Gene therapy delivered for the first time via lumbar puncture showed promise of benefit in slowing progression of giant axonal neuropathy (GAN) in a phase I clinical trial.
The slope of change in motor function went from -7.17 percentage points per year before treatment to -0.54 percentage points with the lowest dose to an improvement of up to 5.32 percentage points per year with higher doses.
That secondary endpoint was a sign of "possible clinical benefit," Carsten G. Bönnemann, MD, of the National Institute of Neurological Disorders and Stroke in Bethesda, Maryland, and colleagues wrote in the . Bayesian analysis showed corresponding posterior probabilities for slowing the slope of progression of neurological decline of 44% to 99% across doses.
Adverse events were common in the trial of 14 children getting intrathecal administration of a self-complementary adeno-associated viral serotype 9 vector containing the GAN transgene (scAAV9/JeT-GAN) at doses of up to 3.5 × 1014 total vector genomes.
But that was as expected for the sick population studied, said co-author Steven J. Gray, PhD, of the University of Texas Southwestern Medical Center in Dallas. "Overall, I think the safety of it is very good ... in line with what's been seen with other gene therapy or better."
GAN is a rare disorder of sensorimotor neuropathy with onset in the first few years of life that progresses to loss of independent ambulation typically by age 10 and death before age 40. It has been considered a candidate for gene therapy because it is caused by biallelic loss-of-function variants in a single gene -- GAN, the gene encoding gigaxonin.
The development of next-generation viral vectors around 2010 was promising for GAN because of their ability to target the brain, Gray told ֱ. Indeed, the second FDA-approved gene therapy was for spinal muscular atrophy using this AAV9 technology via IV injection into infants.
While transformative for those patients, "scaling that approach outside of infants has been problematic," Gray noted. Systemic gene therapy doses largely go to the liver, with little reaching the brain; and the high doses needed for larger patients raise safety concerns. Injecting gene therapy straight into the brain is possible but doesn't distribute well beyond the injection site.
Intrathecal delivery "concentrates the gene therapy to be directed more towards the nervous system tissues, the brain, and the spinal cord" that are impacted by the disease, Gray said. However, "When we were doing this, this general approach had never been done in a human. No one had ever deliberately injected viruses into somebody's spinal fluid."
Their trial gradually ramped up doses as safety was shown in the initial patients, all of whom were over age 6 years and already had significant, genetically-confirmed disease at baseline.
While IV delivery of gene therapy has resulted in fatal complications in some cases, the two deaths in this trial were deemed unlikely to be treatment related. During the median safety observation period of 68.7 months, one died after spinal fusion surgery after an episode of postoperative emesis-induced aspiration followed by anoxemia, resulting in cardiac arrest and multiorgan failure 8 months after dose administration. In keeping with respiratory failure as a leading cause of death in GAN, the second death in the trial was a patient with recurrent pleural effusion in the setting of respiratory insufficiency, who died of respiratory failure at 60 months after gene therapy.
The most common of the 48 serious adverse events were scoliosis (nine cases), urinary tract infection (six cases), and upper respiratory tract infection (five cases). One serious adverse event, fever with emesis, occurred 2 weeks after dose administration; that patient also had grade 3 elevations in C-reactive protein and grade 1 elevation in B-type natriuretic peptide deemed also possibly related to the gene therapy. These events resolved within 48 hours. A grade 3 exacerbation of benign familial neutropenia 3 days after dose administration resolved 8 days later.
The only adverse events that were more likely linked to the treatment -- transient fever and transient presence of white blood cells in spinal fluid -- were not very clinically significant, Gray said.
Postmortem findings showed scAAV9/JeT-GAN vector DNA and GAN transgene expression broadly distributed throughout the nervous system. But, persistently elevated serum and cerebrospinal fluid AAV9 neutralizing antibody levels after treatment suggested that "systemic and probably intrathecal re-administration is most likely precluded in current or future studies," the researchers noted.
One surprise finding was restoration of sensory responses in the upper extremity in some patients.
"Between 6 and 24 months after gene transfer, sensory-nerve action potential [SNAP] amplitudes increased, stopped declining, or became recordable after being absent in 6 participants but remained absent in 8," the researchers reported. "Loss of SNAP amplitude responses in the median and ulnar nerves occurs early in the disorder and, in our anecdotal experience, does not re-emerge."
Gray called the results promising but likely hampered by use of an aggregate measure that looks at symptoms unlikely to be improved in patients with nerves already significantly damaged by the disease.
The proximal trunk, arms, and hands saw a stronger therapeutic benefit, which was not unexpected given that longer nerves like those in the legs are affected earlier, he suggested.
"It's probably not a realistic hope that you would just essentially cure everybody; and the future direction would be to go into younger patients kind of before a lot of the damage is done," Gray told ֱ. "Once the nerves are completely dead, then we can't rescue them. ... If we could go into younger patients and prevent the damage from happening rather than repairing the damage afterward, everyone would expect it would work better." That's possible now that many patients are being identified before age 5.
Another option for the phase III trial that is likely to follow this one (skipping phase II, given that the phase I trial was already in the target population) is to increase the dose further, he said.
Other rare genetic neurological disorders are already traversing the road paved by the pioneering work done with gene therapy in GAN. Two dozen trials have "kind of copy-pasted this trial design ... and you could probably envision hundreds more," Gray said.
The intrathecal method for getting gene therapy into the central nervous system in older children and adults could even reach as far as treatment for Alzheimer's disease, he speculated.
"What this clinical trial did is it basically provided the delivery vehicle and sort of how you're going to deliver the cargo, how you're going to deliver that cargo safely," Gray said. "So now it's just a question of, well, what cargo are you going to have it carry and what diseases are you going to apply that to? And that's really just wide open for imagination and innovation."
Disclosures
The trial was supported by the National Institute of Neurological Disorders and Stroke, Hannah's Hope Fund, Taysha Gene Therapies, and Bamboo Therapeutics–Pfizer.
Bönnemann disclosed relationships with Genethon, Kate Therapeutics, Rocket Pharmaceuticals, Sarepta Therapeutics, Seal Therapeutics, and SOLID Bioscience.
Gray disclosed relationships with Taysha Gene Therapies and the NIH as well as patents related to gene therapy.
Primary Source
New England Journal of Medicine
Bharucha‑Goebel DX, et al "Intrathecal gene therapy for giant axonal neuropathy" N Engl J Med 2024; DOI: 10.1056/NEJMoa2307952.