Retinal Implant for Geographic Atrophy Shows Evidence of Durable Control

NEW ORLEANS — A wireless retinal implant provided durable control of geographic atrophy (GA) for up to 3 years, according to a study reported here.

The improvement in visual acuity at 36 months with the PRIMA system averaged -0.59 logMAR (a gain of about 30 letters), representing an improvement of about 5 letters from the 12-month results. More than 80% of patients reported ability to read letters, numbers, and words at 12 months, and between 24 and 36 months, half of the patients reported ability to perform more advanced tasks, such as using a smart phone, reading a menu, and playing cards. Without device assistance, vision in the eye with the implant remained stable, whereas the untreated eye lost about 1.4 lines of vision.

Serious adverse events (SAEs), such as hypertony, retinal tear, macular hole, and subretinal hemorrhage, were uncommon after the first 6 months and did not affect natural vision, reported Ehsan Rahimy, MD, of Stanford Medicine in Palo Alto, California, at the American Academy of Ophthalmology (AAO) annual meeting.

“The PRIMA system produced durable, clinically meaningful restoration in central vision in patients with geographic atrophy, extending through 36 months,” said Rahimy. “The safety profile is tolerable and manageable. Importantly, adverse events are front loaded, as would be expected after surgery. Almost equally important, having an SAE does not preclude a patient from potentially experiencing significant restoration of central vision.”

The device was approved in Europe in July, and in response to a question during a discussion, Rahimy said he had no data on the number of units sold in Europe or possible safety issues with the device. Manufacturer Science Corporation is on track to seek FDA approval, he said.

Joan Miller, MD, of Mass General Brigham in Boston, said, “I was probably a naysayer about any kind of implant for GA. It just somehow didn’t make sense for me, and you’ve proved me wrong. I think it’s exciting to see this.”

Nonetheless, she noted that seven patients died during the study, although none were considered related to the implant or surgery.

“It just seemed like there were a lot of deaths, and they were not that old [mean age 78.9],” said Miller, who was not involved in the study. “It might be good to have some sort of comparator to know what might have been seen in that population. Not something to worry about, but it seems a little high.”

Rahimy responded that all of the patients had bilateral, end-stage GA and were older than the typical patient considered for GA treatment. In response to another question, he said that four additional dropouts from the study involved patients who had either relocated or for other reasons could not continue follow-up.

“It’s not like you turn on the switch and they have vision,” he said. “There’s a very rigorous postoperative course that they have to go through in terms of training themselves to use these devices.”

PRIMA is a wireless photovoltaic subretinal implant that replaces lost photoreceptor function and restores central vision in areas of retinal atrophy while preserving natural peripheral vision, he explained.

The patient uses augmented-reality glasses that have a front-facing miniature camera that captures a live feed and transfers the information to a processor that converts visual data into pulsed near-infrared light. The light is transmitted through the pupil, absorbed by the implant, and converted into electrical current that stimulates adjacent bipolar cells and subsequently ganglion cells, mimicking the retina’s natural signaling pathway to the brain. The patient can control mode, brightness, and zoom.

Rahimy’s presentation built on the 12-month data reported last year at AAO. The PRIMAvera trial involved 38 patients with advanced GA, enrolled at 17 sites in Europe. Rahimy characterized the trial as a phase III-equivalent pivotal single-arm study. The primary endpoints were efficacy and safety. The efficacy endpoint was clinically meaningful improvement in visual acuity, defined as ≥0.2 logMAR.

Of the original 38 patients, 27 completed 36 months of follow-up. Mortality unrelated to the implant or procedure was the principal cause of attrition (seven patients), which Rahimy said was not unexpected, given the study population’s mean age of 79.

The updated results showed that visual acuity continued to improve, from 0.49 logMAR at 12 months to 0.58 at 36 months.

Throughout the 36 months, 29 SAEs occurred, 24 during the first 6 months. The most common were hypotony (six), subretinal hemorrhage (five, three during the first 6 months), minor retinal tear (four), and macular hole (three).

“Having an SAE did not compromise a patient’s natural vision,” said Rahimy. “When patients had an SAE versus those that did not, their natural vision remained stable. Additionally, SAEs did not impede or restrict the patient’s potential gains from the PRIMA implant, as 68% of patients who did have an SAE still experienced greater than 15 letters of visual gain with the device.”

Longitudinal photography showed no instances of implant migration beyond the GA lesion borders, and postmortem evaluation of two implant eyes showed no signs of significant fibrosis or chronic inflammation, he added.