Skip to content

When you choose to publish with PLOS, your research makes an impact. Make your work accessible to all, without restrictions, and accelerate scientific discovery with options like preprints and published peer review that make your work more Open.


Another Bump in the Road to Gene Therapy?


Corey wouldn’t be able to jump, and land safely, had he not had gene therapy. (Dr. Wendy Josephs)

I am astonished, once again, by the complexity and unpredictability of science.

Last week, a paper in the Proceedings of the National Academy of Science (PNAS) reported that gene therapy to treat a form of blindness called Leber congenital amaurosis type 2 (LCA2) doesn’t stop degeneration of the rods and cones – the photoreceptor cells that provide vision. Gene therapy sends genetic instructions for a protein called RPE65 into a layer of cells that supports the rods and cones – the retinal pigment epithelium, or RPE. The protein is essential for the eye to use vitamin A. And the gene therapy works, so far.

I recently wrote a book about this gene therapy, because the results were so dramatic. On a bright September day in 2008, 8-year-old Corey Haas looked up at a hot air balloon at the entrance to the Philadelphia zoo just 4 days after gene therapy at Children’s Hospital of Philadelphia, and screamed. For the first time, he could see the sun. More than 230 people have had the gene therapy, in four clinical trials, and most of them now have some sight.

Researchers at the Scheie Eye Institute of the University of Pennsylvania (not the same group that worked on Corey) and the Penn School of Veterinary Medicine used imaging technology to track thinning of the photoreceptor layer at several sites in the retina, in people who’ve had the gene therapy and in dogs who pioneered the approach. They compared thinning — a sign of degeneration — in treated and untreated eyes of the same individuals (one eye is treated first). They watched the same retinal regions in the same eyes over time. And they compared treated eyes to eyes from people with LCA2 who didn’t have gene therapy.

The conclusion seems dire: “All three analysis methods supported the conclusion that gene therapy has not modified the natural history of progressive retinal degeneration in the RPE65-LCA patients,” the researchers wrote.

I was puzzled. The researchers had analyzed rods and cones from several areas in the retinas, but I couldn’t tell from several reads of the paper whether they included the sites where the gene therapy had been injected. Unfortunately and inexplicably, the researchers, despite issuing a news release, declined my request for an interview to clarify this important point. But I asked Jean Bennett, MD, PhD, professor of ophthalmology at the F. M. Kirby Center for Molecular Ophthalmology at the University of Pennsylvania, who headed the team that treated Corey, and she agreed that whether or not the researchers assessed the treated areas is crucial information. Perhaps they will read this and weigh in.

Many people, including researchers and affected families, are speculating and hypothesizing about what the findings mean. A big clue comes from the dogs (who, by the way, are cured of a natural blindness and go home with affected families as pets).

Mercury, successfully treated for LCA2 with gene therapy.(Foundation for Retinal Research)

Data from the dozen years that blind sheepdogs have had gene therapy may reveal why the treatment doesn’t save photoreceptors in people. Development differs. Dogs with LCA2 enjoy a time — equivalent to 30 people-years — when their rods and cones are just fine. If the dogs have gene therapy during this period, their rods and cones stay healthy. But gene therapy after the rods and cones begin to die can’t save the cells.

Unlike dogs, in a baby with LCA2 the photoreceptors are already dying. The dogs in the trial may have been too young to exactly mimic the situation in people — they weren’t sick enough because their rods and cones were all still functioning. That’s a big clue, and something researchers had already suspected because of the increasing severity of the disease with age.


When I read the paper, I couldn’t help but think of the novel Flowers for Algernon, by Daniel Keyes, which inspired the film Charly. A young man with mental retardation (the correct term in 1958, when it was written) is chosen for an experimental treatment that has greatly increased the intelligence of a mouse named Algernon. Told through Charlie’s journal entries, the tale traces his restored intelligence, and then its tragic dissipation.

Will the people who’ve had gene therapy for LCA2 experience Charlie’s fate, a temporary gift of normalcy? Nothing suggests this yet, but the new study is disturbing.

Gene therapy sends  working gene into an affected body part, or into cells that are then sent into the body.

Earlier this week I attended the Phacilitate Cell & Gene Therapy Forum in Washington, D.C., and had a chance to seek opinions about the paper, and interviewed experts for Medscape. So I thought a roundup of informed opinions and interpretations might help to put these new, confusing findings into perspective.

Gerald Chader, PhD, chief scientific officer, Doheny Retina Institute:
“The paper proves the value of gene therapy – we can improve vision for at least several years. Saving the photoreceptors long-term is just not as easy as we had originally hoped. I think that the results show us that things are never as simple as initially thought or hoped for. Gene therapy still works, but other measures need to be taken. Maybe we can call these “refinements”? Early treatment may be the key.”

Steven Gray, Ph.D., Research Assistant Professor, Gene Therapy Center, University of North Carolina:
“I don’t see this as a flaw in the natural history analysis of the disease. In essence, it just says that the whole retina wasn’t treated and as a result there is still degeneration. The Leber’s gene therapy as it stands is a treatment, not a cure.”

Samuel Jacobson, MD, PhD, co-author of the PNAS paper, in a news release:
“These unexpected observations should help to advance the current treatment by making it better and longer-lasting.”

Weng Tao, MD, PhD, chief scientific officer at Neurotech USA:
“In the long run we might use a drug to preserve the photoreceptors and another drug (the gene therapy) to replace the defective gene. “

Finley has a severe form of LCA. She is awaiting gene therapy. (Jennifer Pletcher)

David Brint, not a scientist but a board member of the Foundation for Retinal Research, who has a son with LCA5:

“All groups doing gene therapy should look at overall degeneration prior to and after treatment. The jury is still out if gene therapy treatment in younger people will slow or eliminate degeneration. Because degeneration may continue in older populations for RPE65 doesn’t mean it will continue in older gene therapy patients with other gene defects.”

Clinical trials for LCA2 are about to enter phase 3, and gene therapy trials for other forms of LCA, caused by mutations in different genes, are planned. Preclinical work continues, and this work certainly validates the continued experimentation on non-human animals. The new findings of persistent photoreceptor degeneration, as unwelcome as they are, can and will inform future experiments. Researchers now have new questions to ask:

• Will gene therapy provide a more lasting effect in younger patients?
• To up the effect, can researchers deliver more genes in one injection, do multiple injections at once, or administer gene therapy at several times?
• Does rate and location of photoreceptor degradation correlate to loss of visual function?
• Will the degeneration of the rods and cones be fast enough to impair the vision regained with gene therapy? The patients in the PNAS study did not lose visual function.
• Can gene therapy be teamed with photoreceptor-saving drugs, such as anti-oxidants, anti-apoptotic factors, and neurotrophic factors?

Unexpected findings are nothing new in science. That’s why there’s no such thing as “scientific proof.” So researchers may not have been very surprised, but the families seeking gene therapy for their children may have been. And in the case of Leber congenital amaurosis – a collection of rare disorders that lie at the severe end of retinitis pigmentosa — the families are very involved with the research.

Gavin, who has the most recently discovered form of LCA, is a musical prodigy. (Jennifer Stevens)

In fact, I learned about the PNAS paper last week not from the news releases and journal offerings that crowd my inbox, but through Facebook posts from the families whom I got to know when writing my book. You can meet them at Let’s CURE Leber’s congenital amaurosis They’ve raised thousands of dollars to support gene therapy research, and they thrive on the hope that the LCA2 success will be repeated for the other forms of blindness.

And gene therapy for LCA2 IS a success. As Rare Disease Day approaches, let’s hope that this latest bump in the road will lead to true forever fixes – for many diseases.

  1. Comment from Dr. Jean Bennett, sent to PNAS after I posted. IMPORTANT:

    The Cideciyan et al 2013 PNAS paper provides beautiful evidence of stable rescue of retinal and visual function in large animal models and humans and also attempts to tackle the question of long-term effects of gene therapy on the degenerative process.

    This group also looked into the question as to whether the intervention slows the degeneration. This is a very tricky question as it can take many years to observe loss of cells in slowly progressing diseases such as LCA. The authors used non-traditional methods to evaluate the data with hopes of getting a faster answer. They did not present any of the raw data (such as assessment of retinal thickness at or near the injection site) as it was all normalized, reassigned to different ages in order to fit the data into a mathematical model and to a theoretical model of degeneration age and then it was compared to a control population. Thus it is impossible to assess data from individual subjects/eyes. Also, it is not clear that the study had the statistical power to detect a meaningful difference in different regions of the retina. The conclusion that this treatment does not halt degeneration thus may be premature. The take home message should thus be that the treatment appears to work and it also appears to have an enduring effect.

  2. Thanks for this update. What I really want to know is… how is Corey currently doing? How is his vision? What about the other subjects? I am anxiously awaiting information because my own two sons (age 12 and 6) have Usher Syndrome type 1B and I have been following the LCA treatments since I read you book. I have held on to a lot of hope, especially because of Corey. I know about the trial in Oregon for Usher 1B, but of course can not get any sorts of info about how it is progressing until they are ready to report. I wish I could just see something about how Corey is doing after the time has passed.

  3. Thank you for the clarification on this most recently published data. It is very discouraging to have seen such amazing outcomes in LCA patients dashed with one analysis. I was wondering if there are other competing methods of gene tranfer currently in trial. As there are 14 different genetic mutations which lead to LCA, and Dr. Bennett has pioneered the use of AAV technology, has anyone begun seeking different gene therapy vectors? And is there a crossover into other blinding disorders which have a later age of onset such as Chroroideremia?

  4. Thanks Colleen. In addition to the several types of LCA, each caused by mutation in a different gene, there are not at least 9 variants of AAV, the viral vector. I just interviewed Jim Wilson for a Scientific American article, and he pointed out that the AAV used on Corey, LCA2, worked because the application area was so contained and small, but that for other diseases, AAV1, 5, 8 and 9 may actually be much better choices.

  5. Thank you for the information. When does your article become available? I’d like to know more about why AAV1, 5, 8, 9 would be better in other LCA cases. Size of mutation? Because the area would remain the same; intraoccular, right? Or was your discussion including a gamut of other inherited diseases? Thank you so much for this discussion, I am intrigued by the conversation!

  6. AAVs discovered since AAV2 are more efficient, and each one favors certain body parts. LCA2 still works well for eyes because the treated area is small, compared to something like the liver. I’m writing my Scientific American article now — not sure when it will run. It is more about the ups and downs of the field than specific stories but there are a few in there. Not so much on eyes — there’s a great story about salivary glands, and treating cancer with gene therapy.

  7. Thank you for your illuminating blog . I was also a bit surpised by the result of the PNAS paper and also could not easily tell if the thining of the retinal epithelium was affected by the AAV2RPE65 treatment . AAV has become the safest and most stable gene therapy vector and as a teacher and a researcher on gene therapy, I hold up the results with Leber Congenital Amaurosis as a message of hope for the future treatment of hundreds of other rare diseases .

Leave a Reply

Your email address will not be published. Required fields are marked *

Add your ORCID here. (e.g. 0000-0002-7299-680X)

Back to top