PRIMA Eye Implant Gets CE Mark, but Vision Still Depends on Camera Glasses

Science Corporation’s PRIMA retinal prosthesis has moved beyond preliminary trial claims: peer-reviewed results were published in October 2025, and the system received a European CE mark in July 2026. The authorization permits commercial use for severe central vision loss caused by geographic atrophy from age-related macular degeneration, although reimbursement and treatment-site preparations still stand between regulatory clearance and routine access.

The central finding remains significant but narrower than the phrase “restored vision” may suggest. PRIMA produced measurable prosthetic central vision in many trial participants, while the surrounding natural peripheral vision was preserved; it did not recreate normal sight or work as a self-contained chip.

What the European clearance changes

The CE mark allows Science Corporation to market PRIMA in covered European countries. According to TechCrunch’s July 2026 account of the clearance, the company was preparing an initial rollout in Germany while discussing reimbursement with health-care providers.

That distinction matters for prospective patients. Regulatory clearance does not mean the implant is immediately available through every retinal clinic, covered by every insurer or suitable for every cause of blindness. Implantation requires specialist surgery, followed by fitting and training with the external equipment.

The European decision also does not constitute approval in the United States. PRIMA has received FDA regulatory designations intended to facilitate development and review, but those designations are not market authorization. Until the FDA completes the applicable review pathway, the system remains investigational in the US.

What the 38-person study actually demonstrated

PRIMAvera was an open-label, prospective, single-group study involving 38 people with geographic atrophy caused by age-related macular degeneration. Participants had profound central visual impairment, but the retinal cells needed to transmit signals beyond the damaged photoreceptors remained functional.

The peer-reviewed PRIMAvera report states that 32 participants completed the 12-month visual assessment. Of those, 26—81% of the assessed group—achieved the prespecified clinically meaningful improvement of at least 0.2 logMAR when using the system. An analysis accounting for the six missing 12-month assessments estimated a corresponding result of 80% for the full implanted group.

These percentages should not be interpreted as 80% regaining ordinary eyesight. The test measured improvement in visual acuity delivered by the complete PRIMA system compared with baseline, and participants could use its magnification capability. The practical achievement was the recovery of limited central form vision sufficient for tasks such as identifying letters, numbers and words, not restoration of the eye’s original photoreceptors.

The study also lacked a randomized control group. Its strongest comparison was therefore within each participant: prosthetic performance with the system versus baseline and natural vision without it. That design can demonstrate that the apparatus generated usable visual information, but it provides less evidence than a randomized trial for comparing PRIMA with another intervention or rehabilitation program.

The implant is only one part of the visual prosthesis

PRIMA works because the disease targeted in the trial destroys photoreceptors in the macula while leaving parts of the downstream retinal network and optic pathway available. A photovoltaic microarray is placed beneath the damaged central retina, where it bypasses the missing light-sensitive cells and electrically stimulates surviving retinal neurons.

The external glasses capture a scene with a camera. A processor prepares the image, and a projector in the glasses sends patterned near-infrared light into the eye. The implant converts that light into electrical stimulation, so the same optical signal supplies both visual information and power without a cable passing through the eye.

This architecture explains both the achievement and the limitation. The implanted array does not independently see the environment: useful prosthetic vision depends on the camera, projector, processor and correct alignment with the chip. Digital magnification can make letters easier to distinguish, but enlarging an image reduces how much of the scene fits within the available field of view.

The system also adds new visual information inside the central blind area rather than replacing the person’s remaining peripheral sight. In the trial, mean natural peripheral visual acuity after implantation was equivalent to baseline, an important safety finding for people who rely heavily on that residual vision.

Surgery and adverse events remain part of the calculation

PRIMA requires vitreoretinal surgery beneath an already fragile macula. The study recorded 26 serious adverse events among 19 participants; 21 occurred within two months of surgery, and 20 of those early events resolved within two months of onset. The concentration of events around the operation underscores that this is an invasive medical device, not a wearable visual aid.

Six implanted participants did not complete the 12-month assessment: three died, one withdrew and two were unavailable for testing. The published analysis used statistical imputation for those missing outcomes, but the directly observed 81% result applies specifically to the 32 people who underwent the final assessment.

Eligibility is consequently much narrower than “blind patients.” The pivotal evidence concerns profound central loss from geographic atrophy secondary to age-related macular degeneration. It does not establish that the same system can restore vision when the optic nerve is absent, retinal ganglion cells are extensively damaged or blindness has a different neurological cause.

Research is expanding, but the evidence cannot yet be transferred

Science Corporation is now studying whether the platform can help people with other forms of photoreceptor degeneration. A current ClinicalTrials.gov record for PRIMAlia lists a recruiting study that includes inherited macular disorders such as Stargardt disease and retinitis pigmentosa.

That expansion is not evidence that PRIMA already works for those conditions. Different retinal diseases damage different cells, progress at different rates and may leave different amounts of usable neural circuitry. Results from geographic atrophy should therefore remain attached to the population actually studied until the newer trial supplies its own safety and performance data.

The meaningful update is now regulatory as well as scientific: PRIMA has peer-reviewed 12-month evidence and European market clearance. Its real-world impact will depend on surgical capacity, rehabilitation, reimbursement and whether patients find the limited, equipment-dependent central image valuable enough to justify the procedure.

Also read:

• VLMs Can Already Hunt for the “Interesting.” But They’re Still Bad at Walking Away from What They’ve Found
• iPhone 17 Pro Delivered Cooling & Camera Breakthroughs – But iPhone 18 Series Promises Foldables, Variable Apertures & Price Hikes
• Gamma Imagine Makes Social Graphics Fast—but Exports Still Need Checking