ICH E6(R3): What the Standard Contains and How to Navigate It

At a glance

• E6(R3) is structured, not listed: a set of overarching principles, then Annex 1 showing how to apply them to interventional trials, with further annexes anticipated.
• The principles are meant to be read together, not cherry-picked. The guideline says so explicitly.
• Proportionality is the instruction that governs how you read every requirement: process should scale to the importance of the data and the risks involved.
• Duties are allocated by section: IRB/IEC in section 1, investigator in 2, sponsor in 3, data governance in 4. Knowing that map is most of knowing where to look.
• For a US IND trial, GCP is a guideline and 21 CFR Part 312 is law. You satisfy both, and where they differ the regulation is the binding one.

What E6(R3) is, and why its shape matters

Most summaries of ICH E6(R3) hand you the principles as a numbered list and stop. That is a poor guide to using the document, because the principles are not the whole standard and were never meant to be applied in isolation.

The guideline is built in two layers. The overarching principles come first and state what good clinical practice requires in general terms. Annex 1, including its appendices, is then intended to provide information on how the principles can be appropriately applied to clinical trials, and additional annexes may be developed to respond to the needs of interested parties and to address emerging innovations in trial design and conduct (ICH E6(R3), Introduction). That second sentence is worth noticing: the structure is deliberately extensible, which is why the principles are written generally enough to survive new trial designs.

The practical consequence is that a question about what GCP requires usually has a two-part answer. The principle tells you the obligation; the annex tells you what discharging it looks like for your kind of trial.

Read the principles together, not individually

The guideline is explicit that the principles are interdependent and should be considered in their totality to assure ethical trial conduct and reliable results (ICH E6(R3), Principles). This is not throat-clearing. It is a warning against the common failure of quoting one principle to justify a decision that another principle forbids.

The hierarchy is set by the first principle: the rights, safety and well-being of the participants are the most important considerations and should prevail over interests of science and society (ICH E6(R3) Principle 1.1). When two considerations genuinely conflict, that is the tie-breaker, and it is why a protocol departure made to protect a participant sits in a different category from one made for convenience.

A second principle does a lot of structural work and is easy to miss because it reads like contract advice. Agreements should clearly define the roles, activities and responsibilities for the clinical trial and be documented appropriately, and where activities have been transferred or delegated to service providers, the responsibility for the conduct of the trial, including quality and integrity of the trial data, resides with the sponsor or investigator respectively (ICH E6(R3) Principle 10.2). That single sentence is why outsourcing changes who does the work and never who answers for it.

Proportionality: the instruction that changes how you read everything else

If you take one operating idea from E6(R3), take this one. Clinical trial processes and risk mitigation strategies implemented to support the conduct of the trial should be proportionate to the importance of the data being collected and the risks to trial participant safety and data reliability (ICH E6(R3), Principles).

Proportionality is a licence as much as a constraint. It says that a low-risk trial collecting a small amount of non-critical data is not obliged to build the same apparatus as a large first-in-human study. Teams routinely over-read GCP as demanding maximum process everywhere, then resent it. The standard does not ask for that.

It also sets the burden. If you scale process down, the reasoning is what you have to be able to show. Proportionality is a judgement you document, not a default you assume.

Quality by design is the companion idea: it should be implemented to identify the factors, meaning the data and processes, that are critical to ensuring trial quality and the risks that threaten the integrity of those factors (ICH E6(R3), Principles). You decide what is critical first, then aim your process at it.

Where to look for what

Annex 1 allocates duties by party, and knowing the map saves most of the searching.

Section 1, IRB/IEC. Composition, functions, operations, procedures and records of the review body.

Section 2, Investigator. Qualifications and training, resources, responsibilities, communication with the IRB/IEC, protocol compliance, participant care and safety reporting, informed consent, investigational product management, randomisation and unblinding, records and reports. The qualification bar sits at the top: investigators should be qualified by education, training and experience to assume responsibility for the proper conduct of the trial and should provide evidence of such qualifications (ICH E6(R3) §2.1.1). The obligation to produce evidence, not merely to be qualified, is characteristic of the whole document.

Section 2 also carries the data obligation at site level: in generating, recording and reporting trial data, the investigator should ensure the integrity of data under their responsibility, irrespective of the media used (ICH E6(R3) §2.12.1).

Section 3, Sponsor. Trial design, resources, allocation of activities, qualification and training, agreements, investigator selection, oversight, quality management, quality assurance and quality control, noncompliance, safety reporting, investigational product, data and records. The overarching sponsor duty is that the sponsor should ensure that the trial design and trial conduct, the processes undertaken and the information and data generated are of sufficient quality to ensure reliable trial results, trial participants’ safety and appropriate decision making (ICH E6(R3) §3.9.1).

Note the verb. It is an obligation to ensure, discharged through oversight and evidence, not by personally performing the work.

Section 4, Data Governance. Blinding safeguards, the data life cycle, and computerised systems. This is where E6(R3) addresses the machinery that produces the data, and it is the section most often overlooked by teams who assume data questions live in section 3. The sponsor’s side of it is stated there: the sponsor should ensure the integrity and confidentiality of data generated and managed, and should apply quality control to the relevant stages of data handling to ensure that the data are of sufficient quality to generate reliable results (ICH E6(R3) §3.16.1).

Quality control is not a separate department

One point deserves pulling out because it is frequently misread. Quality control should be applied using a risk-based approach to each stage of the data handling, and within clinical trials, monitoring and data management processes are the main quality control activities (ICH E6(R3) §3.11.3).

Monitoring is quality control. It is not an independent check bolted on afterwards, and it is not the same thing as audit. Teams that treat monitoring as a compliance ritual rather than as the mechanism by which data quality is controlled tend to produce monitoring reports that satisfy nobody.

The IRB/IEC section is not just paperwork

Section 1 is the section teams skim, usually because the IRB feels like an external body with its own process. Reading it changes how you treat submissions.

The purpose of an IRB/IEC is to safeguard the rights, safety and well-being of all trial participants, and appropriate consideration should be given to trials that intend to recruit vulnerable participants (ICH E6(R3) §1.2.1). That is the same hierarchy as Principle 1.1, expressed institutionally. The review body exists to apply the first principle independently of the people who want the trial to succeed.

The records obligation is broader than most sites expect. The IRB/IEC should retain all relevant records, including documented procedures, membership lists, lists of occupations and affiliations of members, submitted documents, minutes of meetings and correspondence, in accordance with applicable regulatory requirements, and make them available upon request from the regulatory authorities (ICH E6(R3) §1.5.1). Membership and affiliation lists are part of the record because independence is something the body has to be able to demonstrate, not merely assert.

Protocol compliance is stated flatly

Section 2 contains one of the shortest and least negotiable sentences in the standard: the investigator should comply with the protocol, GCP and applicable regulatory requirements (ICH E6(R3) §2.5.2). There is no proportionality qualifier on it.

Where proportionality re-enters is in how deviations are triaged. The sponsor should determine necessary trial-specific criteria for classifying protocol deviations as important, important deviations being the subset that may significantly impact the completeness, accuracy or reliability of the trial data, or that may significantly affect a participant’s rights, safety or well-being (ICH E6(R3) §3.9.3). Note who holds that duty. Defining what counts as important is a sponsor decision made in advance, not a judgement a site improvises after the fact.

Risk management sits inside quality management

Section 3.10 sets out quality management, and its risk management subsection opens by describing a proportionate approach to the identification and management of risk (ICH E6(R3) §3.10.1). The organising idea is that identifying the critical to quality factors involved in each trial, and the risks associated with those factors, helps ensure efficiency by focusing on activities critical to achieving the trial objectives (ICH E6(R3) §3.10).

That is the same proportionality argument again, now operationalised: decide what is critical, aim effort there, and accept that effort spent elsewhere buys less. The detailed machinery of risk-based quality management, including quality tolerance limits, is a subject in its own right and is covered separately.

How this sits against 21 CFR Part 312

For a trial conducted under a US IND, GCP and the regulations are different kinds of instrument. ICH E6(R3) is a guideline describing expectations; 21 CFR Part 312 is binding law. You do not get to choose between them, and where the regulation is more specific, it governs.

Two Part 312 provisions are worth holding alongside the guideline. The investigator is responsible for ensuring that an investigation is conducted according to the signed investigator statement, the investigational plan, and applicable regulations, and for protecting the rights, safety, and welfare of subjects under the investigator’s care (21 CFR §312.60). And on the sponsor side, the sponsor shall monitor the progress of all clinical investigations being conducted under its IND (21 CFR §312.56).

The second is stated as a flat duty with no proportionality qualifier attached. E6(R3) will tell you how much monitoring, and of what, is proportionate. Part 312 tells you that monitoring happens.

E6(R3) took effect as the operating standard in 2025, replacing R2. What specifically changed between the two versions is a separate subject with its own page, and this one deliberately does not retread it.

Using the document

Three habits make the standard workable rather than oppressive.

Start from the obligation, not the anecdote. When someone asserts that GCP requires a practice, ask which principle or annex section says so. A surprising proportion of “GCP requires” claims turn out to be organisational habit.

Ask what is proportionate before asking what is possible. The document invites you to size process to risk, and that reasoning is worth writing down while you still remember it.

Check who holds the duty. Most disputes about compliance are really disputes about allocation, and the section map above usually settles them faster than argument does.

Sources

• ICH E6(R3) Good Clinical Practice, version R3
• 21 CFR Part 312, Investigational New Drug Application (current as of 4/14/2026)