Ray Therapeutics and the Cells That Refused to Leave

In 2010, Paul Bresge's teenage daughter Tamar was diagnosed with retinitis pigmentosa. The prognosis was blunt: her vision would deteriorate, and there was little to offer. Bresge, who had worked in pharmaceutical manufacturing but was no retinal scientist, began reading the research. He eventually helped found jCyte, a company developing a cell therapy intended to preserve and improve the function of photoreceptors. It was a sensible place to start. Then a more difficult question appeared: what happens to a patient whose photoreceptors are already gone?

That question is the origin of Ray Therapeutics. The Berkeley company, founded in 2021, pursues optogenetics: delivering genetic instructions for a light-sensitive protein into retinal cells that remain after the eye's original light detectors have degenerated. Rather than replace every lost cell, Ray proposes giving some of the survivors a new task. The idea sounds almost mischievous. Biology, unfortunately, does not award points for cleverness. It asks for a measurable result.

The treatment that came first

Bresge's first company matters because it explains the second. jCyte's treatment uses cells that release factors intended to help photoreceptors survive and function. Those photoreceptors are the retina's first responders to light. But retinitis pigmentosa progresses. As they disappear, the logic of keeping them alive becomes less useful to people with advanced disease. Bresge told BioSpace in 2024 that patients without photoreceptors would be unlikely to benefit from jCyte's approach. After jCyte secured a licensing partnership with Santen, he looked for a complementary path.

He found one in research associated with Zhuo-Hua Pan at Wayne State University, now a scientific adviser to Ray. Pan had shown that a light-responsive protein could restore visual responses in mice whose photoreceptors had degenerated. Moving from a mouse retina to human daily life is a formidable distance, but the anatomical insight is plain: advanced retinal disease does not necessarily erase every neuron in the eye. Some downstream circuitry can remain.

“Patients who don't have any photoreceptors will likely not benefit from the [jCell] therapy.”Paul Bresge, in a 2024 BioSpace interview

A new job for a surviving cell

Ray's lead candidate, RTx-015, is designed to make retinal ganglion cells sensitive to light. In a healthy eye, these cells normally pass visual information toward the brain; they are not the original light sensors. RTx-015 uses an intravitreal injection, into the gel of the eye, to deliver a gene for a bioengineered light-sensitive protein. In theory, light could then trigger those cells directly, producing a signal the brain might use.

This is different from a therapy matched to one defective gene. Retinitis pigmentosa can arise from many mutations, but several genetic roads lead to the same lost photoreceptors. Ray's design tries to address that common endpoint. It is also different from an approach that needs those photoreceptors still alive. The company says its engineered protein is sensitive enough for ordinary ambient light, avoiding the light-amplifying goggles used in some optogenetic systems. That is a design claim still being tested in people, not a consumer feature available today.

The distinction has limits. The retina and optic pathway still need enough usable circuitry for a signal to travel. Clinical trial criteria include evidence of remaining retinal ganglion cells or nerve-fiber layer. An eye with no viable downstream cells is a different problem. So is proving that a detectable response becomes useful sight: recognizing an obstacle, navigating a room, or making sense of contrast in ordinary light.

Two programs, two locations in the retina

Ray's pipeline is not simply the same injection with a new disease label. RTx-015 targets ganglion cells for retinitis pigmentosa and other inherited retinal disease. RTx-021 is designed for retinal ON-bipolar cells, a different layer of the visual circuit, in Stargardt disease and geographic atrophy. The intended patients differ too. Retinitis pigmentosa usually erodes peripheral vision before later central loss; Stargardt disease and geographic atrophy damage the macula and central vision. Ray is matching its chosen cell to the geography of the disease.

Ray's immediate users are clinical trial participants, not customers walking into an eye clinic. Its ENVISION study is an open-label, dose-escalation trial of RTx-015. A separate Phase 1/2 study of RTx-021 in Stargardt disease appears in the clinical trial registry. These studies ask first about safety and tolerability, then about early signs of visual function. Ray also established the Vision Research and Assessment Institute to work on standardized tests for people with severe low vision. That matters because a chart on the wall may miss improvements that patients actually use, while an inspiring anecdote is a poor substitute for a repeatable endpoint.

The money buys a test

Ray has funded this work like a biotech rather than a product startup: private investment and research grants pay for manufacturing, clinical operations, regulatory work, and the long follow-up gene therapies demand. A $6 million seed round led by 4BIO Capital was announced in 2022. Novo Holdings led a $100 million Series A in 2023. California's regenerative medicine agency awarded an approximately $8 million grant for RTx-015 in 2025. In April 2026, Janus Henderson Investors led a $125 million Series B intended to support later clinical development and readiness for a potential commercial future. The company has no approved therapy to sell.

The same month brought FDA Regenerative Medicine Advanced Therapy designation and European Medicines Agency PRIME designation for RTx-015. Such designations can provide closer regulatory dialogue and development support. Ray said preliminary results showed improvements in treated eyes across dose levels, but its public announcement did not supply the kind of detailed, controlled efficacy data that would settle how much vision improved, for whom, or for how long. The designations matter; they do not turn an investigational injection into an approved treatment.

The field is crowded with other routes to better vision: mutation-specific gene replacement, therapies meant to protect remaining retinal cells, cell transplantation, and rival optogenetic approaches, including programs at Nanoscope and GenSight. Ray's sharpest claim is the combination of mutation independence, an injection into the eye, and a protein intended to operate without special goggles. Whether that combination proves better for patients will depend on comparative evidence that the market does not yet have.

What the rest of biotech can borrow

There is a useful lesson in the company's origin. Bresge did not abandon an earlier treatment because it was foolish; he noticed the patients its biology could not reach. Ray's strategy begins with that exclusion. Identify the group left outside the mechanism, find what anatomy remains, and build the next experiment around it. The same discipline applies far beyond ophthalmology. So does Ray's attention to measurement: if the hoped-for benefit is more independent movement through daily life, the trial should be able to see that benefit clearly.

Tamar Bresge, according to her father's BioSpace interview, went on to become an artist. Her story cannot stand in for clinical evidence, and Ray's trial participants will decide the scientific story. But it does explain why the company asks a question that sounds almost impolite to established treatment logic. When the first light sensors are gone, what, precisely, is still there? Ray Therapeutics has built a company around finding out.