An investigator-initiated trial (IIT) does not get a pass on ClinicalTrials.gov registration just because there is no pharmaceutical sponsor writing the check. The same federal statute that governs industry trials — Section 801 of the Food and Drug Administration Amendments Act of 2007 (FDAAA 801) and its implementing regulation, 42 CFR Part 11 — applies to a qualifying academic trial exactly the same way, with one operational difference that trips up research offices constantly: in an investigator-initiated trial, the individual investigator is usually also the legal “responsible party,” not a corporate sponsor’s regulatory affairs department. CASRAI’s Clinical Trial Registration and Reporting Compliance guide covers the ICMJE/FDAAA/NIH definitional landscape in general; this guide is the operational companion — who at an academic institution must register, how to determine whether a specific IIT is even covered, and what the registration and results-reporting clock actually requires.
Who counts as the “responsible party”
Under 42 CFR Part 11, the “responsible party” is the person or entity legally accountable for registering a trial and submitting its results. It is either the trial’s sponsor, or a sponsor-designated principal investigator meeting specific criteria. In an industry-sponsored trial the sponsor is almost always a company. In an investigator-initiated trial, the individual who both initiates and conducts the study — and who holds the FDA Investigational New Drug (IND) or Investigational Device Exemption (IDE) number, where one is required — is the sponsor-investigator, defined under 21 CFR 312.3 (drugs/biologics) and 21 CFR 812.3 (devices). An institution cannot itself be a sponsor-investigator; the term applies only to an individual. That means the ClinicalTrials.gov registration obligation, and the civil liability for missing it, attaches to the named investigator personally, even though the university’s research office typically handles the mechanics of account setup and record submission on the Protocol Registration and Results System (PRS). CASRAI’s Sponsor-Investigator entry and the companion guide on conducting investigator-initiated trials under FDA regulations and GCP cover the full dual sponsor/investigator obligation set under 21 CFR Part 312 Subpart D; this page focuses specifically on the ClinicalTrials.gov piece of that obligation.
On a multi-site IIT, only one individual can hold the sponsor role for a given IND or IDE. Other-site investigators take on investigator-side obligations at their own site, but the ClinicalTrials.gov responsible-party role stays with whoever holds the IND/IDE — usually the trial’s originating investigator.
Does this trial even have to register? The “applicable clinical trial” test
Not every academic study of a drug, biologic, or device is a legally covered “applicable clinical trial” (ACT). ClinicalTrials.gov and NIH use a four-part checklist — all four conditions must be “yes” — to determine ACT status for a trial initiated on or after 18 January 2017:
- Study type is interventional — observational studies are never ACTs, regardless of what product they involve.
- The study involves a U.S. FDA-regulated drug, biologic, or device product.
- The trial is not excluded by phase or purpose — for drug/biologic trials, Phase 1 studies are explicitly excluded from the FDAAA definition (they are not ACTs, even though they may still need to be registered to satisfy ICMJE’s broader publication-eligibility rule, which does include Phase 1). For device trials, small feasibility studies whose primary purpose is device feasibility are excluded.
- The trial has a qualifying U.S. connection — at least one U.S. or U.S.-territory site, or an FDA IND/IDE number, or the product is manufactured in and exported from the United States.
This is a genuinely separate test from NIH’s own, broader “clinical trial” definition (covered in CASRAI’s NIH clinical trial definition guide), which is phase-agnostic and can require registration for an NIH-funded Phase 1 or behavioral trial even when that same trial fails the FDAAA ACT test above. A research office managing an academic trial portfolio effectively has to run this determination twice for any NIH-funded study: once against the FDAAA/ACT checklist (does 42 CFR Part 11’s registration and results-reporting mandate apply, with FDA’s civil-penalty authority behind it), and once against NIH’s own funding-conditioned policy (does the NIH Policy on the Dissemination of NIH-Funded Clinical Trial Information apply, enforced through the grant’s terms and conditions rather than FDA enforcement). A trial can be a “yes” on one and a “no” on the other.
Registration timeline
For a trial that is an ACT, registration on ClinicalTrials.gov is due no later than 21 calendar days after the first participant is enrolled. This is a firm regulatory deadline, distinct from and stricter than ICMJE’s editorial standard (registration before first patient consent, as a condition of eventual publication in an ICMJE-following journal). In practice, a well-run academic trial registers well before enrollment opens rather than treating 21 days as a target — late registration is one of the most common, and most avoidable, compliance findings in academic trial portfolios, and registering only after enrollment has begun already forecloses ICMJE-compliant prospective registration for that trial even if the 21-day statutory deadline is technically still met.
Results-reporting timeline
Registration and results reporting are separate obligations with separate clocks. For an ACT, summary results — participant flow, baseline characteristics, outcome measures, and adverse event data — are due on ClinicalTrials.gov no later than 12 months after the trial’s primary completion date (the date the last participant was evaluated for the primary outcome measure), with a possible extension of up to two years in specific circumstances, such as when the studied product is not yet FDA-approved for the studied use and is still under active development. This deadline applies regardless of whether the trial’s results were ever submitted for publication, and regardless of whether the results were positive, negative, or inconclusive — it is a disclosure obligation, not a publication-contingent one.
Form FDA 3674 and NIH’s Certification of Compliance
Two separate certification requirements sit on top of the registration and reporting obligations themselves, and both are easy for an academic sponsor-investigator to overlook:
- Form FDA 3674 must accompany certain FDA submissions — including a new IND, a new drug or device application, and in some cases a new protocol submitted under an existing IND — certifying that the sponsor or sponsor-investigator will comply with (or already has complied with) ClinicalTrials.gov registration requirements. Where a trial is already registered, the form requires the NCT number.
- NIH’s Certification of Compliance is a separate, grants-side requirement: for competing applications supporting an applicable clinical trial, the certification appears on the SF 424/PHS 398 forms, and it must also be renewed in non-competing continuation progress reports, for as long as the award is active. NIH requires this certification whether or not the grantee institution or investigator is the ClinicalTrials.gov responsible party for that trial.
Because an academic sponsor-investigator is frequently completing both an FDA submission and an NIH award simultaneously, both certifications can be due around the same time on the same trial — and neither one substitutes for the other.
What noncompliance actually costs
Three separate enforcement paths apply, and an academic institution can face any or all of them on the same trial:
- FDA civil monetary penalties. FDA can issue a public Notice of Noncompliance (posted on ClinicalTrials.gov itself), which starts a 30-day remediation clock; failing to remediate within that window exposes the responsible party to civil penalties of up to $10,000 per day of continued violation — a statutory base figure subject to periodic inflation adjustment.
- NIH grant-funding actions. Because NIH’s own dissemination policy is enforced through award terms and conditions rather than FDA’s civil-penalty authority, a grantee institution found noncompliant can face funding actions from NIH independent of, and in addition to, any FDA enforcement.
- Loss of publication eligibility. Under ICMJE’s policy, a trial that was not prospectively registered generally cannot be published in a journal that follows ICMJE’s recommendations — a consequence that lands on the investigator’s publication record even where no FDA or NIH enforcement action is ever taken.
Academic institutions are not a hypothetical risk category here. Published compliance research — most notably a widely cited STAT News investigation and a related analysis published in the New England Journal of Medicine examining results-reporting timeliness across sponsor types — has repeatedly found that university and other academic sponsors report results late, or not at all, at substantially higher rates than large industry sponsors, who typically have dedicated regulatory-affairs staff managing PRS submissions as a full-time function. The structural reason is straightforward: an academic sponsor-investigator is usually a working clinician or scientist for whom trial registration is one administrative task among many, without the institutional infrastructure that a pharmaceutical sponsor’s compliance department provides by default.
Practical steps for a research office
- Run the ACT determination at protocol development, not at enrollment. Doing this alongside protocol design and IND/IDE determination (see CASRAI’s IIT FDA regulations and GCP guide) means registration deadlines are known before the IRB approves the study, not discovered after enrollment has already started.
- Identify the responsible party by name, in writing, before the study opens — not as a default assumption that “the department” or “the office” handles it. The sponsor-investigator is personally accountable; a research office can support the PRS submission mechanically, but cannot absorb the legal responsibility.
- Register before enrollment opens, not within the 21-day statutory window, to preserve ICMJE-compliant prospective registration and avoid the common late-registration failure mode.
- Calendar the 12-month results deadline against the primary completion date at the same time registration happens, and build it into whatever system tracks the trial through closeout — not as a separate downstream compliance check discovered a year later. A monitoring plan for the trial is a natural place to anchor this.
- Confirm which certifications apply — Form FDA 3674 for the FDA submission side, NIH’s Certification of Compliance for the grants side — and track both independently rather than assuming one covers the other.
Frequently asked questions
Does a Phase 1 investigator-initiated drug trial need to register on ClinicalTrials.gov?
Not under FDAAA 801/42 CFR Part 11 — Phase 1 drug and biologic trials are explicitly excluded from the “applicable clinical trial” definition. However, if the trial is NIH-funded, NIH’s own broader clinical trial policy can still require registration regardless of phase, and if the results will be submitted to an ICMJE-following journal, ICMJE’s broader definition (which includes Phase 1) will require registration as a condition of publication. Three different rules, three different answers for the same trial.
Who is legally responsible if an investigator-initiated trial is not registered on time?
The “responsible party” — for most IITs, the individual sponsor-investigator who holds the IND or IDE, not the university as an institution. Civil monetary penalties under FDAAA 801 are directed at the responsible party.
Can a university’s research office be the ClinicalTrials.gov responsible party instead of the investigator?
The responsible party must be the trial’s sponsor or a sponsor-designated principal investigator meeting the regulation’s criteria; a sponsor-investigator by definition must be an individual, not an institution. A research office can be designated to handle the PRS submission mechanics on the sponsor-investigator’s behalf, but that does not shift the underlying legal accountability.
Is registering after the first participant is enrolled still legal?
It can still satisfy the 21-day statutory deadline under 42 CFR Part 11 if completed within that window, but it forecloses ICMJE-compliant prospective registration, since ICMJE requires registration at or before first patient consent. A trial registered on day 20 of the FDAAA window is already non-compliant with ICMJE’s publication-eligibility standard.
Does an investigator-initiated trial using donated drug or device supply still count as investigator-initiated?
Yes — what makes a trial investigator-initiated is who holds the IND/IDE and sponsor role, not the funding or supply source. A study can remain investigator-initiated even with company-donated product or a company grant, as long as the company has not taken over IND/IDE sponsorship. See CASRAI’s comparison of investigator-initiated trials and sponsor-led studies for the full distinction.
Related CASRAI resources
See the Clinical Research pillar page for the broader cluster this guide belongs to; the general ICMJE/FDAAA/NIH registration and reporting landscape guide; the investigator-initiated trials FDA regulations and GCP guide for the full sponsor/investigator regulatory obligation set; the NIH clinical trial definition guide; the research registries overview; and the CASRAI Dictionary’s Sponsor-Investigator, Investigator-Initiated Study (IIS), ClinicalTrials.gov, clinical trial registration, and prospective clinical trial registration entries.
