidative stress.

IGT001 has been evaluated in several animal models (rats, mice, and New Zealand White Rabbits). These models are of various genotypes leading to RP, an important underpinning for a study assessing an investigative gene-agnostic approach. In one of those mouse studies (rd10 RP model), IGT001 was found to significantly improve functional vision up to 9 weeks after injection.

InGel believes its technology may be a feasible means to treat patients with other degenerative retinal disease, such as dry AMD. For now, the focus remains on RP. Clinical trials assessing the safety and efficacy of IGT001 in RP patients are slated for 2026, with IND-enabling studies set to start in Q2 2025.

SESSION 5: CLINICAL CELL THERAPY

Developing A Phase 1/2a Trial to Test Safety and Feasibility of an Autologous IPS Cell-Derived Retinal Pigment Epithelium Patch in Age-Related Macular Degeneration Patients
Dr. Kapil Bharti, National Eye Institute

In healthy retinal tissue, photoreceptors are supported by the retinal pigment epithelium (RPE). In eyes with dry age-related macular degeneration (AMD), deteriorating RPE tissue leads to photoreceptor death. If new RPE cells are introduced, they could save the remaining photoreceptors, preventing degeneration of cells that have not yet begun or completed an atrophic process.

Researchers at the National Eye Institute (NEI) hypothesize that transplanting a patch of RPE from induced pluripotent stem cells (iPSCs) — stem cells derived from the patient’s own blood cells (autologous) — could serve as therapy to preserve photoreceptors in patients with dry AMD. This has led to an NEI study investigating whether a reproducible and efficient protocol for manufacturing RPE from autologous iPSCs could be realized.

To achieve this goal, CD34+ blood cells that have been reprogrammed to iPSCs were differentiated into immature RPE cells, which are then matured and fully polarized on a scaffold for 75 days. So far, this process has shown to be reproducible: Mature RPE cells have been generated from more than 60 donors, and 24 labs and 6 companies have been able to mimic the NEI’s protocol and produce similar results. The maturation of iPSC-RPE cells produced in this fashion has been confirmed by an AI algorithm that validates tissue viability.

Transplantation of an iPSC-RPE patch into a pig model with laser-induced RPE trauma has shown that the RPE patch has provided photoreceptor support out to 150 days after transplantation. Other pig models have shown that two patches could be delivered in a single surgical procedure, possibly enabling a single surgery to cover a greater area. Importantly, OCT angiography imaging has shown choriocapillaris (vasculature) regeneration in these pig models. (Vasculature regeneration is critical to survival and function of transplanted RPE cells.) In contrast, pigs that only received vehicle (ie, a scaffold without RPE cells) showed no regeneration of choriocapillaris.

A phase 1/2 clinical trial investigating iPSC-RPE patch implantation in patients with advanced dry AMD (geographic atrophy) has been initiated. The study, called STEM-RPE, will enroll 15 patients in 2 cohorts. Cohort 1 will include 5 patients whose best corrected visual acuity (BCVA) is at least 20/100, and Cohort 2 will include 10 patients with BCVA of at least 20/80. All patients will receive a single 2x4 mm iPSC-derived RPE-scaffold patch implanted underneath the retina. The primary endpoint is the safety and feasibility of patch administration and survival at 1 year; secondary endpoints will assess BCVA and retinal function.

Two-year Follow-up of a Safety Study Using Induced Pluripotent Stem Cell–Derived Retinal Organoid Sheets for Patients with End-stage Retinitis Pigmentosa
Dr. Michiko Mandai, Kobe City Eye Hospital

Implantation of retinal organoid sheets derived from embryonic stem cell (ES) and induced pluripotent stem cells (iPSCs) could be effective for the treatment of end-stage retinitis pigmentosa (RP). Both preclinical and clinical data have underpinned research on this approach.

Preclinical data has demonstrated the consistent maturation of graft photoreceptors after transplantation, host-graft synaptic connectivity, and the recovery of a light response within host retina. Researchers have observed the development of viable photoreceptor outer segments in preclinical settings, and photoreceptor survival times have been documented out to 2 years. And among host bipolar cells, previously retracted dendrites have regrown toward grafted photoreceptor cells in preclinical models — this is evidence that the transplanted photoreceptors are integrating with the host retina.

Functional responses in mice with RP have been observed as well, with host retinal ganglion cells demonstrating responses to light after grafted photoreceptor transplantation. Wild type (unaffected) mice that have learned to associate a light signal with an electric shock will leave a room upon noticing the light signal; mice with RP are unable to associate the light signal with the electric shock. However, after a photoreceptor graft, approximately 30% to 40% of mice with RP learned to associate the light signal with an electric shock and left the room prior to a signaled shock.

Given the encouraging results of the preclinical studies, a clinical study was initiated in 2020 at the Kobe City Eye Hospital in Japan. This study represented the first use of pre-maturation organoid photoreceptor sheets in a clinical trial setting. The study’s primary endpoints were the increase in retinal thickness by graft survival and safety of retina sheets; safety data included rates of rejection and tumor formation. Secondary endpoints included evaluations of protocols for visual function (to be used in future studies) and the efficacy and safety of sheet transplantation surgery. A total of 2 patients with RP and low visual function were enrolled in the study.

Three sheets were implanted into the bleb via a 24-gauge cannula, the eye was filled with silicone oil (which was removed at 2 months), and intravitreal steroids were administered. Systemic cyclosporine was orally administered for 6 months.

Researchers used a letter recognition test to evaluate functional vision. One patient in the study still had central vision and was able to identify all 26 letters at baseline and at every timepoint during follow-up. The other patient in the study started with hand motion vision and was unable to identify any letters prior to surgery. At year 1, the patient identified 7 of 26 letters, but was unable to identify any letters at 2 years.

Overall, the study showed that iPSC-derived retinal organoid sheets could be safely transplanted in 2 humans with RP. Although efficacy must be improved to realize any long-term, sustainable functional vision gains, it should be noted that the grafts survived with no signs of rejection over 3 years in both patients. In the next clinical research phase, investigators aim to transplant genome-edited retinal organoid sheets into a larger area and will assess stability and function of this intervention.

Retinal Pigment Epithelial (RPE) Stem Cell–derived RPE Progenitor Cell Implantation Therapy for Dry Age-related Macular Degeneration (Dry AMD)
Dr. Jeffrey Stern, Neural Stem Cell Institute | Luxa Biotechnology

Degeneration of the retinal pigment epithelium (RPE) leads to geographic atrophy (GA), an advanced stage of age-related macular degeneration (AMD) that leads to central vision loss. Luxa Biotechnology has launched a clinical to determine if a suspension of RPE progenitor (precursor) cells could sufficiently and safely address dying RPE.

Investigators at Luxa Biotechnology have looked at a unique, tissue-specific stem cell source for this potential suspension: adult RPE stem cells (RPESC) from a human donor (deceased). Each donorcould generate over 1000 doses. Two important dynamics have been observed around this approach: the manufacturing process that transforms donor tissue to doses is simple, and RPESC use does not lead to tumor production.

Investigators evaluated suspensions of RPESC cultures aged 2 to 8 weeks. In RCS rat degeneration models, researchers found that RPESCs at 4 weeks of differentiation achieved peak vision rescue and engraftment compared with cultures aged 2 weeks, 3 weeks, and 8 weeks. An IND-enabling vision rescue study found that the vision of rats at various stages of retinal degeneration showed near-normal levels of visual function compared with rats that received sham or no therapy.

An open-label, single-site clinical trial was designed to assess the safety and efficacy of an RPESC payload in patients with GA. Patients were divided into groups based on vision (20/70 to 20/100 and 20/200 to 20/800), and three cohorts will be enrolled in a dose-escalating protocol with doses containing 50,000 cells, 150,000 cells, or 250,000 cells. Each group will contain 3 patients for a total of 18 total patients in the study. The primary endpoints are the number of patients who experience a decrease of less than 15 letters from baseline and overall safety. Anatomic and functional secondary outcomes include microperimetry sensitivity, OCT macular thickness, and GA area.

So far, 5 of 6 patients  have been treated at the 50,000 cells dose , some of whom have outcomes recorded beyond 1 year. Preliminary data are scheduled for release this year. 

First-in-human Study to Evaluate Safety, Tolerability, and Exploratory Efficacy of a Photoreceptor Regeneration Treatment with EA-2353 in Patients with Retinitis Pigmentosa
Dr. Byron L. Lam, Bascom Palmer Eye Institute, University of Miami

Endogena Therapeutics is attempting to address retinitis pigmentosa (RP) by small molecule–directed regeneration of retinal tissue. They theorize that engineered small molecules could activate progenitor cells (stem cells) in the ciliary body (near the iris), and which could migrate toward the retina and differentiate into photoreceptors.

To that end, Endogena developed EA-2353, a small molecule targeting progenitor cells in the ciliary body to promote retinal regeneration. The drug, which is contained in an ophthalmic suspension designed to prolong residence time and effect, is delivered in a single treatment cycle that includes intravitreal injection administered every week for 4 weeks. Preclinical models have demonstrated efficacy of EA-2353 in both chemically induced and genetic retinal degeneration models, with ERG and visual acuity data suggesting that EA-2353 promotes regeneration and recovery of retinal structure and visual function.

A Phase 1/2 first-in-human clinical trial is assessing the safety and efficacy of EA-2353 in patients with RP. The study is enrolling 14 patients over 4 dose-escalating cohorts. Patients have various mutations linked to RP and received therapy in the worse-seeing eye.

No cases of ocular inflammation were related to the EA-2353. Detectable drug particles in the eye were not associated with any adverse events or visual impairment. Of the 7 cases of detectable drug particles, all but 1 case resolved.

Researchers considered the average best corrected visual acuity (BCVA) from the first three visits to be the baseline BCVA. Cohorts 1, 2, and 3 have reached 12 months of follow-up. At 3, 9, and 12 months, respectively, treated eyes experienced gains of approximately 5, 4, and 3 letters from baseline. (5 letters is equivalent to 1 line on an eye chart.) Untreated eyes at 3, 9, and 12 months, respectively, achieved gains from baseline of approximately 4, 3, and 1 letters, respectively.

Low luminance visual acuity (LLVA) gains from baseline at months 3, 9, and 12 were, respectively, approximately 2, 4, and 12 letters. LLVA measurements were virtually unchanged in untreated eyes. Changes in low luminance deficit (LLD), which is calculated by subtracting LLVA letters from BCVA letters, favored treated eyes at months 9 and 12.

Microperimetry was used to assess light sensitivity at various locations across the retina and found that treated eyes had more improving loci than declining loci at 9 months.

Additional data from this Phase 1/2 study are forthcoming, including data from cohorts 4 and 5. So far, researchers have concluded that EA-2353 can be administered safely in patients with RP regardless of genotype and that dosed patients had some structural, functional, and sensitivity improvements. 

OpRegen® Retinal Pigment Epithelium (RPE) Cell Therapy for Patients with Geographic Atrophy (GA): Month 24 Results from the Phase 1/2a Trial
Dr. David Telander, Retinal Consultants Medical Group, Inc | Volunteer Clinical Professor, Ophthalmology, UC Davis

Researchers at Lineage Therapeutics have developed a suspension of retinal pigment epithelium (RPE) cells that are derived from a human embryonic stem cell line. This suspension, called OpRegen, contains differentiated and functional RPE cells. The optimal approach for administering OpRegen to people with GA remains under investigation.

A Phase 1/2a, open-label, single-arm, multi-center, dose-escalating clinical trial assessing the safety and efficacy of a single OpRegen dose for the treatment of geographic atrophy (GA) was initiated. Patients with bilateral GA were divided into 4 cohorts: cohorts 1-3 included patients with best corrected visual acuity (BCVA) at or worse than 20/200 (i.e., legally blind) and GA area of 1.25 mm² to 17.00 mm². Cohort 4 included patients with vision from 20/64 to 20/250 and GA lesion area of 4.00 mm² to 11.00 mm². Cohort 1 dosed 50,000 cells and cohorts 2-4 dosed up to 200,000 cells. All patients underwent a perioperative immunosuppressive regimen. OpRegen was delivered through a subretinal injection to 17 patients and through a suprachoidal injection to 7. The primary endpoint was safety and tolerability of OpRegen, and secondary endpoints included functional and anatomic assessments.

The most frequent ocular AEs observed in the study included conjunctival hemorrhage/hyperemia in 71% of patients and epiretinal membrane formation in 67% of patients. Most of the AEs reported were mild (87%). There were no reported cases of OpRegen rejection, acute or delayed intraocular inflammation, sustained intraocular pressure increase, or AE-based discontinuation of the study.

Among those in cohort 4, BCVA gains in study eyes were sustained through 24 months, with a mean gain from baseline of 7.6 letters at month 12 and a mean gain of 5.5 letters in study eyes at month 24. (5 letters is equivalent to 1 line on an eye chart.)

These anatomic findings loosely correlated with functional changes. Patients in cohort 4 with extensive bleb (treatment) coverage gained a mean of 12.8 letters and 7.4 letters from baseline, respectively, at months 12 and 24. Those with limited bleb coverage gained a mean of 3.9 letters and 3.7 letters from baseline, respectively, at months 12 and 24.

SESSION 6: CLINICAL TRIAL DESIGN: DELIVERY, ENDPOINTS AND REGULATORY

AAV Delivery to the Subretinal Space of Adult and Juvenile Canine Eyes Using the Orbit™ Subretinal Delivery System (OSDS™) and Prototypes
Dr. William Beltran, University of Pennsylvania, School of Veterinary Medicine

Improvements in instrumentation may unlock new approaches to delivering therapy into the subretinal space, leading to optimized dosing, less invasive procedures, and more precise placement of therapeutic payloads. The current class of subretinal injectors used for transvitreal approaches rely on similar designs (i.e., needles and retractable cannulas of different sizes). Surgeons are familiar with these instruments, and engineers successfully integrated them into modern vitrectomy platforms.

Still, transvitreal approaches with subretinal injectors have their limitations, chief among them reflux of therapy into the vitreous cavity that leads to increased risk of immune-mediated responses and risk of epiretinal membrane formation. Additionally, 3-port pars plana vitrectomy (removal of vitreous) is an invasive approach that carries known (but important) risks, and the creation of retinotomy (retinal incision) during transvitreal approaches could lead to potential complications.

The Orbit Subretinal Delivery System (OSDS) from Johnson & Johnson is an innovation that has improved upon the designs of current-generation subretinal injectors. The principal design is the same (i.e., a retractable needle at the end of a cannula), but innovations such as a curved 35-gauge needle and a knob that enables a surgeon to manipulate the cannula are important innovations that distinguish this instrument from other devices. Importantly, two syringes can connect to the OSDS: one that allows a balanced salt solution (BSS) for bleb initiation, and another for delivering a therapeutic bolus.

To use the OSDS, a surgeon creates a sclerotomy (incision in the sclera) and inserts the cannula into the suprachoroidal space. The choroid is the vasculature underneath the retina. With suprachoroidal injection, retinotomies and vitrectomies are avoided. A graduated ribbon enables the surgeon to track how far they have inserted the cannula, which can be visualized directly via the operating microscope. By turning the OSDS’s knob, the surgeon engages the curved 35-gauge needle, which penetrates Bruch’s membrane and the retinal pigment epithelium (RPE) until the subretinal space is reached. A small volume of BSS is injected. Once a bleb is created and visualized, the remaining therapy is administered. 

Approved iterations of the OSDS injectors and four OSDS prototypes were recently evaluated in canine models, thanks in part to a grant from Foundation Fighting Blindness. Canine models aged 10 to 72 weeks were employed in this research. Injection of an AAV5 vector was successful in 13 of 14 eyes.

Improvements to instrumentation for animal models may bear multiple fruits. For now, researchers have a clearer answer as to which OSDS models are best suited for canine surgery, which could lead to improved research information for human retinal disease. While it remains to be seen whether the prototypes used in this research are appropriate for human use, this assessment deepens engineers’ understanding of OSDS device designs and dynamics.

Fully Automated Patient-Tailored Microperimetry: A First Orbit Study Report
Dr. Hendrik Scholl, University of Basel

Mutations in ABCA4, which most frequently lead to Stargardt disease), are the most common cause of inherited retinal disease, and are characterized by early lipofuscin (toxic deposits) accumulation and the development of atrophy. Investigative treatments are mutation agnostic and target the visual cycle, lipofuscin accumulation, or RPE cell metabolism. Investigators have two large-scale disease registries (the ProgStar database and the National Eye Institute ABCA4 protocol) at their disposal, which empower researchers to design effective clinical trials.

ABCA4 retinopathy may result in various phenotypes, ranging from isolated macular dystrophy to cone dystrophy to rod-cone dystrophy. A variant-specific gene editing from the Institute of Molecular and Clinical Ophthalmology Basel (IOB), called the G1961E gene-editing program, may unlock solutions to treating one form of Stargardt disease. However, a variant-specific natural history study must first be conducted to recruit for clinical trials.

Technical challenges in assessing this population’s natural history have recently been addressed by researchers. Means of measuring visual function in Stargardt disease patients include standard microperimetry patterns (points of retinal sensitivity) used to assess inherited retinal dystrophies. However, current microperimetry approaches can’t measure changes over time in a meaningful way.

Patient-tailored microperimetry approaches attempting scotoma (blind spot) boundary–based imaging could work, but they limit analysis to test points within areas at risk of disease progression and do not maximize the number of relevant test points. Improved patient-tailored approaches maximize the number of relevant test points and keeps the overall number of test points in check, but require a technician to adjust the microperimetry grid based on patient fixation, leaving the protocol vulnerable to inaccuracy.

A new, fully automated patient-tailored perimetry system could be the solution to efficient and effective microperimetry capture. First, patients sit for a baseline high-resolution OCT scan, and signs of atrophy are automatically segmented via an AI algorithm. Next, a patient-tailored stimulus grid is automatically generated, after which precise alignment and an optimized staircase strategy are run by the perimeter test with external software. In this automated process, technician performance is no longer a source of error.

Armed with a new automated system for assessing visual function via microperimetry, researchers at the IOB initiated the First Orbit natural history study. The study will, in part, assess the validity of the automated patient-tailored microperimetry system described above. Patients in the study are adults with G1961E-associated Stargardt disease and do not have an ocular history other than cataract surgery. Visual function assessments include visual acuity and contrast sensitivity testing, as well as light- and dark-adapted fundus controlled perimetry testing.

First Orbit is a 2-year study. So far, 20 patients (median age 29.8 years) have been enrolled. The findings from First Orbit are encouraging. The automatic patient-tailored microperimetry system was shown to be highly repeatable, supporting the overall reliability of this workflow.

First Orbit will continue to assess patients out to 2 years. Researchers hope data from this natural history study will guide future clinical trials assessing the safety and efficacy of G1961E gene editing from IOB.

Natural History of Retinitis Pigmentosa Based on Genotype, Vitamin A/E Supplementation, and an Electroretinogram Biomarker
Dr. Jason Comander, Massachusetts Eye and Ear

A clinical trial funded by the Foundation Fighting Blindness that took place between 1984 and 1991 sought to understand whether taking vitamin A and vitamin E affected the progression of retinitis pigmentosa (RP). This large study enrolled 601 patients and followed them for up to 6 years,using ERG to measure disease progression. The study’s conclusion that vitamin A supplementation modestly slowed progression of RP was met with skepticism and criticism. The researchers also concluded that vitamin E supplementation accelerated progression of RP.

The population in this study was selected before the genetic causes of RP were known. Researchers investigating RP felt that this dataset was worthy of deeper exploration and hypothesized that the potential presence and degree of protection from RP progression offered by vitamin A could vary based on genotype—to wit, that some genotypes would experience an outsized benefit while other genotypes would experience little (if any) benefit.

Funded by the Foundation, a team of researchers used the trial data and DNA samples from the original study. Of those samples, 94% were successfully sequenced, and the genetic solution rate was 77% of sequenced samples. With these data, researchers hoped to repeat the original analysis to assess the effects of vitamin A and vitamin E on patients with RP, but this time including genotype.

When researchers re-ran the initial statistical analysis as a means of replicating the original data, they were surprised to find that while vitamin E continued to show a negative effect, vitamin A showed no protective effect for a specific genotype or the study group as a whole. The researchers questioned their own methodology to repeat the statistical analysis. But after running the original analysis dozens of times, they kept reaching the same conclusion.

Researchers then turned to the data pool itself. They found that the original dataset included data from years 5 and 6 of follow-up with original patients—data that was not included in the original data set, as it was collected after a data lock.

A re-evaluation of the original study’s randomization revealed that the patient groups were unbalanced for baseline ERGs, accounting for approximately 1/3 of the treatment effect observed following Vitamin A supplementation. After adjusting for this imbalance, researchers were unable to conclude that vitamin A supplementation was protective against disease progression in RP patients regardless of genotype. They still concluded that vitamin E supplementation should be avoided.

Overall, this reevaluation of established research—powered, in part, by technology that was unavailable at the time of the initial study—provides insights on the relationship between RP and vitamin supplementation, and has unlocked a new framework to understanding RP based on genotyping and on cutting-edge modalities that assess efficacy and cone function.

Navigating the Regulatory Pathways to Advance Gene Therapies for Retinal Degenerative Diseases
Dr. Samarendra Mohanty, Nanoscope Therapeutics

Late-stage retinal degeneration linked to genetic mutations is characterized by the loss of rods and cones. Most gene replacement and gene editing programs seek to intervene before patients reach late-stage retinal degeneration.

The team at Nanoscope Therapeutics, however, hopes to restore visual function via an optogenetics approach in retinal tissue that has late-stage degeneration. In this approach, gene therapy delivers a light-sensitive, multi-characteristic opsin (MCO), a transgene for high-performing opsin. MCO is designed to enable surviving bipolar cells to respond to light in people who have lost all or most of their rods and cones.

MCO therapy, which is delivered via an AAV2 vector, carries many benefits: only a single injection is required, no high-intensity stimulation is needed, and cells become sensitized to all visible light and low light levels. MCO treatment is gene agnostic, and patients with a wide variety of degenerative retinal diseases could find MCO therapy effective.

Establishing reliable and reproduceable manufacturing processes early in a drug’s investigatory timeline is key to meeting regulatory standards. Companies developing gene therapies must stay in contact with regulators to ensure that no surprises arise during the clinical research phase.

For the Phase 2b RESTORE study, which assessed MCO-010 for the treatment of retinitis pigmentosa (RP), Nanoscope Therapeutics designed novel functional vision endpoints  to adequately measure differences between active and control groups. These endpoints include multi-luminance Y-mobility testing (which evaluates a patient’s ability to navigate obstacles at multiple light levels) and multi-luminance shape discrimination testing (which evaluates a patient’s ability to discern the shapes of geometric blocks at multiple light levels). Vision endpoints were measured via the validated Freiburg Visual Acuity Test (FrACT), a validated means of assessing subjective function in patients whose vision is so low that logMAR testing is ineffective.

Patients with RP in RESTORE received low-dose or high-dose MCO-010 or sham. Compared with sham-treated patients at 52 and 76 weeks, patients in the treatment arms experienced statistically significant improvements from baseline in best corrected visual acuity. The company plans to file a biologic license application (BLA) with the US FDA to make the therapy available to patients with advanced RP.

Trial Design and Rationale Behind jCyte Study JC-03: A Randomized, Masked, Placebo-controlled, Adaptive Phase 2/3 Trial of the Efficacy and Safety of a Single Intravitreal Injection of Famzeretcel
Dr. John Pollack, jCyte

Without a previously blazed trail to follow, researchers exploring new approaches to treat rare diseases must accept the adage that “they don’t know what they don’t know.” As such, they must design clinical trials with broad enrollment criteria so that they can learn as much as possible about the relationship between a potential treatment and its interaction with a rare disease. Results from these clinical trials can inform the structure of a pivotal study that will satisfy regulators’ requirements for demonstrating efficacy.

The retinal progenitor cell therapy famzeretcel from jCyte faces this exact situation. A famzeretcel dose contains millions of retinal progenitor cells, which are multipotent stem cells that produce neurotrophic factors to improve cone function and reduce cone cell death. The therapy is delivered via intravitreal injection, is gene agnostic, and appears to be durable up to 12 months. The Phase 2b clinical trial for the treatment of retinitis pigmentosa (RP) is completed.

The inclusion criteria for this study were broad. Adult patients with RP, regardless of genotype, and best corrected visual acuity (BCVA) 20/80 to 20/800 were enrolled, so long as they did not have any other visually significant eye disease. This study—the largest ever for a new medication to treat RP—was designed to inform a planned pivotal trial. Patients (N = 83) were randomly assigned to high-dose famzeretcel, low-dose famzeretcel, or sham. Overall, 96% of patients (80 of 83) stayed in the study for 12 months, which is when the primary endpoint of mean change in BCVA from baseline was assessed.

Because this phase 2b study would inform a future pivotal trial, researchers performed test-retest variability for endpoint measurements and performed a separate endpoint validation study. They found that two factors increased test-retest variability in BCVA assessments: reduced central fixation reliability and interocular BCVA disparities of more than 15 letters. A closer look at the latter factor revealed that among patients with large interocular BCVA disparities, the range of letters within which 95% of patients fell was nearly 18 letters, whereas in patients with smaller interocular BCVA disparities (ie, ≤15 letters), the range was approximately 5.6 letters. This led researchers to conclude that they could reliably measure a difference of at least 6 letters using BCVA if interocular BCVA disparities were less than 15 letters.

One surprising finding from the phase 2b study assessing famzeretcel for RP: some sham patients with less than 15 letters of interocular BCVA disparity experienced small-but-noteworthy BCVA gains from baseline at 12 months. It is believed that this is the result of a neurologic-ophthalmic phenomenon documented in the literature wherein patients with bilateral retinal degenerative disease will experience spontaneous and variable improvements in untreated eyes.

At 1 year among patients with interocular BCVA disparities of less than 15 letters, 10% of sham-treated eyes experienced gains of at least 10 letters, and 5% of eyes experienced gains of at least 13 letters. No sham patients experienced gains of more than 15 letters. Among patients who received high-dose famzeretcel, 50%, 38%, 31%, and 19% gained from baseline at least 10 letters, at least 13 letters, at least 15 letters, and at least 20 letters, respectively, at 1 year. These differences were statistically significant across all groupings.

Study investigators used these BCVA findings to guide the design of a forthcoming phase 2/3 pivotal study, and will enroll patients with interocular BCVA disparities of less than 15 letters. Other findings that will inform that trial include enrolling patients with at least a 12° central visual field (who have rod-cone ratios optimized for greater potential BCVA and contrast sensitivity improvements) and patients with at central subfield thicknesses of at least 130 µm (which indicates a level of cone survival high enough for famzeretcel to have a therapeutic effect).

Patients in the phase 2/3 pivotal study assessing famzeretcel for RP will be randomly assigned to high-dose famzeretcel, low-dose famzeretcel, or placebo control; researchers are prepared to enroll a second group of patients if an adaptive analysis at 3 months hits predefined benchmarks indicative of an underpowered study. The primary endpoint will be BCVA gain of at least 15 letters at 12 months; secondary endpoints will include a range of structural (eg, gains ≥10 letters) and functional endpoints (eg, contrast sensitivity responder analysis).