An IND safety report splits into two clocks, not one

This blog has already covered what can stop a study FDA has already cleared to run, in the clinical hold a sponsor doesn't get to negotiate, and the two sets of obligations one person can owe at once, in a sponsor-investigator's full duties under both roles. IND safety reporting is the duty that feeds both of those mechanisms their evidence. The rule, at 21 CFR 312.32, obligates a sponsor to tell FDA about a suspected adverse reaction on two different clocks, not one, depending on how serious the event is and how expected.

Seven days for the worst combination, fifteen for the rest

21 CFR 312.32 splits expedited IND safety reporting by severity and by expectedness together, not by either one alone. An unexpected fatal or life-threatening suspected adverse reaction gets the short clock: initial notice to FDA within seven calendar days of the sponsor's receipt of the information, typically by phone or fax, followed by a complete written report shortly after. Other serious and unexpected suspected adverse reactions — and certain other findings, such as results from another study suggesting a significant risk in people exposed to the drug — get fifteen calendar days. A reaction that is merely serious, or merely unexpected, doesn't by itself tell a sponsor which clock applies; it takes both facts together to know whether a report is due in a week or two.

Expectedness is judged against the investigator's brochure, not the label

A reaction is “unexpected” under the rule when its nature or severity isn't consistent with the current investigator's brochure, or with the protocol if no brochure exists yet. That's a narrower and more specific test than it sounds. A reaction can be well known to the clinical team, discussed in published literature, and still count as unexpected for reporting purposes if the IB on file at the time doesn't describe it — because the comparison runs against a specific, version-controlled document, not against general medical knowledge or against whatever the eventual label will say once the drug is approved. A sponsor relying on “everyone already knows this can happen” is answering the wrong question.

Seriousness is a defined list, not a feeling

The rule's seriousness criteria are also closed rather than open-ended: death, a life-threatening event, inpatient hospitalization or its prolongation, a persistent or significant disability, a congenital anomaly, or an intervention required to prevent one of those outcomes. A reaction that is uncomfortable, unusual, or clinically notable but doesn't fit one of those categories isn't a serious adverse reaction for this rule's purposes, however much it concerns the treating investigator. The two filters — a closed list for seriousness, a document-specific test for expectedness — are what decide the clock, and a sponsor that applies either one loosely will end up reporting on the wrong one, or not at all.

The sponsor owes the investigators, not only FDA

The same rule that sets the sponsor's clock to FDA also obligates the sponsor to get relevant safety information to every participating investigator, and through them to the reviewing IRBs, in a form and manner that keeps the people running the study current on what the sponsor has reported. That notice doesn't have to repeat every individual case report line for line — amendments to the rule were written specifically to let sponsors summarize accumulating evidence in an IND safety report rather than forward each unconfirmed case separately, after over-reporting under the older approach had been burying investigators and IRBs in individual reports that mostly turned out not to show a causal signal. The obligation is still real; it's just aimed at the aggregate picture a site needs to keep consenting subjects informed, not at every single unconfirmed event.

A safety tracker has to hold the IB version it compared against

Because the expectedness call is only as good as the document it was measured against, a sponsor's safety database needs a field most adverse-event trackers skip: which edition of the investigator's brochure, and which date, supported each individual determination. A brochure revised six months later to add a reaction doesn't reopen the earlier call — that determination was correct against the document that existed when it was made — but a tracker with no record of which version was in force at the time has no way to show that the original call was reasonable at all, which is exactly the record an inspector or a sponsor's own auditor will ask to see first.

Where this meets the file

An IND safety reporting tracker that keeps the applicable brochure edition as its own required field alongside the seriousness and expectedness determination — not just the reaction and the date — is previewed in the launch catalog alongside this week's other clinical templates. If your program tracks that determination differently, the shelf takes that correction directly.