An FDA Reversal, A Sham Surgery, And 41,000 Americans Running Out Of Time

For half a year, uniQure had a written agreement with the FDA on what its Huntington's disease gene therapy needed to show. Then the agreement disappeared. Then a new one emerged that asks dying patients to be randomized to nothing.

For 41,000 Americans living with Huntington's disease, there has never been a single approved treatment that slows the progression of their illness. Not one. By June 2025, after seven years of clinical work, four FDA expedited designations, and a December 2024 written agreement that the agency reaffirmed in June 2025, uniQure had FDA alignment to submit its gene therapy AMT-130 for accelerated approval based on its existing trial data. In November 2025, with the company's Phase 1/2 trial showing a 75% slowing of disease progression at three years, the FDA reversed itself. By March 2026, the agency was demanding a years-long Phase 3 sham-surgery-controlled trial. The UK regulator, looking at the same data, is moving toward approval. The cost of the disagreement is measured in years of life patients do not have.

What Huntington's Disease Actually Does

Huntington's disease is autosomal dominant. If a parent carries the mutation, each child has a 50% chance of inheriting it. The mutation is on a single gene called HTT, and it produces a misshapen protein called mutant huntingtin that gradually destroys neurons in the basal ganglia and cortex. Symptoms typically begin between the ages of 30 and 50. The mean disease duration after onset is 17 to 20 years. The mean age at death is 51 to 57.

Patients lose motor control. They develop chorea - involuntary, dance-like movements that progress until the body cannot be quieted. They lose cognitive function. They lose the ability to swallow. They lose the ability to speak. Roughly 7% of patients die by suicide, the second-leading cause of death after pneumonia. Families watch the same disease take parents, then siblings, then children. There is no approved disease-modifying treatment. There is no cure. More than 150 years after George Huntington first described the disease in 1872, and more than three decades after the huntingtin gene was identified in 1993, there has never been an approved disease-modifying treatment. The available treatments do not slow the disease. They manage symptoms while the disease finishes its work.

What AMT-130 Is

uniQure is a Dutch-American gene therapy company. AMT-130 is its lead program for Huntington's disease. The therapy uses an adeno-associated virus to deliver a microRNA into the brain, where it lowers production of the mutant huntingtin protein at the source. The treatment is administered through MRI-guided neurosurgery in a procedure that lasts eight to ten hours, during which the catheter delivers the therapy directly into the striatum - specifically the putamen and caudate nucleus, the regions most severely affected in early Huntington's - through small holes drilled in the skull. It is given once. The intent is that lowering the mutant protein for the rest of the patient's life will slow or halt the neurodegenerative cascade.

The Phase 1/2 trial began in 2019. As of April 2025, 45 patients had been treated. The September 2025 three-year data, drawn from the high-dose cohort of 12 patients evaluated at 36 months, showed a statistically significant 75% slowing of disease progression on the composite Unified Huntington's Disease Rating Scale (p=0.003) compared to a propensity-score-matched external control drawn from the Enroll-HD natural history database, an observational study comprising more than 20,000 Huntington's disease patients. The composite was not the only endpoint that moved. Treated patients also showed a 60% slowing of decline on the Total Functional Capacity scale, a 59% slowing of decline on the Total Motor Score, and a 113% relative improvement on the Symbol Digit Modalities Test, a measure of cognitive processing speed. Cerebrospinal fluid neurofilament light - a biomarker of active neurodegeneration - declined 8.2% from baseline at the three-year mark, after an initial transient spike attributable to the surgical procedure. The medical community responded with what one peer-reviewed editorial in The Lancet characterized as renewed hope after decades of failed trials. uniQure's share price more than doubled. The Huntington's research community treated the result as the most significant clinical signal in the history of the disease.

The Designations The FDA Granted, In Order

Before the reversal, the FDA itself agreed - repeatedly, formally, and in writing - that AMT-130 was on an expedited path. Each of these designations is a substantive regulatory determination, not a routine acknowledgment.

In November 2024, after a Type B meeting, the FDA agreed in writing that data from uniQure's ongoing Phase 1/2 trials, compared to a natural history external control, could serve as the primary basis for a Biologics License Application under the Accelerated Approval pathway. The cUHDRS scale was accepted as an intermediate clinical endpoint. The cerebrospinal fluid neurofilament data was accepted as supportive evidence of clinical benefit. uniQure's chief medical officer called it "an important milestone for the Huntington's disease community." The company began BLA readiness activities. In June 2025, the agreement was formalized in further written guidance. Investors, patients, and clinicians proceeded on the reasonable assumption that the FDA had said what the FDA meant.

The Reversal, In Sequence

In late February 2026, FDA Commissioner Marty Makary appeared on CNBC and, while not naming uniQure, described "a product where the researchers drilled a burr hole, literally a hole in people's skulls, to inject intrathecally into the ventricle a therapy." He said the trial "found no benefit" and characterized it as one of the drugs the FDA had been "pressured to approve." uniQure's share price fell more than 30% in a single day. Several analysts noted the technical details Makary used did not all match AMT-130 - the Huntington's therapy is not delivered intrathecally into the ventricle - but the market interpreted the description as a reference to AMT-130 regardless. Investor confidence in the program has not recovered.

A week later, on March 2, uniQure released minutes of its January 30 Type A meeting with the agency, confirming what the market had feared: the FDA had formally rejected the previous accelerated approval pathway and was now requiring a prospective, randomized, double-blind, sham-surgery-controlled Phase 3 trial. The shares fell another 35-40% on the news, briefly trading near $10. The company had aligned with the FDA. The FDA had un-aligned with itself. Years had been added to the development timeline. No documented rationale for the reversal was provided.

The Sham Surgery Question

A sham-surgery-controlled trial randomizes patients with a fatal disease into two groups. One group receives the active therapy: an MRI-guided eight-to-ten-hour neurosurgical procedure under general anesthesia, with a catheter inserted into the brain through small holes drilled in the skull, delivering AMT-130. The other group receives the sham: general anesthesia, with one to three nicks in the scalp, designed to make the patient unable to tell which group they were in. According to the FDA, the sham procedure would last roughly thirty minutes. According to uniQure and the Huntington's patient community, the design asks dying patients to undergo a years-long study in which half of them get nothing but anesthesia and a scar.

On a March 2 conference call, uniQure leaders stated plainly that a sham-controlled study "could impose significant risks and burden to patients - some might even consider this trial design to be unethical." Senator Ron Johnson of Wisconsin, who chairs the Senate Permanent Subcommittee on Investigations, called the FDA's request "bureaucratic idiocy." Johnson asked the obvious question: "You're expecting people to go through sham surgeries where they get holes drilled in their heads?" Former senior FDA official Janet Woodcock, in an interview with The New York Times, called the agency's decision on AMT-130 "truly evil." That phrase came from someone who spent more than three decades at the FDA, including as the agency's acting commissioner.

The FDA's defense, articulated by an unnamed senior official widely reported to be then-CBER director Vinay Prasad, was that the sham procedure would be brief, would only require nicks rather than burr holes, and was scientifically necessary because the previous data was not sufficient. The official said the agency only demands randomized data when, in his words, "the will to believe is strong" and "the possibility you are fooling yourself is high." He pointed to the 2021 Phase 3 failure of tominersen - a different therapy from Roche and Ionis Pharmaceuticals that, like AMT-130, was designed to lower huntingtin protein - as evidence that randomized controlled trials can produce different results than observational comparisons. The tominersen reference is the FDA's strongest argument: a prior Huntington's therapy, with a related target, did fail in a placebo-controlled study after early signals had appeared promising. The official also told reporters the agency requires placebo-controlled trials to ensure a product is not, in his word, "snake oil." That word choice, applied to a one-time gene therapy that has shown a statistically significant 75% slowing of progression on the composite scale, plus consistent slowing across motor, cognitive, and biomarker endpoints, was widely characterized in the rare disease community as inflammatory. uniQure and outside scientists have noted significant mechanistic differences between AMT-130 and tominersen: tominersen was an antisense oligonucleotide delivered intrathecally on a recurring basis to lower huntingtin throughout the central nervous system; AMT-130 is a one-time AAV-delivered RNA interference construct that targets the striatum directly and is intended to lower huntingtin locally for the rest of the patient's life. Whether those differences are sufficient to justify a different evidentiary standard is the empirical question at the heart of the dispute.

On the central question of what the FDA had actually agreed to, the senior official issued a flat denial. Asked directly whether the agency had ever agreed that AMT-130's Phase 1/2 data with an external control could serve as the primary basis for accelerated approval, the official said: "No, I've asked our people to look for any written or verbally transcribed or minutes of a meeting that document such a promise, and there is no such promise made." The official went further, characterizing uniQure's clinical comparison as "a distorted or manipulated comparison in the mind of FDA," and saying that as a general rule the agency "will always say, well, we have to see the data when we get it." The position the FDA was articulating, in other words, was not just that the data was insufficient. It was that there had never been an agreement of the kind uniQure was describing in the first place.

uniQure has rejected this characterization in unusually direct language. The company has called the FDA's description of the regulatory history "highly irregular, unprecedented, and incomplete or entirely incorrect," and has pointed to its November 2024 alignment letter and June 2025 follow-up correspondence as documenting the framework the agency now denies ever existed. The company's position is supported by its own SEC filings, which disclosed the alignment as a material fact to shareholders at the time, and by independent analysts who covered the original announcements as definitive regulatory milestones, not as preliminary discussions. Several of those analysts have publicly noted that the language uniQure used in late 2024 and mid-2025 - "FDA has agreed," "FDA has aligned," "FDA has accepted" - is inconsistent with the picture the agency's spokesperson is now painting. The dispute is, at its core, a factual one about what was said in meetings between the FDA and the company over a 14-month period, and the contemporaneous documentation that uniQure has cited in support of its account. uniQure called the agency's description of its position "highly irregular, unprecedented, and incomplete or entirely incorrect." What the FDA has not addressed publicly is that uniQure used a sham control in approximately 10 patients earlier in the AMT-130 program and then stopped using it, precisely because of ethical concerns about asking patients with a fatal disease to undergo a procedure with no therapeutic benefit. According to a clinical analysis published by chemistry blogger and longtime industry observer Derek Lowe at Science.org, roughly 10% of those original sham-arm patients developed a blood clot during the procedure, which involved 10 to 12 hours of general anesthesia. After unblinding, the patients originally randomized to the sham arm were allowed to cross over and receive the actual therapy. The company's March 2 statement that a sham-controlled trial is potentially unethical is not a new objection. It is the same objection uniQure already acted on, years earlier, with the agency's knowledge, after a documented serious adverse-event signal in the prior sham cohort. HHS spokesperson Andrew Nixon used social media to dispute the company's characterization. To one user objecting that sham surgery for Huntington's disease "isn't logical," Nixon replied: "Why don't you read up on your history? Or maybe you're too old and forget." A week later, on March 12, 2026, Nixon posted again from his verified @AndrewNixonHHS account, replying to FDA Commissioner Marty Makary on a thread about the AMT-130 controversy. His tweet, in full: "The swamp continues their single-product lobbying campaigns." The 37,800 people who saw the post had no way to know whether the "swamp" in question was uniQure's shareholders, the Huntington's Disease Society of America, the patient families who had delivered a petition asking the agency to honor its prior guidance, the bipartisan senators investigating the agency, or the rare disease community broadly. The U.S. Department of Health and Human Services, on the official account of its press secretary, characterized advocates for the first disease-modifying treatment ever shown to slow Huntington's disease as members of the swamp.

The Vinay Prasad Problem

Vinay Prasad, M.D., is a hematologist-oncologist who built a public profile criticizing what he viewed as low-quality evidence supporting drug approvals. In May 2025, he succeeded Dr. Peter Marks as director of the FDA's Center for Biologics Evaluation and Research, the office that oversees vaccines, blood products, and gene therapies. Marks had been a leading voice for the use of accelerated approval and natural history controls in rare disease drug development; Prasad arrived with a documented record of skepticism toward both. By July 2025, he had been forced out following a high-profile dispute over Sarepta Therapeutics' Duchenne muscular dystrophy gene therapy Elevidys. He was reinstated within a month. By March 2026, he was leaving again. According to H.C. Wainwright analysts, "CBER rejected at least five cell and gene therapies that we believe could have received approval under prior FDA officials" during his tenure.

Throughout that period, Prasad expressed reservations about external control comparisons - the methodology the FDA had previously agreed to accept for AMT-130. Asked by Fierce Biotech how the agency's rejection of AMT-130 squared with the FDA's own newly unveiled "plausible mechanism pathway" - which explicitly endorsed natural history controls for individualized rare-disease therapies - Prasad pointed to the fact that AMT-130 is not a bespoke therapy tailored for each patient. Critics noted that this distinction was not articulated in the plausible mechanism pathway guidance itself. uniQure's therapy meets every criterion the FDA listed in its draft guidance: it targets a defined genetic cause, has shown clinical benefit, and uses well-characterized natural history as a comparator. The agency's own framework, on its face, would have allowed the approval. The agency declined to apply it.

The UK Path

The UK Medicines and Healthcare products Regulatory Agency reviewed the same data. According to uniQure, the MHRA reached out to the company proactively, and on May 1, 2026, uniQure announced it would submit AMT-130 for marketing authorization in the United Kingdom in the third quarter of this year. The submission will be based on the same three-year analysis from the Phase 1/2 program that the FDA has called insufficient. UK regulators have raised no objection to natural history controls in this context. They have not requested a sham-surgery Phase 3 trial. They are processing the submission. Leerink Partners called the UK development "an important symbolic win."

The practical implication is significant. If MHRA approves AMT-130 in the UK while the FDA is still demanding a years-long Phase 3 trial in the United States, American Huntington's patients will face the choice between traveling to the UK for treatment - paying out of pocket, with no insurance coverage in most cases, and likely no aftercare reimbursement at home - or waiting for the FDA. For patients who are already symptomatic, the wait may exceed their disease horizon. For pre-symptomatic carriers who have tested positive for the HTT mutation, every year of regulatory delay is a year of irreversible neuronal loss occurring before any treatment can be administered.

Two Approaches, Same Data

Who Is Pushing Back

Senator Ron Johnson, Republican of Wisconsin and chair of the Senate Permanent Subcommittee on Investigations, has launched a formal investigation into the FDA's recent rare disease rejections, including AMT-130. Johnson is requesting copies of the agency's complete response letters and is considering calling Commissioner Makary to testify before the subcommittee. Johnson is a long-standing advocate of expanded access to experimental therapies; he sponsored the Right to Try Act in 2018, and last year personally intervened to help restore patient access to Sarepta's Duchenne therapy after the FDA tried to pull it. "The stories are so outrageous," Johnson told Bloomberg News. "It just appears that they're looking for excuses to say no."

A February 26, 2026 hearing of the U.S. Senate Special Committee on Aging on rare disease drug development drew testimony from physicians, advocates, and patients' families about "inconsistently applied approval pathways, disregarded surrogate endpoints, and late-stage reversals of negotiated trial designs." A March 2026 Senate hearing chaired by Senator Rick Scott of Florida heard from a Harvard neurologist who described his decade of correspondence with the FDA over the spinocerebellar ataxia drug troriluzole as "like talking to a brick wall." The Huntington's Disease Society of America delivered a petition to the FDA asking it to "honor its prior guidance and expedite review" of AMT-130.

Janet Woodcock, who served as Acting FDA Commissioner from January 2021 to February 2022 and spent more than three decades at the agency, made a separate public statement to CNN about the agency's handling of the case. She told the network that the FDA should at least review uniQure's therapy and hold an advisory committee meeting to allow public discussion. "For such a terrible disease and people who volunteered to have this therapy put into their brains," Woodcock said, "they owe it to the community to review and have a public discussion. They promised radical transparency." The FDA's response, delivered by the same anonymous senior official on the same March 5 background call, was to publicly chastise her: "Dr. Woodcock is an esteemed regulator," the official said. "There are many commenters who I don't expect better from, but Dr. Woodcock I do expect better from." A sitting FDA official, on a press call, taking the agency's former Acting Commissioner publicly to task by name. There is no precedent for this in modern FDA history.

The Alliance for Regenerative Medicine, the trade group representing the gene therapy industry, has stated bluntly that uniQure's situation is "another clear sign that the Agency has increased the regulatory requirements for rare disease gene therapies, despite saying for months that it was committed to regulatory flexibility." Bioethics professor Holly Fernandez Lynch of the University of Pennsylvania, asked about the FDA's practice of attacking uniQure on background calls with reporters, said: "The fact that you are sharing information with the press without standing behind it either suggests that you know you should not be talking to the press or that you don't really stand behind your opinion. Why are we debating this in the court of public opinion? The stakes are too high."

The procedural failure Woodcock identified is the most concrete charge against the agency. In cases where data are complicated or where the FDA itself is unlikely to have deep internal expertise, the agency historically convenes an advisory committee, or adcomm, of outside experts to provide independent input. Patients, doctors, and other interested parties get to comment on the record. The science gets debated in public. The FDA has not done this for AMT-130. Several other companies have publicly requested adcomm meetings, and at least three have said the FDA initially told them to prepare for a meeting and then retracted that guidance. On the same March 5 background call where the senior FDA official attacked uniQure and singled out Woodcock by name, the official also dismissed the broader pool of outside scientists who serve on FDA advisory panels, saying they "do not do their homework" and may be motivated by financial conflicts of interest. The agency's position is that an advisory committee is unnecessary. The agency's former Acting Commissioner, multiple senators, the rare disease community, and at least three other companies that have requested adcomms believe an advisory committee is precisely what is required.

The Inconsistency That Will Not Go Away

The FDA, under the same leadership now blocking AMT-130, has unveiled what it calls the "plausible mechanism pathway" to accelerate the development of individualized gene therapies for rare diseases. The pathway explicitly endorses natural history controls, exactly the methodology the FDA agreed to use for AMT-130 and then disavowed. The agency has granted four expedited designations to AMT-130 specifically: Fast Track in 2019, Orphan Drug, RMAT in 2024, and Breakthrough Therapy in 2025. Each designation is supposed to mean the FDA recognizes both the unmet need and the substantial preliminary evidence of effectiveness. Each was granted on the same data the agency now describes as unable to demonstrate "any therapeutic benefit."

The FDA has, separately, acted with extraordinary regulatory flexibility on individual cases when politically convenient. In 2024, after Sarepta's Elevidys was linked to two patient deaths and the agency requested a market withdrawal, Senator Johnson personally lobbied Commissioner Makary and texted President Trump; the agency reversed itself and allowed the drug to remain available for ambulatory patients. In early 2026, the agency initially rejected Moderna's mRNA flu vaccine and then, within a week, reversed itself to allow review pending an additional study. The agency that can move that quickly when it chooses to has not chosen to for Huntington's patients. There is no Sarepta-style reversal in sight. The Right to Try framework that Johnson championed does not apply to therapies the FDA has not approved through any pathway. Patients have no Plan B in the United States.

What This Costs, In Years

Mean disease duration in Huntington's is 17 to 20 years from symptom onset. The mean age at death is 51 to 57. Every year added to the AMT-130 development timeline is a year in which a substantial fraction of currently-symptomatic patients will lose function they cannot recover and a smaller fraction will die. A sham-surgery Phase 3 trial of the size and duration the FDA has indicated it requires would, on standard assumptions, take three to five years from initiation to BLA filing, plus additional review time. For a patient who is currently 50 years old and four years into symptoms, that is the entire remainder of their statistical life expectancy.

There is a separate population of pre-symptomatic carriers - the roughly 200,000 Americans who have tested positive for the HTT mutation but have not yet developed clinical disease. The biological premise of AMT-130 is that lowering mutant huntingtin slows the cascade. The earlier the intervention, the more neurons survive, the longer the patient retains function. For pre-symptomatic carriers, every year of regulatory delay is a year of cumulative neuronal loss occurring with a known mechanism and no available intervention. The Phase 1/2 trial three-year results are the strongest disease-modifying signal ever generated in Huntington's. The FDA's position is that this signal is insufficient evidence to allow access while a confirmatory trial runs. Patients are being asked to wait. Many of them do not have the time.

The Question The FDA Will Not Answer

In the year since the FDA reversed its written agreement with uniQure, the agency has not provided a documented rationale for the change. It has briefed reporters anonymously. Its commissioner has criticized the therapy on cable television without naming the company. Its CBER director, before resigning, defended the demand for sham surgery on a background call. Its public affairs office has used social media to dispute patient advocates and to suggest, in response to a critical reply, that the user was "too old and forget" the relevant history. None of this is a rationale. None of it is responsive to the simple question Senator Johnson and the Huntington's community have been asking for six months: what changed?

If the answer is that the data was always insufficient, then the FDA's November 2024 alignment letter, the June 2025 follow-up correspondence, and the four expedited designations granted under both the previous and current administrations were all errors. If the answer is that the data is now sufficient by some standards but not by the agency's newly applied standards, then the FDA needs to explain what those new standards are, why they were not articulated when the alignment letter was issued, and how they apply to AMT-130 differently than to therapies that have received accelerated approval on similar evidence. If the answer is that the agreements uniQure has cited never existed in the form the company described - the FDA's current position - then the agency needs to explain how four expedited designations were granted, how a publicly disclosed alignment letter was issued, and how an SEC-filed material disclosure was made about a regulatory pathway that the agency now says was never agreed to. If the answer is that an individual official changed his mind, the agency needs to say so. There has been no such acknowledgment.

The Bottom Line

For 41,000 Americans living with Huntington's disease and the 200,000 more who carry the genetic mutation, AMT-130 is the first therapy in more than 150 years that has shown durable disease-modifying activity in a controlled trial. According to uniQure, supported by its public regulatory disclosures and SEC filings at the time, the FDA agreed in writing in November 2024 and reaffirmed in June 2025 that the existing trial data could support accelerated approval. According to the FDA - speaking through an unnamed senior official on a March 2026 background call - no such agreement was ever made. What is undisputed is that the agency reversed course on the pathway uniQure had been pursuing for over a year, granted four expedited designations along the way, and is now demanding a years-long sham-surgery trial that the company, multiple senators, the rare disease community, and at least one former senior FDA official have characterized as ethically and practically untenable. The UK regulator, looking at the same data, is processing the submission.

The cost of bureaucratic reversal is not measured in shareholder value, although uniQure has lost roughly half a billion dollars in market capitalization across the affected trading sessions. The cost is measured in patient-years - in the people for whom every additional regulatory month is a measurable, irreversible loss of cognitive and motor function, and in the people who will reach the end of their statistical disease horizons before the trial the FDA is now requiring is even completed. Those people exist. They are named. They have families. They have read the same data the FDA read. They have read the same agreement the FDA signed and unsigned. They are asking what changed. They deserve an answer.