For sponsors conducting clinical trials under Good Clinical Practice, the transition from ICH E6(R2) to E6(R3) represents the most substantial revision to the GCP framework in nearly a decade. Understanding precisely where sponsor obligations have shifted—whether in quality management, oversight of service providers, or data governance—is essential for updating standard operating procedures, training programs, and contractual arrangements ahead of regulatory submissions.
The analysis below maps the key areas in which E6(R3) modifies, expands, or restructures sponsor responsibilities relative to E6(R2), drawing on E6(R3)'s own descriptions of what it modernised and replaced. It covers structural architecture changes, risk- and quality-based approaches, investigator and vendor oversight, monitoring strategy, computerised systems, and safety reporting obligations.
Want Rhizome's help on your own question? Try it for free.
Sponsor responsibilities under ICH E6(R3) versus E6(R2): what changed
The headline change: a principles-plus-annexes architecture
The single most consequential change for sponsors is structural. E6(R3) is organised around overarching Principles plus Annexes rather than one static list of GCP instructions. The Principles and Annex 1 are intended to reflect and replace the content and scope of E6(R2), while Annex 2 is added for non-traditional trials 84. The consolidated contents confirm this layout: Introduction, Principles of ICH GCP, Annex 1, and Annex 2 83, and the guideline states additional annexes may be developed to address emerging innovations 76.
Two cross-cutting design choices flow from this and reshape how every sponsor obligation is read:
- Media-neutral and technology-agnostic. The guideline is "intended to be media neutral to enable the use of different technologies," and the Principles are "designed to be flexible and applicable to a broad range of clinical trials" 76. Technology use should be adapted to participant characteristics and the specific trial design, with digital health technologies such as wearables and sensors given as examples 76.
- Fit-for-purpose and proportionate as an operating rule. GCP should be tailored to the trial design and the technologies used, and a proportionate, risk-based approach supports evolving designs without compromising participant rights, safety, well-being, or reliability of results 6768. The principles "may be satisfied using differing approaches" and applied to fit the trial's purpose 76.
For sponsors, this means the R3 obligations are outcome-based expectations to be scaled to trial risk and complexity, rather than a fixed checklist tied to paper-era processes.
Sponsor responsibility changes by domain
1. Quality management and quality by design
E6(R3) makes quality management an explicit, structured sponsor responsibility built on quality by design. Under section 3.10 the sponsor must implement an appropriate quality management system across all stages of the trial and adopt a proportionate, risk-based approach that incorporates quality into the design of the trial and prospectively identifies critical-to-quality (CtQ) factors, the attributes likely to have a meaningful impact on participant rights, safety and well-being and on reliability of results 154. Quality is to be built into scientific and operational design rather than inspected in afterward, with CtQ factors identified prospectively (section 6.2) 158159. The range and extent of oversight measures should be fit for purpose and tailored to trial complexity and risk 154155. This codifies, as sponsor duties, the quality-by-design and CtQ concepts that were not part of the original core GCP model.
2. Risk management as a defined lifecycle
R3 sets out a full risk-management cycle the sponsor must run, section by section under 3.10.1 110111:
- 3.10.1.1 Risk identification: identify risks to CtQ factors before initiation and throughout conduct, considering the processes and systems used, including computerised systems, consent, randomisation, blinding, IMP administration, data handling, and service-provider activities 110111.
- 3.10.1.2 Risk evaluation: assess risks and existing controls by likelihood, detectability, and impact on participant protection and result reliability 110111.
- 3.10.1.3 Risk control: apply proportionate controls, and where relevant set pre-specified acceptable ranges such as quality tolerance limits at the trial level; if exceeded, evaluate whether there is a systemic issue 110111.
- 3.10.1.4 Risk communication: document and communicate risks and mitigations to those who must act on them 110111.
- 3.10.1.5 Risk review: periodically review controls to confirm they remain effective, accounting for emerging knowledge 152153.
- 3.10.1.6 Risk reporting: summarise and report important quality issues, including exceeded acceptable ranges and remedial actions, in the clinical trial report 152153.
3. Oversight of service providers and delegated duties
R3 tightens and clarifies the sponsor's accountability when work is outsourced. Any trial-related activity transferred to a service provider must be documented in an agreement, and anything not specifically transferred remains the sponsor's responsibility (3.6.4) 5960. The sponsor may transfer any or all activities, but ultimate responsibility for participant rights, safety and well-being and for data reliability stays with the sponsor (3.6.6) 5960. The sponsor must assess suitability and select providers (3.6.7), provide the protocol and documents needed and have access to SOPs and performance metrics for selection and oversight (3.6.7 to 3.6.8), and ensure appropriate oversight of important activities including those further subcontracted (3.6.9) 5960. Service-provider activities must be conducted per relevant GCP (3.6.10), and providers must report incidents that could affect participant safety or results (3.6.6) 5960. Where the sponsor arranges activities that fall under the investigator's responsibility, it must ensure the provider is suitable, while the investigator retains oversight of that provider (3.8) 6566.
4. Monitoring: risk-based and centralised, by design
R3 frames monitoring as a risk-based activity rather than a default of exhaustive on-site source data verification. The sponsor determines the extent and nature of monitoring from identified risks, considering objective, design, complexity, blinding, participant numbers, the IMP, current safety knowledge, and endpoints (3.11.4.1) 106. Monitoring may be on-site and/or remote, and frequency is set and modified using accumulating knowledge 106. Centralised monitoring (3.11.4.2), an evaluation of accumulated data by qualified sponsor staff, can complement and reduce site monitoring or stand alone, and can target site selection and processes 106. The sponsor develops a monitoring plan (3.11.4.3) tailored to safety, data-quality and reliability risks, describing the strategy, activities of all parties, methods and tools and their rationale, giving particular attention to participant safety and endpoints, and referencing the sponsor's policies and procedures 106.
5. Data governance and computerised systems
R3 introduces a dedicated data-governance framework as a sponsor responsibility spanning the full data life cycle. The sponsor should ensure systems and processes are designed proportionate to risks to participants and to reliability, and documented, covering confidentiality of participant data, fit-for-purpose management of computerised systems, safeguarding essential elements such as randomisation, dose adjustments and blinding, and supporting key decisions such as data finalisation before analysis, unblinding, allocation to analysis sets, design changes, and IDMC activities (section 4) 104105. The data life cycle is enumerated (4.2.1 to 4.2.8: capture, metadata and audit trails, review, corrections, transfer/migration, finalisation, retention and access, destruction) 104105, and computerised systems carry their own subsections (4.3.1 to 4.3.8: procedures, training, security, validation, release, failure, technical support, user management) 102103. This structured, life-cycle treatment of data integrity and computerised-system governance is a substantive expansion of the sponsor's explicit duties.
6. Safety assessment and reporting
R3 keeps the sponsor responsible for ongoing safety evaluation and organises the obligations under section 3.13. The sponsor is responsible for ongoing safety evaluation of the IMP, with the Investigator's Brochure (or current scientific information) as the basis of assessment, and must aggregate and review safety information in a timely way, including events before IMP administration such as during screening, and update the protocol, IB, consent materials and related documents as needed (3.13.1) 135. Reporting duties (3.13.2) include submitting safety updates and periodic reports, expediting SUSARs to authorities per applicable requirements and ICH E2A, judging expectedness against the reference safety information, reporting to investigators and IRBs/IECs with appropriate urgency, reporting urgent safety issues without undue delay, and using alternative reporting arrangements only where prospectively agreed with authorities and described in the protocol 133134. Section 3.13.3 requires prompt action on immediate hazards, determination of causes, and remedial action 133134.
7. Investigational product management
Under section 3.16-related provisions, the sponsor supplies the IMP to investigators/institutions (and, where appropriate and permitted, directly to participants), only after required IRB/IEC and regulatory approval, with safeguards for product integrity, protocol use and participant safety 116. The sponsor ensures handling and storage instructions covering receipt, handling, storage, dispensing, retrieval of unused product, and return or authorised disposition 116; supports blinding, including a process to prevent and detect inappropriate unblinding and a mechanism for rapid emergency identification while protecting other assignments 116; and, for significant formulation changes, has the supporting studies (for example stability, dissolution, bioavailability) before use 116. Products should be managed per specifications and protocol, with processes for handling, shipping, storage, dispensing, return, and destruction or disposal 121122.
8. Protocol, trial design and participant engagement
R3 pushes design quality and feasibility onto the sponsor's plate. The protocol should be clear, concise and operationally feasible, minimise unnecessary complexity, and mitigate or eliminate important risks, incorporating input from relevant interested parties where appropriate (Appendix B) 100101. The sponsor must ensure design, conduct and the data generated are of sufficient quality to support reliable results and sound decisions, and provide the protocol and an up-to-date Investigator's Brochure to potential investigators with sufficient review time (3.7 and 3.9) 9597. R3 also introduces explicit sponsor-facing language on early patient and community engagement: involving patients and advocacy groups early can help refine the design, address the logistics of participation, and assess suitability of digital health technologies used in decentralised elements (3.1.1) 87.
9. Non-traditional trials: Annex 2 (new)
Annex 2 has no counterpart in E6(R2). It was added because of increased use of varied operational approaches, design elements and data sources, and it addresses trials with decentralised elements, pragmatic elements, and/or real-world data 6768. Decentralised elements are activities conducted outside the investigator's location (home visits, local healthcare centres, mobile units, remote video, and DHTs for visits, procedures and data collection); pragmatic elements integrate usual clinical practice, such as simplified protocols and streamlined data collection 6768. Annex 2 defines RWD as patient health-status data from sources outside trials (EHRs, registries, claims), distinguishes primary collection from secondary use, and requires such data to be fit for purpose when used for endpoints or as external controls 7475. Annex 2 is expressly not comprehensive and not an endorsement of any specific methodology 6768.
How to read these changes as a sponsor
The net effect is a shift from prescriptive, process-level GCP to an accountability model where the sponsor designs quality in, runs a documented risk-management cycle, and scales oversight, monitoring, and data governance to trial-specific risk, all in a technology-neutral frame that explicitly accommodates decentralised and real-world-data trials 76676884. The specific sponsor clauses have been reorganised and, in several areas (quality by design, the risk cycle, data governance and computerised systems, and Annex 2), materially expanded relative to the original core GCP model.
Limitations
This analysis is grounded in the E6(R3) documents available in the ICH corpus. The full E6(R2) consolidated text is not present in this dataset, so the change statements rely on E6(R3)'s own description of what it replaced (Principles plus Annex 1 reflecting and replacing R2, Annex 2 as new) rather than a line-by-line diff against the R2 source 8483. For a clause-level comparison of any single requirement, confirm the corresponding E6(R2) wording against the R2 document directly.