On June 3, 2026, the ICH Assembly's regulatory members endorsed Annex 2 to ICH E6(R3) under Step 4. In the EU, CHMP followed with final adoption on June 25, 2026, setting a coming-into-effect date of January 15, 2027. The trade coverage since has mostly treated this as one more entry in the long E6(R3) rollout — base Principles and Annex 1 already final, already covered on this site, now an annex on top. That framing undersells what Annex 2 actually is: a defined, bounded set of additions for exactly three trial-design situations, not a general refresh of good clinical practice.

Why this is a different article from the base E6(R3) piece

The base E6(R3) guideline — the Principles plus Annex 1 — restructured GCP around a quality-by-design logic: build quality into the protocol and the critical-to-quality factors up front rather than inspecting for it afterward. That guideline is trial-design-agnostic; it applies whether the trial is a traditional single-site study or something more distributed. Annex 2 does not revisit any of that. It answers a narrower question the base guideline left open: what does a sponsor specifically need to address when the trial itself departs from the traditional model — when some or all study procedures happen outside an investigator site, when the design is pragmatic rather than explanatory, or when real-world data sources do some of the evidentiary work a traditional case report form would otherwise do.

Jun 3, 2026
ICH Assembly regulatory members endorsed Annex 2 under Step 4.
Jun 25, 2026
CHMP's final EU adoption date for Annex 2.
Jan 15, 2027
The EU coming-into-effect date — the planning horizon for any trial enrolling decentralised or pragmatic elements past that point.

The three situations Annex 2 actually covers

ICH scoped Annex 2 deliberately rather than writing a general-purpose modernization document, and it said so directly: the clinical trial ecosystem will keep evolving, and the annex does not try to anticipate every future conduct model. What it does cover is established current practice in three categories.

  • Decentralised elements. Procedures and data collection that do not require the participant to attend a traditional investigator site for every visit — remote assessments, local labs or imaging, direct-to-participant drug shipment, and the oversight questions each of those raises for data integrity and participant safety.
  • Pragmatic trial design. Studies run closer to routine clinical-practice conditions — broader eligibility, less protocol-driven deviation from standard care — than a traditional explanatory trial, and the GCP considerations that shift when the trial is deliberately less controlled by design.
  • Real-world data as evidence. Using sources such as electronic health records, registries, and claims databases as part of the trial's data, which raises its own questions about data provenance, reliability, and how pre-specifying real-world-data analyses interacts with the annex's expectations.
  • What is deliberately not covered. Annex 2 does not create a fourth, separate GCP regime for "modern" trials generally — a fully traditional, single-site, explanatory trial with no real-world-data component has no new Annex 2 obligations to track.
Annex 2 is not a rewrite of GCP for the decentralised era. It is three specific additions, laid on top of the Principles and Annex 1 a sponsor already has to satisfy, for the specific design choices that trigger them. Why scope, not novelty, is the right frame for Annex 2

What to do before January 2027

For most sponsors this is not an urgent rewrite of an existing protocol. It is a planning input for protocols still in design, and a readiness question for clinical quality assurance and vendor-oversight teams that will be running decentralised or pragmatic elements after the EU effective date. The practical sequence is to identify which upcoming protocols actually use one of the three covered elements, confirm the clinical investigation procedures and vendor contracts for those elements speak to Annex 2's expectations specifically, and treat a protocol with no decentralised, pragmatic, or real-world-data component as unaffected rather than defaulting every GCP document to a rewrite it does not need.

A readiness check before the EU effective date
  1. Flag protocols by element, not by date alone. A trial finalized today that enrolls past January 15, 2027 only needs an Annex 2 review if it actually uses decentralised, pragmatic, or real-world-data elements.
  2. Separate this from the base E6(R3) work already done. If the Principles and Annex 1 quality-by-design work is already in place, Annex 2 is additive scope, not a redo.
  3. Check vendor and site agreements for the decentralised-element language specifically. This is where Annex 2's expectations most often require a contractual or procedural update, not just a protocol edit.
  4. Watch for regional adoption gaps. EU timing is confirmed; confirm FDA's and other regions' adoption status directly before assuming the same effective date applies outside the EU.

Frequently asked questions

Is ICH E6(R3) Annex 2 already in effect?

Annex 2 was endorsed under Step 4 by the ICH Assembly's regulatory members on June 3, 2026. In the EU, CHMP adopted it on June 25, 2026, with a coming-into-effect date of January 15, 2027. Sponsors planning trials with decentralised, pragmatic, or real-world-data elements that will run past that date should treat the annex as the applicable standard now, not a future consideration, since protocol and vendor-oversight design decisions made today determine whether the trial is ready.

Does Annex 2 replace the base E6(R3) Principles and Annex 1?

No. Annex 2 is additive. It is meant to be read together with the E6(R3) Principles and Annex 1 — which, in the EU, already came into effect on July 23, 2025 — not as a standalone or superseding document. A trial with decentralised or pragmatic elements still has to satisfy the base Principles and Annex 1's traditional-trial expectations; Annex 2 adds the considerations specific to the non-traditional elements on top of that foundation.

What trial elements does Annex 2 actually cover?

Three categories: decentralised elements (remote or local data collection and procedures that do not require the participant to attend a traditional investigator site for every visit), pragmatic trial design (studies run closer to routine clinical-practice conditions than a traditional explanatory trial), and the use of real-world data from sources such as electronic health records, registries, and claims databases as part of the trial's evidence. ICH has been explicit that the annex does not attempt to anticipate every future trial-conduct model; it covers what is established practice today.

Sources & further reading

  1. ICH. E6(R3) Annex 2: Guideline for Good Clinical Practice — Annex on decentralised elements, pragmatic trial design, and real-world data sources. Endorsed under Step 4 by the ICH Assembly, June 3, 2026. database.ich.org
  2. European Medicines Agency. ICH E6 Good clinical practice — scientific guideline (Annex 2 status, CHMP adoption June 25, 2026, effective January 15, 2027). ema.europa.eu

This article is provided for general informational purposes and reflects the regulatory landscape as of November 2026. It is not legal or regulatory advice. Confirm current ICH E6(R3) Annex 2 adoption status in your applicable region, including FDA and other non-EU regulators, with the relevant agency or qualified counsel before relying on a specific effective date.