A new study has found that a light-sensitive injection into the eye can trigger early vision-related signals in people with severely damaged retinas.
The finding raises a precise possibility: even after the eye’s main light sensors fail, surviving cells may still carry signals toward the brain.
Eye injection trial in humans
Inside the first human trial, six adults with advanced inherited vision loss received experimental injections in their eyes.
Following those treated eyes over 30 days, Dr. Robert J. Casson, an ophthalmologist at Royal Adelaide Hospital in Adelaide, Australia, documented no serious safety problems after dosing.
Some participants showed changing light perception or functional vision scores, but those signals were uneven and faded over time.
That boundary keeps the result promising but unfinished, because everyday vision still requires proof in larger, controlled trials.
Retinas and vision
The trial focused on a group of inherited eye disorders called retinitis pigmentosa that can worsen over years.
Usually, rod cells – retinal cells that help low-light vision – fail first, so people lose night vision before daytime detail breaks down.
Later, cone cells, which support color and detail, suffer too, narrowing side vision before central tasks weaken.
At late stages, clear parts of the eye may still work while the usual light sensors cannot.
How the eye injection works
KIO-301, a light-reactive drug candidate, aims at surviving cells below the damaged light sensors.
Those cells include retinal ganglion cells – eye cells that send signals to the brain – which can remain after severe disease.
Light changes the shape of a small chemical, called an azobenzene photoswitch, that flips between two forms.
Any effect likely runs through ion channels – tiny protein openings for charged particles – but human cells were not directly tested for that mechanism.
Safety came first
For ABACUS-1, safety carried more weight than proof that anyone could see better.
Clinicians gave single intravitreal injections, which place medicine into the eye’s jelly-like center, at gradually higher doses.
Across all treated eyes, the team found no serious harm, no toxic reaction, no drug-related eye inflammation, or structural retinal damage.
Mild discomfort and a temporary eye pressure rise fit with known injection effects, rather than being a clear drug injury.
Signals reached brain
The most intriguing signals came from tests that watched whether light-related activity reached the brain.
Brain scans – called functional magnetic resonance imaging – watched blood-oxygen changes, a sign that nearby nerve cells needed more fuel.
After dosing, light-linked changes appeared in visual brain regions in some participants, especially during the first two to three days.
Those scans cannot prove useful sight, but they showed the drug may have reached a living pathway.
Limitations to eye injection trial
A six-person study can catch obvious safety problems, but it cannot settle whether a treatment works.
Everyone knew they had received the drug, so expectation and repeated testing could influence reports or task performance.
No comparison group received an inactive injection during ABACUS-1, leaving no clean way to separate ordinary variation.
Because the study lasted 30 days, it also could not show whether repeat dosing remains safe over months.
Benefits remain unproved
Daily vision is harder to measure than a flash of brain activity or a laboratory task.
Participants completed walking, window, room exit, and door tasks under controlled lighting, with performance changing across visits.
Quality-of-life scores changed during the study period, although scores varied enough to resist a firm conclusion.
“The results of the trial provide evidence of short-term ocular safety and feasibility, while underscoring that larger, controlled studies are still needed to determine whether any functional changes translate into reliable, everyday vision benefit,” summarized Casson.
Targeting location, not genes
Many vision therapies work only when doctors know and target a specific broken gene.
KIO-301 aims below the damaged photoreceptors – light-sensing cells called rods and cones – rather than the mutation that injured them.
That design could help people with different genetic causes, and possibly other retinal diseases, if later trials confirm benefit.
For now, the claim remains limited to feasibility because ABACUS-1 did not test broad disease groups.
Next trial tests durability
A larger Phase 2 study will now test whether repeated KIO-301 doses can improve function more reliably.
Random assignment compares treated participants with a control group, reducing the chance that hope gets mistaken for effect. Higher and repeated doses may also reveal whether light responsiveness lasts longer than the early window seen before.
Clear gains in everyday movement, object finding, or light awareness would matter more than isolated scan changes.
The result now rests on one careful point: damaged retinas may still contain cells that a drug can activate safely.
Future trials must show whether that activity becomes useful sight, how often injections are needed, and who is most likely to benefit.
The study is published in Nature Medicine.
NOTE – This article was originally published in Earth and can be viewed here
