Protocol Amendment

A protocol amendment is a formal, sponsor-initiated change to — or clarification of — a research protocol that has already received Institutional Review Board (IRB) or Independent Ethics Committee (IEC) approval. ICH E6(R2) Section 1.60 defines it simply as “a written description of a change(s) to, or formal clarification of, a protocol.” What makes something an amendment rather than something else is timing and sequence: an amendment is proposed, reviewed, and approved before the changed procedures take effect. That prospective sequence is the entire point — it is what separates an amendment from a protocol deviation, which is discovered and documented only after an unplanned departure has already happened.

The core rule: no implementation before approval, with two narrow exceptions

ICH E6(R2) Section 4.5.2 states the investigator’s obligation directly: no deviation from, or change to, the protocol should be implemented without prior agreement from the sponsor and documented approval from the IRB/IEC, except in two circumstances:

• Immediate hazard: a change is necessary to eliminate an apparent immediate hazard to trial subjects. The safety action comes first; sponsor and IRB/IEC notification follows promptly.
• Logistical or administrative changes only: the change involves purely operational details — ICH E6(R2)’s own examples are a change of monitor or a change of telephone number — that do not affect subject safety, rights, or welfare, or the scientific integrity of the trial.

US regulations for FDA-regulated human subjects research state the same rule from the IRB’s side of the desk. 21 CFR 56.108, governing IRB functions and operations, requires IRBs to have written procedures ensuring that changes in approved research may not be initiated without IRB review and approval, except where necessary to eliminate an apparent immediate hazard to subjects. The Common Rule (45 CFR 46) imposes parallel obligations on IRBs overseeing federally funded non-FDA-regulated human subjects research. In both frameworks, the default is: submit, get approval, then implement — not implement, then report.

Two separate approval tracks: the IRB/IEC and the regulator

For an IND-regulated drug trial in the United States, a protocol amendment travels two parallel paths that are easy to conflate:

• IRB/IEC review — required for any amendment before the changed procedures reach subjects, per ICH E6(R2) 4.5.2 and 21 CFR 56.108.
• FDA notification — separately, 21 CFR 312.30 requires the IND sponsor to submit a protocol amendment to FDA whenever it makes a substantial change to an existing protocol, adds a new protocol, or adds a new investigator. 21 CFR 312.30 recognizes three amendment types, each prominently labeled as such in the submission: New Protocol (a new protocol plus a description of the most clinically significant differences from the prior one), Change in Protocol (a description of the change with reference to the submission that contained the original protocol), and New Investigator (the added investigator’s qualifications and a reference to the already-submitted protocol).

IRB/IEC approval and FDA notification are not substitutes for one another — a sponsor amending an IND-regulated protocol generally needs both, and neither track’s paperwork discharges the other’s obligation. Non-FDA-regulated protocols (e.g., most federally funded social/behavioral research reviewed only under the Common Rule) only need the IRB track.

Full-board review vs. expedited review of an amendment

Not every amendment requires the same level of IRB scrutiny. Under the expedited-review categories set out in 45 CFR 46.110 and the parallel FDA provision at 21 CFR 56.110, an IRB chair or a designated reviewer — rather than the convened full board — may review minor changes to previously approved research during the period for which approval is already authorized. What counts as “minor” is an IRB judgment call, but the general dividing line most Human Research Protection Programs apply is whether the change affects the risk-benefit profile, the eligibility criteria, the informed consent process or document, or the scientific validity of the trial. A vendor swap with no assay or handling change is a common example of an expeditable amendment; adding a study arm, changing a primary endpoint, broadening eligibility, or revising consent language are the kinds of changes that typically require full-board review, precisely because they touch the things expedited review is not meant to cover. See Expedited Review (IRB) for how that determination is made more generally.

Substantial vs. non-substantial amendments: the EU Clinical Trials Regulation classification

Outside the FDA framework, sponsors running trials in the EU work under a different, explicitly two-tier classification. Regulation (EU) No 536/2014, Article 2(2)(13) defines a substantial amendment as any change to an aspect of the trial, made after the original authorisation decision, that is likely to have a substantial impact on the safety or rights of subjects, or on the reliability and robustness of the trial data. Examples treated as substantial under Regulation 536/2014 guidance include changes to the investigational medicinal product’s formulation or manufacturing, added toxicology data, a new study arm or dose level, a materially different eligibility criterion, or a change of coordinating investigator. A non-substantial amendment is everything else — administrative or clarifying changes that do not meet that impact threshold, comparable in spirit to the ICH E6(R2) ‘logistical or administrative’ exception described above. The consequence of the classification is procedural: only substantial amendments must go through formal Member State (re-)authorisation via the Clinical Trials Information System (CTIS) before they can be implemented; non-substantial amendments do not.

Notification and review timelines: FDA vs. EMA/EU

The US and EU frameworks differ not just in classification but in how the clock runs:

• FDA (21 CFR 312.30): the regulation requires the sponsor to submit the protocol amendment to FDA, but does not itself impose a fixed waiting period before the changed protocol can be implemented — IRB/IEC approval, not a separate FDA clock, is what gates implementation in the ordinary case. The one built-in exception is 21 CFR 312.30(b)(2)(ii): a change made to eliminate an apparent immediate hazard may be implemented immediately, provided FDA is subsequently notified by protocol amendment and the reviewing IRB is notified per 21 CFR 56.104(c) — the regulation text does not itself specify a fixed number of days for that subsequent FDA notification.
• EU CTR / CTIS (Regulation 536/2014, Article 22): a substantial amendment requires affirmative Member State authorisation before implementation, on an explicit statutory clock. The Member State(s) concerned must report their assessment within 38 days of the validation date. That period may be extended by up to 31 days if the Member State requests additional information from the sponsor; the sponsor’s response is capped at a period the Member State sets, itself capped at 12 days; and after receiving that response the Member State completes its assessment within a further 19 days. In short: EU substantial amendments are authorization-gated with a defined multi-week statutory timeline, while FDA protocol amendments are submission-gated, with IRB/IEC approval (not an FDA response clock) controlling when the change can actually take effect.

Sponsors running multi-region trials should not assume the two regimes are interchangeable: a change classified as expeditable under 45 CFR 46.110 / 21 CFR 56.110 in the US may still meet the EU CTR’s substantial-amendment threshold and require CTIS authorisation in EU sites, and vice versa. See ICH GCP (Good Clinical Practice) for the harmonized framework both regimes sit within.

What commonly triggers an amendment

• Adding, removing, or modifying a study arm or dose level
• Changing inclusion/exclusion (eligibility) criteria
• Revising the informed consent form or process
• Changing the primary or secondary endpoints, or the statistical analysis plan in a way that affects the protocol itself
• Adding or changing a study site or a principal investigator
• Modifying the visit schedule, procedures, or assessments in a way that affects subject burden or risk
• Changes prompted by accumulating safety data reviewed by a Data Safety Monitoring Board

Amendment vs. deviation: why the distinction matters operationally

The distinction is not academic paperwork — it drives which compliance workflow applies. An amendment is a change to the protocol’s own text, planned in advance and gated by prospective IRB/IEC (and, where applicable, FDA) approval; nothing changes on the ground until that approval is documented. A protocol deviation is an unplanned departure from the protocol as currently approved, identified after it has already occurred, and assessed retrospectively for its effect on subject safety and data integrity. Sites and sponsors that blur the two risk two different failure modes: treating a deviation as though it needs sponsor “approval” before it can be documented (deviations are, by definition, already in the past), or treating a planned change as a deviation and implementing it before the IRB/IEC has actually approved it — which is itself a compliance finding, not a workaround. For the mechanics of drafting the protocol document an amendment ultimately revises, see CASRAI’s guide How to Write a Research Protocol for IRB Submission.

Related CASRAI terms

Protocol Deviation · ICH GCP (Good Clinical Practice) · IRB (Institutional Review Board) · Expedited Review (IRB) · ICH E6(R2) · Informed Consent · IRB/REC Approval Process · How to Write a Research Protocol for IRB Submission