RWD and RWE in EAEU pharmaceutical registration: new rules from two 2024-2025 EEC recommendations
Imagine an orphan drug for a disease affecting only a few hundred patients nationwide. Conducting a classic randomized trial is physically impossible. Until recently, regulators faced a choice with no good answer: either demand data that cannot be obtained, or register the drug nearly blind.
In October 2024, the EEC Council adopted a recommendation that changes this situation. Data from hospital systems, patient registries, insurance databases, and even wearable devices can now officially enter a registration dossier as full-fledged evidence of a drug’s efficacy and safety.
Let’s break this down: what happened, where this data comes from, and what to do with it in practice.
Why Clinical Trials Are Not Enough
Randomized Clinical Trials (RCTs) are considered the gold standard of evidence-based medicine. Strict patient selection, control groups, and blinded designs ensure purity of results, but create a structural problem: the patient in a real pharmacy looks very different from a trial participant.
RCTs exclude people with comorbidities, the elderly, pregnant women, and those taking multiple other medications. The result is that efficacy gets studied in a carefully curated population, and then everyone wonders why the findings don’t hold in real clinical practice. Rare adverse reactions are a separate issue: if a side effect occurs in one out of ten thousand patients, detecting it in a 500-person trial is statistically impossible. Several serious safety problems were identified only after registration, once millions of people were already taking the drug.
Real-World Data (RWD) and the evidence derived from it — Real-World Evidence (RWE) — close exactly these gaps. The FDA and EMA have worked with RWE for several years: the EMA has adopted the ICH E6(R3) standard oriented toward digital data sources, while the FDA is in the process of implementation. The EAEU, until recently, operated without a clear regulatory foundation for such data.
Two Documents That Changed the Rules
EEC Council Recommendation No. 1 dated 18.10.2024 established general approaches to the collection, analysis, and use of RWD in the EAEU. This is a strategic document: it defines the principles for working with data, recognizes permissible sources, and sets a legal framework through Articles 3, 4, 6, and 12 of the Agreement on Unified Principles of Medicinal Product Circulation dated December 23, 2014.
EEC Board Recommendation No. 15 dated 10.06.2025 is the practical toolkit. The Guideline on General Aspects of Using RWD entered into force on July 10, 2025, and for the first time at Union level formalized requirements for data quality, study design, and evidence generation procedures.
Both documents are recommendations rather than mandatory regulations: binding Union-level rules do not yet exist. Practice shows, however, that EEC recommendations in pharmaceuticals tend to become de facto requirements quickly, particularly during dossier review.
Sources of Real-World Data
Recommendation No. 1 officially recognized six categories of RWD sources.
Patient Registries are considered the most structured source. A registry may be organized by disease (all patients with a specific diagnosis) or by drug (all those receiving a certain treatment). They enable long-term observation of therapy outcomes across large populations.
Electronic Health Records (EHRs) reflect the real patient journey: diagnoses, prescriptions, laboratory results. The main challenge is that different systems rarely communicate well with each other, and data entry errors are common.
Insurance Data covers medical care cases paid through compulsory or voluntary health insurance. Excellent for analyzing hospitalization rates and resource consumption, but lacking in detailed clinical depth.
Wearable Devices (wearables) — fitness trackers, medical sensors, smartphone apps — provide continuous monitoring in ambulatory settings. Regulatory use requires technical validation of each device.
Patient-Reported Outcomes (PRO) are obtained through questionnaires and surveys completed by patients themselves. This is the only way to capture subjective experiences: pain, fatigue, quality of life. The limitation is significant: results depend heavily on how carefully the patient responds.
Among the six recognized sources, the most unconventional for the regulatory field is social media and open platforms. Regulators permit their use for identifying safety signals, but verification requirements are the highest of all categories.
One important detail: using Artificial Intelligence for automated collection and primary analysis of RWD is officially permitted. This opens the door to processing datasets that would otherwise be unmanageable.
| Source | Strengths | Limitations |
|---|---|---|
| Patient Registries | High quality, structured | Expensive to maintain, incomplete coverage |
| EHRs | Large samples, real practice | System incompatibility, entry errors |
| Insurance Data | Scale, economic transparency | Lacks detailed clinical data |
| Wearable Devices | Continuous monitoring | Device validation required |
| PRO | Patient perspective | Subjectivity, risk of incomplete responses |
| Social Media | Rapid signal detection | Complex verification |
How the Regulator Evaluates Data Quality
Not all data from an EHR or registry qualifies for a dossier. The Guideline approved by Recommendation No. 15 introduced the concept of fitness-for-purpose and splits evaluation into two directions.
Relevance asks whether the data can answer a specific scientific question. Five parameters are checked:
- Population matching: does the source population align with the drug’s target audience by age, diagnosis, and comorbidities.
- Exposure: can it be precisely established that the patient received this specific drug, at what dose, and for how long.
- Endpoints: are there clearly defined, measurable outcomes in the data.
- Covariates: is information available on confounding factors for correct interpretation.
- Observation period: is it long enough to capture the necessary events.
Reliability asks a different question: how much can you trust the data collection process itself? Assessment covers three aspects. Data accrual examines how data was entered into the system and whether Standard Operating Procedures (SOPs) existed at the primary collection stage. Quality assurance requires checks for logical consistency and completeness, along with the ability to verify data against primary medical records. Data integrity means protection against unauthorized changes: without an audit trail — a log of all database operations — data will not be accepted for regulatory review.
The applicant must describe all known limitations of the chosen source in the dossier. This is not a formality: experts will assess whether the company understands the weaknesses of its own data and interprets them correctly.
Research Designs Based on RWD
Recommendation No. 15 divides all studies into two types.
Interventional studies are conducted under Good Clinical Practice (GCP) rules. RWD is used here to form external control groups or to monitor outcomes in pragmatic clinical trials — where patients receive treatment under real-world conditions rather than in a specialized trial setting.
Observational (non-interventional) studies are conducted under Good Pharmacovigilance Practice (GVP) rules. These include cohort studies, case-control studies, and cross-sectional studies. The researcher does not intervene in the treatment process and only records what occurs.
A separately recognized approach is the Synthetic Control Arm. Archival data on patients who did not receive the drug serve as a control group. This is particularly valuable for orphan diseases, where recruiting a control group through randomization is difficult for both ethical and practical reasons.
All RWE studies must have a documented protocol before data collection begins. Retrospective analysis planning after results are obtained is not accepted by regulators.
Two typical limitations must be addressed. Bias covers patient selection errors and informational distortions. Confounding arises when a variable linked to both the treatment and the outcome affects results simultaneously. For correction, researchers use the propensity score — a statistical method for balancing comparison groups across a set of characteristics.
A strict rule worth remembering: RWD/RWE is prohibited for marketing or promotional purposes. Scientific and regulatory use only.
Where RWE Fits in the Drug Lifecycle
At registration, RWE most often justifies benefit for rare diseases or supplements RCT data when classic trials are ethically infeasible. RWE can also become a mandatory post-registration commitment under conditional authorization: a company receives approval but commits to collecting real-world data and submitting it within a defined timeframe.
For label extensions, real-world usage data — including off-label use (prescriptions not covered by the official Summary of Product Characteristics) — can support the addition of a new indication, provided the data source is highly reliable and the study design is sound.
In pharmacovigilance, RWE is already indispensable. Post-Authorization Safety Studies (PASS) and Post-Authorization Efficacy Studies (PAES) are direct obligations of Marketing Authorization Holders (MAHs). Monitoring rare adverse reactions that only emerge after use by tens of thousands of patients is impossible without real-world data.
Where the EAEU Currently Lags
Honesty requires acknowledgment: the regulatory framework is in place, but a technological gap remains.
The EMA has adopted ICH E6(R3), which is built around digital data sources (eSource — electronic primary data entered directly into computer systems without a paper intermediary). The FDA is in the process of implementation. The EAEU, by contrast, is only completing its transition to the previous version, ICH E6(R2), per EEC Council Decision No. 63 dated 01.08.2025, effective March 10, 2026. The R3 version is not expected in the Union before 2027.
Most requirements currently remain recommendations, and mandatory Union-level regulation of observational studies is still being developed. In practice, this means approaches to RWE evaluation may still vary between the competent authorities of different Member States.
The third challenge is infrastructure. Even with rules in place, applying them is difficult when hospital information systems are incompatible with each other and patient registries cover only a small fraction of the real population.
None of this diminishes the value of the adopted recommendations. Companies that begin building RWD competencies now will be better positioned when these rules become binding.
What to Do
Audit your available data sources. For each drug in your portfolio, identify which registries or databases contain real-world usage information. Start with state registries (oncology, cardiology, rare diseases) and insurance databases. The key question to answer early: is usable data actually there?
Seek preliminary consultation from an expert organization. Recommendation No. 15 explicitly provides for this option at the study planning stage. Feedback on design before data collection begins is far cheaper than redoing a completed study after regulatory comments.
Develop SOPs for RWD work. If you plan to use EHR or registry data, you need documented verification and quality control processes. Without them, data will not pass the reliability check during review.
Include RWE in your post-authorization study plan. For drugs with conditional authorization or PASS/PAES obligations, this is already a regulatory requirement. For the rest of your portfolio, early data collection is a strategic asset.
Monitor regulatory developments. The EAEU transition to ICH E6(R3) is expected after 2027. This will change requirements for digital data sources. Companies studying the international standard now will be ready for adaptation ahead of competitors.
Two EEC documents have established the methodological foundation for real-world data in the EAEU — against the backdrop of active market unification and deadline pressure for dossier compliance. For companies already working through the EAEU registration transition, RWE offers an additional tool for justification where traditional data falls short.
Regulatory Framework
1. EEC Council Recommendation dated 18.10.2024 No. 1 «On General Approaches to the Development of Regulation of Medicinal Product Circulation in the EAEU Regarding the Collection, Analysis, and Use of Real-World Clinical Practice Data»
2. EEC Board Recommendation dated 10.06.2025 No. 15 «On the Guideline on General Aspects of Using Real-World Clinical Practice Data in the Framework of Medicinal Product Circulation» (entered into force 10.07.2025)
3. EEC Council Decision dated 03.11.2016 No. 78 «On the Rules for Registration and Expert Review of Medicinal Products for Medical Use» (as amended 2024–2025)
4. EEC Council Decision dated 01.08.2025 No. 63 (Transition to ICH E6(R2) effective 10.03.2026)
5. Agreement on Unified Principles and Rules for the Circulation of Medicinal Products within the EAEU dated 23.12.2014