Gene therapy side effects prompt search for new delivery methods

SAN FRANCISCO — Just a few years since the first gene therapies became available, there has been a shift among small biotechs and large pharma companies about how to best deliver the edited genes, in particular regarding the viruses and other vectors that insert gene therapy payloads into cells.

The goal of these new delivery techniques is a more precise mechanism less likely to produce harmful side effects, and the shift is likely to ripple through the sector.

Last week, for example, Pfizer announced plans to “externalize” some early-stage gene therapy programs; in particular, those that rely on viruses that have been modified to deliver the therapy to the inside of cells, but can also trigger safety issues when the immune system spots a foreign invader.

And on Friday, the European Medicines Agency asked healthcare providers to conduct liver screening tests on patients treated with Novartis’ gene therapy Zolgensma, after the company reported in August that two individuals died after taking the drug.

While solving the delivery puzzle will be gene editing’s next step, research is still nascent. There will be bumps along the way, as the next generation of biopharmas takes shape. There are other pieces left to solve too — some gene therapies still don’t work even when they appear safe.

Investigators are still digging into how the Zolgensma deaths occurred, but said the ultimate cause of the deaths was acute liver failure. Zolgensma’s genes are delivered through an adeno-associated virus, or AAV, which in many cases target the liver. The vectors, however, are popular for their ability to insert edits outside liver cells, a feat that hasn’t been clinically reproduced by other delivery methods.

Pfizer’s hemophilia A and Duchenne therapies, which also use adeno-associated viral delivery, were also previously placed on FDA hold, though both have since been lifted. The company plans to continue with development of those therapies, which are in later stages than the ones it will give up, as well as a treatment for hemophilia B. Pfizer declined to comment for this story.

David Liu

Not all are ready to write off first-generation gene therapies, though. The Broad Institute’s David Liu, who has founded or co-founded several genetic medicine biotechs, said the pessimism around adeno-associated viruses is premature.

“I certainly don’t think the current pessimism towards AAV should be interpreted as a general harbinger of AAV’s potential to continue to deliver important medicines,” Liu said in an email. “While first-generation AAVs are not without their drawbacks, they remain one of the only clinically validated macromolecular delivery methods that can access multiple organs beyond the liver. And next-generation AAVs are showing promise in monkeys and I’m hopeful that their clinical readouts will also be encouraging.”

Non-viral approach

Regardless, investors are turning toward other, newer gene therapy delivery technologies. While much of the early interest in the field revolved around viral delivery, Tessera Therapeutics, a Flagship-launched biotech, is developing a tool it describes as “gene writing.” The company aims to not only use non-viral methods but also to deliver to cells outside the liver.

Michael Severino

If everything works as it should, it would allow for integrating DNA without cutting it or relying on repair pathways to stitch it back together, Tessera CEO Michael Severino told Endpoints News. And it uses lipid nanoparticles (LNPs) to deliver the gene editors, instead of AAVs.

“We’re using an extension of that [LNP] technology,” Severino said. “In other words we’re modifying it. You can’t just take that plug and play, but you can expand it, you can extend it.”

At last week’s JP Morgan conference, Tessera touted preclinical data from non-human primate studies showing 30% delivery to hematopoietic stem cells (those that haven’t left the bone marrow) and 45% to T cells. Any eventual drug would still have to clear several hurdles, however, and translating preclinical results to humans thwarts research frequently.

Tessera’s approach is akin to prime editing, using similar biological mechanisms to edit genes. Where they differ, Severino writes in an email, is “prime editing systems can write small segments of DNA. With RNA Gene Writers, we can make entire gene additions and precise edits.”

The hope, like many newer genetic medicine biotechs, is to avoid the immune responses seen with viral vectors, because human bodies have evolved to treat foreign DNA as an attack, Severino said. Tessera’s LNPs could insert the edits in non-liver cells more safely than virus-based delivery, he added.

Tessera’s ambitious undertaking has caught the attention of blue-chip investors like SoftBank, and the company has raised more than $600 million since its founding in 2018. (Endpoints also previously highlighted co-founder Jake Rubens as part of its 20 Under 40 series.) Whether it will bear fruit is up in the air, and Severino declined to shed light on when researchers might start the first human testing.

“It’s not far off, but we’ve not commented on specific timing,” he said.

Efficacy issues

Sometimes, genetic medicines do prove safe, but still don’t work. Not every first-generation drug like Zolgensma and Spark Therapeutics’ Luxturna — which use AAVs but provide a copy of a missing gene rather than edit genes inside a living person — makes it to the finish line.

Few biotechs embody this struggle more than Editas Medicine, where Liu is a co-founder, and which is on its third CEO since 2019 after multiple clinical disappointments.

Editas’ latest chief executive, former Sarepta CMO Gilmore O’Neill, announced last week the biotech would enact a strategic shift to prioritize an ex vivo therapy for sickle cell disease and beta thalassemia, called EDIT-301, two months after saying it would seek to partner out its now-former lead program, EDIT-101. Layoffs affecting about 20% of employees came with the move.

Though that ex-lead drug, which used AAV, had a clean safety profile, only three of 14 patients responded to the therapy. Just one out of 12 individuals who possessed one copy of the affected gene, rather than two, achieved a “clinically meaningful” response, per a November announcement.

When Editas revealed its first data cut in September 2021, experts previously told Endpoints the therapy’s safety data were a significant step forward for the field as a whole. But EDIT-101 still ended up a bust, highlighting how safety isn’t the only piece of the puzzle.

Those data, in addition to results from other companies, have helped provide a foundation even if the drug didn’t work, O’Neill said.

Gilmore O’Neill

“We now have multiple proofs-of-concept for that technology within 10 years,” O’Neill said. “If you actually look at the history of biotech, that really is actually a remarkable success.”

Liu agreed, and emphasized there are some next-generation technologies he’s excited about such as engineered virus-like particles (he’s published at least one paper on this) that can combine the strengths of both viral and non-viral delivery. But he also noted the challenges of trying to develop new tech, saying there have only been “a handful of fundamentally different ways” to edit the human genome since the DNA double-helix structure was discovered in 1953.

“The recipe for long-term success in this space is no secret; it takes a perfect storm of innovative and rigorous foundational science, talented people capable of building and sustaining an outstanding culture, and constant execution and innovation at a very high level,” Liu said.

therapeutics
gene therapy
stem cells
cmo
gene editing
biotech
pharma
healthcare
fda
research