A gene therapy paired with camera goggles made 7 of 10 blind people more sensitive to light, according to a study published in October in the New England Journal of Medicine1. All 10 had retinitis pigmentosa, an inherited disease that slowly destroys the light-sensing cells of the eye1. Six of them improved enough for the researchers to call the change clinically meaningful, but none regained normal sight1. Science News summed it up as giving “some blind people limited sight”2.
This is the first time the approach has worked in a group of patients rather than in a single person1. The work was led by José-Alain Sahel of the University of Pittsburgh and Botond Roska of the Institute of Molecular and Clinical Ophthalmology Basel (IOB)1.
How the treatment works
In retinitis pigmentosa, the retina’s photoreceptors (the cells that turn light into signals) gradually stop working1. Faults in more than 100 different genes can cause the disease, which makes fixing it one gene at a time hard1.
So the team did not try to repair the photoreceptors. Instead it used optogenetics, a method that adds a light-sensitive protein to cells that do not normally respond to light1. The field of optogenetics was recognized by the 2026 Nobel Prize in medicine, Science News noted2.
Each person got one injection into the worse-seeing eye1. The injection holds a harmless virus, an adeno-associated virus or AAV, used as a delivery truck for a gene1. That gene makes a protein called ChrimsonR inside retinal ganglion cells, the nerve cells that pass visual signals to the brain and that often survive even late in the disease1.
ChrimsonR responds to amber light, so patients wear special goggles1. A camera films the scene, a processor turns changes in it into pulses of amber light, and those pulses switch on the treated cells1. Amber light was chosen in part because it is safer for retinal cells than the blue light used in other optogenetic research3. In the trial registry the gene therapy is listed as GS030-DP and the goggles as GS030-MD4.
What the trial found
The reported gains, as summarized in coverage of the study, were:
- Light sensitivity rose in 7 of 10 people, by a factor of 2.0 to 62.31.
- Six crossed the threshold the researchers count as clinically meaningful, roughly a fourfold gain1.
- Of the 8 people who completed tests of everyday-style tasks, 4 got better at detecting or locating objects1.
- Recordings of brain activity showed signals consistent with visual information reaching the visual cortex, the part of the brain that processes sight1.
The limits were just as clear. Patients still could not read or recognize faces1. The IOB said that people who spent more time training with the goggles tended to do better, which suggests rehabilitation would be part of any future treatment1.
Safety and what it does not show yet
The trial, called PIONEER, was mainly designed to test safety1. The registry lists it as an early study (Phase 1/2a, the first stages of testing in people) in which doses were raised step by step, everyone knew they were getting the treatment, and there was no randomly chosen comparison group4. Its main goal is to measure safety and tolerability after one year4.
Nine of the 10 participants had a total of 34 eye-related side effects: 23 mild, 10 moderate and one severe1. The severe one was a temporary blockage of the main artery feeding the retina right after injection, which cleared within minutes after eye drops were given1.
Several things remain unknown:
- How well it works. With 10 people and no comparison group, the study cannot show how large the true benefit is. The authors say larger trials are needed to confirm both safety and efficacy1.
- Who it can help. The approach needs a working optic nerve, the cable that links eye and brain, so it cannot help when that connection is damaged1.
- How long effects last. The registry lists the study as still active, though no longer recruiting, with its main data collection due to finish in October 20274. No results had been posted to the registry4.
Who is behind it
The study was funded by GenSight Biologics, a French biotech company1, which is also listed as the trial’s sponsor4. Because the company is developing the therapy, it has a financial stake in the results. The trial was first registered in October 2017 and started in September 2018, and it enrolled 10 people4.
Sahel and Roska have worked together on this for years. When they reported their first patient in 2021, the work was described as the result of 13 years of effort by a team that included the Paris Vision Institute, the Quinze-Vingts hospital, the University of Pittsburgh, the IOB, StreetLab and GenSight Biologics3. Sahel is chair of ophthalmology at the University of Pittsburgh, and Roska is founding director of the IOB and a professor at the University of Basel3.
How it fits the field
The new results build on that 2021 case, in which one man treated with the same method could see and count objects1. That patient’s results appeared in Nature Medicine3. He started training with the goggles nearly five months after his injection, and about seven months later began reporting signs of visual improvement3.
His gains were modest and depended on the goggles. He touched a large notebook on a table in 36 of 39 tries (92% of the time) but found a smaller staple box only 36% of the time, and he correctly counted glass tumblers 63% of the time3. Without the goggles, he could not do these tasks3. A computer program reading his brain waves could tell with 78% accuracy whether a tumbler was on the table3. Roska said this helped confirm that the brain activity was tied to a visual object, “and therefore that the retina is no longer blind.”3
Other groups are taking very different routes to artificial vision, including Neuralink’s brain implant, Blindsight1. Optogenetics avoids brain surgery and is meant to work no matter which gene caused the disease1. Back in 2021, Sahel described who might one day qualify:
“Importantly, blind patients with different kinds of neurodegenerative photoreceptor disease and a functional optic nerve will potentially be eligible for the treatment. However, it will take time until this therapy can be offered to patients.”3
What to watch next
For now the therapy is experimental and not approved anywhere1. The next signs to watch are the final one-year safety results from the PIONEER trial, due by its listed completion date in October 20274, and whether GenSight Biologics or the research teams announce a larger trial with a comparison group, which the authors say is needed1. Another open question is how much structured training with the goggles adds, since people who trained more tended to do better1.
