On September 4th, 2026, the FDA sent Biohaven a letter that stopped new patients from joining the opakalim epilepsy trials. When Biohaven disclosed it in a regulatory filing days later, headlines used words like "hit," "hold," and "pause." If you or your child is one of the 600+ people already enrolled in those studies, or in any trial that suddenly makes the news this way, the headlines answer none of the questions that actually matter to you. Is my drug dangerous? Do I have to stop? Is the trial over?
The honest answers, in order: probably not, no, and almost certainly not. The longer answers are worth understanding, because clinical holds are a normal part of how drug development works, they follow rules that are public, and the data on how they resolve is far more reassuring than the coverage around them. This guide walks through all of it, using the opakalim hold and 3 famous predecessors as the map.
What is an FDA clinical hold?
A clinical hold is the FDA's formal mechanism for pausing research while a question gets answered. It is defined in federal regulation, and the definition is worth reading in the original because it is shorter than most headlines about it:
“Clinical hold is an order issued by FDA to the sponsor to delay a proposed clinical investigation or to suspend an ongoing investigation.”
U.S. Code of Federal Regulations, 21 CFR 312.42
The key word is question. The FDA issues a hold when it believes participants may be exposed to an unreasonable risk, or when it simply lacks the information needed to rule that risk out. Those are different situations, and the second one is far more common. A hold is not a verdict that a drug is harmful. It is the agency saying: prove to us this specific concern is not a problem before we let more people be exposed to it.
Holds happen at every stage. Some INDs (the application to start human testing) are held before the first patient ever enrolls. Others, like opakalim's, arrive mid-stream, after hundreds of patients have been dosed for years.
Full vs. partial clinical hold, the difference that matters most
This distinction does more work than any other fact in this article. A full clinical hold stops everything: no new enrollment, and everyone currently on the study drug stops taking it. A partial clinical hold stops only what the FDA names in its letter, most often new enrollment or a specific study arm, while everything else continues.
The opakalim hold is partial. New enrollment is paused, but every one of the 600+ participants already randomized keeps receiving the drug, and Biohaven says the fully enrolled pivotal study remains on track to read out data in the second half of 2026 (Biohaven regulatory filing, reported by BioPharma Dive on September 10th, 2026). Read that again if you skimmed it: the FDA reviewed a new safety signal and decided that people already taking opakalim should keep taking it. When the agency believes current participants are in danger, it stops dosing. It did not.
Why the FDA paused enrollment in Biohaven's opakalim trials
The trigger was not something that happened to a patient. Rodent studies of opakalim identified a metabolite (a substance the body produces while breaking the drug down), and the FDA said there was not enough information to judge its risk to people (BioPharma Dive, September 10th, 2026). New enrollment stays paused until Biohaven delivers more animal data on that metabolite.
Animal findings are not the usual trigger. In a 2023 analysis of 33 cell and gene therapy holds announced from 2020 to 2022, 23 followed problems seen in patients and 3 asked for more preclinical (lab and animal) data. Species really do differ: a metabolite that accumulates in a rat can be efficiently cleared in humans, and the entire point of the additional studies the FDA requested is to establish which case this is. We covered opakalim's science and its race with a competing Kv7 drug in our August deep dive, which now carries an update about this hold. Biohaven's company page tracks all of its rare disease programs.
How often do clinical holds get lifted?
Better odds than the headlines suggest. Researchers analyzed clinical holds affecting cell and gene therapy programs and published the results in the journal Molecular Therapy Methods & Clinical Development in 2023. Their findings are the best public numbers we have:
Two details from that research matter for anyone watching the opakalim situation. A study of 33 cell and gene therapy holds announced from 2020 to 2022 found that holds over adverse events took an average of about 6.5 months to lift, and holds over manufacturing data took about 8.4 months (Wills et al., 2023). None of this guarantees any individual outcome, and about 15% were still in place when the study ended. Programs do die on hold, usually when the answer to the FDA's question turns out to be the bad one, or when the company runs out of money to find out.
Clinical holds that made headlines, and what happened next
The recent history of rare disease research is full of holds that terrified patient communities and then resolved. It also includes harder stories. All 3 of these are worth knowing, because together they show the full range of what a hold can become.
- bluebird bio's sickle cell gene therapy, 2021
- In February 2021, trials stopped after a participant developed what looked like leukemia possibly caused by the therapy. Investigation showed the gene-delivery vector was very unlikely to be responsible, and a second suspected cancer case turned out to be a misdiagnosis. The FDA lifted the hold that June, 4 months later. That same therapy is now Lyfgenia, an approved treatment we compared to Casgevy here.
- Novartis's intrathecal Zolgensma, 2019 to 2021
- The closest parallel to opakalim's situation. The FDA partially held trials of spinal-injection Zolgensma over nerve-cell damage seen in animal studies. Novartis answered with a comprehensive primate study addressing every issue the agency raised, and the hold lifted in 2021. The program continued into the Phase 3 STEER trial, and the FDA approved it as Itvisma on November 24, 2025 for people with SMA aged 2 and older.
- Sarepta's Elevidys, 2025
- The hard case. After 2 patient deaths from acute liver failure, and a third death in a related trial, the FDA asked Sarepta to halt Elevidys shipments entirely in July 2025. About 10 days later, on July 28, the FDA recommended that shipments resume for ambulatory patients, whose risk profile differed, while the pause for non-ambulatory patients remained. In November 2025 the FDA added a boxed warning for acute liver injury and acute liver failure and limited the label to patients who can walk. Former FDA biologics chief Peter Marks, who had championed the therapy's approval, said the pause "makes sense" (Endpoints News, July 2025). Safety systems sometimes deliver hard answers, and this was one.
Three holds, 3 outcomes: fully cleared and now an approved therapy, cleared after 2 years of additional animal work, and partially sustained because the risk was real. That range is the honest picture. The system is not designed to reassure anyone. It is designed to find out.
What to ask your study team when your trial is paused
If your trial goes on hold, your study coordinator is required to be reachable, and the informed consent you signed entitles you to material new information about the study. Most sites are good about proactive calls. If yours is not, call them. These are the questions worth asking, roughly in order:
Is this a full or partial hold, and what exactly did the FDA pause? Do I keep taking the study drug, and if not, is there a tapering plan? What was the specific finding, in plain language? Was it in patients or in animals? What is the sponsor doing to answer the FDA's question, and what is the expected timeline? Will my scheduled visits and monitoring continue? If the hold is not lifted, what happens to my access to the drug, and what are my treatment alternatives? Who calls me when something changes?
The communication gap nobody regulates
Here is the uncomfortable part. FDA regulations govern what a sponsor must tell the agency and what investigators must tell ethics boards. No regulation sets a deadline for telling the patients in a trial, in plain language, that their program is on hold. In practice, many participants learn about a hold the way investors do: from a press release written for Wall Street, or from a news alert on their phone. Patient advocates have pushed sponsors toward direct, same-day participant communication, and some companies now do it well. Sarepta, to its credit, published community letters alongside its investor updates during the Elevidys pause. Until that becomes standard, assume the burden of asking falls on you, and know that your study team owes you answers.
One more thing worth saying plainly. A clinical hold, for all the fear it generates, is evidence of a system doing its job while the stakes are highest. The alternative to occasional pauses is not faster cures. It is finding problems after approval, in thousands of patients instead of hundreds of volunteers. The families who enroll in trials deserve both the protection and the straight story about what it means.
What is a clinical hold in a clinical trial?
A clinical hold is an FDA order that delays or suspends a clinical trial, defined in federal regulation 21 CFR 312.42. The FDA issues one when it believes participants may face unreasonable risk or when it needs more information to rule that risk out. A hold pauses some or all trial activity until the sponsor answers the FDA's specific questions.
What is the difference between a full and partial clinical hold?
A full clinical hold stops all trial activity, including dosing of current participants. A partial clinical hold stops only what the FDA specifies, most often new enrollment or a particular study arm, while existing participants continue treatment. Biohaven's September 2026 opakalim hold is partial: enrollment paused while 600+ enrolled patients keep receiving the drug.
How long do FDA clinical holds last?
Published research on cell and gene therapy programs found that 80% of analyzed clinical holds were lifted, with a median resolution time of about 5 months. In a study of 33 cell and gene therapy holds, the average hold that was lifted lasted 6.2 months. Holds over adverse events averaged about 6.5 months and manufacturing holds about 8.4 months. Holds ranged from 2 to 19 months, and about 15% were still in place at the end of 2022.
Does a clinical hold mean the drug is dangerous?
Not by itself. A hold means the FDA has a specific unanswered question about risk, which is different from a finding of harm. Some holds are triggered by animal findings that may not apply to people, and about 80% of the holds in one published analysis were lifted. Some holds do reflect real risks, as with Sarepta's Elevidys in 2025, which is exactly why the FDA asks first and lets trials resume after.
Can I keep taking the study drug during a clinical hold?
It depends on the type of hold. Under a partial hold limited to new enrollment, current participants typically continue dosing as normal. Under a full hold, dosing stops for everyone, and drugs that should not be stopped abruptly get a tapering plan. Your study coordinator can tell you which situation applies within a day of the hold becoming public, and you should ask.