September will be 25 years since the first gene therapy experiment, and FDA approval is finally in sight. Several gene therapies are approaching the finish line, awaiting results from comparisons to existing therapies and analyses of long-term efficacy.
Among the contenders are:
• lipoprotein lipase deficiency (already available in Europe)
• childhood cerebral adrenoleukodystrophy (late clinical trials)
• SCID-X1 (better than allogeneic stem cell transplant; more boys got better and did so faster)
• Leber congenital amaurosis type 2 (RPE65 blindness mutation; late clinical trials, long-term effects published soon)
Because thousands of single-gene disorders are theoretically candidate for gene therapy, yet only a few companies are shepherding treatments towards commercialization, I’m intrigued by how they choose their targets.
Applied Genetic Technologies Corp (AGTC) is one such company that has carved its niche in “orphan ophthalmology,” focusing on a trio of single-gene disorders in which the photoreceptors do not degenerate: X-linked retinoschisis, achromatopsia, and X-linked retinitis pigmentosa. Five experts in the use of adeno-associated virus (AAV) to deliver genes founded the company.
AGTC’s vectors introduce functional versions of mutated genes, enabling the targeted cells to produce the crucial proteins whose production the mutation impairs. The company recently announced collaboration with 4D Molecular Therapeutics to develop new AAV vectors.
(A note concerning the recent flurry of reports on genome editing using CRISPR and other technologies: Gene therapy as envisioned since the first trial in 1990 delivers functional genes that supplement the actions or inactions of mutant genes. It doesn’t replace mutant genes as genome editing does. Some reports mix these up.)
I spoke recently with AGTC President and CEO Sue Washer.
WHY DID AGTC CHOOSE THESE THREE CONDITIONS?
Several reasons. These diseases are very well understood at the cellular and DNA levels, and we know how a missing protein affects vision. Robust animal models have the same genetic defect as human patients. Screening and testing efficacy is straightforward. Unlike other orphan drug spaces in which companies and researchers spend a lot of time figuring out clinically meaningful endpoints to negotiate with regulators and do tests reliably, with ophthalmology, you know the endpoints: visual acuity, visual field, and contrast sensitivity.
WHAT IS X-LINKED RETINOSCHISIS, AND HOW CAN GENE THERAPY HELP?
Using OCT (optical coherence tomography) you can see the layers of specialized cells in the eye. In a patient with XLRS, the layers are pulled apart because a protein is not there to hold them together. Because the layers of the retina are not talking to teach other, electrical signals when photons hit can’t get to the back of the eye, even though the photoreceptors function.
XLRS affects 35,000 males in the US and Europe. All patients have a mutation in the RS1 gene that produces structural proteins in the extracellular matrix that form complexes that interact with cell surface molecules. The only treatment is off-label use of carbonic anhydrase inhibitors – glaucoma drops. These dehydrate the back of the eye so the retinal layers lay down on each other, but results are anecdotal and it doesn’t always help visual acuity.
By intravitreal injection of AAV with the correct copy of the gene, transduced cells in the macula and fovea (the area of densest photoreceptors) send the protein into the space and pull the layers of the retina back together. AAV supports secretion of the normal protein for the life of the cell because retinal cells don’t turn over. We expect human phase 1/2 clinical trial data by the end of the year. The first few cohorts will be over age 18, but once the maximum tolerated dose is determined, we’ll expand to age 6.
Oliver Sachs’ book “The Island of the Colorblind” made achromatopsia famous. (A typhoon in 1780 decimated the population of the Pingelapese people on an eastern Caroline island, and when the population grew anew from a few surviving individuals, up to 10% of them became blind in infancy. (It’s a classic population bottleneck.)
Achromatopsia is more than colorblindness. In typical X-linked colorblindness a man has normal visual acuity but can’t see red or green. In achromatopsia all 3 cone types have no function and the person only has rod vision, seeing in black and white and shades of grey. (It is autosomal recessive.) When the lights are on, the person is completely blind. A person with vision going to the bathroom in the middle of the night and turning on the light can see because the cones turn on. In achromatopsia, they don’t. People are severely photophobic, legally blind, and even in a normally lit office building wear heavily tinted glasses. Outside they wear goggles.
In the US and Europe 28,000 people have achromatopsia. Half have mutations in the CNGB3 gene (cyclic nucleotide-gated channel type B3) and another 25% in the CNGA3 gene. Both encode proteins that form channels in cones through which photons enter. In achromatopsia, photons won’t trigger the cascade, interrupting the visual pathway. We are developing gene therapy for each type. A proprietary engineered promoter only allows gene expression in the cell membrane of the photoreceptors.
WHAT IS THE STATUS OF GENE THERAPY FOR X-LINKED RETINITIS PIGMENTOSA?
RP is a disease class caused by at least 150 different gene mutations. X-linked RP accounts for 10% of cases, and about 90% of them, or 20,000 people, have the RPGR gene mutation. RPGR (retinitis pigmentosa GTPase regulator) is a protein that helps the phototransduction cascade from the inner to the outer segments of the photoreceptors.
The disease affects the rods initially, but the cones over time. It is progressive, starting with nightblindness and then constriction of the visual field until by the 50s or 60s there is only tunnel vision. Later in life people lose central vision as well. We’re working with a dog model to deliver AAV and improve visual function in the dog’s eyes and are beginning dosage studies in non-human primates in preparation for a toxicity study. We will file an investigational new drug application next year.
WHAT ARE THE BEST MODEL ORGANISMS TO STUDY EYE DISEASES?
Lower mammals – mice, dogs, and pigs — have retinal cells that use the same phototransduction pathway as primates, but the structure of the eye is different. They have no macula or fovea or inner limiting membrane (which separates the retina from the vitreous body), as primate eyes do. Mice have tiny eyes, highly disorganized retinas, and cell types that eventually spread throughout the retina.
When we select the capsid (viral protein coat), promoters (genetic controls), and physical delivery methods for human clinical trials of a gene therapy, we need to screen in non-human primates. Capsid and promoters that work astoundingly well in mice, dogs and pigs don’t work well in primates. So we need 2 sets of data: the lower animal model to see if the protein goes to the right place and can have a clinical effect, and primate data to show that we can get the protein to the right place. Having these two sets of data significantly improves chances of success in human clinical trials.
TWO LITTLE GIRLS
With all of the pieces that must fit exactly right for delivery of a gene to have a therapeutic effect, it isn’t surprising that gene therapy has taken a quarter of a century to get off the ground. In 2 weeks, DNA Science will revisit two incredible families about to embark on the gene therapy journey, if they can overcome potential problems posed by the immune system.
Eliza O’Neill and her family have been in self-imposed quarantine in their home in South Carolina for nearly a year, to keep her virus-free so that she might be selected for a clinical trial of gene therapy to treat Sanfilippo syndrome A.
Hannah Sames is not among the first children to participate in the clinical trial for giant axonal neuropathy that has just begun, and that her family largely funded, because she doesn’t make the missing protein. Her immune system might reject healing genes.
Stay tuned …