ICH E6(R3) as the Common Floor

Part of Regulatory Affairs and the Jurisdiction Map: The Global GCP Baseline.

About 6 minutes. Free, in your language, no account needed.

Why this matters

There is a reason a coordinator trained in Helsinki can work a trial in Nairobi without relearning the job. Most of the technical expectations for how a clinical trial is run are the same almost everywhere, and they are the same because of one document.

ICH E6, Good Clinical Practice, is the international standard for designing, conducting, recording and reporting trials that involve human participants. Revision 3 restructured it substantially. The principles moved into a core document, the operational detail moved into Annex 1 for interventional trials, and Annex 2 addressed additional considerations including the use of real world data.

For a regulatory professional this document is the floor. Knowing what sits on the floor, and knowing what does not, is the difference between a defensible global position and an assumption.

What the baseline actually settles

E6(R3) settles the expectations you can carry across a border without checking:

That responsibilities are assigned and documented. The sponsor's duties and the investigator's duties are distinct, and delegation is recorded. A person performs a trial task because someone with authority permitted them to, in writing, from a date.

That participants consent, informed and freely. The consent must be given before any trial-specific procedure, documented, and renewed when what a participant was told stops being true.

That records support what is claimed. Data should be attributable, legible, contemporaneous, original and accurate, with the further expectations of completeness, consistency, endurance and availability. This is the principle behind almost every data finding an inspector writes.

That risk drives effort. A proportionate, risk-based approach is expected rather than uniform intensity everywhere. Quality is built into design rather than inspected in afterwards, and issues that matter to participant safety and data reliability get the attention.

That oversight survives delegation. A sponsor may delegate duties to a contract research organization or any other vendor. It does not thereby delegate responsibility, and it must be able to evidence that it oversaw what it delegated.

These are not country-specific. A sponsor that fails any of them fails everywhere.

What the baseline does not settle

This is the part that gets people into trouble, because the baseline is comprehensive enough to feel complete.

It does not say who authorizes your trial. E6 refers to the regulatory authority and the ethics committee without telling you which body that is in Kenya, or what it is called, or how to reach it.

It does not give you a clock. National law does. An expedited safety report has a seven day or fifteen day clock depending on the report and the jurisdiction, and E6 does not supply the number that binds you.

It does not decide who may consent for someone who cannot consent for themselves. That is national law, and it varies substantially. Regulation (EU) No 536/2014 addresses incapacitated subjects, minors, pregnant and breastfeeding women, cluster trials and emergency situations in Articles 30 to 35; a different country answers those questions under a different instrument, or in some cases has not answered some of them at all.

It does not tell you what you may import, or how. Investigational product crossing a border is a customs and licensing question governed by national law and often by a different agency from the one that authorized the trial.

It is not itself law anywhere. This is the point most often missed. E6(R3) is a guideline adopted by the members of the International Council for Harmonisation. It acquires legal force only when a jurisdiction implements it, and the manner of implementation differs.

Implementation, and why the manner matters

Three jurisdictions can all be described as "following ICH GCP" and mean three different things.

Some adopt it directly as national guidance. Some transpose it into national regulation with national modifications, so the text that binds you is the national text and the differences are real. Some have their own long-standing instrument that predates or parallels ICH and is substantively similar without being identical.

CRLN's register records the governing regime per country for exactly this reason, and it names the national instrument where one has been verified. Kenya, for example, is recorded as applying E6(R3) as national guidance while being neither an ICH member nor an observer, with the Pharmacy and Poisons Act, Cap 244 and the 2022 clinical trials guidelines as the national instrument. That is a materially different statement from "Kenya follows ICH GCP," and it is the one you can defend.

In practice

A sponsor's global standard operating procedure states that expedited safety reports go to authorities within fifteen calendar days. It cites E6 as its basis.

The procedure is wrong in a specific and dangerous way. Fifteen days is the outer clock for most serious unexpected reports in several major jurisdictions, but a suspected unexpected serious adverse reaction that is fatal or life threatening carries a seven day clock in the European Union and in others. A procedure that says fifteen everywhere will be late, by eight days, on precisely the reports that matter most.

The regulatory lead's fix is not to change fifteen to seven. It is to replace a single global number with a per-jurisdiction table, because the correct answer is that there is no single number, and a procedure that implies there is will fail somewhere.

What this looks like when it goes wrong

A sponsor writes "conducted in accordance with ICH GCP" in a protocol running in eleven countries and treats that sentence as its regulatory strategy.

It is a true sentence and an empty one. It describes the floor. It says nothing about who authorized the trial in each country, on what instrument, under what conditions, or what each authority requires that the floor does not.

When an inspector in one of those eleven countries asks which national requirements were identified and how the sponsor satisfied them, "we followed ICH GCP" is not an answer. It is a statement that the question was never asked.

Key takeaways