From living RPE monolayers to wireless implants and gene-agnostic optogenetics, Monday’s ARVO-cosponsored symposium focused on therapies built to restore usable vision, not just slow the slide.
Orlando trades on sunshine and spectacle, and this program brought both to the retina. In the Therapeutic Approaches to Vision Restoration symposium, chair Dr. Wei Li (USA) convened four teams advancing different routes to the same goal: rebuilding the RPE with living monolayers, converting images to current with a photovoltaic chip, sensitizing surviving retinal circuits with optogenetics and replacing lost photoreceptors with self-powered nanowire arrays.
Covering both living tissues and engineered hardware, the session felt more Space Coast than theme park, pairing careful surgical playbooks with maturing safety data and patient-level gains that suggest these ideas are leaving the launchpad and taking flight, one carefully chosen eye at a time.
READ MORE: Rewiring Sight with Nanowires: Vision Restored and Infrared Detected in Primate Models
RPE Comeback Plan
Starting the session, Dr. Amir H. Kashani (USA) walked us through the long road from early autologous retinal pigment epithelium (RPE) grafts that triggered proliferative vitreoretinopathy (PVR) to today’s engineered monolayers on defined scaffolds for dry age-related macular degeneration (AMD). He noted that intravitreal cell suspensions were not effective. Subretinal suspensions produced scattered pigment, which led teams to pivot toward true sheets and more controlled delivery. “This truly requires cell-based therapy to restore structure and hopefully function,” he said.
Modern trials place differentiated RPE as a continuous layer in a hydrodissected pocket under intraoperative OCT guidance.1,2 Safety has been steady, with no tumor signal, low inflammation and good implant stability. In a California series, surgical refinements reduced early hemorrhage and vision trends were encouraging: “Of 15 patients, up to 27% improved by at least one line of visual acuity at three years; 60% had stable or improved vision, while unimplanted eyes declined as expected.”1
Donor eye pathology two years post-implantation showed pigmented, polarized RPE expressing RPE65 with rhodopsin co-localization, a sign that phagocytic function may be present. Next-generation work uses a biodegradable Poly(lactic-co-glycolic acid) (PLGA) scaffold to host a single-cell monolayer and targets geographic atrophy (GA) borders where photoreceptors remain. The field still chases perfect coverage and durability, but the direction is set. Build the layer, keep it alive, let the retina do the rest.
READ MORE: Retina Roundup at AAO 2025: Fresh data, Familiar Challenges and the Future of Eye Care
Photovoltaic pixels
Dr. Ralf Hornig (Germany) presented PRIMA, a subretinal photovoltaic implant for central vision loss from GA. Smart glasses capture the scene, an on-body processor prepares the signal and near-infrared light projects it onto a 2 × 2 mm chip that is 30 μm thick. The chip holds 378 pixels, each a small solar cell that stimulates inner retinal neurons. As he summarized, “Unlike previous visual prostheses, PRIMA provides shape perception—patients draw and recognize letters as expected.”
The surgical steps will be familiar to vitreoretinal teams: create a subretinal bleb, deliver the implant through a retinotomy, flatten with perfluorocarbon and position the device before gas. Rehabilitation begins around week four. In the PRIMAVERA pivotal study of 38 eyes, peripheral native vision stayed stable while central function improved with the implant.3 “Mean visual acuity improvement at 12 months was 25.5 letters; one patient gained 59 letters.” Many participants now use the system at home for labels, manuals, cards and low-vision books. Safety findings tracked mainly to surgery rather than to the device.
Next steps include tighter pixel spacing, a wider field and slimmer glasses with an integrated projector. He advised moving forward in a practical direction by keeping surgery familiar, maintaining structured rehab and continuously refining the hardware.
Turning bipolar cells into light catchers
Dr. Samarendra Mohanty (USA) and Dr. Stephen Tsang (USA) pitched a gene-agnostic optogenetic strategy for advanced retinitis pigmentosa (RP) and Stargardt disease. Nanoscope’s multi-characteristic opsin (MCO) platform uses an adeno-associated virus (AAV) vector to transduce more than 70% of bipolar cells with a broad-spectrum opsin. The response spans 400–700 nm with four orders of dynamic range, so patients do not need high-intensity light or scanning systems. As Dr. Mohanty explained, “Our goal is a disease-agnostic and gene-agnostic optogenetic therapy,” a practical way to step around the genetic diversity of inherited retinal diseases (IRDs).
Targeting bipolar cells preserves upstream retinal processing and increases potential “pixel count” compared with ganglion-cell approaches. Early studies back the choice. In a dose-ranging phase 1–2 trial for RP, the higher dose produced about a half-logMAR gain and informed a phase 2b design that reached significance at weeks 52 and 76, with durability out to three years and a clean safety profile.4,5
Dr. Tsang shared data from a six-patient phase 1–2 Stargardt cohort, where macula-predominant phenotypes appeared to respond best. Several eyes gained double-digit ETDRS letters and selected patients reached five lines with a single intravitreal injection. As he noted, “Some achieved up to five lines improvement in acuity—about 32 ETDRS letters—with a single intravitreal injection.”6
READ MORE: Foundation Fighting Blindness to Host a Mental Health Webinar for Eye Care Professionals
Self-powered arrays
Presenting work from Dr. Y. Zhang, Dr. Chunhui Jiang (China) described a subretinal nanowire array that behaves like a dense bed of artificial photoreceptors. The device uses gold–titanium dioxide nanowires about 1 nm in diameter and 2.5 μm long, packed at a higher density than native photoreceptors. The design generates current directly in response to light, so there is no belt pack, cable or goggles. As Dr. Jiang put it, “The arrays are self powered, with no external power or goggles required.”
In macaques, implants were stable at one year without detachment or proliferative responses, and animals completed dim-light tasks. Early human data from a China-based study add cautious optimism. Two long-blind patients received a 2 × 2 mm, 200 μm thick chip using standard subretinal delivery. Postoperative inflammation was mild and transient. Imaging confirmed stable placement, steady intraocular pressure and resolution of early edema by month three.7
Functionally, both patients improved from no light perception to consistent stimulus detection with a blue-leaning response profile. They followed floor cues in navigation tasks and screen-based discrimination improved over days. “Patient 1 reached 100% accuracy on shape and direction detection within days; Patient 2 improved from 30% to 100% over a week.” Daily-life gains varied, but the signal supports continued enrollment as the team refines materials and surgical steps.
READ MORE: From Latency to Lineages: Reading Viral Signals in the Eye at AAO 2025
Reimagining sight in the Sunshine State
What unfolded Monday morning in Orlando felt less like a gadget parade and more like a blueprint for rebuilding vision. Stem cell sheets that behave like RPE, photovoltaic chips that sketch letters, optogenetic switches that recruit bipolar cells and self-powered arrays that remove the need for external gear, all pointed in the same direction.
The hurdles are real in manufacturing, durability, rehabilitation and equitable access. Yet the signal is getting stronger across models, trials and early patient stories. Florida knows a good launch when it sees one. After this morning, it is hard not to believe that some of these approaches are already clearing the tower.
Editor’s Note: The American Academy of Ophthalmology Annual Meeting 2025 (AAO 2025) was held October 17-20, 2025, in Orlando, Florida. Reporting for this story took place during the event. This content is intended exclusively for healthcare professionals. It is not intended for the general public. Products or therapies discussed may not be registered or approved in all jurisdictions, including Singapore.
References
- Humayun MS, Clegg DO, Dayan MS, et al. Long-term follow-up of a phase 1/2a clinical trial of a stem cell-derived bioengineered retinal pigment epithelium implant for geographic atrophy. Ophthalmology. 2024;131(6):682-691.
- Kashani AH, Lebkowski JS, Rahhal FM, et al. One-year follow-up in a phase 1/2a clinical trial of an allogeneic RPE cell bioengineered implant for advanced dry age-related macular degeneration. Transl Vis Sci Technol. 2021;10(10):13.
- Holz FG, Le Mer Y, Muqit MMK, et al. Subretinal photovoltaic implant to restore vision in geographic atrophy due to AMD. N Engl J Med. 2025 Oct 20.
- Nanoscope Therapeutics. Nanoscope Therapeutics announces positive top-line results from randomized controlled trial of MCO-010 for retinitis pigmentosa. Available from: https://nanostherapeutics.com/2024/03/26/nanoscope-therapeutics-announces-top-line-results-from-ph2-trial-of-mco-010-for-retinitis-pigmentosa/ Accessed on October 21, 2025.
- Nanoscope Therapeutics. Nanoscope presented positive 2-year randomized, controlled trial results of MCO-010 for retinitis pigmentosa. https://nanostherapeutics.com/2024/10/31/nanoscope-presented-positive-2-year-randomized-controlled-trial-results-of-mco-010-for-retinitis-pigmentosa/ Accessed on October 21, 2025.
- Lam BL, Zak V, Gonzalez VH, et al. Safety and efficacy of MCO-010 optogenetic therapy in patients with Stargardt disease in USA (STARLIGHT): an open-label multi-center Ph2 trial. EClinicalMedicine. 2025;87:103430.
- Yang R, Zhao P, Wang L, et al. Assessment of visual function in blind mice and monkeys with subretinally implanted nanowire arrays as artificial photoreceptors. Nat Biomed Eng. 2024;8(8):1018-1039.