Research Spotlight · Macular Degeneration
Dry macular degeneration slowly wipes out the layer of support cells under the retina. Researchers are now growing replacement cells from stem cells and placing them under the retina by surgery. Here is what the 2026 trial data show, where the evidence is still thin, and the one kind of "stem cell" eye injection that has already blinded people.
A 72-year-old woman with macular degeneration walked into a stem cell clinic in the United States. She could still read the 20/30 line with one eye.
She got an injection of "stem cells" made from her own fat. Into both eyes.
A year later, she was close to blind. So were two other women treated at the same clinic. Their story was published in the New England Journal of Medicine in 2017.
Now here is the strange part. That same year, the same journal published a very different eye study. Japanese scientists had grown a sheet of retinal cells from a patient's own reprogrammed skin cells and slipped it under her retina. The sheet was still in place a year later.
Two stem cell eye stories. Same year. Same journal. Opposite lessons.
Nearly a decade later, the careful version of this science has moved into larger trials. This post walks through what the research actually shows as of October 2026, and how to tell real eye cell research from the kind that ended up in that 2017 paper.
What Is Dry Macular Degeneration, and Why Are Stem Cells Being Studied?
Age-related macular degeneration (AMD) damages the macula. That is the small center of the retina you use to read, drive, and see faces.
There are two forms. Wet AMD is when leaky blood vessels grow under the retina. Doctors treat it with drug injections that block a growth signal called VEGF. Dry AMD is more common and slower. Its late stage is called geographic atrophy, or GA. Patches of retina simply die off and spread over time.
At the center of the problem is a thin layer of cells called the retinal pigment epithelium, or RPE. Think of the RPE as the support crew for your light-sensing cells. It feeds them, cleans up their waste, and keeps them alive. When RPE cells die, the light-sensing cells above them die too. That is what leaves blind spots.
In 2023 the FDA approved two drugs for geographic atrophy, pegcetacoplan (Syfovre) and avacincaptad pegol (Izervay). Both are injected into the eye on a schedule, and both are designed to slow the growth of the damaged patches. Neither replaces lost cells.
So researchers asked a simple question. If the RPE layer is dying, can you grow new RPE cells in a lab and put them back?
How Do Stem Cell Treatments for Macular Degeneration Work?
This is where most people get confused, so let's be precise.
The serious trials do not inject "stem cells" into the eye. They start with pluripotent stem cells, either embryonic stem cells or induced pluripotent stem (iPS) cells, which are adult cells reprogrammed back to an early state. Scientists then turn those cells into fully formed RPE cells in the lab. The finished product is RPE, not stem cells.
Then a retina surgeon does a vitrectomy, removing the gel inside the eye, and places the cells under the retina. Sometimes the cells go in as a liquid suspension. Sometimes they go in as a thin sheet or patch. Patients usually take immune-suppressing drugs for a period around surgery.
That is a hospital operation with real surgical risk. It is nothing like a shot in a clinic. If you want the background on the cell types, we explained iPS cells in our post on Japan's first iPS cell therapy approvals, and embryonic-stem-cell-derived therapy in our post on bemdaneprocel for Parkinson's.
What Did the First Human Studies Show?
The first wave of studies was tiny. Their job was to answer one question: can you put these cells in a human eye without causing harm?
2015: Embryonic-stem-cell RPE in 18 patients
A U.S. team transplanted RPE cells made from human embryonic stem cells into 9 people with dry AMD and 9 with Stargardt disease, an inherited form of macular degeneration. Over a median of 22 months, they saw no signs of abnormal cell growth or rejection from the transplanted tissue. The side effects that did happen were linked to the eye surgery and the immune-suppressing drugs. Vision improved in 10 treated eyes and dropped by more than 10 letters in one. These were open-label studies with no control group, published in The Lancet.
2017: One patient, one iPS cell sheet
In Japan, Masayo Takahashi's team made iPS cells from two patients' skin and grew them into RPE sheets. One patient with wet AMD received a sheet. At one year the sheet was intact. The patient's vision had not improved or worsened, and there was swelling in the retina. One person. A feasibility test, not a treatment result.
2018: Two patches in London and California
In London, two patients with severe wet AMD received an embryonic-stem-cell RPE patch on a synthetic backing. Both gained vision over 12 months, 29 and 21 letters. In California, a similar patch called CPCB-RPE1 went into 4 of 5 enrolled patients with advanced dry AMD. None of the implanted eyes lost more vision during the interim analysis, and one gained 17 letters.
Notice the numbers: 18, 1, 2, 5. Interesting signals. Nowhere near proof.
What Are the Latest OpRegen Results?
OpRegen is the furthest-along RPE cell therapy for dry AMD. It is made by Lineage Cell Therapeutics from embryonic stem cells and is now being developed by Roche and Genentech. The FDA has given it Fast Track and Regenerative Medicine Advanced Therapy (RMAT) designations. Those speed up FDA interaction. They are not approval.
The Phase 1/2a trial (NCT02286089) treated 24 patients with geographic atrophy in one eye. Cohorts 1 to 3 had advanced disease and very poor vision. Cohort 4 had 12 patients with less advanced disease. The trial was open-label, with no placebo or sham group. The untreated fellow eye served as a rough comparison.
In May 2026, the team presented 3-year results for Cohort 4 at the Foundation Fighting Blindness Retinal Therapeutics Innovation Summit:
- The 10 Cohort 4 patients who completed 3 years averaged +6.2 letters on the eye chart in the treated eye, up from +5.5 letters at 2 years.
- In the 5 patients whose damaged area was most fully covered by the cells during surgery, the average was +9.0 letters.
- On retinal scans, the RPE layer area held above baseline in those 5 treated eyes (+1.9 mm² at 3 years) while it shrank in their untreated fellow eyes (−3.8 mm²).
Researchers say the scans show areas where an RPE layer reappeared, along with features linked to recovery of light-sensing cells. They say this suggests the cells may support the remaining retina. That is their interpretation of imaging in a handful of eyes. It is worth taking seriously. It has not been tested against a control group.
What about safety?
An earlier report of all 24 patients (ARVO 2022) found no unexpected adverse events. But eye surgery is not free. Among 17 patients treated through vitrectomy, 15 developed an epiretinal membrane, a layer of scar-like tissue on the retina. Most were mild to moderate. Three were severe enough to need surgical peeling. Two patients had retinal detachments, which were repaired. In the 7 patients treated by a different route, 3 developed abnormal blood vessel growth, which was treated or managed with anti-VEGF injections.
How to read these numbers
- 15 letters is three lines. On the standard ETDRS eye chart, 5 letters is one line. A +6.2 average is a bit more than one line.
- No control group. Without a sham or placebo arm, nobody can say how much of the change came from the cells versus the surgery, the attention, or chance.
- Small subgroups. The +9.0 figure is 5 people. One patient's result moves that average a lot.
- Conference data, sponsor-reported. The 3-year results were presented at a meeting and released by the company. They are not yet in a full peer-reviewed paper.
What comes next for OpRegen
Genentech is running GAlette (NCT05626114), a Phase 2a, open-label, single-arm study of up to 60 patients at 17 sites in the U.S. and Israel. Its main goals are about surgery: how often the cells get delivered to the right spot, and how safe the delivery is at 3 months. Genentech also plans to test new surgical delivery devices. ClinicalTrials.gov lists primary completion in 2031. In other words, the controlled efficacy answer is still years away.
What Is Eyestem Doing in India?
Eyestem, a company in Bengaluru, makes Eyecyte-RPE, an RPE cell suspension grown from iPS cells. Its Phase 1/2a trial (NCT06394232) is running at three Indian eye centers.
In Phase 1, 9 patients with geographic atrophy and poor vision (20/200 or worse) received 100,000, 200,000, or 300,000 cells under the retina, with oral immune-suppressing drugs. At ARVO in May 2026, the team reported the cells were well tolerated. Eight patients showed lasting vision gains, averaging 13.3 letters, at a point when 4 of the 9 had reached one year of follow-up. An epiretinal membrane seen in 4 eyes did not get worse.
At the EURETINA congress in October 2026, co-investigator Ritesh Narula described the Phase 1 gain as about 11 letters on average, holding at around 18 months. He said the Phase 1 results have been submitted for publication.
The Phase 2 part is randomized: per the ARVO abstract, 22 patients get 200,000 cells and 11 are followed as untreated controls. About 13 had been treated by early October. One developed endophthalmitis, a serious eye infection, and one developed a macular hole. The team plans to seek a U.S. trial in 2027.
Again, the honest summary: 9 patients, no control group in Phase 1, and an early, partial look at Phase 2. Encouraging. Early.
How Do the Main Approaches Compare?
| Approach | What goes in the eye | Key study and size | What it found |
|---|---|---|---|
| OpRegen (Lineage, Genentech) | Embryonic-stem-cell-derived RPE cell suspension, placed under the retina by surgery | Phase 1/2a, 24 patients, open-label | Cohort 4 (n = 10) averaged +6.2 letters at 3 years; epiretinal membranes common; Phase 2a ongoing |
| Eyecyte-RPE (Eyestem) | iPS-cell-derived RPE cell suspension, placed under the retina by surgery | Phase 1, 9 patients; randomized Phase 2 enrolling | Average gains of roughly 11 to 13 letters reported at meetings; not yet peer-reviewed |
| CPCB-RPE1 (California) | Embryonic-stem-cell RPE on a synthetic patch | Phase 1/2a interim, 5 enrolled, 4 implanted (2018) | No implanted eye lost more vision; one gained 17 letters |
| Takahashi iPS sheet (Japan) | Patient's own iPS-cell-derived RPE sheet | 1 patient (2017) | Sheet intact at 1 year; vision unchanged |
| Clinic "stem cell" eye injections | Unpurified cells from fat, injected into the eye's gel | Case report, 3 patients (2017) | Severe vision loss in all three; 20/200 to no light perception at 1 year |
What Happened to the Three Women Treated at That Clinic?
Back to where we started, because this part matters.
The three women, ages 72 to 88, had macular degeneration. Before treatment, their best vision ranged from 20/30 to 20/200. At one U.S. clinic, each had cells taken from their own fat and injected into the gel of both eyes. One paid $5,000. The clinic had a study listed on ClinicalTrials.gov, but the paperwork given to at least one patient did not mention a trial.
Within days, problems started. Bleeding inside the eye. High eye pressure. Retinal detachments. A dislocated lens. A year later, their vision ranged from 20/200 to no light perception. The doctors who wrote up the cases, at Bascom Palmer Eye Institute and elsewhere, suspected the cell preparation itself, not routine injection technique.
Compare that with the trials above. Those use one defined cell type, made under controlled lab conditions, delivered by retina surgeons under the retina, in one eye, with regulators watching and every complication reported. The clinic did none of that.
Same words, "stem cells for macular degeneration." Completely different things.
What Are the Honest Objections?
- The studies are small. The largest RPE cell trial so far treated 24 people. That is not enough to know how well this works or what the uncommon risks are.
- There is no randomized efficacy result yet. Eyestem's Phase 2 has an untreated control group, but results are early. GAlette has no control group.
- The surgery has its own risks. Scar membranes, retinal detachment, infection, and macular holes have all been reported. See our overview of stem cell therapy side effects for the general picture.
- Much of the newest data is sponsor-reported. Conference talks and company releases come before full peer review. They can change.
- Gains vary. In OpRegen, results looked best when the cells covered more of the damaged area. That points to how much depends on surgical delivery.
None of that makes the field less exciting. Replacing a dead layer of cells in a living human eye was science fiction 20 years ago. It just means the right word today is promising, not proven.
Does This Research Apply to Other Stem Cell Products?
No. Each of these trials tests one specific product: a particular cell type, made by a particular lab, at a particular dose, placed in a particular spot. RPE cells grown from pluripotent stem cells are not the same as mesenchymal stem cells (MSCs) from fat, bone marrow, or umbilical cord tissue. Results from RPE trials say nothing about MSC products, and the reverse is also true.
Regulation matters too. In the U.S., cell products being tested to treat a disease are studied as investigational products in registered trials. For the bigger picture, see our guide to stem cell therapy and FDA approval.
Where The Stem Cell Club Fits
We are not part of any of these trials and have no connection to Lineage, Genentech, Roche, or Eyestem.
We offer umbilical cord Wharton's Jelly MSC options sourced from an FDA-registered U.S. lab. They are a different product from anything described above. We do not offer eye injections, and none of our offerings are intended for macular degeneration or any eye disease. Lab registration covers facility and manufacturing standards. It is not FDA approval of a therapy. Our offerings are not FDA-approved to diagnose, treat, cure, or prevent any disease.
If you have macular degeneration, the person to talk to is a retina specialist. If you want to look into a stem cell eye trial, the legitimate path runs through ClinicalTrials.gov and a retina specialist who can tell you whether you qualify.
Questions to Ask Before You Believe Any "Stem Cells for Macular Degeneration" Claim
- What cell is it, exactly? RPE cells grown from pluripotent stem cells, or something else?
- Where does it go? Under the retina by a retina surgeon, or injected into the eye's gel at a clinic?
- Is it a real trial? Is there an NCT number, an ethics board approval, and do you pay nothing to take part? Legitimate trials rarely charge.
- Are both eyes treated at once? The trials above treated one eye. Treating both eyes at once removes your safety margin.
- Has it been published? Peer-reviewed results, or testimonials and before-and-after claims?
We keep a longer checklist here: questions to ask any stem cell clinic.
Frequently Asked Questions
Is there an FDA-approved stem cell therapy for macular degeneration?
No. As of October 2026, no stem cell therapy is FDA-approved for macular degeneration. Cell therapies that replace retinal pigment epithelium (RPE) cells are being studied in clinical trials. For geographic atrophy, the late stage of dry AMD, the FDA approved two drugs in 2023, pegcetacoplan (Syfovre) and avacincaptad pegol (Izervay), which are designed to slow the growth of damaged areas.
How does stem cell therapy for dry AMD work?
The leading approach grows retinal pigment epithelium (RPE) cells from embryonic or induced pluripotent stem cells in a lab. A retina surgeon then places those cells under the retina during a vitrectomy, aiming to replace the support cells that die in geographic atrophy. The finished product is RPE cells, not stem cells, and it is still investigational.
What did the OpRegen trial show?
OpRegen's Phase 1/2a trial (NCT02286089) treated 24 patients with geographic atrophy, with no control group. In May 2026, the team reported that 10 patients with less advanced disease averaged +6.2 letters on the eye chart at 3 years, and 5 patients with the most complete cell coverage averaged +9.0 letters. Epiretinal membranes were common after surgery and two retinal detachments were repaired. A Phase 2a study of up to 60 patients is ongoing.
What is Eyecyte-RPE?
Eyecyte-RPE is an RPE cell suspension made from induced pluripotent stem cells by Eyestem in India. In a 9-patient Phase 1 study, the team reported at medical meetings in 2026 average vision gains of roughly 11 to 13 letters. The results are not yet peer-reviewed, and a randomized Phase 2 with an untreated control group is underway.
Are stem cell eye injections at clinics safe?
Serious harm has been reported. In 2017 the New England Journal of Medicine described three women, ages 72 to 88, who received injections of fat-derived "stem cells" into both eyes at a U.S. clinic. They developed bleeding, retinal detachments, and other complications, and one year later their vision ranged from 20/200 to no light perception.
What are the risks of RPE cell transplants?
The procedure requires eye surgery under the retina and usually immune-suppressing drugs. In the OpRegen trial, epiretinal membranes were common and some needed surgical peeling, and two patients had retinal detachments that were repaired. In Eyestem's Phase 2, one patient developed a serious eye infection and one developed a macular hole. With small trials, uncommon risks are not yet known.
How can I find a legitimate macular degeneration stem cell trial?
Search ClinicalTrials.gov for geographic atrophy or macular degeneration cell therapy studies and talk with a retina specialist about whether you might qualify. Legitimate trials have a registration number and ethics approval, usually treat one eye, and rarely charge participants.
Does The Stem Cell Club offer stem cell therapy for macular degeneration?
No. The Stem Cell Club does not offer eye injections, and none of our offerings are intended for macular degeneration or any eye disease. We are not affiliated with the trials described here. Our umbilical cord Wharton's Jelly MSC offerings are a different product and are not FDA-approved to diagnose, treat, cure, or prevent any disease.
Want a Straight Answer About Regenerative Options?
Our free consultation is honest about what the research shows and what it doesn't. If something isn't a fit for you, we'll say so.
Schedule a Free Consultation → Or call (435) 281-2999Educational content only, not medical advice. Individual results vary. Stem cell offerings are not FDA-approved to diagnose, treat, cure, or prevent any disease. FDA registration of the source lab pertains to facility and manufacturing standards only and does not constitute FDA approval. This article discusses investigational products in published studies and registered clinical trials. The Stem Cell Club is not affiliated with these studies, and their findings do not apply to our services. Talk with your eye doctor before starting or changing any treatment.
OpRegen 3-year results: Lineage Cell Therapeutics. OpRegen for Dry AMD (Retinal Therapeutics Innovation Summit 2026 highlights, May 2026), and press release. ClinicalTrials.gov NCT02286089 (Phase 1/2a, 24 enrolled) and NCT05626114 (GAlette Phase 2a).
OpRegen safety, all 24 patients: Ho AC, Banin E, Barak A, et al. Safety and Efficacy of a Phase 1/2a Clinical Trial of Transplanted Allogeneic Retinal Pigmented Epithelium (RPE, OpRegen) Cells in Advanced Dry AMD. Invest Ophthalmol Vis Sci. 2022 (ARVO abstract).
Eyestem: Battu R, Narula R, Chawla R, et al. Results of Phase 1/2a Trial of Eyecyte-RPE for the Treatment of Geographic Atrophy associated with AMD. ARVO 2026 abstract. ClinicalTrials.gov NCT06394232. Ophthalmology Times, EURETINA 2026: Eyestem interim Phase 2 results, October 5, 2026.
Early RPE studies: Schwartz SD, et al. Human embryonic stem cell-derived retinal pigment epithelium in patients with age-related macular degeneration and Stargardt's macular dystrophy. Lancet. 2015;385:509–516. doi:10.1016/S0140-6736(14)61376-3. PMID 25458728. Mandai M, et al. Autologous induced stem-cell-derived retinal cells for macular degeneration. N Engl J Med. 2017;376:1038–1046. doi:10.1056/NEJMoa1608368. PMID 28296613. da Cruz L, et al. Phase 1 clinical study of an embryonic stem cell-derived retinal pigment epithelium patch in age-related macular degeneration. Nat Biotechnol. 2018;36:328–337. doi:10.1038/nbt.4114. PMID 29553577. Kashani AH, et al. A bioengineered retinal pigment epithelial monolayer for advanced, dry age-related macular degeneration. Sci Transl Med. 2018;10:eaao4097. doi:10.1126/scitranslmed.aao4097. PMID 29618560.
Clinic injections: Kuriyan AE, Albini TA, Townsend JH, et al. Vision loss after intravitreal injection of autologous "stem cells" for AMD. N Engl J Med. 2017;376:1047–1053. doi:10.1056/NEJMoa1609583. PMID 28296617.