AAV Gene Therapy: What It Will Take To Get Through the Wilderness

When Luke first asked for my perspective on the state of AAV gene therapy over the summer, he asked me “Will AAV ever deliver on its early promise? Be safe? Be commercially successful?”.

After ruminating for a while, I settled on the Stockdale paradox (as chronicled in Jim Collins’ Good to Great) as the best description of where we stand.

The Stockdale paradox states:

“You must never confuse faith that you will prevail in the end — which you can never afford to lose — with the discipline to confront the most brutal facts of your current reality, whatever they might be.”

This acceptance of the brutal facts, combined with faith of prevailing in the end, was with me heading into this year’s Alliance for Regenerative Medicine Meeting on the Mesa. Meeting there with colleagues helped me distill what has to happen next.

The key takeaway? It’s execution time. Gene therapy developers need to show we can turn this impressive science into a meaningful business. Plain and simple.

The most brutal (perceived) facts

The early days of gene therapy teased the possibility of a steady stream of transformative treatments. Surely, amazing, life-changing therapies would emerge for patients and financial rewards would inevitably flow to those innovators who could simply develop the right copy/paste science. We would hop from gene to gene, swapping genetic payloads into viral vectors and collectively work our way through the whole genomic atlas of rare disease. Talent and capital coalesced around this enticing opportunity.

The reality has not proven nearly so straightforward. Safety concerns and varying degrees of efficacy have sown seeds of doubt. When some safe and effective gene therapies have struggled commercially, and a revolving door of regulators have occasionally moved the goalposts and sent mixed messages about what’s required, you have what can charitably be called a rough patch for the field of cell and gene therapy. Even as the biotech market rebounded in 2026,  gene therapy appears to have lagged.

Why?

For starters, the gene therapy business model operates by different rules. More on that in a minute.

The biotech industry is sustained by a large financial feedback loop. It rewards and incentivizes innovation that brings value to patients and to society. When the system works, discoveries from academic labs become biotech startups backed by venture capital, grants, angel investors and such. Those investors earn returns when strong data enables IPOs, acquisitions, or licensing deals. Society benefits from medical innovations that should save healthcare costs and improve lives over time. (Peter Kolchinsky makes this point well in his excellent book “The Great American Drug Deal.”)

When expected commercial returns are less clear however, the system stalls: large companies hesitate to license or acquire, public investors step back, and early investors lose hope for returns and the appetite to fund new companies.

Gene therapy poses a particular challenge to the biopharma business model that typically depends on recurring revenue streams from regularly re-dosed pills or injections. The opportunity to realize a commercial return for gene therapy occurs just one time per patient for an adeno-associated viral (AAV) vector delivered gene therapy.

This means that payors are being challenged to adapt to a business model which relies on front-loading payment for outcomes that can range from durable life-changing benefits to more incremental, shorter-term improvements in quality of life. And like other medicines, some patients may respond more strongly than others, creating uncertainty around population-level outcomes from studies with a small sample size.

Insurers are being asked to pay million-dollar per patient price tags upfront, uncertain whether the benefits may vary, and also wary that patients may switch insurers every few years. This possibility can allow a another payor to reap the rewards of a patients treatment, without having to pay for it. Insurers confronted with a patient in need of an expensive gene therapy have a perverse incentive to consider delaying, resisting or hoping another healthcare system will front the cost of the life-changing gene therapy.

That’s not the only business challenge for gene therapy developers. Companies who develop new therapies for rare genetic diseases capture near-term revenues by treating the existing patient population. Once those people get their one-time therapy, longer term revenues are mainly driven by the incident patient population – the number of patients who are born and/or are newly-diagnosed each year. The nature of some rare diseases means that pool can become quite small over time.

For larger companies looking for multi-billion dollar products to replace the franchise products losing patent protection,  a single gene therapy addressing a rare disease is unlikely to produce the investment math needed to become another Keytruda. This can temper interest from some of the larger pharma players, limiting the pool of buyers for programs and companies, cooling the market in turn.

The peak of elevated expectations

Early expectations were buoyed by the success of the first gene therapy approved for Spinal Muscular Atrophy (SMA), Novartis’ onasemnogene abeparvovec-xioi (Zolgensma). With the potential to deliver transformative benefit to young children with a crippling and deadly condition, Zolgensma is on track to exceed $1 billion in revenue for a sixth consecutive year.

Investors who hoped to bet on the next Zolgensma have been left wanting. The most oft-cited cautionary examples are gene therapies for Hemophilia A and B that have fallen far short of their pre-launch expectations.

As humans we reach for simple explanations to make sense of a complex world. “Well SMA was the outlier and hemophilia is the norm”. The complicated truth of the matter is that one needs to dig deeper to unearth the success factors behind a commercial launch.

Hemophilia is complicated by strong injectable biologic competitors that serve the needs of many patients. A one-time treatment also challenges the economic incentives at US infusion centers. Patients are often hesitant to switch.

Compare and contrast this situation with Sarepta Therapeutics’ delandistrogene moxeparvovec-rokl (Elevidys) for Duchenne Muscular Dystrophy (DMD). The one-time gene therapy was introduced to the market in 2023. The initial sales trajectory was strong – demonstrating that a combination of a motivated community living with a significant unmet medical need, urgency to treat and support of payors who saw value in intervention came together in spite of concerns about whether Duchenne families would hold off treatment in favor of next-generation offerings.

In 2025 two deaths of wheelchair-bound DMD patients who developed acute liver failure after treatment temporarily halted dosing of the drug and limited the eligible patient population. The product’s withdrawal, new boxed warning, and re-introduction to the US market made headlines. Fewer people noticed a more encouraging development that happened in parallel. Sarepta’s partner outside the US, Roche, reported over $200 million in ex-US sales for the first half of 2026, despite lacking an approved Marketing Authorisation Application (MAA) in the European Union. That reinforces the significant commercial potential that exists beyond the US.

To frame this differently, the modality does not predetermine commercial success or failure, rather it’s the needs of the community and the assessment of value in treatment that determines commercial success. (See the Alliance for Regenerative Medicine’s excellent 2026 ARM State of the Industry deck for more on this point).

So, if we buy into the thesis that new product rollouts in AAV gene therapy CAN be successful, then what other factors need to be in play to turn “can be successful” into “are successful”?

The risk/benefit balance

Fatalities in clinical trials or in the marketplace should never be overlooked or trivialized. For AAV gene therapies, there has long been an intense focus on safety dating back to 1999 and the Jesse Gelsinger tragedy.

By now, an expanding body of evidence suggests  these fatalities have not been completely random, but rather stem from a constellation of factors that are discreet and may well be addressable.

Emerging consensus centers on factors such as total viral vector dose, immune response/immune suppression regimen, patient or disease specific factors such deletions of exons 8 or 9 in DMD, or pre-existing hepatobiliary disease in X-linked Myotubular Myopathy, and potential product quality attributes.

The gene therapy field has been taking important steps to understand how to avoid exposing patients to risks that may exceed benefits and finding ways to create potentially safer and more effective treatments.

Next-generation gene therapies for DMD employing engineered capsids appear to be effectively de-targeting the liver, while delivering strong transduction in skeletal and smooth muscle where you want the gene expressed. Proficiency with localized delivery and increased tissue selectivity to address regions of the CNS, the heart and other organs has reduced the amount of viral vector needed. Scientists can now accurately express just the right amount of the MECP2 protein on a cell-by-cell basis in patients living with Rett Syndrome without over-expression that drives toxicity.

We’re seeing the emergence of a second wave of AAV approaches achieve clinical proof of concept. While it would be premature to we have solved all safety challenges, it’s also inappropriate to write the field off as unsafe .

On the “benefit” side, we are witnessing a steady stream of clinical data. In addition to those conditions cited above, we’re seeing unprecedented effects in diseases as diverse as Limb Girdle Muscular Dystrophies, Dravet Syndrome, OTOF-related hearing loss, X-linked retinitis pigmentosa, Huntington’s disease, Friedreich Ataxia, and many more.

Are all these programs perfect? No. But that’s not the point. The takeaway is that gene therapy is chipping away at previously intractable conditions.

Regulatory uncertainty – becoming more certain?

Much has been written about the changing times at FDA, in the Center for Biologics Evaluation and Research (CBER) and the Office of Therapeutic Products (OTP), that have directly impacted the cell and gene therapy sector. Contrary to the perception that companies are looking for more permissive regulation; consistency and predictability are typically the most important factors most CEOs seek.

Adaptability to the specific requirements of each indication can be critical here as well. Not all conditions are suitable for double-blind placebo or sham-controlled studies; not all diseases have robust natural history data available for use as a comparator.  If a pathway is agreed to by regulators one year and a regulator reverses course the following year, how will any management team invest the resources necessary to complete a development program in support of a new medicine?

Consider the case of Uniqure. The company received notably conflicting advice on potential pathways to approval during a series of leadership changes at FDA resulting in wild swings in their stock price and market perception. The company  has recently aligned on a path to filing a Biologics License Application to market its gene therapy for Huntington’s Disease in the US. The recent appointment of Karim Mikhail as director of CBER along with  gene therapy approvals in Glycogen Storage Disease 1a, Sanfilippo Syndrome type A (MPSIIIA) bode well for the field.

Regulatory uncertainty is not restricted to the United States – differing evidentiary requirements on either side of the Atlantic can further complicate matters. However, encouraging signs are emerging in 2026 as we are seeing the FDA stand by some of its prior commitments and just this week the agency released a new white paper and initiative to create a new framework for rare disease review.

If the FDA follows through and clears some of the new products for sale based on prior guidance, then it’s critical that that companies seize the opportunity to demonstrate the commercial potential of these new medicines. Eventually, that will persuade investors to come back in large numbers. One last factor has been the recent strength of the marketplace for the Priority Review Voucher program, set up to incentivize investment in R&D for rare pediatric diseases. Per the RTW Institute PRV tracker there have been 9 PRV sales in 2026 alone (including the two recent voucher sales by Ultragenyx) all at around the $200 million per voucher mark. Three of these sales have gone to developers of AAV gene therapies, proving an additional much-needed incentive for the rare disease space.

The prognosis

So, having faced the brutal reality, do I have faith that we will prevail in the AAV field? Yes. I do. It may not be exactly in the manner we’ve become accustomed to in other fields and with other modalities, but the preponderance of breakthroughs seen in patients combined with a clearer view of what commercial factors may be needed for the next successful therapy lead me to conclude that we have a strong chance.