Gene Therapy Given Directly to Spinal Canal Might Be Safer With ‘Silencing’ Step

By taking advantage of a natural process of gene silencing, a new gene therapy approach appears to prevent the toxicity to dorsal root ganglion (DRG) — a specific cluster of sensory neurons — seen in non-human primates during gene therapy studies for neurological disorders, researchers report.

The approach successfully protected the primates’ DRG from excessive activity of the introduced gene — known as a transgene — and subsequent toxicity, without affecting the transgene’s activity elsewhere in the nervous system.

DRG inflammation and toxicity was observed in non-human primates after spinal canal injection of AVXS-101 — approved as Zolgensma when given as an intravenous infusion — prompting a partial hold of a Phase 1/2  trial (NCT03381729) to allow for further investigation.

That trial, called STRONG, is testing intrathecal (spinal canal) injection of this gene therapy in children with spinal muscular atrophy (SMA) between 6 months and 5 years old.

The study, “MicroRNA-mediated inhibition of transgene expression reduces dorsal root ganglion toxicity by AAV vectors in primates,” was published in the journal Science Translational Medicine.

“We believe that this new approach could improve safety in gene therapy universally,” as well as in SMA, Juliette Hordeaux, PhD, the study’s first author at University of Pennsylvania’s Perelman School of Medicine (Penn Medicine) said in a press release.

“This approach could be used to design other gene therapy [carriers] to repress transgene [activity] in the cell types that are affected by the toxicity and not others, which is critical, because you need [such activity] everywhere else to effectively treat the disorder,” added Hordeaux, who is also senior director of translational research for Penn Medicine’s gene therapy program.

Gene therapy works to deliver a functional version of a gene to correct or replace a faulty gene within specific cells in the body. Most therapy approaches today use a modified and harmless adeno-associated virus (AAV) as a “carrier” for the working gene, a vehicle to transport it to a target cell.

But previous studies, including primate work by Penn researchers, showed that AAV-based gene therapies targeting the central nervous system (CNS; brain and spinal cord) can damage the DRG, a cluster of neurons of the spinal nerve that bring sensory information from the periphery to the spinal cord.

Notably, DRG toxicity was observed in studies regardless of the therapy’s route of administration (directly into the bloodstream or into the spinal canal). No reports of such toxicity in humans, including children treated in STRONG, are known.

Conventional immunosuppressive regimens were ineffective in preventing this toxicity, strongly indicating that an excessive immune response was not its cause. This led Hordeaux and her Penn colleagues to evaluate whether the damage was related to excessive levels of the transgene’s product, which could cause cellular stress.

To test their idea, they took advantage of RNAi, a natural process of gene silencing, in which microRNA (miR) molecules bind to a specific messenger RNA (mRNA), targeting it for destruction and preventing the production of that protein. (mRNA is the molecule derived from DNA and used as a template for protein production.)

Since the miR183 complex is specifically produced in sensory tissues and organs such as dorsal root ganglion, the researchers introduced miR183’s sequence targets at the end of the transgene sequence in an AVV. With this, any mRNA produced from the transgene would be destroyed in DRG neurons by the naturally present miR183, preventing protein production in these cells.

Researchers then compared the effects of administering AAV with a transgene containing — or not containing — miR183’s sequence targets into the cerebrospinal fluid (the fluid that bathes the brain and spinal cord) of non-human primates.

Introducing miR183’s sequence targets in the transgene were found to significantly reduce its mRNA levels and subsequent toxicity in DRG neurons, without affecting the transgene’s mRNA levels elsewhere in the primate’s brain.

Steroids given to primates treated with  unmodified AAVs, in contrast, failed to alleviate DRG damage, despite their known anti-inflammatory and immunosuppressive effects. This ineffectiveness was “consistent with the proposal that immune system activity does not mediate this neuronal toxicity,” the researchers wrote.

“We were concerned about the DRG [toxicity] that was observed in most of our [non-human primate] studies,” said James M. Wilson, MD, PhD, the study’s senior author, and gene therapy program director and a professor of medicine and pediatrics at Penn Medicine.

“This modified [viral] vector shows great promise to reduce DRG toxicity and should facilitate the development of safer AAV-based gene therapies for many CNS diseases,” Wilson added.

Novartis’ gene therapy Zolgensma, when given directly into the bloodstream, is currently available for use in newborns and toddlers up to age 2 with any type of SMA in the U.S. and Japan, and to those with almost all types who weigh up to 21 kilograms (about 46 pounds) in Europe.

To meet a U.S. Food and Drug Administration (FDA) request for a “pivotal confirmatory study” of the gene therapy’s use with older SMA patients, who would be treated via intrathecal (IT) injection, Novartis plans to launch a new AVXS-101 IT trial.

This administration route is favored for those beyond toddler age, as it is thought to better target the motor neurons damaged by the disease.

According to the company, this IT trial cannot include U.S. sites until the hold on STRONG is lifted.