A gene therapy trial that recently led to dramatic benefits for babies born with a fatal neuromuscular condition has raised hopes for using a similar approach to treat other diseases. But a new animal study suggests that the high doses of gene-carrying viruses used in such treatments may not always be as safe as the human clinical trial indicated.
In the new research, the disclosure of which briefly sent the stock prices of several gene therapy companies plummeting yesterday, researchers injected a handful of young monkeys and pigs with many copies of adeno-associated virus 9 (AAV9), a normally harmless virus that infects neurons and is increasingly being used to ferry therapeutic genes into cells to treat neuromuscular diseases. Within days, some of the animals developed severe liver and neuron damage.
The results have garnered attention in part because they come from the lab of gene therapy researcher James Wilson at the University of Pennsylvania, who led a 1999 trial in which a teenager died from an immune reaction to a gene therapy vector. Based on these latest animal results, Wilson is urging researchers planning to treat patients with high-dose AAV9 gene therapy to look for similar side effects.
Despite that warning, he and others in the gene therapy field say that doesn’t mean ongoing clinical trials with the gene carrier should be halted. “This is a big deal, potentially,” says Terence Flotte, a gene therapy researcher at the University of Massachusetts School of Medicine in Worcester and editor-in-chief of Human Gene Therapy, which published the new study online on 29 January. However, Flotte writes in a commentary, the community should not “overreact.”
Among the so-called vectors used in gene therapy to deliver DNA, AAV9 has proved particularly good at spreading through neural tissue and at high doses can even cross the blood-brain barrier to reach spinal neurons and the brain. In the first human trial with AAV9 published last November, 15 babies with a severe form of spinal muscular atrophy, a neurodegenerative disease that typically kills by age 2, received an infusion of the viruses carrying a missing gene called SMN. Most of the babies can now sit up, and two are walking.
But in Wilson’s study, three young rhesus macaque monkeys that received a high-dose intravenous infusion of a similar AAV9 carrying SMN developed signs of liver toxicity, and one had to be euthanized. Three piglets given the same treatment had motor neuron damage, could no longer walk, and were also euthanized. These reactions don’t seem to involve an immune response to the AAV9's protein shell or the gene it's carrying, unlike side effects seen in previous AAV studies, Wilson’s group reported.
The lab plans to publish similar side effects for another AAV9 variant carrying a different gene that was tested in one monkey, suggesting the results apply broadly, they write. Wilson and colleagues note that at least five planned or ongoing clinical trials are using high doses of AAV9 or other AAV variants to get genes into muscle or neural cells. These studies and other safety tests in animals “should include careful monitoring for similar toxicities,” they write. As his findings were nearing publication earlier this month, it became public that Wilson recently resigned from the board of Solid Biosciences, which is planning to use intravenous AAV9 gene therapy to treat Duchenne muscular dystrophy. (A clinical trial for that treatment is under scrutiny by the U.S. Food and Drug Administration [FDA], effectively halting, for now, its use of high AAV9 doses.)
Flotte notes several reasons the new results may not apply to the human trials: The AAV9 injected by Wilson’s team isn’t the same type used in the clinical trials; the study used a human version of SMN gene, which could make it toxic to the animals; and the researchers didn’t rule out the possibility that the treatment was contaminated. In their commentary, Flotte and a co-author write that AAV9 gene therapy trials should continue because they are potentially saving lives. “Let’s study this, let’s not ignore it,” Flotte adds. An FDA spokesperson declined to comment on the study.
Wilson also does not want to “derail anything at this point.” He agrees that the new experiments differed in significant ways from previous studies of AAV9 safety. Moreover, because labs and vector production facilities use different methods to measure the AAV9 dose, it’s difficult to compare doses across studies. “I just hope this is useful information” that will inform researchers developing new gene therapies, Wilson says.
Gene therapy researcher Jude Samulski of the University of North Carolina in Chapel Hill, who has never seen such toxic effects in his primate studies with AAV9, predicts that many labs will now request Wilson’s vector and repeat his experiments. “If this is truly a concern, it needs to be reproducible,” he says.