Seeing Is Believing

In the last few years, a handful of small but provocative clinical studies have demonstrated that gene therapy can improve vision in patients with a rare type of inherited blindness called type 2 Leber congenital amaurosis (LCA2). Now that the clinical benefits from just one injection of the therapeutic gene have been shown to persist for several years, researchers are hoping to expand this approach to treat more common retinal diseases.

Studies of the use of gene therapy to correct defective genes responsible for disease found the approach to be safe and effective in animal models as well as in a limited number of clinical studies. But while genetically engineered viral vectors are adept at penetrating cells and efficiently transferring genes, they also have the potential to elicit negative effects, such as damage to immune responses, that can cause illness or death. Retinal gene therapy trials have relied on viral vectors, and so far aggressive challenges from the immune system have not been reported, perhaps due to the immune-privileged status of the eye. Nevertheless researchers continue to search for better and safer gene delivery vectors, including nonviral vectors such as peptides with cell-penetrating properties, to deliver genes in preclinical ocular gene therapy studies.

“It's a very exciting time for the field of ocular gene therapy that's been close to 20 years in the making,” noted Jean Bennett, MD, PhD, professor of ophthalmology at the University of Pennsylvania Medical School, Philadelphia.

In the early 1990s—around the same time that researchers were identifying many mutations that caused various blinding diseases—Bennett's group demonstrated that it was possible to use recombinant viruses to deliver genes safely and stably to retinal cells. This work set the stage for testing gene rescue in various animal models of eye disease by the end of the decade. Bennett's pioneering work in hereditary blindness in mice and dogs provided a foundation for applying gene therapy to treat LCA in children and adults.

Comprising a group of degenerative diseases of the retina, LCA is the most common cause of congenital blindness in children. One genetic form, LCA2, is caused by a mutation in the retinal pigment epithelium–specific 65-kD protein gene (RPE65). The RPE65 protein is required to keep light-sensing photoreceptor cells—the rods and cones of the retina—in operating order. LCA2 is rare, affecting only about 2000 people in the United States, but it is untreatable and causes blindness early in life.

“It's a devastating diagnosis for parents because they’ve been basically told there is nothing you can do,” said Bennett. “So it was very exciting when the mutation causing the disease in the dog was identified.”

The researchers inserted a healthy copy of the RPE65 gene into a genetically engineered adeno-associated virus and treated 3 affected dogs. Within 2 weeks of treatment, “dramatic” results were seen in the dogs, who were able to navigate with little problem, said Bennett.

Importantly, the results from a single injection have persisted in the first dog, treated 10.5 years ago. Bennett noted that the cells in the retina are terminally differentiated at birth, meaning that they do not divide and the gene does not get diluted out. “It's an ideal target tissue,” she said.

So far Bennett's group and a few others have published results from phase 1 clinical trials in participants with LCA2. Bennett's study, carried out at the Children's Hospital of Philadelphia, enrolled 12 people aged 8 to 44 years, 5 of whom were children aged 8 to 11. Each participant received 1 injection in the subretinal area of only 1 eye.

All participants showed improved aspects of their retinal and visual function within about a month, and no significant safety concerns arose (preliminary study: Maguire AM et al. N Engl J Med. 2008;358[21]:2240-2248; and 1.5-year follow-up: Simonelli F et al. Mol Ther. 2010;18[3]:643-650). While the adults showed benefit, the most dramatic results were in the younger children. These findings are not surprising, said Bennett, given that there is a degenerative component of this disease; as the person ages, cells die off, leaving fewer cells to resuscitate.

The ideal treatment would be to deliver this therapy between 6 months and 3 years, said Bennett, and in their next study, the group will enroll children aged 3 years or older. She and others are also looking into the safety of providing treatment to the other eye, which all the patients have been requesting.

At the same time, investigators at the University of Pennsylvania and the University of Florida, Gainesville, carried out another clinical trial for LCA2, sponsored by the National Eye Institute at the National Institutes of Health in Bethesda, Md. These researchers took an approach similar to that of Bennett's group, and their results were equally promising (Hauswirth WW et al. Hum Gene Ther. 2008;19[10]:979-990; Cideciyan AV et al. Proc Natl Acad Sci U S A. 2008;105[39]:15112-15117).

William Hauswirth, PhD, professor of ophthalmology at the University of Florida in Gainesville, one of the leaders of this study, noted that his group sought to quantify the vision gain in the study participants by looking not only at how much function the participants gained but also where in the retina this gain was located. He attributes this approach to his colleague Samuel Jacobson, MD, PhD, professor of ophthalmology at the University of Pennsylvania, who has worked for 4 decades to cure vision diseases. Hauswirth and his colleagues found that the only portion of the retina that showed improvement was the area that received the gene therapy treatment, which demonstrates that the treatment was responsible for the benefit.

In addition to the visual improvement that occurred soon after treatment, some patients experienced additional gains between 9 and 12 months afterward. “We believe that this [finding] involves cortical adaptation to the therapy—the brain needs time to adapt in order to use the new function,” said Hauswirth. He explained that the researchers treat an area in the retina somewhat off center of the fovea—the central part of the retina where the sharpest vision normally occurs. What seems to be happening between 9 and 12 months is that some participants begin using this treated area—the pseudo-fovea—to focus and may even switch between the 2 regions unconsciously.

In a recent preclinical gene therapy study in which Hauswirth and colleagues restored color vision in color-blind squirrel monkeys, improvement did not occur until 5 to 6 months after treatment (Mancus K et al. Nature. 2009;461[7265]:784-787). This therapeutic delay, along with that observed in the LCA study, suggests that researchers should be alert for delays in treatment response in future gene therapy trials, Hauswirth said.

Looking ahead, Hauswirth noted that a human foveal disease called blue cone monochromacy, which renders those with the disease almost blind, is perfectly suited to follow-up of his work with color-blind monkeys. In addition, he and other researchers hope to move forward soon in treating other genetic forms of LCA.

Researchers are also exploring gene therapy as a means to treat other eye diseases. For example, a phase 1 clinical trial is currently under way for the wet form of age-related macular degeneration using a gene therapy approach to stop blood vessel growth under the retina (http://www.clinicaltrials.gov/ct2/show/NCT01024998?term=NCT01024998&rank=1). The approach uses a viral vector to deliver a gene encoding a protein that binds to and inactivates the vascular endothelial growth factor molecule, which stimulates vessel growth. This mechanistic action is similar to that of the antibody drug ranibizumab, which currently is used to treat the disease; however, the antibody treatment requires monthly injections into the eye, whereas the gene therapy approach theoretically requires only one long-lasting ocular injection.

The workhorses of gene therapy—viruses—deliver genes effectively to tissues in vivo because that is exactly what they have evolved to do over hundreds of millions of years. But they are not without their limitations.

Rajendra Kumar-Singh, PhD, associate professor of ophthalmology at Tufts University School of Medicine, in Boston, investigates the use of gene therapy to treat retinal diseases, and he has focused on solving the problems that viral vectors can cause. Kumar-Singh and his colleagues recently developed a nonviral nanoparticle that duplicates a viral vector's ability to deliver therapeutic genes to retinal cells. This PEG-POD nanoparticle consists of DNA conjugated to a small peptide for ocular delivery (POD) of DNA linked to polyethylene glycol (PEG), which prevents nanoparticle aggregation.

Recently Kumar-Singh and his colleagues showed that they could delay progression—at least temporarily—of light-induced retinal degeneration in a mouse model using this nonviral vector carrying a gene for the protein GDNF (glial cell line–derived neurotrophic factor), which protects photoreceptor cells in the eye (Read SP et al. Mol Ther. doi: 10.1038/mt.2010.167 [published online August 10, 2010]). Injection of the therapeutic nanoparticle into the subretinal space in the mouse eye significantly reduced light-induced apoptosis of the photoreceptor cells and improved eyesight in the mice temporarily.

As Kumar-Singh noted, the majority of retinal diseases, including retinitis pigmentosa and age-related macular degeneration, have a common pathway of photoreceptor loss through apoptosis, and the disease model of retinal degeneration they used is a general one, rather than being specific for one disease. In addition, the use of the gene for a growth factor makes this approach more generic and potentially applicable to a wide spectrum of retinal diseases, he said.

Kumar-Singh and his team are working to improve the efficiency and longevity of this approach before moving into clinical studies.