Clinical Trial Close-Out Checklist: A Dependency-Gated Sequence Built on GCP and 21 CFR Part 312

At a glance

• Site close-out is a sequence of dependency-gated reconciliations, not a single visit and not a flat alphabetical document list. You cannot lock data with open queries, and you cannot dispose of investigational product before accountability is reconciled and signed off.
• The close-out visit (COV) is one event inside the close-out process. The process starts before the visit (the readiness gate) and continues long after it (the retention clock), and inspections find their findings in that post-COV tail.
• Investigational product accountability and disposition are governed: the investigator/institution holds the IP records, and the sponsor assures return or authorizes alternative disposition. Under 21 CFR 312.59 only the sponsor can authorize an alternative to return.
• The record-retention clock does not start at COV day. Under 21 CFR 312.62 it runs from marketing-application approval (or from discontinuation plus FDA notification), and ICH E6(R3) adds a “whichever is longest” rule tied to sponsor notification. Those two clocks differ, and you must satisfy both.
• The investigator owes a final report to the sponsor and a summary of outcome to the IRB/IEC. Neither is the close-out visit report, and missing either is a documented gap.

Most published close-out checklists hand you the CASPER-style alphabetical pile of documents and stop at the visit. That is where audits start, not where they end. This piece treats close-out as what it actually is in practice: a chain of reconciliations where each link gates the next, and where the obligations that outlive the visit are the ones that fail inspections. Every line below names the GCP or Part 312 obligation it satisfies, so each item is defensible when an inspector asks “why.”

Close-out visit vs. close-out process: why the distinction changes your timeline

Conflating the two is the single most common planning error. The close-out visit is a discrete monitoring event. ICH E6(R3) §3.11.4.5.2 lists site close-out among the monitor’s site-management activities, and at item (j) it specifies what the monitor confirms during close-out: the arrangement for retention of the essential records and the final accountability of the investigational product, including return and destruction or alternative disposition where appropriate. That is a confirmation activity, which means the work it confirms must already be done.

The close-out process is the chain that produces those confirmable facts: query resolution, safety follow-up closure, IP reconciliation, essential-document completeness, and the final reports to sponsor and IRB/EC. Treat the COV as a milestone you earn by completing upstream reconciliations, and your timeline becomes a sequence with gates. Treat it as a date on a calendar, and you will arrive at the visit with open dependencies you cannot close in a day.

The sibling topics here are essential-documents / TMF completeness, protocol-deviation handling, and AE/SAE reporting. Each feeds the close-out gate, and each is covered in its own piece.

The readiness gate: what must be true before you schedule the COV

Do not schedule the close-out visit until the upstream dependencies clear. The gate has three load-bearing conditions.

Data and query lock. You cannot meaningfully lock a database while material queries are open. ICH E6(R3) §2.12.5 requires the investigator to ensure the accuracy, completeness, legibility and timeliness of data reported to the sponsor, and §2.12.6 requires reported data to be consistent with the source records or the discrepancies explained. Open, unexplained discrepancies are by definition unresolved data, so query closure is a precondition for a defensible lock, not a post-visit cleanup task.

Safety follow-up status. Adverse-event and serious-adverse-event follow-up must be brought to its protocol-defined endpoint before you can assert the safety record is complete. Under 21 CFR 312.64(b), an investigator must immediately report to the sponsor any serious adverse event and must record and report nonserious adverse events on the protocol’s timetable. Outstanding follow-up means the safety dataset is still moving, and a moving dataset cannot be reconciled at close-out.

IP reconciliation started. Investigational-product accountability should be substantially reconciled before the visit, because the visit confirms final accountability rather than performing it from scratch. ICH E6(R3) §2.10.4 requires the investigator/institution (or a delegated pharmacist) to maintain records of delivery, inventory, use by each participant, and return to the sponsor and destruction or alternative disposition of unused product, including dates, quantities, and batch numbers. Those records are what the monitor reviews at the visit; they must exist beforehand.

The sequenced close-out checklist (phased, with owner and reg reference per line)

The centerpiece. Read the columns in order: a task cannot start until its gating dependency clears. “Sponsor-monitor” is the sponsor/CRA side; “site” is the investigator/institution including the site coordinator; “data mgmt” is the sponsor’s data-management function.

Phase	Task	Owner	Gating dependency	Governing reg cite
Pre-COV readiness	Resolve and close all material data queries	Site + data mgmt	None (entry condition)	ICH E6(R3) §2.12.5, §2.12.6
Pre-COV readiness	Bring AE/SAE follow-up to protocol endpoint; reconcile safety log	Site	Queries on safety data closed	21 CFR 312.64(b)
Pre-COV readiness	Start IP accountability reconciliation (delivery, dispensing, returns)	Site (pharmacist under investigator oversight)	None (entry condition)	ICH E6(R3) §2.10.1, §2.10.4
Pre-COV readiness	Confirm essential records / TMF are current and under site control	Site	Query and safety closure	ICH E6(R3) §2.12.11
COV day	Confirm essential-record maintenance and retention arrangement	Sponsor-monitor	Site records current	ICH E6(R3) §3.11.4.5.2(c), (j)
COV day	Confirm final IP accountability; decide return vs. alternative disposition	Sponsor-monitor + site	IP reconciliation complete	ICH E6(R3) §3.11.4.5.2(j); 21 CFR 312.64(a)
COV day	Verify AE reporting met protocol/GCP/regulatory timeframes	Sponsor-monitor	Safety follow-up closed	ICH E6(R3) §3.11.4.5.2(e)
Post-COV tail	Investigator returns unused IP to sponsor or disposes per authorized route	Site + sponsor	Accountability signed off	21 CFR 312.64(a), 312.59
Post-COV tail	Investigator provides sponsor an adequate final report	Site	Participation complete	21 CFR 312.64(c)
Post-COV tail	Investigator/institution provides IRB/EC a summary of outcome	Site	Trial completion at site	ICH E6(R3) §2.13
Post-COV tail	Archive essential records; assign and document custodian	Site + sponsor	All above complete	ICH E6(R3) §2.12.12, §2.12.13; 21 CFR 312.62

The table is deliberately ordered so that each post-COV item depends on a COV-day confirmation, which in turn depends on a readiness-gate reconciliation. That is the dependency chain a flat list hides.

Investigational product: reconciliation, return vs. destruction, who signs off

IP disposition is where the lines of responsibility matter most, because the wrong party authorizing destruction is a finding.

Responsibility for IP management, including accountability, handling, dispensing, administration and return, rests with the investigator/institution per ICH E6(R3) §2.10.1; the sponsor may facilitate it (forms, systems, distribution) but does not assume it. So the records of disposition are the site’s to hold, and §2.10.4 requires those records to cover delivery, inventory, per-participant use, and return to the sponsor and destruction or alternative disposition of unused product, with dates, quantities and batch numbers.

The authorization to deviate from return, however, is the sponsor’s. Under 21 CFR 312.59, the sponsor shall assure the return of all unused supplies from each investigator whose participation is discontinued or terminated, and the sponsor may authorize alternative disposition only if that disposition does not expose humans to risks from the drug. The investigator’s own obligation, at 21 CFR 312.64(a), is to return unused supplies to the sponsor when the investigation is terminated, suspended, discontinued, or completed, or otherwise provide for disposition under 312.59. Read together: the default is return; on-site destruction is an exception the sponsor authorizes in writing, and the site executes and documents it. Do not let a site destroy IP on its own initiative because it is convenient, and do not destroy anything before final accountability is reconciled and signed off. That sequence (reconcile, then sponsor-authorize, then dispose, then document) is the decision point, and inverting it is a classic close-out error.

The close-out visit report: what it must capture and the sign-off chain

ICH E6(R3) requires monitoring activities to be reported. Under §3.11.4.6, reports of monitoring activities should include a summary of what was reviewed, a description of significant findings, conclusions, and the actions required to resolve them, including follow-up on resolution of items left open in previous reports. The close-out visit report is the monitoring report for the close-out visit, so it inherits that content standard.

A practical, SOP-grade COV report outline that satisfies the §3.11.4.6 content standard:

• Visit identification: protocol, site, date, monitor, attendees (sponsor-monitor and site staff present).
• Scope reviewed: essential-document/TMF completeness, data/query status, AE/SAE reconciliation, IP accountability.
• Final IP accountability: reconciled quantities, return shipments, and any sponsor-authorized alternative disposition (cross-reference the authorization).
• Significant findings and outstanding actions: each with an owner and a target resolution, plus status of any items carried from prior monitoring reports.
• Retention arrangement: where essential records will be archived and who the named custodian is.
• Confirmation of reports due: investigator final report to sponsor, summary of outcome to IRB/EC.
• Sign-off chain: monitor prepares; the report follows the sponsor’s procedures for monitoring-report review and approval.

The report is not a substitute for the investigator’s final report to the sponsor or the summary of outcome to the IRB/EC. Those are separate obligations, covered next.

The post-COV tail: record retention, archiving, and final reports

This is the section the ranking checklists skip, and it is where inspections concentrate.

The retention clock, and a real tension to manage. The retention clock does not start on COV day. Under 21 CFR 312.62(c), an investigator shall retain the required records for 2 years following the date a marketing application is approved for the drug for the indication investigated; or, if no application is filed or it is not approved for that indication, until 2 years after the investigation is discontinued and FDA is notified. The sponsor side mirrors this: 21 CFR 312.57(c) requires the sponsor to retain records for 2 years after a marketing application is approved, or 2 years after shipment and delivery for investigational use is discontinued and FDA has been notified. So the clock is pinned to regulatory milestones, not to the visit.

ICH E6(R3) frames the same retention duty differently, and the difference is worth stating plainly rather than smoothing over. Section §2.12.12 requires the investigator/institution to retain the essential records for the required retention period in accordance with applicable regulatory requirements, or until the sponsor informs the investigator/institution that these records are no longer needed, whichever is the longest. That introduces a sponsor-notification trigger and a “whichever is longest” rule that the 21 CFR 312.62 two-year floor does not contain. The two instruments make different demands on the same point: 312.62 sets a definite minimum tied to marketing approval or discontinuation-plus-notification, while E6(R3) §2.12.12 can extend retention beyond that minimum until the sponsor affirmatively releases the records. The defensible posture is to satisfy both: never destroy below the 312.62 floor, and never destroy before sponsor release under E6(R3). Do not reconcile these into a single tidy duration; honor the longer of the two on every record.

Custodian and archiving. Retention is meaningless without a named custodian. ICH E6(R3) §2.12.13 requires the investigator/institution to keep the sponsor informed of the name of the person responsible for maintaining the essential records during the retention period, for example when the site closes or an investigator leaves. Archiving is the sponsor’s duty too: §3.16.3 requires the sponsor (or subsequent owners of the data) to retain the sponsor-specific essential records in accordance with applicable regulatory requirements.

Final reports. Two distinct reports close the loop. Under 21 CFR 312.64(c), the investigator shall provide the sponsor with an adequate report shortly after completing participation in the investigation. Separately, ICH E6(R3) §2.13 directs that upon completion of the trial the investigator/institution should provide the IRB/IEC with a summary of the trial’s outcome and, if applicable, provide the regulatory authority with any required reports. Track both on the post-COV checklist; they are easy to defer and easy for an inspector to notice missing.

Where teams get it wrong (common inspection findings at close-out)

• Destroying IP before accountability is signed off, or letting the site authorize destruction. Return is the default under 21 CFR 312.59 and 312.64(a); alternative disposition is a sponsor authorization, not a site convenience. Reconcile first, authorize second, document always.
• Starting the retention clock at COV day. It runs from marketing approval or discontinuation-plus-FDA-notification under 21 CFR 312.62, and E6(R3) §2.12.12 can push it longer. Pinning it to the visit under-retains records and invites a finding.
• Treating the COV report as the only close-out report. It does not discharge the investigator’s final report to the sponsor (312.64(c)) or the summary of outcome to the IRB/EC (E6(R3) §2.13).
• Locking data over open queries. A lock asserted while material discrepancies are unexplained contradicts E6(R3) §2.12.5 and §2.12.6, and it unwinds later when the queries surface.
• No named records custodian. E6(R3) §2.12.13 requires the site to tell the sponsor who holds the records through the retention period; “the site has them somewhere” is not an arrangement.

None of this certifies a study as compliant. Good sequencing and complete records enable the sponsor and investigator to meet their obligations; the responsibility to meet them stays with the sponsor and the investigator. Use the phased table as the spine of your close-out SOP, and make every line traceable to the obligation it satisfies.

Sources

• ICH E6(R3) Good Clinical Practice (version r3, adopted 06 January 2025) — https://www.ich.org/page/efficacy-guidelines
• 21 CFR Part 312, Investigational New Drug Application (current as of 4/14/2026) — U.S. Food and Drug Administration