Podcast recap: The state of AAV gene therapy and where it’s headed

On the most recent episode of The Genetics Podcast, host Patrick Short took a solo deep dive into gene therapy, with a focus on therapies based on delivery via adeno-associated virus (AAV), to explore recent breakthroughs and setbacks. With several major biopharma exits, ongoing safety debates, and a wave of next-generation delivery platforms in development, this episode explored where AAV stands today and what needs to change for it to scale.

Why this conversation matters

Gene therapy is at a turning point. It has already transformed outcomes for diseases like spinal muscular atrophy (SMA), yet headlines about liver toxicity, halted trials, and high manufacturing costs show the steep climb still ahead. Patrick set out to clarify what’s driving these shifts: what’s working, what isn’t, and which innovations could bring AAV, and gene therapy as a whole, into its next chapter.

The foundation: gene therapy as “gene transplant”

Patrick frames AAV delivery as a type of gene transplant. Instead of replacing an organ, the aim is to introduce a functional copy of a gene into cells to restore a missing or defective protein. Each therapy is built from three core components:

• A vector to carry DNA into cells
• A promoter that ensures expression in the right tissue
• The transgene (referred to as the “payload”)

Unlike integrating vectors, AAV stays outside the genome, which reduces the risk of insertional mutations. However, this also means the therapy may fade over time in dividing cells.

Where AAV shines and where it struggles

One of the clearest successes so far is Zolgensma, an AAV9-based treatment for SMA. Delivered early, it can dramatically change the course of disease by restoring the function of the SMN1 protein. Importantly, two other non-viral therapies, Biogen’s Spinraza (an antisense oligonucleotide) and Roche’s Evrysdi (a small molecule), show that understanding disease biology can yield multiple viable modalities for the same target.

By contrast, Duchenne muscular dystrophy (DMD) has proven far more complex. Sarepta’s ELEVIDYS uses an AAVrh74 vector to deliver a shortened “micro-dystrophin” gene, but reports of liver toxicity and immune-related complications have underscored the field’s biggest challenge: the same viral properties that make AAV effective at reaching tissues can also provoke strong immune reactions, especially at the high doses required for muscle and systemic delivery.

The core challenges

Safety has been a major focal point in AAV therapy recently. Liver toxicity remains the most common and serious side effect, driven by immune responses to both the viral capsid and the massive amount of vector administered.

Durability is another challenge. Because AAV doesn’t integrate into the genome, its effect can wane as tissues grow or regenerate, particularly in pediatric indications where cells divide rapidly.

Finally, cost and scalability continue to hinder the acceleration of next-gen therapies. Manufacturing AAV at clinical or commercial scale remains resource-intensive. A large fraction of viral particles produced are empty capsids, which add to immune load without therapeutic benefit.

The combination of safety risks, technical complexity, and small target populations has made gene therapy a difficult business case for large companies, even when the science is strong.

Evolving the delivery toolkit

Despite these challenges, innovation around delivery is accelerating. Scientists are engineering new AAV capsids that are more tissue-specific and less immunogenic, using both directed evolution and machine learning to refine them.

Beyond AAV, lipid nanoparticles (LNPs) are being explored for DNA, RNA, and even CRISPR payloads. They’re less likely to provoke immune responses and could enable repeat dosing, though they currently struggle to reach organs outside the liver. Other delivery systems (lentivirus, adenovirus, and synthetic virus-like particles) offer additional options, each with its own trade-offs in durability, safety, and payload size.

A new regulatory and clinical reality

For ultra-rare programs, the traditional Phase 1–2–3 path rarely fits. Teams instead move through small, adaptive trials that start with first-in-human safety and expand as data accumulate. Patrick emphasized that fit-for-purpose endpoints (such as functional milestones, cognition, or caregiver-reported measures) are crucial for these slowly progressive diseases, where measurable change can take years. Regulators have shown flexibility, but programs still need strong quantitative data alongside compelling patient narratives.

What progress will depend on

Patrick outlined several areas where progress could unlock the next phase of growth for gene therapy:

• Safer vectors and immunomodulation strategies to reduce liver impact
• Better manufacturing yields and analytical control to lower cost
• Durability strategies, from redosing with new capsids to complementary RNA or biologic therapies
• Earlier diagnosis through newborn screening, enabling lower-dose intervention
• Pragmatic regulation that reflects the realities of ultra-rare disease development

The outlook ahead

Patrick expects momentum in immunoprivileged tissues, such as the eye, where immune barriers make AAV safer and more durable. As manufacturing efficiency improves and delivery systems diversify, gene therapy’s reach will expand to more common genetic subtypes in neurology and cardiology.

In the long run, he predicts a modality-blended future: gene replacement, RNA therapeutics, and gene editing coexisting. Each therapy can be deployed at the right moment in a patient’s disease course. The key will be pairing strong biology with scalable technology and learning from both the successes and the setbacks of the first generation.

Listen to the full episode below.