The site initiation visit is done. The investigational product is delivered, the neurosurgical team is trained, and the first patient is dosed. Then someone in clinical operations opens the enrollment tracker and does the arithmetic: 200 patients, 2:1 randomization, 36-month endpoint, rare disease population. Sites I work with in rare neurological indications would call that timeline a planning assumption. The FDA would call it a confirmatory commitment. The gap between those two descriptions is where trials collapse.
uniQure’s ifezuntirgene inilparvovec, now carrying a May 1, 2027 PDUFA action date under Accelerated Approval, is exactly that case study. The FDA accepted the BLA after an earlier reversal in which the agency initially signaled the Phase I/II data was insufficient, then changed course in November 2025. What emerged from that regulatory negotiation is a confirmatory trial design that checks every box the FDA wants to see, and creates a set of operational problems that the protocol’s risk register almost certainly underweights.
What the Design Actually Requires from Sites
The Phase I/II program enrolled 26 patients in the U.S. study and 13 in the European open-label extension, with participants randomized to AMT-130 at low dose, high dose, or a sham procedure. That patient count was workable for a small-site, specialist-center model. The confirmatory trial target of 200 patients is a different category of problem. In Huntington’s disease, the eligible population is already narrow: early-manifest HD, confirmed CAG repeat expansion, functional status tight enough to generate a measurable TFC signal over 36 months but not so advanced that the patient cannot tolerate neurosurgical delivery.
The Total Functional Capacity scale, validated across data from more than 15,500 HD patients in the Enroll-HD natural history dataset, measures decline across five domains: occupation, finances, domestic chores, activities of daily living, and care level. TFC is the right endpoint for a disease with this trajectory. But a 36-month TFC endpoint means every enrolled patient must complete structured assessments at defined intervals across three years, and in a progressive neurological disease, the patient who consented at baseline may have a very different functional profile by month 24. Dropout in this population carries a clinical meaning that dropout in an oncology trial does not: some patients will deteriorate to the point where participation becomes impossible, and that is not a protocol deviation. It is the disease doing exactly what the endpoint is designed to measure.
That creates a retention paradox that site operations teams need to plan for explicitly. The 2:1 active-to-SOC randomization ratio, roughly 133 treated patients, 67 on standard of care, concentrates the high-burden visits on the larger arm. Neurosurgically delivered gene therapy requires post-procedural monitoring, MRI follow-up, neurological assessments, and safety reporting that standard-of-care comparators do not. A site carrying 10 treated patients is running a very different coordinator workload than a site carrying 10 SOC patients, and both counts are small enough that losing two patients to withdrawal or disease progression meaningfully disrupts the site’s data contribution.
The Regulatory Approval Gene Therapy Sites Are Not Designed Around
The FDA has acknowledged, in draft guidance on cell and gene therapy development, that traditional randomized controlled trials may not always be feasible for rare disease CGT products and has outlined alternatives including natural history comparators and single-arm designs. uniQure’s confirmatory design chose the more rigorous randomized path, which is why the BLA moved forward. But rigorous for the FDA and feasible for sites are not the same calculation.
Standard review timelines run 10 months from BLA acceptance. The May 1 PDUFA date tracks that clock. What does not appear on that clock is the 36-month confirmatory data collection window that runs in parallel with commercial launch if approval comes through. This is the structural feature of Accelerated Approval that sites consistently underestimate: approval and confirmatory obligation arrive simultaneously, which means the sponsor is running a commercial program and a clinical operations program against the same patient population, often at the same sites, with documentation requirements that belong to both.
Any coordinator who has worked a post-approval confirmatory study knows the pressure this creates. The site’s commercial liaison is talking to the neurology department about treatment access. The CRA is in the monitoring room reviewing source documentation for the confirmatory protocol. The PI is managing a patient who was randomized to SOC and is now asking about commercial access to the gene therapy that just got approved. The ICF that patient signed 18 months ago needs an amendment. The IRB turnaround on that amendment adds days the enrollment timeline does not have. None of this appears in the phase-specific risk register that was filed with the protocol.
The global CGT CRO market, estimated at $1.9 billion in 2025, reflects how much specialized infrastructure rare-disease gene therapy trials require. But market size does not translate to site-level capacity. The neurosurgical centers capable of delivering intrathecal or stereotactic gene therapy are a short list, and those same centers are fielding competing protocol requests from every other sponsor in the HD and broader neurological gene therapy space. Site activation timelines at surgical centers routinely run longer than 90 days from contract execution to first-patient-in, because the credentialing, pharmacy setup, and procedure training requirements have no analogue in a standard Phase III tablet study.
What Operators Need to Lock Down Before Enrollment Opens
For sponsors building the operational plan behind a confirmatory gene therapy design, the 36-month endpoint demands a retention infrastructure that gets funded in the original site budget, not negotiated as an amendment two years in. That means per-patient budgets that account for the full visit schedule across three years, including the administrative burden of annual ICF reconsent in jurisdictions that require it and the coordinator time for SAE narrative writing when a serious adverse event occurs in a patient who is 28 months post-dose. SAE reporting clocks run from awareness, not from dose date, and a neurosurgical gene therapy program will generate events that require careful causality assessment across a long follow-up window.
Across networks running rare neurological trials, the retention strategies that actually work at the three-year mark share one feature: a dedicated patient liaison role that is funded separately from the CRC’s visit-based workload. When a coordinator’s time is fully consumed by scheduled assessments and data entry, the relationship work that keeps a Huntington’s patient and their caregiver engaged across three years of a progressive disease falls off the schedule. It falls off because nobody put a line item on it.
For sites accepting randomized patients into the SOC arm, the operational question is equally concrete: what is the protocol amendment plan when commercial approval arrives and the SOC patient wants to cross over? The Phase I/II U.S. program did allow four control patients to cross over to treatment. If the confirmatory design includes a pre-specified crossover provision, sites need an amendment-ready IRB submission package that can move within 30 days of approval. If it does not include one, sites need a conversation script for the PI that does not constitute protocol deviation counsel.
The May 1 PDUFA date is the visible milestone. The 36-month data lock that follows approval is the operational commitment that determines whether the confirmatory trial actually delivers what the FDA accepted the BLA to require. Sites that plan for that endpoint now, with funded retention infrastructure and amendment-ready documentation, will be the ones still holding complete data files when the confirmatory submission deadline arrives.
References
- Genetic Engineering News, “FDA Sets May 1 PDUFA Action Date for uniQure’s Huntington’s Disease Gene Therapy”
- Investing.com, “FDA Accepts uniQure’s Huntington’s Disease Gene Therapy Filing”
- HDSA, “uniQure Announced They Are Resuming the Patient Enrollment at the Higher Dose Level in AMT-130 Trial”
- PMC / Enroll-HD, TFC Validation Study in Huntington’s Disease, 15,527 patients
- Pharmaceutical Technology, “uniQure to Seek Approval of Huntington’s Gene Therapy After FDA U-Turn”
- Wilson Sonsini, “FDA Releases Three Draft Guidances to Facilitate Development of Cell and Gene Therapies”
- FDA CBER, “Biologics License Applications (BLAs): CBER-Regulated Products”
