Light sensitivity rose in 7 of 10 blind people given an optogenetic gene therapy and special goggles

In 2021 researchers reported that a blind man could see and count objects after an experimental gene therapy, with the help of a pair of camera-equipped goggles. It was one patient. The same team has now published results for ten.

The report appeared in the New England Journal of Medicine on October 7, two days after the Nobel Prize in medicine went to the scientists who developed the underlying method, optogenetics. Seven of the ten participants became more sensitive to light, and four of the eight tested could do simple visual tasks with the goggles on. Nobody regained anything like ordinary sight. The trial was designed to show that the treatment was safe enough to keep testing, and that is the claim its authors make.

How the optogenetic therapy and goggles work

The disease in question destroys the eye’s light detectors. Retinitis pigmentosa (RP) encompasses a group of inherited retinal dystrophies characterized by the primary degeneration of rod and cone photoreceptors, in the words of a standard review. Photoreceptors are the cells at the back of the eye that turn light into nerve signals. Rods handle dim light and cones handle color and detail.

The usual course is slow. RP typically manifests with night blindness in adolescence, followed by concentric visual field loss, meaning the field of view closes in from the edges. It is not rare for an inherited disease: the same review calls it a leading cause of visual disability, with a worldwide prevalence of 1:4000.

Most attempts at treatment try to save or repair the photoreceptors. That is hard when the fault can lie in any of dozens of genes. The review counts disease-causing variants in more than 80 genes.

Optogenetics goes around the problem. Photoreceptors pass their signals to a second layer of cells, the retinal ganglion cells, whose fibers form the optic nerve. Those cells can survive after the photoreceptors are gone. The therapy makes them respond to light themselves. As Science News described it, rather than repair the retina’s damaged light detectors, it gives that ability to retinal ganglion cells that normally relay visual signals to the brain.

The tool is a protein called ChrimsonR, a light-sensitive protein derived from algae. The gene for it is carried into the cells by a harmless virus. In the trial, each participant received a single intravitreal injection of an adeno-associated viral (AAV) vector encoding the red-shifted channelrhodopsin ChrimsonR in the worse-seeing eye. Intravitreal means into the jelly that fills the eyeball. A vector is the carrier, and channelrhodopsin is the family of light-sensitive proteins.

Treated cells respond to amber light, and supplying it is the job of the goggles. A camera in the custom goggles detected changes in brightness in what was in front of a study participant, and the goggles projected those changes onto the retina as pulses of amber light, activating the treated cells.

Because the approach does not depend on which gene caused the disease, it could in principle apply to anyone with the condition whose ganglion cells and optic nerve are intact.

Who took part in the gene therapy trial

The study, known as PIONEER, enrolled 10 participants with blindness due to advanced retinitis pigmentosa. They were treated in three steps of rising dose and then an extension group, a plan the registry sets out as 3 dose escalation cohorts (low, medium and high dose) with 3 subjects per cohort followed by an extension cohort. Cohort here means a group of participants.

To qualify, people had to have very little vision left and enough of the target cells remaining. The registry requires a relatively preserved ganglion cell layer volume and retinal nerve fiber layer thickness, measured by a scan of the retina.

The trial was open-label. Everyone received the therapy and everyone knew it. There was no untreated group for comparison, although the untreated eye was covered during testing.

Its purpose was narrow. The paper states it plainly: the primary outcome was safety. A trade report on the announcement repeats the point from the other side, saying the study was not designed to establish the efficacy of GS030, the therapy’s code name.

How safe the optogenetic therapy was

Side effects in the eye were common and mostly minor. A total of 34 ocular adverse events occurred among 9 of the 10 participants, including 23 mild events and 10 moderate events. Ocular means of the eye, and an adverse event is any medical problem recorded during a trial.

One was serious. The paper describes a transient occlusion of the central retinal artery that occurred immediately after intravitreal injection and resolved within minutes after eye drops were given. An occlusion is a blockage. The central retinal artery supplies the retina with blood, and a lasting blockage would threaten what vision remains.

Nothing was reported beyond the eye. According to the announcement, there was no systemic adverse event determined to be related to the study drug or to the intravitreal injection, systemic meaning affecting the body as a whole.

The authors’ verdict is hedged with its own caveat. Within the limits of this small study, they write, the injection combined with the goggles was safe.

How much light sensitivity and vision returned

The trial measured benefit in three ways, each further from the laboratory than the last.

Light sensitivity. Participants were tested for the dimmest flash they could detect, with the goggles on and the untreated eye patched. Light sensitivity increased in 7 of the 10 participants, with increases ranging from a factor of 2.0 to a factor of 62.3. Six of them passed the bar the researchers used for a meaningful gain.

That bar deserves a note. It was set at a fourfold rise in sensitivity, and the paper says plainly that the definition was based on evidence from the literature and not prespecified in the protocol. A threshold chosen after the results are in is weaker evidence than one fixed beforehand.

Tasks. A smaller number were tested on what they could do. Four of the 8 participants who completed visual behavioral testing showed positive changes in the ability to perform certain tasks while using the light-stimulating goggles: detecting, localizing or touching objects, or determining the orientation of bars. Science News gave everyday versions of the same result, reporting that four were able to find a doorway or follow a line while wearing the goggles.

Brain activity. In some participants the team recorded electrical activity over the visual part of the brain. Those recordings showed changes consistent with the transmission of optogenetically evoked visual signals to the visual cortex. The point of that check is to show that the responses came from vision and not from guesswork or other cues.

What did not happen is as clear. No patient gained the ability to read visual words or see faces.

Botond Roska of the Institute of Molecular and Clinical Ophthalmology Basel, one of the senior authors, pitched the result at that level. “This is a proof of concept that optogenetics can bring back some visual activity,” Roska told Science News.

How the therapy compares with other work on inherited blindness

The new paper extends the team’s own earlier case. Science News describes it as building on a 2021 report that the same approach helped one blind man see and count objects. Ten patients is still a small number, but it suggests the first was not a one-off.

Other groups are pursuing the same strategy with different proteins. In April a team in China reported an early trial of a therapy called UGX-201. Nine patients with advanced nonsyndromic RP were enrolled, nonsyndromic meaning the disease affects the eyes alone. No serious side effects were recorded over a year. Three of the nine could not perceive light at all before treatment. All three regained the ability to perceive light in the treated eye, and two still had it a year later. That trial too was open-label with no comparison group.

What the 10-person trial cannot show about vision

Whether the gains are real improvements. With no comparison group, practice and expectation cannot be separated from the treatment. People tested repeatedly on the same tasks tend to improve. The brain recordings argue against that explanation in the participants who had them, but only some did.

How large the benefit is. The trial was not built to measure it. The threshold for a meaningful gain was chosen afterward, and the task results come from eight people.

What it means in daily life. Detecting a bar or locating a doorway in a test room is a long way from walking down a street. The gains depend on the goggles, which turn the scene into a single color of light.

Rare harms. One of ten participants had a brief arterial blockage. Ten people cannot show how often that would happen, or what else might.

Who could benefit. The therapy needs surviving ganglion cells and a pathway to the brain. Jose-Alain Sahel of the University of Pittsburgh, a study author, told Science News that the approach requires a working optic nerve to carry signals to the brain.

How long it lasts. The registry lists the study as still active, and the main safety outcome was set at one year.

Who paid. The paper states that the trial was funded by GenSight Biologics, which is developing the therapy. It is not on sale. The trade report carries the company’s own statement that GS030 has not received marketing authorization in any jurisdiction.

The authors end where a safety trial should. Additional research is needed to further assess safety and efficacy, they write.

What can be done for a person with an inherited eye disease depends on the gene involved and the stage it has reached, which is assessed by their doctor.

Ten people blind from retinitis pigmentosa received an optogenetic gene therapy in one eye, and seven became more sensitive to light while four of eight tested could locate objects or doorways with special goggles, in a company-funded safety trial with no comparison group in which nobody regained the ability to read or recognize faces.