A tiny wireless chip under the retina, paired with special glasses, helped most people in a major trial read again after advanced macular degeneration had wiped out their central vision.
The device is called PRIMA — short for photovoltaic retina implant microarray. In results published October 20, 2025, in the New England Journal of Medicine, a Stanford Medicine–led European trial (PRIMAvera) showed that 26 of 32 patients assessed at one year had a clinically meaningful gain in visual acuity, and 27 could read. With zoom and contrast tools built into the glasses, some reached acuity roughly equivalent to 20/42.
As an Amazon Associate I earn from qualifying purchases. If you buy through links on this page, I may earn a commission at no extra cost to you.
What geographic atrophy takes away
Age-related macular degeneration (AMD) damages the macula — the small central patch of the retina used for reading, faces, and fine detail. In the advanced “dry” form known as geographic atrophy, light-sensing photoreceptor cells in that center die off over years. Peripheral vision often remains, but the middle of the visual field becomes a blank or distorted zone.
More than 5 million people worldwide live with geographic atrophy, according to the trial’s background and Stanford Medicine’s report. Until now, no approved therapy restored useful central vision once those photoreceptors were gone. Drugs and lifestyle steps can sometimes slow related disease, but they do not put reading vision back.
What often remains, even in advanced atrophy, are the retinal neurons that normally relay signals from photoreceptors toward the optic nerve and brain. PRIMA is designed to talk to those remaining cells.
How the chip and glasses work together
PRIMA has two main parts. A camera on a pair of glasses captures the scene in front of the wearer. That image is projected, in near-infrared light, onto a wireless chip about 2 by 2 millimeters wide that surgeons place under the retina where photoreceptors have died.
The chip is photovoltaic: each of its tiny pixels turns the projected infrared light into a small electrical pulse. Those pulses stimulate the surviving retinal neurons in the pattern of the camera image. Infrared is used on purpose so the implant’s signal does not confuse the patient’s remaining natural photoreceptors outside the atrophic patch.
Because the chip makes its own current from light, it needs no cable out of the eye and no implanted battery. Patients can keep using their natural peripheral vision at the same time as the prosthetic central image — something earlier wired eye prostheses struggled to offer cleanly. Palanker has stressed that merging those two streams helps with orientation and navigation, not only with reading a chart in a clinic.
Daniel Palanker, PhD, of Stanford Medicine, who co-led the work and spent roughly two decades developing the approach, put the milestone this way: previous prostheses mostly restored light sensitivity, not true form vision — the ability to see shapes and patterns. “We are the first to provide form vision,” he said in Stanford’s announcement.
What the PRIMAvera trial measured
PRIMAvera (ClinicalTrials.gov NCT04676854) was an open-label, multicenter European study sponsored by Science Corporation. Lead author Frank Holz, MD, of the University of Bonn, and co-senior author José-Alain Sahel, MD, of the University of Pittsburgh, joined Palanker and collaborators at sites including Moorfields Eye Hospital in London.
Thirty-eight people older than 60 with geographic atrophy due to AMD and vision worse than about 20/320 in the study eye received an implant. Training with the glasses typically began four to five weeks after surgery. Performance improved over months of practice — Palanker compared that learning curve to cochlear implants.
At 12 months, 32 participants were assessed (three had died, one withdrew, two were unavailable). Among those 32:
- 26 of 32 (81%) gained a clinically meaningful improvement in visual acuity, defined as at least 0.2 logMAR — roughly two extra lines on a standard eye chart.
- Mean improvement was about five lines; one participant gained twelve lines (Stanford Medicine news).
- 27 could read with the system.
- With digital zoom (up to about 12×) and contrast controls on the glasses, some reached acuity equivalent to roughly 20/42.
Participants used the device in daily life for books, food labels, and subway signs. About two-thirds reported medium to high satisfaction. Importantly, natural peripheral acuity stayed similar to baseline after implantation — the surgery did not trade away the side vision patients still had.
Side effects and safety context
Nineteen participants had serious adverse events related to the procedure or device path — including ocular hypertension (high pressure in the eye), peripheral retinal tears, and subretinal hemorrhage (bleeding under the retina). None were life-threatening. Most clustered early after surgery, and almost all resolved within about two months, according to Stanford Medicine and the NEJM report.
That is still surgery inside the eye. The trial shows a clear benefit for many carefully selected patients, not a casual outpatient gadget. Moorfields surgeons involved in the UK arm have emphasized that trained vitreoretinal teams can implant the chip in under two hours, which matters if the therapy is eventually offered more widely after regulatory review.
Black-and-white for now — denser chips next
Today’s PRIMA image is black-and-white, without intermediate gray shades. Palanker’s group is working on software for full grayscale — patients rank face recognition just behind reading on their wish list, and faces need gray levels.
Resolution today is set by 100-micrometer pixels (378 pixels on the 2×2 mm chip). A next-generation design already tested in rats aims for pixels around 20 micrometers and on the order of 10,000 pixels per chip. Palanker has said a 20-micrometer chip could support roughly 20/80 native prosthetic acuity, and with electronic zoom approach something closer to 20/20 for reading tasks.
Researchers also want to test the approach in other blindness caused by lost photoreceptors, not only geographic atrophy from AMD.
What this does — and does not — mean yet
PRIMAvera is strong evidence that a wireless subretinal photovoltaic implant can restore form vision and reading in people who had essentially no useful central sight. It is not, by itself, a supermarket cure or an immediate U.S. clinic appointment for everyone with AMD. Availability depends on regulators, manufacturing, surgeon training, and long-term follow-up. Science Corporation is developing the commercial system; the NEJM paper and the Stanford and Moorfields explainers are the place to start for primary facts.
The trial also enrolled a narrow group: older adults with geographic atrophy in both eyes, very poor acuity in the study eye, and an atrophic patch large enough for the implant. Results should not be stretched to early AMD, wet AMD with active leakage, or other eye diseases without new studies. Even within the eligible group, people needed months of rehab to get the most from the glasses — much like learning to hear with a cochlear implant.
For readers living with macular disease, the practical takeaway is still straightforward: after decades of “light perception only” prostheses, a chip the size of a few grains of rice, plus glasses that beam infrared patterns, has let most trial patients read words again. That is a real shift — measured in eye-chart lines, daily tasks, and patient-reported use — not just a lab curiosity.
Further reading
The Eye Book: A Complete Guide to Eye Disorders and Health — Clear, evidence-based guide to eye exams, common disorders, macular degeneration, and low vision — written for patients and non-specialists.
Macular Degeneration: The Complete Guide to Saving and Maximizing Your Sight — Practical overview of AMD, risk factors, nutrition, low-vision strategies, and living well when central vision fades.