Retinal Gene Therapy: Long-term Results Less Than Expected

Ricki Lewis, PhD

May 04, 2015

Photoreceptors continue to degenerate years after gene therapy for the childhood blinding disorder RPE65-associated Leber's congenital amaurosis (RPE65-LCA) despite continued clinical benefit in some patients, according to two studies published online May 3 and 4 in the New England Journal of Medicine.

Initial results were encouraging and even dramatic in reports from patients. However, the new findings indicate that the underlying pathology, detectable with various measures even if patients report good vision, continues.

"These papers demonstrate that gene therapy works, and we need to learn better the right doses and circumstances for maximal benefit," Eric Pierce, MD, PhD, director, Ocular Genomics Institute and Berman-Gund Laboratory for the Study of Retinal Degenerations at the Massachusetts Eye and Ear Infirmary, Boston, told Medscape Medical News. Dr Pierce was not involved in the two trials currently reported, but did work on the third LCA gene therapy trial.

Three Trials Showed Early Efficacy

Mutations in at least 19 genes cause LCA by disrupting reactions of the visual cycle. RPE65-LCA disables an enzyme (retinal pigment epithelium [RPE]-specific protein 65 kDa) that catalyzes the reaction in the RPE that produces 11-cis retinal. Because rods require the enzyme, the condition severely impairs night vision from birth. Cones, which have an alternate source of 11-cis retinal, degrade more slowly.

Since 2008, three independent clinical trials have demonstrated the safety and efficacy of introducing functioning RPE65 genes into the RPE via subretinal injection. All three trials used adeno-associated virus 2 as the vector but differed slightly in transgene design. The introduced gene augments function but does not replace a mutant gene.

Photoreceptor Diminution

Samuel G. Jacobson, MD, PhD, from the Scheie Eye Institute at the University of Pennsylvania, Philadelphia, and colleagues report long-term follow-up of three patients, ages 23, 21, and 16 years at the start, 4.5 to 6 years posttreatment. The other 12 patients in the trial did not meet follow-up criteria (4.5 years minimum, one subretinal administration outside the fovea of one eye, no postoperative complications, no cataracts, maintained foveal fixation, and ability to undergo psychological and physical tests).

In all three evaluable patients, the researchers saw continued photoreceptor degeneration.

The researchers assessed visual sensitivity (the ability to see different light intensities after dark adaptation, expressed as a two-dimensional retinal map) and photoreceptor layer thickness, as seen with optical coherence tomography. Thinning indicates areas of photoreceptor degradation.

Visual sensitivity for each of the three patients peaked at 3 years, 1 to 3 years, and 1 year and then diminished. Optical coherence tomography showed thinning at the same rate in treated and untreated areas of the retina.

"The loss of visual function at later times after treatment is in keeping with progression of the degenerative process," the researchers conclude, and hypothesize that waning expression of the transgene may explain the finding. They suggest using the extent of retinal degeneration at baseline to guide prognosis and using retinal thickness to classify patients as mild, moderate, or severe.

However, outside experts raised questions about the newly reported data, noting that the study does not consider patient reports of effect on quality of life. "Would a change in the microperimetry measurements even be noticeable to a patient? Do the subjects feel that they have benefitted? Can they navigate in dimmer environments?" asked Jean Bennett, MD, PhD, the F. M. Kirby Professor of Ophthalmology at the Perelman School of Medicine and leader of the third RPE65-LCA study at Children's Hospital of Philadelphia in Pennsylvania.

The team did not analyze the entire injected region or account for the normal slow loss of sensitivity over time, Dr Bennett added. They also consulted historical data about disease progression "when they could have used the contralateral eye as a control. The disease progresses at different rates in different people," she told Medscape Medical News.

Dosage Effect

In the second article, J. W. B. Bainbridge, MD, PhD, from the University College London Institute of Ophthalmology, University College London, United Kingdom, and colleagues analyzed visual function over the course of 3 years for 12 patients who were aged 6 to 23 years at the trial's start. Four patients received a low dose (1 × 1011 vector genomes) and eight a high dose (1× 1012). The researchers assessed visual acuity, contrast sensitivity, color vision, and spectral sensitivities and used optical coherence tomography and electroretinography imaging.

Six patients experienced improved retinal sensitivity, with peak response at 6 to 12 months. Two high-dose patients showed 10 to 100 times improvement in rod sensitivity, and one high-dose patient had an increase in cone sensitivity that peaked at 12 months. However, rod and cone function did not change on electroretinography from baseline. Three patients had a decline in visual acuity, two of whom reported poorer vision.

Of the eight high-dose patients, five had inflammation or an immune response, whereas none of the four at lower dose did. Four of those five responses were mild, but the other had neutralizing adeno-associated virus 2 antibodies against the capsid.

The adverse effects were likely a result of the inefficient promoter, Dr Bennett said. "Bainbridge et al used a transgene that isn't expressed well, so they had to use high levels of expression to get a response, and that causes toxicity in a huge volume," she explained.

The Bainbridge group also found that the natural animal model for the disease, the Briard sheepdog, maintains improved vision longer than humans after gene therapy. Photoreceptors in dogs degenerate much later than they do in children, and expression of RPE65 is 2.5-fold greater in humans than in dogs, suggesting more severe disease in humans.

"More efficient delivery of RPE65 at an appropriately early point in disease progression will have a greater effect on retinal function and will better protect against progressive degeneration," the researchers write.

Is the Gene Therapy Glass Half Empty or Half Full?

In an accompanying editorial, Alan F. Wright, MD, PhD, from the MRC Institute of Genetics and Molecular Medicine at the University of Edinburgh, United Kingdom, writes that the two studies suggest that "without a highly efficient vector delivery system and sufficient surviving retinal pigment epithelium and photoreceptors, treatment success will be transient."

Dr Bennett is more optimistic, based on what she has seen in her own patients. Her study, which used a somewhat different gene promoter than the other two trials, has not found the evidence of photoreceptor degeneration that the other two did. "I wouldn't say one administration of treatment is going to totally stop the disease in its tracks. Who cares how many photoreceptors you have if you're still seeing? The numbers don't mean anything if the person is still benefiting from the intervention."

Kristin Smedley, the mother of two boys who are profoundly blind with a different form of LCA, agrees: "Until other therapies are available, gene therapy is the best hope right now for my boys. Even if this approach only restores a fraction of vision, that fraction could mean no longer needing a cane to navigate or being able to read print instead of Braille."

The Foundation Fighting Blindness and the National Eye Institute provided funding for these studies. Dr Bainbridge and two coauthors hold financial interests in Athena Vision Ltd. Another two coauthors are inventors of Progressor software, for which they receive royalties. Another coauthor has a financial interest in AGTC Inc. The other authors, the editorialist, and the commentators have disclosed no relevant financial relationships.

N Engl J Med. Published online May 3, 2015. Jacobson full text. Published online May 4, 2015. Bainbridge full text, Editorial full text


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.