AAV Gene Therapy Recalibrates

This past month, researchers announced the use of gene therapy to restore hearing in both children and adults with OTOF-related deafness. (The OTOF gene encodes otoferlin, a protein essential for transmitting auditory signals from the inner ear to the brain.) After a single injection of a functional OTOF gene into the inner ear, participants began regaining hearing within one month.¹ Scientists also announced promising results from a gene therapy trial in patients with hemophilia B, caused by insufficient functional clotting factor IX (FIX). Again, a single infusion restored FIX levels and prevented the need for intervention over a five-year period.²

These successes, and many others, rely on adeno-associated virus (AAV), perhaps the most actively investigated gene therapy vehicle to date. Over the past decade, AAV-based therapies have achieved landmark outcomes, offering one-time treatments for previously intractable genetic disorders. Yet despite these achievements, concerns over long-term safety, immunogenicity, complex manufacturing, and steep costs have prompted a reassessment of AAV’s future. Rather than signaling the end, this moment marks a recalibration that could pave the way for safer, more accessible gene therapies.

The rise of AAV

Initially discovered as a contaminant in adenovirus lab preparations (thus the name adeno-associated virus), AAV consists of a small icosahedral protein shell surrounding a 4.8-kilobase single-stranded DNA genome. Because it requires co-infection to replicate, rarely integrates into the host genome, and can establish long-term gene expression, AAV emerged as a leading method of clinical gene delivery. In nature, AAVs exhibit vast genomic diversity, with at least 12 distinct serotypes—each showing a preference for specific tissues, making them useful for targeting certain organs.³

Recombinant AAV (rAAV), first cloned in the 1990s, has all viral protein-coding genes removed to make room for therapeutic cargo.⁴ To date, eight rAAV-based gene therapies have received FDA approval. The first, Luxturna, approved in 2017, treats retinal dystrophy and resulting blindness (See Table below).

A decade of breakthroughs and setbacks

In a major shift, Pfizer recently announced it is exiting the gene therapy space. This includes discontinuing its hemophilia B treatment, Beqvez, just over a year after FDA approval, citing low demand. The one-time treatment had a list price of $3.5 million. (Pfizer will instead prioritize its antibody-based therapy, Hympavzi.) Vertex, Takeda, and Roche, to name a few more, have also scaled back AAV programs.

These moves reflect a growing recognition of the challenges in translating AAV’s preclinical potential into commercial success. AAV manufacturing is costly and complex, requiring large-scale viral vector production under stringent quality control. Even approved therapies face low uptake due to high costs and lingering safety concerns.

While AAV gene therapy is efficient and relatively safe at lower doses, high-dose treatments carry risks of severe, and sometimes fatal, immune-related toxicities. Alongside the achievements of the past month came tragedy: a 51-year-old man became the third patient to die after receiving Sarepta Therapeutics’ AAV gene therapy for Duchenne muscular dystrophy (DMD), Elevidys, in a phase 1 study involving non-ambulatory patients. All deaths appear to be related to liver complications. (The same drug, approved in 2024 for ambulatory DMD patients, remains the only gene therapy specific to this disease.)

In response, Sarepta added a black box warning outlining the risk, and then on July 22, voluntarily paused all shipments of Elevidys in the U.S. Less than a week later, on July 28, the FDA recommended the voluntary hold be removed and shipments continued for ambulatory patients with DMD.⁵ (Less than one business day after, investigations were announced into the death of a fourth patient—an 8 year old child—this death was later attributed to a viral infection). These cases serve as a painful reminder that although gene therapy is making great strides, it is still very much in its experimental stages.

Immunogenicity is real but context matters

AAV’s greatest limitation lies in its interaction with the immune system. Many individuals carry neutralizing antibodies (NAbs) from natural infections, which can block therapy. Similarly, patients who receive AAV gene therapy typically cannot be redosed due to an adaptive immune response.

That said, context matters. Some serotypes are relatively low in the population. Additionally, low positive NAb titers may not prevent therapeutic benefit—at least in hemophilia B trials.⁶ More serious immune complications, such as liver inflammation, thrombocytopenia, dorsal root ganglia toxicity, and death, are more likely with high-dose systemic delivery. By contrast, local delivery allows for lower doses and less immune exposure.

Immune-privileged sites like the eye and CNS are therefore leading AAV targets. Luxturna, delivered subretinally, has a strong safety profile. Similarly, in late 2023, the FDA approved Kebilidi, the first U.S. gene therapy delivered directly into the brain, for AADC deficiency—a rare, fatal disorder.

Researchers are actively exploring strategies to reduce immunogenicity, including capsid engineering to avoid common antibodies, short-term immunosuppression, in vitro complement testing, and localized delivery methods to limit peripheral exposure.

A path forward for AAV, in the right setting

While some companies are backing away, others are forging ahead. Novartis’s Zolgensma for spinal muscular atrophy has treated thousands of patients across the world and shows no signs of going anywhere. And Roche will continue its ocular and CNS AAV-related applications.

Behind these efforts, the field is recalibrating and focusing on diseases and delivery methods where AAV’s strengths align with therapeutic needs. New stealth capsid variants can evade immune detection and improve targeting. Contract development and manufacturing organizations (CDMOs) are helping scale up production and improve vector quality and yield. By outsourcing manufacturing, developers can focus on research and clinical progress.

Meanwhile, alternative delivery technologies are advancing. mRNA therapies and lipid nanoparticle (LNP) delivery (accelerated by COVID-19 vaccine development) offer transient gene expression, redosing potential, and immune modulation. Lentiviral vectors, widely used in ex vivo therapies like CAR-T, offer some advantages over AAV: they integrate into the genome, are retained through cell division, and allow larger payloads with lower immunogenicity. These innovations can also complement AAV. For instance, CRISPR paired with AAV offers precise genetic editing with durable effects.

While AAV may no longer be viewed as the universal gene therapy solution, it still remains a powerful tool when used in the right context. With more precise strategies and ongoing innovation, the next wave of AAV therapies will likely continue to offer hope in safer and more sustainable ways.

Table. FDA-Approved AAV-Based Gene Therapies

References

1.Qi J, Zhang L, Lu L, et al. AAV gene therapy for autosomal recessive deafness 9: a single-arm trial. Nat Med. Published online July 2, 2025. 

2. von Drygalski A, Gomez E, Giermasz A, et al. Completion of phase 2b trial of etranacogene dezaparvovec gene therapy in patients with hemophilia B over 5 years. Blood Adv. 2025;9(14):3543-3552. 

3. Zhao, Q., Peng, H., Ma, Y. et al. In vivo applications and toxicities of AAV-based gene therapies in rare diseases. Orphanet J Rare Dis 20, 368 (2025).  [Table 1: https://ojrd.biomedcentral.com/articles/10.1186/s13023-025-03893-z/tables/1]

4. Wang JH, Gessler DJ, Zhan W, Gallagher TL, Gao G. Adeno-associated virus as a delivery vector for gene therapy of human diseases. Signal Transduct Target Ther. 2024;9(1):78. Published 2024 Apr 3. d

5. Press Release, July 28, 2025. Sarepta Therapeutics

6. Ulrike M. Reiss; Getting closer to hemophilia gene therapy for all? Blood Adv 2025; 9 (14): 3629–3630.