n a Phase 3 gene therapy trial intended to improve vision among patients with Leber hereditary optic neuropathy, recipients gained somewhat better sight in both eyes even though only one was treated. The results and an investigation into possible explanations for the findings were published December 9 in Science Translational Medicine.
The paper “has very strong clinical implications that a single injection maybe is enough for bilateral effects,” says Thomas Corydon, who studies ocular gene therapy at Aarhaus University in Denmark and was not involved in the work.
The onset of Leber hereditary optic neuropathy (LHON) is sudden. Patients—usually young men—start losing vision at the center of one eye. Within months, the other eye follows, leaving them legally blind. The disease is caused by a point mutation in the mitochondrial genome that leads to dysfunction and death of retinal ganglion cells, the axons of which make up the optic nerve. About 70 percent of patients have the same mutation, known as MT-ND4.
“If you're going to start somewhere, it makes sense to tackle this variant,” says Patrick Yu-Wai-Man, an ophthalmologist at the University of Cambridge in the United Kingdom. He and his collaborators, including teams from GenSight Biologics and a group led by University of Pittsburg Medical Center ophthalmologist José-Alain Sahel, as well as other groups, previously showed that the point mutation could be corrected in animal models and in cell culture using gene therapy.
It’s difficult to get genetic material into the mitochondrial genome because mitochondria have two membranes, an outer and inner membrane, Yu-Wai-Man explains. In the clinical trial, he, Sahel, and colleagues overcame this hurdle by injecting an AAV vector containing a wildtype copy of the ND4 gene with an added mitochondrial-targeting sequence—a strategy that had already been shown to correctly direct the protein product of ND4 and other mitochondrial genes to the organelle.
Each of 37 patients received the therapeutic virus via a single injection into the vitreous fluid within one eye six to 12 months after the onset of vision loss. They also got a sham treatment in the other eye: a surgeon pressed the eye with a blunt cannula to simulate an injection.
“We thought that, if there was going to be an effect, it would be isolated to that eye and then the other one would be the perfect internal control,” Yu-Wai-Man tells The Scientist. “But as it turns out, that wasn’t the case.”
With a slight delay in the sham-treated eye, both eyes started to improve. By 96 weeks after treatment, 29 of the patients had gained visual acuity in both eyes and reported increases in their quality of life.
“Patients do improve, but, even with the treatment, they still function at a very low level,” says Byron Lam, an ophthalmologist at the University of Miami who was not involved in the study. Most of the subjects were still near legal blindness at the end of the study.
To determine how the bilateral effect might be happening, Yu-Wai-Man and colleagues injected the therapeutic virus into one eye of three monkeys. Three months later, they found viral DNA in the noninjected eye and optic nerve. This raises the possibility that the viral vector supplies the wildtype protein in the untreated eye, but it’s not firm proof.
Finding viral DNA in the untreated eye in primates is “a little short of being definitive because DNA expression alone doesn’t prove that you’re getting a therapeutic effect. Detecting DNA doesn’t mean there is mRNA expression or protein production,” says Mark Pennesi, an ophthalmologist at Oregon Health & Science University who did not participate in the work.
Previous work has shown that there could “be transneuronal spread of the vector, but we also need to keep a critical mind and think that there might be other explanations,” agrees Yu-Wai-Man. It could be that injecting the vector in one eye leads to some form of localized inflammation that induces mitochondrial biogenesis, thus making the mitochondria work better, he adds. Another option is that improvement in one eye leads to reorganization in the part of the brain that interprets signals from the eye, which could enhance vision overall.
“Clearly, further investigations are needed to understand the underlying mechanisms of how the interocular diffusion of viral DNA vector occurs and whether there are other mechanisms by which the optic nerves directly communicate,” Bin Li, an ophthalmologist at Tongji Hospital in China who was not involved in the study, writes in an email to The Scientist. Li explains that his group has also reported that material injected in one eye can reach the other optic nerve.
These findings have implications for how this type of research should be performed in the future, he writes. They’ve shown that “contralateral sham-treated eyes cannot serve as ‘true internal control’ for clinical studies.”
“When you read this paper, you get a little excited, and then in some ways, you get a little disappointed, because it does look like there’s some kind of positive effect with this treatment—that it does do something more than what would happen with just the natural history of the disease. Unfortunately, the results are confounded by the fact that you treat one eye, but then there is improvement in the untreated control eye,” Pennisi tells The Scientist. “The question then really becomes . . . why did you get that result?”
Along with academic collaborators, Yu-Wai-Man, who consults for GenSight Biologics, will continue to explore this question as they focus on ongoing clinical trials of this therapeutic.
P. Yu-Wai-Man et al., “Bilateral visual improvement with unilateral gene therapy injection for Leber hereditary optic neuropathy,” Science Translational Medicine, doi:10.1126/scitranslmed.aaz7423, 2020.
Correction (December 14): The story has been updated to remove mention of a company that was not involved in the work and to specify which fluid compartment in the eye was injected. The Scientist regrets the error.