No clinical hold for an IDE. Only disapproval.

A drug sponsor running an IND study can get a call from FDA at any hour of any day in the trial, telling it to stop. That call is a clinical hold under 21 CFR 312.42, and a sponsor moving from a drug program to a device program sometimes goes looking for its counterpart in an IDE file — a mid-study order FDA itself issues to pause the investigation. Part 812 does not have one. It has something structured differently, and the difference is easy to misread as a gap.

What a clinical hold actually is

312.42 defines the tool precisely: “A clinical hold is an order issued by FDA to the sponsor to delay a proposed clinical investigation or to suspend an ongoing investigation.” One order can reach one or several investigations under the same IND. On a proposed study, the drug cannot be given to a single subject until the hold lifts. On an ongoing study, enrollment of new subjects stops immediately, and subjects already on the drug generally have to come off it, unless FDA specifically permits otherwise for their safety. The sponsor cannot proceed again until FDA notifies it the hold is lifted; if the sponsor disagrees with the reasons, it can seek reconsideration under 312.48. The order, the stop, and the appeal all run through FDA.

What 812.30 gives FDA instead

On an IDE application, 812.30(a) gives FDA exactly three outcomes: approve the investigation as proposed, approve it with modifications, or disapprove it. A disapproval order has to contain a complete statement of FDA’s reasons and tell the sponsor it may request a hearing under Part 16. For an IDE FDA has already approved, the agency’s own lever to intervene mid-study is withdrawal of that approval under 812.30(b) — on one of five grounds: noncompliance with Part 812 or the Act, an untrue or omitted material statement in the application or a report, a sponsor’s failure to answer an FDA information request on time, a finding that the risks are not outweighed by the benefits or that the investigation is scientifically unsound, or a catch-all for when continuing is otherwise unreasonable. That is FDA’s equivalent intervention — but it travels on the disapproval track, with its own stated reasons and hearing right, not a phone call that lifts on FDA’s later say-so.

Where the mid-study stop actually comes from

The device regulation puts the corresponding duty somewhere else entirely: 812.46(b)(2) requires the sponsor, not FDA, to act once it determines that an unanticipated adverse device effect presents an unreasonable risk to subjects. The sponsor has to terminate the affected investigations, or the parts of them that present the risk, no later than five working days after making that determination and no later than fifteen working days after first learning of the effect. Resuming afterward needs the reviewing IRB’s approval, and FDA’s too if the stop was triggered under this provision. FDA finds out; it does not have to be the one who calls first.

A clinical hold is FDA calling the sponsor. Withdrawing IDE approval is FDA writing a letter — and until that letter exists, finding the unreasonable risk is the sponsor’s job, not FDA’s.

The practice

Do not search an IDE file for the word “hold” and read its absence as a clean record. Check three things on their own terms instead: whether FDA has issued an approve-with-modifications or disapproval order against the current IDE under 812.30(a), whether FDA has withdrawn approval on any of the five 812.30(b) grounds, and whether the sponsor’s own safety monitoring has ever triggered a 812.46(b)(2) unreasonable-risk termination that FDA was notified of after the fact rather than one it ordered first. None of the three will turn up under a search built for the drug-side vocabulary.

This is regulatory intelligence and method, not a determination about any specific study — whether a given IDE’s history reflects any of these actions is a question for your regulatory and clinical teams, reading the actual correspondence file.