A gene injection and goggles helped some blind patients sense objects

A one-time gene injection into the eye, used together with special camera goggles, was given safely to 10 people with an advanced inherited form of blindness and improved some of their vision, researchers reported on October 7, 2026, in the New England Journal of Medicine.¹ The study was led by scientists at the University of Pittsburgh School of Medicine with international collaborators.¹ Seven of the 10 patients could detect dimmer light afterward, and some got better at finding and reaching for objects while wearing the goggles. None regained normal sight or the ability to read.¹

What happened

The patients have retinitis pigmentosa, a condition that destroys the light-sensing cells in the retina, the thin layer of tissue at the back of the eye that sends images to the brain. It affects more than 1.5 million people worldwide. It usually starts with poor night vision, then side vision narrows, and it can end in severe vision loss or blindness.¹ Because the retina is part of the central nervous system, it has little ability to repair itself.²

All 10 participants were legally blind with little or no remaining vision.¹ Each received one injection into the worse-seeing eye. A modified virus carried instructions for making a light-sensitive protein called ChrimsonR into retinal ganglion cells, the nerve cells that pass visual signals toward the brain and that often survive the disease.³ This approach is called optogenetics: changing cells so they make proteins that respond to light.¹ STAT noted that optogenetics was just recognized with a Nobel Prize.⁴

The protein alone is not enough. Patients wear goggles with a built-in camera. A portable processor turns the camera’s view into patterns of light, and a small projector in the glasses shines those patterns back into the eye at the colors that switch on ChrimsonR.¹ The light is amber, and the goggles adjust to outside lighting, according to the lead researchers.³ The treatment has been developed under the name GS030 by the French company GenSight Biologics.⁵

What the study measured

The trial was mainly designed to check safety, but the team also tested light sensitivity and the ability to detect objects. Patients practiced with the goggles, and their untreated eye was covered during tests.³

• Dim light: the researchers measured the faintest light each person could see through the goggles before and after treatment. Seven improved, by anywhere from about 2 times to 62 times. Six reached roughly a fourfold gain or more, which the team counted as meaningful for patients.³
• Objects: eight people did tests with and without the goggles, trying to detect, locate and touch a notebook or a small box of staples, and to tell which way a bar was pointing. Four consistently did better at detecting the notebook with the goggles over sessions spanning about a year to a year and a half. Others showed no consistent improvement.³ The university’s announcement described the four as improving across multiple real-world visual tasks over months to years.¹
• Brain signals: using EEG, a test that records brain activity through the scalp, a team led by Marlene Behrmann found evidence that visual signals reached the brain’s vision center when participants looked at objects.¹ These recordings were made in five selected participants, four of whom had already responded in the object tests, and some of the estimates were statistically uncertain.³

“These results show that even in people with profound vision loss, the visual system retains a remarkable capacity to process new information.”

That was how José-Alain Sahel, chair of ophthalmology at Pitt and first author, put it.¹

Side effects

Researchers recorded 34 eye-related side effects in nine of the 10 people, mostly mild or moderate. Seven developed inflammation inside the eye, which cleared with treatment and left no lasting effects. One serious event was a brief blockage of the main artery feeding the retina right after the injection; it cleared within minutes after medication.³ The university also listed short-lived rises in eye pressure among the side effects.¹

There were 70 side effects elsewhere in the body, six of them severe. The lead investigator, overseen by an independent safety board, judged none of these to be caused by the gene therapy or the injection. Three mild ones were linked to the goggles.³ “Safety was very good and is not a limitation,” Sahel and Botond Roska told ScienceAlert, while the paper’s authors cautioned that the safety data come from only 10 people.³

What it does not show yet

The study was not randomized and had no comparison group, so it cannot show for certain that the treatment, rather than practice or chance, caused the gains.³ The cutoff for a “meaningful” light-sensitivity gain came from earlier research and was not set in advance in the trial plan, and the rules for calling someone a responder were not set in advance either.³ People who trained more with the goggles tended to do better, but the data were also consistent with no link at all.³

Follow-up varied. Four people completed five years, several others were followed for two to four and a half years, and two withdrew earlier.³ The researchers said the differences between patients may reflect how well the gene reached the retina, how many retinal cells were left, or how much training each person had.³

“What is still missing is the ability to recognize faces and to read; greater sensitivity of the optogenetic sensor would also be valuable.”

Sahel and Roska said they are working on those limits.³

Who is behind it

GenSight Biologics funded the trial, helped design it and managed the data.³ Sahel and Roska, the study’s corresponding authors, were both described as co-founders of GenSight when the first patient’s results were announced in 2021.⁵ Roska leads the Institute of Molecular and Clinical Ophthalmology Basel in Switzerland.² In September 2026 the two received the António Champalimaud Vision Award for their work on restoring vision.²

The new paper builds on a 2021 Nature Medicine report of a single blind patient who regained some ability to perceive, locate, count and touch objects with the goggles.⁵ That person’s results are included in the new group of 10.³ The trial, called PIONEER, was planned in 2021 as a first-in-human study testing increasing doses, with 12 to 18 participants expected.⁵

How it fits the field

Retinitis pigmentosa can be caused by changes in any of more than 100 genes, which makes building a separate gene fix for each cause slow and costly.¹ Except for a subset of patients with one specific genetic change who can get gene therapy, there is no cure, and care focuses on slowing the loss.⁴ The optogenetic approach aims to work no matter which gene caused the disease, by giving surviving nerve cells a new way to sense light.¹ Sahel said it could potentially help other blinding diseases where light-sensing cells are lost but ganglion cells survive.²

Other groups are trying different routes to the same goal. In 2023, Kiora Pharmaceuticals began dosing patients in Australia with a light-sensitive drug, not a gene therapy, meant to act as a reversible “photoswitch” in the eye.⁶ For a different eye disease, Biohack recently covered an early test of a gene therapy that aims to make eye cells act younger.

What to watch next

The reporting and the authors agree that more research is needed to weigh benefits against risks.³ The Pitt team describes this as one of several complementary strategies it is pursuing for end-stage retinal disease.² Key questions are whether a larger, controlled trial follows, whether more sensitive light sensors can improve results, and whether patients can move from detecting objects toward recognizing faces or reading.³