var _paq = window._paq = window._paq || []; /* tracker methods like "setCustomDimension" should be called before "trackPageView" */ _paq.push(['trackPageView']); _paq.push(['enableLinkTracking']); (function() { var u="https://rwrregs.matomo.cloud/"; _paq.push(['setTrackerUrl', u+'matomo.php']); _paq.push(['setSiteId', '1']); var d=document, g=d.createElement('script'), s=d.getElementsByTagName('script')[0]; g.async=true; g.data-privacy-src='//cdn.matomo.cloud/rwrregs.matomo.cloud/matomo.js'; s.parentNode.insertBefore(g,s); })();

RWR INSIGHTS | Europe’s Emerging Health Data and Innovation Framework — How the AI Act, EHDS, GDPR, HTA Regulation, Pharma Package and Biotech Act Are Reshaping Real-World Research

RWR CONTEXT

For real-world research, the combined effect of the AI Act, EHDS, GDPR, EDPB scientific research guidance, HTA Regulation, Pharma Package and Biotech Act is a major shift from opportunistic use of data toward regulated, transparent and decision-grade evidence generation.

The opportunity is significant. EHDS should improve access to health data. HTA reform should create clearer evidence expectations. The pharmaceutical reform and Biotech Act should increase demand for lifecycle evidence. AI may improve feasibility, phenotyping, signal detection and analytics.

However, the bar for using RWD will continue to rise.

Organisations will need to demonstrate that data are lawful, suitable, representative, secure, interoperable and methodologically fit for purpose. AI-enabled research will need stronger validation and monitoring. HTA-facing evidence will need to address relative effectiveness and patient-relevant outcomes. Regulatory-facing evidence will need stronger protocol discipline, transparency and data-quality justification.

The EDPB draft scientific research Guidelines are particularly important because they clarify the GDPR conditions under which RWR can use personal data for scientific research, including secondary use, broad consent, further processing, safeguards and accountability [2]. This makes them a critical companion to EHDS implementation rather than a separate data-protection issue.

The practical message is that RWE strategy in Europe can no longer sit only within epidemiology, medical affairs or health economics. It needs to be integrated across regulatory, HTA, data privacy, AI governance, clinical development, pharmacovigilance, market access and digital health.

MAY 2026 –The European regulatory and policy landscape for health data, medicines, biotechnology, artificial intelligence (AI), data protection and health technology assessment (HTA) is changing rapidly.

Several major EU frameworks are now beginning to converge:

      • The General Data Protection Regulation (GDPR)
      • The draft EDPB Guidelines on processing personal data for scientific research purposes
      • The European Health Data Space (EHDS) Regulation
      • The EU Artificial Intelligence Act
      • The EU Health Technology Assessment Regulation
      • The reform of EU pharmaceutical legislation
      • The proposed European Biotech Act

At first glance, these may look like separate legal and policy developments. In practice, they are increasingly connected. Together, they are creating a more structured European ecosystem for health data access, scientific research, AI-enabled analysis, real-world evidence (RWE), regulatory decision-making, HTA, lifecycle oversight and health innovation.

For real-world research, the direction of travel is clear: Europe is moving toward wider and more systematic use of health data, but within a more formalised governance environment.

Why This Matters for Real-World Research

Real-world research is no longer sitting at the edge of regulatory and market access strategy.

It is becoming central to how Europe intends to support innovation, assess medicines and technologies, regulate AI-enabled health tools, evaluate relative effectiveness, monitor safety, improve access and strengthen healthcare resilience.

The opportunity is significant. The EHDS should improve discoverability and access to electronic health data for secondary use. The HTA Regulation is creating more structured EU-level clinical assessment processes. The pharmaceutical reform places greater emphasis on access, unmet medical need, supply, and lifecycle management. The Biotech Act aims to strengthen Europe’s biotechnology and biomanufacturing ecosystem. The AI Act sets rules for trustworthy AI, including in high-risk health settings.

At the same time, the EDPB draft scientific research guidelines show that the EU is also clarifying how GDPR applies to research use of personal data. This is particularly important for secondary use of health data, registries, AI-enabled research and real-world evidence generation.

The overall message is that data use will need to be lawful, transparent, secure, proportionate, explainable, interoperable and methodologically robust.

For sponsors, CROs, academic researchers, registry holders, data partners, HTA teams and AI developers, this means RWE strategy will need to become more integrated across regulatory, data protection, clinical development, pharmacovigilance, market access, AI governance and digital health functions.

The Regulatory Layers Are Beginning to Connect

Each framework plays a different role:

      • GDPR remains the foundation for personal data protection. It governs lawful basis, special category health data, transparency, data minimisation, pseudonymisation, anonymisation, international transfers and research safeguards [1].
      • The EDPB draft scientific research guidelines are the GDPR interpretation layer. They clarify how GDPR applies to scientific research processing, including further processing, broad consent, public interest, legitimate interest, transparency, data subject rights, safeguards and controller accountability [2].
      • EHDS is the health-data access layer. It creates a sector-specific framework for access to and secondary use of electronic health data for purposes including research, innovation, policy-making, patient safety, personalised medicine, statistics and regulatory activities [3].
      • The AI Act is the AI governance layer. It sets harmonised rules for AI systems, including requirements for high-risk AI systems that may affect health, safety or fundamental rights [4].
      • The HTA Regulation is the evidence-use layer. It creates EU-level joint clinical assessments and joint scientific consultations, increasing the need for evidence that can support relative effectiveness assessment and future HTA decision-making [5][6].
      • The EU pharmaceutical reform is the medicines lifecycle layer. It is the most significant overhaul of the EU medicines framework in more than two decades and is intended to modernise authorisation, innovation, access and supply-related elements of the EU pharmaceutical system [7][8].
      • The proposed European Biotech Act is the innovation and competitiveness layer. It aims to strengthen the EU biotechnology and biomanufacturing sector by supporting innovation, simplifying regulatory processes, improving competitiveness and maintaining high safety, ethics and sustainability standards [9][10].

The combined effect is a new European operating environment in which data access, evidence generation, AI governance, regulatory assessment, HTA and innovation policy are increasingly interdependent.

GDPR: The Baseline Governance Framework

GDPR continues to be the starting point for any real-world research involving personal data.

For RWR, GDPR matters because most health data used in observational research will either be personal data or require careful assessment to determine whether data are anonymised, pseudonymised or still indirectly identifiable.

The GDPR research framework allows processing for scientific research purposes, but only with appropriate safeguards for the rights and freedoms of data subjects. These safeguards include technical and organisational measures, particularly to support data minimisation [1].

In practice, GDPR affects:

      • Lawful basis for processing
      • Processing of special category health data
      • Transparency information for data subjects
      • Pseudonymisation and anonymisation strategies
      • Data minimisation
      • Data retention
      • Data protection impact assessments
      • Controller / processor arrangements
      • International transfers
      • Secondary use governance

GDPR does not prevent real-world research. However, it means RWR must be governed, documented and justified.

EDPB Scientific Research Guidelines: The GDPR Interpretation Layer

The draft EDPB Guidelines 1/2026 on processing personal data for scientific research purposes should be viewed as a key part of the emerging European research-data framework.

Adopted for public consultation in April 2026, the Guidelines are intended to clarify how GDPR applies to scientific research processing and respond to difficult questions raised by controllers, processors and supervisory authorities [2].

For real-world research, the Guidelines are particularly important because they address several issues that routinely arise in secondary use of health data, registry-based studies, biobanks, AI-enabled research and public-private research collaborations.

These include [2]:

      • What qualifies as scientific research
      • When further processing for research may be compatible with the original purpose
      • How broad consent and dynamic consent may be used
      • When public interest or legitimate interest may support research processing
      • How special category health and genetic data should be handled
      • Transparency expectations for research participants and data subjects
      • Data subject rights in research contexts
      • Controller responsibility and accountability
      • Appropriate safeguards for scientific research processing

The Guidelines also reinforce the importance of using anonymised or pseudonymised data where possible, and using directly identifiable data only where necessary and proportionate. They identify safeguards such as ethical oversight, secure processing environments, privacy-enhancing technologies, publication safeguards and confidentiality arrangements [2].

This is highly relevant to the EHDS because improved access to health data will not remove GDPR accountability. Instead, EHDS access pathways and GDPR research safeguards will need to operate together.

Organisations preparing for EHDS-enabled secondary use should therefore also review the EDPB Guidelines when updating RWE governance, data-access procedures, transparency materials, DPIAs, controller / processor role assessments and data minimisation strategies.

EHDS: From Fragmented Access to Structured Secondary Use

The EHDS is likely to be one of the most important long-term developments for real-world research in Europe.

The Regulation establishes a common European framework for electronic health data and is designed both to improve individuals’ access to and control over their health data, and to enable secondary use of electronic health data for purposes that benefit society [3].

Secondary use purposes include research, innovation, policy-making, health threats preparedness, patient safety, personalised medicine, official statistics and regulatory activities [3].

This is directly relevant to real-world research.

Once implemented, EHDS should support more structured access to health data across Member States, including through Health Data Access Bodies, data permits, data catalogues and secure processing environments.

For RWR teams, this could support:

      • Multinational observational studies
      • Registry-based research
      • Feasibility assessments
      • Natural history studies
      • Post-authorisation safety studies
      • Post-authorisation effectiveness studies
      • Comparative effectiveness research
      • Health services research
      • AI model development and validation
      • Regulatory studies
      • HTA evidence generation

However, EHDS will not simply make data “freely available”. It will create a regulated access framework. Researchers and sponsors should expect stronger expectations around permitted purpose, public interest, data minimisation, secure processing, transparency, data-source metadata and controls on re-identification.

The EDPB Guidelines are important here because they clarify how GDPR research concepts should be understood alongside emerging health-data access frameworks. EHDS may improve access, but GDPR accountability will remain central to lawful and trustworthy research.

AI Act: AI-Enabled RWR Will Need Stronger Governance

AI is becoming increasingly important in real-world research.

AI and machine learning may be used for cohort identification, phenotype development, endpoint extraction, natural language processing, imaging analysis, risk prediction, signal detection, patient stratification, data linkage, synthetic controls and decision-support tools.

The AI Act adds a new layer of governance to this environment.

The Regulation establishes harmonised rules for AI systems in the EU and sets out requirements for high-risk AI systems, including systems that may affect health, safety or fundamental rights [4].

This matters for real-world research because AI-enabled tools used in health research may sit at the intersection of several frameworks at once:

      • GDPR for personal data processing
      • EDPB scientific research interpretation for research safeguards
      • EHDS for access to electronic health data
      • AI Act for model governance and high-risk AI requirements
      • MDR / IVDR where the AI system is a medical device
      • HTA expectations where AI-enabled evidence supports reimbursement
      • Pharmaceutical regulation where AI-enabled evidence supports regulatory decision-making

For RWR teams, the practical impact is likely to include greater scrutiny of:

      • Training data provenance
      • Data quality and representativeness
      • Bias and fairness
      • Model validation
      • Transparency and explainability
      • Human oversight
      • Performance monitoring
      • Change control
      • Documentation
      • Accountability for model outputs

AI-enabled RWR will need to be treated as a governed evidence-generation process, not simply an analytical shortcut.

HTA Regulation: RWE Must Address Decision-Relevant Questions

The HTA Regulation creates an EU framework for joint clinical assessments and joint scientific consultations [5][6].

This matters because regulatory approval and HTA decision-making often require different evidence.

Regulators may focus on quality, safety, efficacy, benefit-risk and specific regulatory questions. HTA bodies are more likely to focus on relative effectiveness, appropriate comparators, patient-relevant outcomes, subgroups, durability of effect, treatment sequencing, healthcare-resource impact and uncertainty relevant to reimbursement.

The HTA Regulation does not make RWE mandatory in every assessment. However, it increases the importance of planning evidence that can support both regulatory and HTA needs.

For real-world research, this means RWE may become increasingly important for:

      • Burden of disease
      • Natural history
      • Treatment pathways
      • Current standards of care
      • Comparator selection
      • External control arms
      • Long-term outcomes
      • Subgroup evidence
      • Real-world effectiveness
      • Adherence and persistence
      • Quality of life
      • Healthcare resource utilisation
      • Post-launch evidence generation

The practical implication is that RWE planning should begin earlier in development. Waiting until after authorisation to design real-world studies may be too late if evidence is also needed to support HTA, pricing, reimbursement or managed access.

EU Pharmaceutical Reform: Lifecycle Evidence Becomes More Important

The reform of EU pharmaceutical legislation is the most significant overhaul of the EU medicines framework in more than 20 years [7].

While the final legal details are being implemented, the direction of travel is clear: the reform is intended to modernise the EU medicines system, support innovation, improve access, address unmet medical need, strengthen supply resilience and update the regulatory framework for a changing development environment [7][8].

For RWR, this is important because many of these policy objectives depend on real-world data.

Real-world evidence may support:

      • Unmet medical need assessments
      • Rare disease and orphan medicine development
      • Post-authorisation safety monitoring
      • Post-authorisation effectiveness studies
      • Conditional or exceptional evidence packages
      • Real-world utilisation and access monitoring
      • Supply and shortage impact assessment
      • Treatment pathway analysis
      • Comparative effectiveness
      • Evidence generation for under-represented populations

The pharmaceutical reform should therefore be viewed alongside EMA’s broader RWE activities, including regulatory use of real-world data, DARWIN EU, data-quality expectations, registry-based studies and non-interventional study guidance.

The practical message for sponsors is that RWE is likely to become more important across the entire lifecycle, from early development and scientific advice through authorisation, HTA, pharmacovigilance and post-authorisation evidence generation.

Biotech Act: Innovation Will Need Evidence Infrastructure

The proposed European Biotech Act is less directly focused on real-world research than EHDS or HTA, but it is strategically relevant.

The proposal is intended to strengthen Europe’s biotechnology and biomanufacturing sector, support innovation, simplify regulatory processes, improve competitiveness, facilitate access to finance and maintain high safety, ethics and sustainability standards [9][10].

For RWR, this matters because many biotechnology innovations raise evidence-generation challenges.

These may include:

      • Small patient populations
      • Rare diseases
      • Advanced therapies
      • Personalised medicines
      • Platform technologies
      • Long-term safety monitoring
      • Manufacturing changes
      • Complex endpoints
      • Biomarker-defined populations
      • Real-world follow-up after early access or conditional approval

Biotech innovation will increasingly require evidence infrastructure that can support long-term follow-up, registries, treatment-outcome tracking, manufacturing comparability, safety monitoring and real-world effectiveness.

The Biotech Act therefore connects to EHDS, AI, HTA and pharmaceutical reform by increasing the strategic importance of Europe’s ability to generate decision-grade evidence from health data.

The Key Intersections

The practical impact comes from how these frameworks overlap:

      • Data Access and Data Protection
        • EHDS aims to improve access to health data for secondary use. GDPR sets the baseline safeguards for personal data processing. The EDPB Guidelines clarify how GDPR applies to scientific research processing.
        • This means RWR teams may gain clearer routes to access data, but will also need stronger governance, documentation and safeguards.
        • The key question will not simply be: “Can we access the data?”
        • It will be: “Can we access and use the data lawfully, transparently, securely and proportionately for this specific research purpose?”
      • Scientific Research and Secondary Use
        • The EDPB Guidelines are particularly relevant because they address further processing for scientific research purposes, broad consent, legal bases, safeguards and transparency.
        • This directly intersects with EHDS because much EHDS-enabled research will involve secondary use of existing health data.
        • For RWR, the practical issue will be ensuring that data access under EHDS is supported by GDPR-compliant research governance. That includes data minimisation, pseudonymisation, secure processing environments, transparency, ethical oversight and clear accountability for controllers and processors.
      • AI and Health Data
        • EHDS may improve access to health data for research and innovation. The AI Act governs how AI systems are developed, deployed and monitored. GDPR and the EDPB Guidelines govern the personal-data processing underpinning many AI-enabled research activities.
        • This creates a direct intersection between health data access, scientific research safeguards and AI governance.
        • RWR teams using AI for endpoint extraction, cohort selection, disease progression modelling, safety signal detection or predictive analytics will need to demonstrate that data and models are fit for purpose, validated and appropriately governed.
      • Regulatory Evidence and HTA Evidence
        • The Pharma Package and EMA RWE initiatives support regulatory use of real-world evidence. The HTA Regulation creates more structured EU-level evidence expectations for relative clinical assessment.
        • This means RWE strategies need to be designed for multiple decision-makers.
        • A study designed only for regulatory purposes may not answer HTA questions. A study designed only for HTA may not meet regulatory expectations for data quality, bias control or protocol discipline.
        • Integrated evidence planning will become increasingly important.
      • Innovation and Lifecycle Oversight
        • The Biotech Act and Pharma Package are designed to support innovation. EHDS, GDPR, EDPB guidance, the AI Act and HTA Regulation create governance and evidence expectations.
        • This is not a contradiction. It is the emerging European model.
        • Innovation is being supported, but increasingly within a lifecycle framework requiring transparency, evidence generation, safety monitoring, data governance and post-market evaluation.
      • Fragmented Data and European-Scale Evidence Infrastructure
        • EHDS, HTA, AI governance, EU data catalogues and EMA RWE initiatives all point toward more structured, interoperable and reusable evidence systems.
        • This will increase the importance of:
            • Metadata quality
            • Common data models
            • Data catalogues
            • Federated analytics
            • Secure processing environments
            • Data-source fitness-for-purpose assessments
            • Study registration
            • Protocol transparency
            • Reproducibility
            • Auditability

RWE will need to become more standardised without losing the flexibility required for different research questions.

What This Means in Practice

For organisations conducting real-world research in Europe, the combined impact is substantial.

Key practical implications include:

      • RWE strategy should be developed earlier in the product lifecycle
      • Data access planning should account for EHDS and national implementation pathways
      • GDPR governance remains essential and should be integrated with EHDS readiness
      • The EDPB Guidelines should be reviewed when updating scientific research governance
      • AI-enabled research tools will need model governance, validation and documentation
      • HTA evidence needs should be considered alongside regulatory evidence needs
      • Registries and data partners should prepare for stronger metadata and interoperability expectations
      • Sponsors should strengthen data quality and fitness-for-purpose assessment processes
      • Protocols should be more transparent about data sources, bias, limitations and analytical assumptions
      • Multinational RWD studies should prepare for more federated and secure processing models
      • Long-term follow-up and lifecycle evidence will become more important, particularly for rare disease, ATMPs, biotech products and AI-enabled interventions

What Should Sponsors and CROs Do Now?

The immediate priority is not to treat these developments as separate compliance projects.

Organisations should build an integrated European evidence strategy covering:

      • Data Governance: GDPR, EDPB scientific research guidance, EHDS, transparency, consent / lawful basis, data minimisation, secure processing and cross-border access.
      • Evidence Planning: Regulatory, HTA, pharmacovigilance, market access and post-authorisation evidence needs.
      • Data Quality: Fitness-for-purpose assessment, metadata, completeness, representativeness, provenance and linkage capability.
      • AI Governance: Model purpose, training data, bias, validation, explainability, human oversight and performance monitoring.
      • Lifecycle Monitoring: Long-term safety, real-world effectiveness, utilisation, shortages, access and patient outcomes.
      • Operational Readiness: SOPs, templates, data partner contracts, study registration, data access applications, secure analytics environments and governance boards.

This is not just a legal exercise. It is an operating model change.

Real-World Research Context

For real-world research, the combined effect of the AI Act, EHDS, GDPR, EDPB scientific research guidance, HTA Regulation, Pharma Package and Biotech Act is a major shift from opportunistic use of data toward regulated, transparent and decision-grade evidence generation.

The opportunity is significant. EHDS should improve access to health data. HTA reform should create clearer evidence expectations. The pharmaceutical reform and Biotech Act should increase demand for lifecycle evidence. AI may improve feasibility, phenotyping, signal detection and analytics.

However, the bar for using RWD will continue to rise.

Organisations will need to demonstrate that data are lawful, suitable, representative, secure, interoperable and methodologically fit for purpose. AI-enabled research will need stronger validation and monitoring. HTA-facing evidence will need to address relative effectiveness and patient-relevant outcomes. Regulatory-facing evidence will need stronger protocol discipline, transparency and data-quality justification.

The EDPB draft scientific research Guidelines are particularly important because they clarify the GDPR conditions under which RWR can use personal data for scientific research, including secondary use, broad consent, further processing, safeguards and accountability [2]. This makes them a critical companion to EHDS implementation rather than a separate data-protection issue.

The practical message is that RWE strategy in Europe can no longer sit only within epidemiology, medical affairs or health economics. It needs to be integrated across regulatory, HTA, data privacy, AI governance, clinical development, pharmacovigilance, market access and digital health.

Conclusion

Europe is building a more connected health data and innovation ecosystem.

The AI Act, EHDS, GDPR, EDPB scientific research guidance, HTA Regulation, pharmaceutical reform and proposed Biotech Act do not all regulate the same thing. However, they increasingly intersect around the same strategic objective: enabling health innovation and data use while maintaining trust, safety, transparency and evidence quality.

For real-world research, this is a defining moment.

RWE is becoming more important, but also more regulated. Data access may improve, but governance expectations will increase. AI may create new analytical opportunities, but model accountability will become more important. HTA and regulatory evidence needs will increasingly need to be planned together.

The direction of travel is clear: Future real-world research in Europe will need to be EHDS-ready, GDPR-compliant, EDPB-aware, AI Act-aware, HTA-relevant, regulator-ready and capable of supporting lifecycle evidence generation across the full product pathway.

References

1. European Parliament and Council – Regulation (EU) 2016/679: General Data Protection Regulation (GDPR).
Link: https://eur-lex.europa.eu/eli/reg/2016/679/oj/eng

2. European Data Protection Board – Guidelines 1/2026 on processing of personal data for scientific research purposes. Adopted on 15 April 2026; adopted version for public consultation.
Link: https://www.edpb.europa.eu/system/files/2026-04/edpb_guidelines_202601_scientificresearch_en.pdf

3. European Parliament and Council – Regulation (EU) 2025/327 on the European Health Data Space.
Link: https://eur-lex.europa.eu/eli/reg/2025/327/oj/eng

4. European Parliament and Council – Regulation (EU) 2024/1689 laying down harmonised rules on artificial intelligence (Artificial Intelligence Act).
Link: https://eur-lex.europa.eu/eli/reg/2024/1689/oj/eng

5. European Parliament and Council – Regulation (EU) 2021/2282 on health technology assessment.
Link: https://eur-lex.europa.eu/eli/reg/2021/2282/oj/eng

6. European Commission – Joint Scientific Consultations under the EU HTA Regulation.
Link: https://health.ec.europa.eu/health-technology-assessment/implementation-regulation-health-technology-assessment/joint-scientific-consultations_en

7. European Medicines Agency – Reform of the EU pharmaceutical legislation.
Link: https://www.ema.europa.eu/en/about-us/what-we-do/reform-eu-pharmaceutical-legislation

8. Council of the European Union – The pharma package: new EU rules on medicines.
Link: https://www.consilium.europa.eu/en/policies/pharma-pack/

9. European Commission – Biotechnology: European Biotech Act and supporting documents.
Link: https://health.ec.europa.eu/biotechnology_en

10. European Parliamentary Research Service – European Biotech Act: EU legislation in progress. April 2026.
Link: https://www.europarl.europa.eu/RegData/etudes/BRIE/2026/785708/EPRS_BRI(2026)785708_EN.pdf

RWR INSIGHTS | Europe’s Emerging Health Data and Innovation Framework — How the AI Act, EHDS, GDPR, HTA Regulation, Pharma Package and Biotech Act Are Reshaping Real-World Research2026-06-12T11:53:06+00:00

EU | Cybersecurity and Medical Device Rules Converge — What NIS2 Means for MDR, IVDR, AI Act, EHDS, and Real-World Research

Please login to view this page.

EU | Cybersecurity and Medical Device Rules Converge — What NIS2 Means for MDR, IVDR, AI Act, EHDS, and Real-World Research2025-11-09T10:51:38+00:00

RWR Insights | What’s in a name? The humble (confusing) non-interventional study and the anti-definition

RWR CONTEXT

The FDA definition for non-interventional studies is intuitively actionable. We, as researchers, can (confidently) classify our study based on the parameters provided, which in turn allows us to identify the applicable regulations and guidelines and build our study playbook to describe what we need to do (regulatory requirements) and what we should consider (regulatory considerations) depending on the intended use of the RWE we generate.

4 MARCH 2024 – In his latest Guest Column, Stuart McCully (Real-World Research Ltd) discusses the differing definitions of non-interventional studies in Europe and the US, and the need for clarity when discussing these studies in real-world evidence (RWE) generation.

Read the full article HERE.

RWR Insights | What’s in a name? The humble (confusing) non-interventional study and the anti-definition2024-03-10T16:14:33+00:00

RWE 201 | A Landscape Analysis of Regional RWE Frameworks – The European Health Data Space and DARWIN-EU

RWR CONTEXT

The European Union (EU) is making significant strides in advancing healthcare through a series of interconnected initiatives. The EU4Health Program aims to bolster health systems, focusing on crisis preparedness and disease prevention. The European Health Data Space is an effort to ensure secure access and exchange of health data across EU countries, enhancing healthcare quality and research. Complementing this, the Data Governance Act seeks to foster trust and facilitate data sharing for societal benefits. The European Medicines Agency (EMA)’s 2025 Vision for Real World Evidence (RWE) and the EMA RWE Framework to Support Regulatory Decision Making are pivotal in integrating real-world data (RWD) and RWE into regulatory processes, improving drug development and monitoring. The EU’s Action Plan for RWD & RWE further emphasizes the use of real-world healthcare data in policy and decision-making. Finally, the incorporation of RWD/RWE into the new EU Medicines Regulations marks a significant shift towards evidence-based, data-driven approaches in the pharmaceutical sector, aiming to enhance patient outcomes and healthcare efficiency across the EU.

OCTOBER 2023 – In our real world evidence (RWE) 201 series we have been exploring the regional regulatory and data access frameworks that have been implemented to support access to clinical experience data (real world data | RWD), which is crucial if we are to use this RWD to generate RWE.

Our RWE 201 posts can be accessed here: https://rwr-regs.com/rwe-201/ 

The European Union (EU) is making significant strides in advancing healthcare through a series of interconnected initiatives. The EU4Health Program aims to bolster health systems, focusing on crisis preparedness and disease prevention. The European Health Data Space is an effort to ensure secure access and exchange of health data across EU countries, enhancing healthcare quality and research. Complementing this, the Data Governance Act seeks to foster trust and facilitate data sharing for societal benefits. The European Medicines Agency (EMA)’s 2025 Vision for Real World Evidence (RWE) and the EMA RWE Framework to Support Regulatory Decision Making are pivotal in integrating real-world data (RWD) and RWE into regulatory processes, improving drug development and monitoring. The EU’s Action Plan for RWD & RWE further emphasizes the use of real-world healthcare data in policy and decision-making. Finally, the incorporation of RWD/RWE into the new EU Medicines Regulations marks a significant shift towards evidence-based, data-driven approaches in the pharmaceutical sector, aiming to enhance patient outcomes and healthcare efficiency across the EU.

The following posts provide more details and visual summaries:

  • EU – EU4Health Program [Link]
  • EU – The European Health Data Space [Link]
  • EU – The Data Governance Act [Link]
  • EU – EMA’s 2025 Vision for RWE [Link]
  • EU – EMA RWE Framework to Support Regulatory Decision Making [Link]
  • EU – EU’s Action Plan for Real-World Data (RWD) & RWE [Link]
  • EU – RWD/RWE is Embedded into the New EU Medicines Regulations [Link]
RWE 201 | A Landscape Analysis of Regional RWE Frameworks – The European Health Data Space and DARWIN-EU2023-12-04T15:35:56+00:00

RWR Insights | Regulatory Considerations for Non-Interventional Study Protocols

RWR CONTEXT

Both RWE and clinical trials play critical roles in healthcare research. While clinical trials provide the highest level of evidence for determining a treatment’s efficacy, RWE studies complement this by providing evidence on real-world effectiveness and long-term safety.

Real-world evidence (RWE) study protocols and clinical trial protocols both outline the design and conduct of a study. However, they are distinctly different in several ways given the differences in objectives, methodologies, settings, and populations involved in clinical trials versus RWE studies.

The HARPER framework is a valuable resource for researchers and clinicians who are planning or conducting RWE studies. The framework can help to ensure that protocols are well-designed and will produce high-quality evidence.

Real-world evidence (RWE) study protocols and clinical trial protocols both outline the design and conduct of a study. However, they are distinctly different in several ways given the differences in objectives, methodologies, settings, and populations involved in clinical trials versus RWE studies.

[1] Objectives: The main objective of a clinical trial is to evaluate the efficacy and safety of a medical intervention in a controlled environment, usually by comparing it to a placebo or standard treatment. On the other hand, RWE studies typically aim to understand how an intervention works in routine clinical practice, often focusing on outcomes such as long-term effectiveness, side-effects, quality of life, and cost-effectiveness.

[2] Study Design and Methodology: Clinical trials, especially phase III, are predominantly randomized controlled trials (RCTs) where subjects are randomly assigned to the intervention or control group to minimise bias. They follow a pre-specified protocol and are conducted under tightly controlled conditions. RWE studies, on the other hand, are typically observational in nature and analyse data from sources like electronic health records (EHRs), claims databases, or patient registries.

[3] Setting: Clinical trials are conducted in specific, controlled environments and follow a strict protocol. RWE studies are conducted in routine clinical practice settings, making them more representative of ‘real-world’ conditions.

[4] Population: Clinical trials often have strict inclusion and exclusion criteria, resulting in a relatively homogeneous group of participants. This can limit the generalisability of the results. RWE studies, in contrast, involve broader, more diverse populations (including those often excluded from trials like the elderly, people with multiple co-morbidities, etc.), making the findings more generalisable to everyday practice.

[5] Data Collection: In clinical trials, data collection is rigorous, detailed, and specific to the trial endpoints. Adverse events are actively sought and documented. RWE studies primarily rely on existing data sources such as EHRs, patient registries, or insurance claims data. This can potentially lead to incomplete or inaccurate data.

[6] Intervention: In clinical trials, the intervention (dosage, frequency, duration, etc.) is pre-specified and strictly monitored. In RWE studies, interventions reflect routine clinical practice and may vary widely.

[7] Follow-up: Clinical trials have a defined follow-up period while RWE studies can often provide information on long-term outcomes, given they use data from routine clinical practice over longer periods.

Despite these differences, both RWE and clinical trials play critical roles in healthcare research. While clinical trials provide the highest level of evidence for determining a treatment’s efficacy, RWE studies complement this by providing evidence on real-world effectiveness and long-term safety.

HARPER PROTOCOL TEMPLATE

Regulatory agencies, health technology assessors, and payers are increasingly interested in studies that make use of real-world data to inform regulatory and other policy or clinical decision-making. However, concerns over the credibility of real-world evidence studies have led to calls for more transparency on the design and conduct of RWE studies.

A joint task force between ISPE and ISPOR created a harmonized protocol template for RWE studies that evaluate a treatment effect and are intended to inform decision-making [1]. The HARPER template provides clarity, structure, and a common denominator regarding the level of operational detail, context, and rationale necessary in a protocol.

HARPER = HARmonized Protocol Template to Enhance Reproducibility of hypothesis evaluating real-world evidence studies on treatment effects

Link: https://onlinelibrary.wiley.com/doi/10.1002/pds.5507 

Four protocol templates were identified for RWE studies: 

      1. The European Medicines Agency’s (EMA) Guideline on Good Pharmacovigilance Practices (GVP) Module VIII – post-authorisations safety studies (PASS) template,
      2. ISPE’s guidelines for good pharmacoepidemiology practice (ISPE GPP) section on protocol development, 
      3. The National Evaluation System for health Technology (NEST) protocol guidance, and
      4. The Structured Template and Reporting Tool for Real World Evidence (STaRT-RWE).

The HARPER protocol contains nine sections, including a title page, abstract, and a table for amendments and updates. Each section includes structured free text, a structured table, or a figure, and a free-text section to lay out context and rationale for scientific choices.

The study design diagram shows the context and rationale for the study setting, time 0 (index date), inclusion criteria, exclusion criteria, variables, exposure, outcome, follow up, covariates, sensitivity analyses, data sources, metadata, and software used in the study.

The data sources section includes a free text component followed by a structured table for specifying data sources. The data sources section can also include a detailed evaluation of the fitness-for-purpose of data source options.

Overall, the HARPER framework is a valuable resource for researchers and clinicians who are planning or conducting RWE studies. The framework can help to ensure that protocols are well-designed and will produce high-quality evidence.

References

1. Wang, SV, Pottegård, A, Crown, W, et al. HARmonized Protocol Template to Enhance Reproducibility of hypothesis evaluating real-world evidence studies on treatment effects: A good practices report of a joint ISPE/ISPOR task force. Pharmacoepidemiol Drug Saf. 2023; 32( 1): 44- 55. doi:10.1002/pds.5507

Link: https://onlinelibrary.wiley.com/doi/10.1002/pds.5507 

RWR Insights | Regulatory Considerations for Non-Interventional Study Protocols2023-08-04T13:06:33+00:00

RWR Insights | GDPR and the Secondary Use of Existing Data

RWR CONTEXT

GDPR is a facilitator of the secondary use of large healthcare data in the EU.  However, there are currently limitations and challenges at a national level due to differences in the interpretation, for example, of the requirements for explicit consent.

Further work is needed on issues regarding compatible processing of RWD (secondary use of existing data) in the absence of consent or where data were gathered to form a patient record (e.g., processing compatible with the original purpose).

Secondary use of health data is the processing of health data for purposes other than the initial purposes for which the data were collected. This approach is becoming increasingly popular in real-world research (research that uses real world data to generate real world evidence) because of the large amounts of data that are available through various sources, (e.g., electronic health records, administrative databases, and social media), and the availability of AI-powered analytical tools [1].

In many cases, secondary data analysis can provide valuable insights and answer research questions that would otherwise be difficult or impossible to answer with primary data collection. For example, researchers can use existing data to study disease trends, evaluate the effectiveness of health interventions, and identify risk factors for various health outcomes.

In the world of clinical research we often refer to secondary data as ‘real world data (RWD)’ to distinguish it from data generated through clinical trials.  As per FDA guidance, real-world data are data relating to patient health status and/or the delivery of health care routinely collected from a variety of sources. Examples of RWD include data derived from electronic health records, medical claims data, data from product or disease registries, and data gathered from other sources (such as digital health technologies) that can inform on health status [2].

Traditionally, existing healthcare data are collected from medical records and processed to provide insights in to the safety and effectiveness of drugs etc. In Europe, we call these types of studies retrospective non-interventional studies.  These are protocol-defined studies that require local regulatory approvals and can only collect data that was collected before the start of the study.  This self-limits the usefulness of the research, especially given that the healthcare data will continue to be generated.  For these reason, the emphasis is moving from ‘retrospective’ to ‘secondary use of existing data’ which can be both retrospective and prospective.  The research is still non-interventional (or observational) because there are no healthcare interventions that impact the clinical management of the patient.

Under the General Data Protection Regulation (GDPR – Regulation EU/2016/676), personal data (especially health and genetic data – Article 9) must be collected and processed lawfully, fairly, and transparently, and individuals have the right to be informed about how their data is being used. This means that researchers should obtain explicit and informed consent from individuals to use their personal data for research purposes and the data should be pseudonymized or anonymized to protect individuals’ privacy [3].  

The requirement for explicit informed consent from each individual can become problematic when the intention is to process (analyse) very large healthcare datasets, such as electronic health records in the context of scientific research.  This is where GDPR becomes a facilitator, rather than the hindrance it was thought it would be when it was first implemented.

One of the key ways that the GDPR supports the secondary use of health data for research is through the concept of “legitimate interests”. Article 6(1)(f) of the GDPR allows for the processing of personal data if it is necessary for the legitimate interests of the data controller or a third party, provided that those interests do not override the fundamental rights and freedoms of the data subject. Scientific research can be considered a legitimate interest, provided that appropriate safeguards are in place to protect individuals’ rights and freedoms.  In addition, the GDPR includes provisions that specifically address the use of health data for scientific research. For example, Article 9(2)(j) allows for the processing of special categories of personal data, such as health data, for scientific research purposes, provided that appropriate safeguards are in place.  Whereas, Article 89(1) provides for further processing of existing data for scientific research when appropriate safeguards such as pseudonymisation, no longer permits the identification of data subjects [3].

GDPR indicates that personal data should be gathered for an identifiable purpose or purposes and not further processed for incompatible purposes. Therefore, processing for purposes that are compatible with the purpose of the original gathering and processing of the data are permitted. In addition, the GDPR goes further to indicate that further processing for research purposes are compatible with the original purpose. In the case of the GDPR, this is very positive for RWD processing. However, it is not without difficulties (Section 4.4.4 of the draft CIOMS Real-World Data and Real-World Evidence in Regulatory Decision Making)[4]. 

Currently, in the context of scientific research, GDPR (especially Article 89(1) is interpreted and implemented differently at the national level.  As per recent European Commission reports, more harmonisation of the implementation of GDPR is required at the national Member State level [5] [6].  

See visual example below.

This is particularly relevant to the proposed European Health Data Space and the creation of a federated network of health data hubs that will facilitate access to secondary health data, especially for research purposes (HealthData@EU) [7].

As per the recent draft CIOMS report, there is a strong argument that the processing of RWD only works where data subjects have trust and confidence in the institutions and individuals who process data that relate to them, and therefore a strong personal data protection regime is essential to the acceptance and operation of RWD processing. As noted above, further work is needed on issues regarding compatible processing of RWD (secondary use of existing data) in the absence of consent or where data were gathered to form a patient record (Chapter 5 of the draft CIOMS Real-World Data and Real-World Evidence in Regulatory Decision Making)[4].

References

1. World Health Organisation (WHO) – Meeting on Secondary Use of Health Data (13 December 2022)

Link: https://www.who.int/europe/news-room/events/item/2022/12/13/default-calendar/meeting-on-secondary-use-of-health-data

2. FDA – Real-World Evidence

Link: https://www.fda.gov/science-research/science-and-research-special-topics/real-world-evidence 

3. Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation)

Link: https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A02016R0679-20160504&qid=1687865819117

4. Real-World Data and Real-World Evidence in Regulatory Decision Making. CIOMS Working Group report. Geneva, Switzerland: Council for International Organizations of Medical Sciences (CIOMS), 2023

Link: https://cioms.ch/wp-content/uploads/2020/03/CIOMS-WG-XIII_6June2023_Draft-report-for-comment-1.pdf 

5. European Commission, Consumers, Health, Agriculture and Food Executive Agency, Hansen, J., Wilson, P., Verhoeven, E., et al., Assessment of the EU Member States’ rules on health data in the light of GDPR, Publications Office, 2021

Link: https://data.europa.eu/doi/10.2818/546193 

6. Study on the appropriate safeguards under Article 89(1) GDPR for the processing of personal data for scientific research  – Final Report  – EDPS/2019/02-08 (August 2021)

Link: https://edpb.europa.eu/system/files/2022-01/legalstudy_on_the_appropriate_safeguards_89.1.pdf 

7. European Commission – Proposal for a regulation – The European Health Data Space (May 2022)

Link: https://health.ec.europa.eu/publications/proposal-regulation-european-health-data-space_en  

5. Astellas – U.S. Food and Drug Administration Expands Indication for PROGRAF® for Prevention of Organ Rejection in Adult and Pediatric Lung Transplant Recipients (20 July 2021)
Link: https://newsroom.astellas.us/2021-07-20-U-S-Food-and-Drug-Administration-Expands-Indication-for-PROGRAF-R-for-Prevention-of-Organ-Rejection-in-Adult-and-Pediatric-Lung-Transplant-Recipients?_ga=2.73980498.1553566477.1627827053-1302835671.1627827053

RWR Insights | GDPR and the Secondary Use of Existing Data2023-06-29T09:38:22+00:00

RWR Insight | The Difference Between De-Identified and Pseudo-Anonymised Data

RWR CONTEXT

A tangible example of how real world evidence (RWE) can be used to support label extensions for existing drugs.

Note the FDA’s emphasis on:

“This approval reflects how a well-designed, non-interventional study relying on fit-for-purpose real-world data (RWD), when compared with a suitable control, can be considered adequate and well-controlled under FDA regulations”

Hopefully, we will see similar approvals in Europe and the rest of the World

16 JULY 2021 – Today, the U.S. Food and Drug Administration (FDA)[1] approved a new use for Prograf[2] (tacrolimus) based on a non-interventional (observational) study providing real-world evidence (RWE)[3] of effectiveness. FDA approved Prograf[2] for use in combination with other immunosuppressant drugs to prevent organ rejection in adult and pediatric patients receiving lung transplantation[1].

Prograf[2], originally approved to prevent organ rejection in patients receiving liver transplants, was later approved to prevent organ rejection for kidney and heart transplants as well. The drug has also been routinely used in clinical practice for patients receiving lung transplants. Today’s action marks the first approval of an immunosuppressant drug to prevent rejection in adults and pediatric patients who receive lung transplants. Prograf[2] is the only approved immunosuppressant drug product for this population[1].

This approval reflects how a well-designed, non-interventional study relying on fit-for-purpose real-world data (RWD)[3], when compared with a suitable control, can be considered adequate and well-controlled under FDA regulations. Specifically, the non-interventional study supporting approval for this new indication used RWD from the U.S. Scientific Registry of Transplant Recipients (SRTR)[4], supported by the Department of Health and Human Services. The data were collected on all lung transplants in the U.S. and were supplemented by information from the Social Security Administration’s Death Master File as a trusted repository of mortality data. A dramatic improvement in outcomes was observed among lung transplant patients receiving Prograf[2] as part of their immunosuppression medications compared to the well-documented natural history of a transplanted drug with no or minimal immunosuppressive therapy[1].

In addition to the RWE from the non-interventional study, randomized controlled trials of Prograf[2] used in other solid organ transplant settings provided confirmatory evidence of effectiveness. Additional clinical trial evidence from research publications supports the independent contribution of Prograf[2] as part of a multi-drug immunosuppressive regimen[1].

Prograf[2] should only be prescribed by physicians experienced in immunosuppressive therapy and management of organ transplant and patients receiving the drug should be managed in facilities equipped and staffed with adequate laboratory and supportive medical resources. Prograf[2] is associated with increased risk of developing lymphoma and other malignancies and is associated with increased susceptibility to bacterial, viral, fungal, and protozoal, including opportunistic infections[1].

FDA granted the approval to Astellas Pharma US, Inc[5].

References

1. FDA Approves New Use of Transplant Drug Based on Real-World Evidence (16 July 2021)
Link: https://www.fda.gov/drugs/drug-safety-and-availability/fda-approves-new-use-transplant-drug-based-real-world-evidence

2. PROGRAF (tacrolimus) – Highlights of Prescribing Information
Link: https://www.accessdata.fda.gov/drugsatfda_docs/label/2021/050708s053,050709s045,210115s005lbl.pdf

3. FDA – Real-World Evidence
Link: https://www.fda.gov/science-research/science-and-research-special-topics/real-world-evidence

4. U.S. Scientific Registry of Transplant Recipients (SRTR)
Link: https://srtr.transplant.hrsa.gov/

5. Astellas – U.S. Food and Drug Administration Expands Indication for PROGRAF® for Prevention of Organ Rejection in Adult and Pediatric Lung Transplant Recipients (20 July 2021)
Link: https://newsroom.astellas.us/2021-07-20-U-S-Food-and-Drug-Administration-Expands-Indication-for-PROGRAF-R-for-Prevention-of-Organ-Rejection-in-Adult-and-Pediatric-Lung-Transplant-Recipients?_ga=2.73980498.1553566477.1627827053-1302835671.1627827053

RWR Insight | The Difference Between De-Identified and Pseudo-Anonymised Data2023-06-29T07:51:45+00:00

RWR Insights| Real World Evidence (RWE) 101 – Primary Data vs Secondary Data

RWR CONTEXT

This is the first in our new RWE 101 series in which we explore and explain the fundamentals of real world evidence, specifically the differences, advantages, disadvantages and limitations of primary data versus secondary data.

Primary data and secondary data are two types of data used in research. The main difference between the two is that primary data is collected directly from the source, while secondary data is collected from sources that have already collected the data (i.e., secondary use of existing data).

Primary data is original data that is collected for a specific research project. This type of data can be collected through various methods, including surveys, interviews, observations, and experiments. Primary data is collected with a specific research objective in mind, and the data is usually more focused and targeted than secondary data.

On the other hand, secondary data is data that has already been collected by someone else for a different purpose. This type of data can be collected from a wide variety of sources, including healthcare organisations, government agencies, academic institutions, and commercial organizations. Secondary data can be used to supplement primary data or to answer research questions that are not directly related to the original research objective.

There are advantages and disadvantages to both types of data. Primary data is more likely to be accurate and relevant to the specific research question being studied, but it can also be more time-consuming and expensive to collect. Secondary data is generally less expensive and easier to access, but it may not be as accurate or relevant to the specific research question being studied.

In general, researchers will use a combination of primary and secondary data to address their research questions and achieve their research objectives.

Secondary Use of Existing Data

Secondary use of existing data refers to the practice of analyzing data that was collected for a different purpose than the current research question. This approach is becoming increasingly popular in real-world research because of the large amounts of data that are available through various sources, such as electronic health records, administrative databases, and social media.

In many cases, secondary data analysis can provide valuable insights and answer research questions that would otherwise be difficult or impossible to answer with primary data collection. For example, researchers can use existing data to study disease trends, evaluate the effectiveness of health interventions, and identify risk factors for various health outcomes.

A current, well published example is DARWIN EU®, the Data Analysis and Real-World Interrogation Network, which recently celebrated its first year of establishment. The platform aims to generate real-world evidence (RWE) to support the decision-making of EMA scientific committees and national competent authorities [Link] [1].

DARWIN EU® has initiated its first four studies using real-world data (RWD) from across Europe to better understand diseases, populations and the uses and effects of medicines.  These first four studies start to demonstrate the benefits of DARWIN EU®. The use of a common data model, standardised analytics and agile processes allow faster performance of studies, increased capacity, and lower costs. The design and conduct of these first studies have also supported the establishment of analytical pipelines and processes. The studies were not linked to individual medicines currently under evaluation procedures but selected based on previous procedures and requests for RWE from EMA committees [1].

According to recent DARWIN EU® news [1], the use of a common data model, standardised analytics and agile processes allow faster performance of studies, increased capacity, and lower costs. Additionally, secondary data analysis can allow researchers to study topics that may not have been feasible to study with primary data collection due to ethical or practical limitations.

Some of the challenges associated with secondary use of existing data in the context of electronic health data in the EU, relate to determining the regulatory requirements for data access in the country of interest e.g., GDPR compliance + health research regulation compliance.  We’ll explore this more later in the year and provide you examples and use cases.

There are also potential limitations to secondary data analysis, such as the lack of control over the quality and accuracy of the data, and the potential for biases and confounding factors that were not accounted for in the original data collection. Therefore, researchers must carefully evaluate the suitability of existing data for their research question and take steps to address any limitations or potential biases in the data.

References

1. European Medicines Agency – DARWIN EU® has completed its first studies and is calling for new data partners (28 March 2023)

Link: https://www.ema.europa.eu/en/news/darwin-eur-has-completed-its-first-studies-calling-new-data-partners

RWR Insights| Real World Evidence (RWE) 101 – Primary Data vs Secondary Data2023-06-29T07:51:16+00:00

RWR Insights | USA – Lessons Learned from FDA Reviews of External Control Arms

RWR CONTEXT

In February 2023, the FDA published a draft guidance document on considerations for the design and conduct of externally controlled trials and external control arms.  The aim being to help address and prevent the common mistakes and methodological design flaws that limit the acceptability of external controls for regulatory decision making.

Here we explore some recent examples of external control arms reviewed by the FDA, highlight the limitations identified, and provide recommendations (Lessons Learned) for optimising the acceptability of external controls for regulatory decision making.

FEBRUARY 2023 – On the 1 February 2023 the FDA published draft guidance on ‘Considerations for the Design and Conduct of Externally Controlled Trials for Drug and Biological Products’. We have provided a summary of this new draft FDA guidance earlier in the report…Read More.

Here we explore some recent examples of external control arms reviewed by the FDA, highlight the limitations identified, and provide recommendations (Lessons Learned) for optimising the acceptability of external controls for regulatory decision making.

Based on the 3 case studies below, the FDA identified the following common limitations to the acceptability of external controls for regulatory decision making:

      • Post-hoc analysis (lack of prior FDA review of the protocol and SAP)
      • Selection Criteria Issues (selection bias, misclassification, and confounding)
      • Index Date Issues (immortal time bias)
      • Comparability Issues (lack of comparability, missing data, confounding bias)
      • Limited Sample Size (Lack of statistical power due to insufficient real world cohort size)
      • Real world data not fit for purpose

First, let’s set the foundation and have a quick look at the purpose of  a control group in a clinical trial and the trends and uses of external control arms.

What is the Purpose of a Control Group?

      • Control groups have one major purpose: to allow discrimination of patient outcomes (for example, changes in symptoms, signs, or other morbidity) caused by the test treatment from outcomes caused by other factors, such as the natural progression of the disease, observer or patient expectations, or other treatment (Section 1.2 of ICH E10) [2]. 
      • The control group experience tells us what would have happened to patients if they had not received the test treatment or if they had received a different treatment known to be effective (Section 1.2 of ICH E10) [2]. 
      • In most situations, a concurrent control group is needed because it is not possible to predict outcome with adequate accuracy or certainty (Section 1.2 of ICH E10) [2]. 
      • A concurrent control group is one chosen from the same population as the test group and treated in a defined way as part of the same trial that studies the test treatment, and over the same period of time (Section 1.2 of ICH E10) [2]. 
      • The test and control groups should be similar with regard to all baseline and on-treatment variables that could influence outcome, except for the study treatment (Section 1.2 of ICH E10) [2]. 
      • Failure to achieve this similarity can introduce a bias into the study (Section 1.2 of ICH E10) [2]. 
      • Bias here means the systematic tendency of any aspects of the design, conduct, analysis, and interpretation of the results of clinical trials to make the estimate of a treatment effect deviate from its true value (Section 1.2 of ICH E10) [2]. 
      • Randomization and blinding are the two techniques usually used to minimize the chance of such bias and to ensure that the test treatment and control groups are similar at the start of the study and are treated similarly in the course of the study. Whether a trial design includes these features is a critical determinant of its quality and persuasiveness (Section 1.2 of ICH E10) [2]. 

External Control Arms – Trends and Uses

      • Accelerated approval programs for high morbidity and high unmet need diseases have driven the use of single-arm studies, studies that do not include placebo or active comparator arms (ie, no concurrent control), for drug development (Jaksa et al., 2022) [3]. 
      • External control arms (ECAs) generated from real world data (RWD) are emerging to contextualize single-arm trial data by exploring what would happen if single-arm trial patients did not receive the study drug (Jaksa et al., 2022) [3].
      • Single-arm studies speed up patient access to innovative treatments because they often require fewer patients than randomized controlled trials (RCTs) and use intermediate or surrogate endpoints (e.g., objective response rate [ORR]) (Jaksa et al., 2022) [3].
      • Oncology drug development has been increasingly relying on single-arm studies: from 1992 to 2017, 67% of the Food and Drug Administration’s (FDA) accelerated approvals were based on single-arm trials (Jaksa et al., 2022) [3].
      • Similar trends have been observed in health technology assessment (HTA) submissions. From 2000 to 2016, 22 submissions to the National Institute for Health and Care Excellence (NICE) in the UK were based on non-randomized data, and oncology drugs accounted for more than half of these submissions (Jaksa et al., 2022) [3].

Case Study #1 – Rozlytrek (Entrectinib) for the Treatment of ROS1-Positive, Advanced Non-Small Cell Lung Cancer (NSCLC)

 

FDA (NDA 212725) = Priority Review. Breakthrough Therapy. Orphan Drug Designation. Accelerated Approval (15 August 2019) [4] [5].

This study report reviewed by the FDA contained a comparative analysis of time to treatment discontinuation (TTD), progression free survival (PFS), and overall survival (OS). It compared patients with ROS1-positive NSCLC receiving entrectinib in three single arm clinical trials (ALKA, STARTRK-1, STARTRK-2) and patients with ROS1-positive NSCLC receiving crizotinib in the real world captured by the Flatiron Health Analytic Database [6].

Is the Crizotinib RWE arm sufficient to establish the natural history of disease for ROS1-positive NSCLC?

FDA Review [7]:

      • The crizotinib arm is unlikely to be generalizable to the entire population of patients with ROS1-positive NSCLC
      • The generalizability of the RWE control arm was limited by the low rate of ROS1 testing in clinical practice and resultant sensitivity (estimated as 15%-30%) and the high proportion of community-treated patients in the selected data source.
      • Examination of baseline characteristics demonstrates that the crizotinib arm is not sufficiently comparable to the entrectinib clinical trial population.

FDA Review Conclusions [8]:

      • While the crizotinib population identified may be representative to patients who currently receive treatment for ROS1-positive NSCLC in the community setting, it is not generalizable to the entire ROS1-positive NSCLC population and it is not generalizable to patients enrolled in entrectinib clinical trials. 

Does the study methodology provided allow for a comparison of treatment outcomes between the entrectinib arm and crizotinib arm in this study? 

FDA Review [7]:

      • The study identified substantial differences in time to treatment discontinuation (TTD), progression-free survival (PFS), and overall survival (OS) between study arms, all favoring the entrectinib arm
      • Differentially implemented study eligibility criteria, resultant differences in baseline criteria, and limitations in statistical modeling due to low sample size make it difficult to determine what proportion of the observed differences in rates of clinical outcomes are due to imbalances in study populations at baseline (i.e. selection bias) versus differential treatment effects of the study drugs = This limits comparison of study arms. 
      • Additionally, despite a well-done attempt at defining treatment outcomes, there were limitations. TTD is complicated by treatment beyond disease progression, PFS is limited by missingness in radiographic imaging within electronic medical record data, and OS may be more subject to bias from baseline imbalances

FDA Review Conclusions [8]:

      • This study report is not adequate to allow a robust comparison of treatment outcomes between crizotinib and entrectinib study arms.

List of key external control arm limitations identified by the FDA [9]:

      • Post-hoc analysis (lack of prior FDA review of the protocol and statistical analysis plan)
      • Selection Bias
          • This is the greatest threat to study validity for the comparison of study arms. Substantial differences in baseline covariates were observed. While this is a generally well-done study report, it is unlikely these differences can be overcome with the provided analyses [10]
      • Missing Data Among Covariates and Missing Covariates
          • The Applicant did not try to replicate all the study eligibility criteria in this RWE protocol, likely because data to implement them are missing for many inclusion and exclusion criteria in the crizotinib RWE arm [10]. 
          • It would have been useful for the Applicant to evaluate all eligibility criteria to the extent possible, especially baseline laboratory data [10]. 
          • It is noteworthy that ECOG was missing in 55.1% of patients in the crizotinib arm [10]
      • Statistical Modelling…limited by sample size
      • Measure of Study Outcomes
          • This study report provides a generally acceptable definition of study outcomes given limitations of available data. It does have limitations. Time to Treatment Discontinuation (TTD) is complicated by treatment beyond disease progression, Progression Free Survival (PFS) is limited by lack of radiographic imaging in EMR data, and Overall Survival (OS) may be more subject to bias from baseline imbalances [10].
          • This study report is not adequate to allow a robust comparison of treatment outcomes between crizotinib and entrectinib study arms [8].

 

Impact on Rozlytrek (Entrectinib) Approval = Label limitations. The resulting FDA label of Rozlytrek excluded Time to Treatment Discontinuation (TTD), Progression Free Survival (PFS), and Overall Survival (OS) outcomes, and only referenced improvements in Overall Response Rate (ORR) [11]. 

Case Study #2 – Xpovio (selinexor)  for the Treatment of Adult Patients with Relapsed or Refractory Multiple Myeloma (RRMM)

FDA (NDA 212306)(Xpovio) = Regular Review. Orphan Designation. Accelerated Approval (3 July 2019) [12] [13]

In support of the NDA 212306 for selinexor, the Applicant submitted the results of a phase 2b, open label, single arm clinical trial (STORM) and an external control arm from analyses using retrospectively collected electronic health record (EHR) data (Study KS-50039 ) – below [12]:

Methodological Issues Identified by the FDA [14]: 

      • Selection Criteria Issues (selection bias, misclassification, and confounding)
          • Substantial differences in the inclusion and exclusion criteria for the STORM and the external control (Flatiron Health Analytical Database (FHAD) cohorts) are likely to result in selection bias, misclassification, and confounding. 
          • For example, the Applicant cited real-world overall survival (OS) of patients with penta-exposed, triple-class refractory MM as 3.5–3.7 months; however, patients with less than 4 months life expectancy were excluded from STORM. An exclusion criterion for minimal life expectancy was not implemented for the FHAD population. Differences in selection criteria between the study arms systematically ensure that the STORM cohort will have longer expected OS compared to FHAD cohort  
      • Index Date Issues (immortal time bias)
          • Systematic differences in how the index date was defined may have resulted in biased results (immortal time bias).
          • The definition of the index date has a direct effect on the length of the observed survival time intervals. 
          • The index date to start assessment of overall survival, for both the STORM trial and FHAD, was the date upon which a patient failed his or her last treatment. Using this index date, some FHAD patients could have exhausted all treatment options and could not be indexed at their next treatment (FDA note that 27/64 FHAD patients had no subsequent treatment and should have been excluded from the study). 
          • However, in STORM, all patients must survive until randomization (initiation of selinexor) by design. Thus, person-time between failure of the prior therapy and randomization is “immortal” by design in STORM. It is unknown how many months of immortal time this represents on average. 
      • Comparability Issues (lack of comparability, missing data, confounding bias)
          • oIn addition to difference in inclusion and exclusion criteria, additional factors result in a lack of comparability between the FHAD and STORM cohorts. 
          • RWD analysis compares patients in STORM, who are sufficiently healthy to enroll in a clinical trial, versus patients in FHAD who may or may not receive additional therapy. 
          • Patients who have failed their current treatment but do not receive another treatment are likely have a lower expectation for overall survival. 
          • oPatients who would likely have been more similar to the STORM cohort were explicitly excluded from the FHAD cohort.
          • FHAD had different prior treatment histories, and differential distributions of ECOG scores and missing data. 
          • Imbalances between treatment groups were not adequately accounted for in the design or analysis phases, which likely resulted in confounding bias, primarily favoring survival for the STORM cohort. 

FDA Review Conclusions [15]:

      • To enhance transparency and facilitate evaluation of validity, FDA requires submission of study protocols and statistical analysis plans (SAP) prior to study initiation. Pre-specification of study protocols and SAPs can preclude unplanned multiple testing and analyses, which may inflate Type I error probability and lead to spurious or un-reproducible findings. In support of NDA 212306 for selinexor, the Applicant submitted analyses using retrospectively collected electronic health record (EHR) data. However, neither the protocol or SAP for the selinexor RWD analysis was submitted to FDA prior to the conduct of the study. FDA was made aware of Study KS-50039 upon receiving the final study report on October 6, 2018 [15]. 
      • Given the methodological limitations discussed above, we conclude that the evidence generated from the RWD analysis is not adequate to provide context or comparison for the overall survival observed in the STORM patients. This conclusion is based on the lack of comparability between the STORM and FHAD treatment groups. Furthermore, FDA’s analysis finds that post-hoc strategies to create greater comparability across cohorts were inadequate and resulted in very limited sample size and unstable estimates [15].
      • Due to major methodological issues (including immortal time bias, selection bias, misclassification, confounding, and missing data), the FDA does not consider these results adequate to support regulatory decision making [15].

Impact on Xpovio (selinexor) Approval = The Oncology Drug Advisory Committee (ODAC) members voted in favor of delaying approval until results of the randomized phase 3 BOSTON trial are available [16].

Case Study #3 – Abecma (idecabtagene vicleucel/ide-cel) for the Treatment of Multiple Myeloma

FDA (BLA 125736)(Abecma) = Priority Review. Orphan Designation. Breakthrough Therapy Designation. Regular Approval (26 March 2021) [17] [18].

Idecabtagene vicleucel was approved for fourth-line relapse or refractory multiple myeloma by the FDA in 2021. The clinical evidence included a phase 2b, single-arm, multi-centre clinical trial (MM-001) and an external control arm (ECA) generated from real world data across multiple sources including clinical sites, registries, and a research database (Real world evidence study NDS-MM-003) . The goal of the external control arm was to provide an estimate of overall remission rate (ORR) in patients receiving at least 3 previous myeloma regimens [3] [19].

Study MM-001 – Enrolled 140 subjects and 127 were infused with conformal ide-cel [19]. 

Study NDS-MM-003 (Retrospective Observation Study using Real-World Data) – A global non-interventional retrospective study (NDS-MM-003) to compare the outcome of MM-001 study with a real-world cohort of relapsed and refractory myeloma patients treated with standard therapies. Patient level data from clinical sites, registries and research database was collated into a single data model using data cut off of 30 October 2019. Subjects in the eligible RRMM cohort received approximately 90 different treatment regimens predominantly as a combination of 3 or more drug regimens [20]. 

FDA Review Comments to the External Control Arm (Study NDS-MM-003) [20]:

      • The Agency communicated concerns about the real-world evidence (RWE) study (NDS-MM-003) which was being conducted to provide an indirect comparison of effectiveness of bb2121.
      • Issues with the RWE study include selection of a population which may not be comparable to subjects enrolled in Study MM-001 due to missing baseline patient characteristics, missing or absent data on efficacy assessments which may bias the outcomes and heterogeneity of real world data from different databases that will be collated for analysis. 
          • Selection criteria issues
          • Missing data
          • Real world data not fit for purpose
      • Given the methodological limitations discussed above, we conclude that the evidence generated from the RW analysis is not adequate to provide context or comparison for the outcome of MM-001 study.
          • Significant amount of missing data
          • Differences in follow up and response assessment of subjects from these different sources may impact the interpretability of the study results. 
          • Significant heterogeneity in the RWE population limits its utility as a control arm.
          • The follow up schedule for response assessment in RWE and clinical trial myeloma patients may be different. This can result in potential bias in the estimate of duration of response. 
          • Different response assessments for clinical trial vs RWE patients
          • The efficacy results from the RWE study population are uninterpretable as compared to efficacy evaluable population determined by the Agency (N=100) and based on FDA adjudicated efficacy results.
      • While it reiterates the challenges of an appropriate choice of a treatment in the control arm and supports the approach of considerations for a single arm study design in support of a primary study intended for marketing purposes, an alternative approach may be to consider a randomized controlled trial with investigator’s choice of treatment from pre-specified therapeutic options as the control arm.

FDA Review Conclusions for Abecma [20]:

      • Given the methodological limitations discussed above, we conclude that the evidence generated from the external control arm (Study NDS-MM-003) is not adequate to provide context or comparison for the outcome of MM-001 study. 

Conclusions and Recommendations

The FDA identified the following common limitations in the external control arms of the 3 case studies above:

      • Post-hoc analysis (lack of prior FDA review of the protocol and SAP)
      • Selection Criteria Issues (selection bias, misclassification, and confounding)
      • Index Date Issues (immortal time bias)
      • Comparability Issues (lack of comparability, missing data, confounding bias)
      • Limited Sample Size (Lack of statistical power due to insufficient real world cohort size)
      • Real world data not fit for purpose

Recommendations for optimising the acceptability of external controls for regulatory decision making, include:

      • Apply the guidance provided in the new draft FDA guidance on the design and conduct of externally controlled trials [1]
      • Design the external control to emulate the preferred randomised controlled trial – use a “target trial approach”
      • Use appropriate Real world data sources to optimise comparability, relevance and generalizability of the control group population
      • Seek early FDA review of the protocol and statistical analysis plan (a priori rather than post hoc)
      • Assess and limit bias using the appropriate methodological/ statistical designs
      • Use the START-RWE templates to demonstrate the what, how and why of the data sources, curation and analysis [21] [22]
      • List the limitations of the real world data/ real world evidence in the protocol

References

1. FDA Draft Guidance – Considerations for the Design and Conduct of Externally Controlled Trials for Drug and Biological Products (February 2023)

Link: https://www.fda.gov/regulatory-information/search-fda-guidance-documents/considerations-design-and-conduct-externally-controlled-trials-drug-and-biological-products 

 

2. ICH E10 – Choice of Control Group in Clinical Trials (July 2000)

Link: https://database.ich.org/sites/default/files/E10_Guideline.pdf 

3. Ashley Jaksa, Anthony Louder, Christina Maksymiuk, Gerard T. Vondeling, Laura Martin, Nicolle Gatto, Eric Richards, Antoine Yver, Mats Rosenlund. A Comparison of Seven Oncology External Control Arm Case Studies: Critiques From Regulatory and Health Technology Assessment Agencies. Value in Health, 2022. ISSN 1098-3015, doi.org/10.1016/j.jval.2022.05.016. (25 June 2022)

Link: https://www.valueinhealthjournal.com/article/S1098-3015(22)02004-6/fulltext?_returnURL=https%3A%2F%2Flinkinghub.elsevier.com%2Fretrieve%2Fpii%2FS1098301522020046%3Fshowall%3Dtrue  

4. FDA approves entrectinib for NTRK solid tumors and ROS-1 NSCLC (August 2019) 

Link: https://www.fda.gov/drugs/resources-information-approved-drugs/fda-approves-entrectinib-ntrk-solid-tumors-and-ros-1-nsclc  

5. Drugs@FDA: FDA-Approved Drugs – Rozlytrek (NDA 212725)

Link: https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=overview.process&ApplNo=212725 

6. Section 3.1 – Study Overview (page 7) – CDER – Review of Study Report No WO40977: Comparative analysis of ROS1-positive locally advanced or metastatic non-small cell lung cancer between patients treated in entrectinib trials and crizotinib treated patients from real world data (11 July 2019)

Link: https://www.accessdata.fda.gov/drugsatfda_docs/nda/2019/212725Orig1s000,%20212726Orig1s000OtherR.pdf  

7. CDER – Review of Study Report No WO40977: Comparative analysis of ROS1-positive locally advanced or metastatic non-small cell lung cancer between patients treated in entrectinib trials and crizotinib treated patients from real world data (11 July 2019)

Link: https://www.accessdata.fda.gov/drugsatfda_docs/nda/2019/212725Orig1s000,%20212726Orig1s000OtherR.pdf  

8. Section 6 – Recommendations (page 28) – CDER – Review of Study Report No WO40977: Comparative analysis of ROS1-positive locally advanced or metastatic non-small cell lung cancer between patients treated in entrectinib trials and crizotinib treated patients from real world data (11 July 2019)

Link: https://www.accessdata.fda.gov/drugsatfda_docs/nda/2019/212725Orig1s000,%20212726Orig1s000OtherR.pdf   

9. Section 8 – Appendix (page 30) – CDER – Review of Study Report No WO40977: Comparative analysis of ROS1-positive locally advanced or metastatic non-small cell lung cancer between patients treated in entrectinib trials and crizotinib treated patients from real world data (11 July 2019)

Link: https://www.accessdata.fda.gov/drugsatfda_docs/nda/2019/212725Orig1s000,%20212726Orig1s000OtherR.pdf  

10. Section 4 – Discussion (page 27) – CDER – Review of Study Report No WO40977: Comparative analysis of ROS1-positive locally advanced or metastatic non-small cell lung cancer between patients treated in entrectinib trials and crizotinib treated patients from real world data (11 July 2019)

Link: https://www.accessdata.fda.gov/drugsatfda_docs/nda/2019/212725Orig1s000,%20212726Orig1s000OtherR.pdf  

11. Learnings from three FDA decisions on ECA submissions in oncology, Aetion (November 2019)

Link: https://aetion.com/evidence-hub/learnings-from-three-fda-decisions-on-eca-submissions-in-oncology/  

12. NDA 212306: Selinexor – Oncologic Drugs Advisory Committee Meeting – Introductory Comments – February 26, 2019

Link: https://www.fda.gov/media/121670/download 

13. Drugs@FDA: FDA-Approved Drugs – Xpovio (NDA 212306)

Link: https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=overview.process&ApplNo=212306 

14. Section 7.2.6 (KS-50039 (Retrospective observational study using real-world data)) (page 84) – FDA – NDA/BLA Multi-disciplinary Review and Evaluation, NDA 212306, XPOVIO® (selinexor) (July 2019) 

Link: https://www.accessdata.fda.gov/drugsatfda_docs/nda/2019/212306Orig1s000MultidisciplineR.pdf 

15. 1.Section 7.2.6 (KS-50039 (Retrospective observational study using real-world data)) (page 84) – FDA – NDA/BLA Multi-disciplinary Review and Evaluation, NDA 212306, XPOVIO® (selinexor) (July 2019) 

Link: https://www.accessdata.fda.gov/drugsatfda_docs/nda/2019/212306Orig1s000MultidisciplineR.pdf 

16. Section 1.2 (Conclusions on the Substantial Evidence of Effectiveness ) (page 16) – FDA – NDA/BLA Multi-disciplinary Review and Evaluation, NDA 212306, XPOVIO® (selinexor) (July 2019) 

Link: https://www.accessdata.fda.gov/drugsatfda_docs/nda/2019/212306Orig1s000MultidisciplineR.pdf 

17. FDA – ABECMA (idecabtagene vicleucel) (21 April 2021)

Link: https://www.fda.gov/vaccines-blood-biologics/abecma-idecabtagene-vicleucel 

18. FDA – Summary Basis for Regulatory Action – Abecma (26 March 2021)

Link: https://www.fda.gov/media/147627/download 

19. FDA – BLA 125736 (Abecma) – Statistical  Review (27 July 2020)

Link: https://www.fda.gov/media/147781/download 

20. Section 9.2 (Aspect(s) of the Clinical Evaluation Not Previously Covered – Study NDS-MM-003 (Retrospective Observation Study using Real-World Data))(page 113) – FDA BLA Clinical Review Memorandum – STN 125736/0 (27 July 2020)

Link: https://www.fda.gov/media/147740/download 

21. Wang S V, Pinheiro S, Hua W, Arlett P, Uyama Y, Berlin J A et al. STaRT-RWE: structured template for planning and reporting on the implementation of real world evidence studies BMJ 2021; 372 :m4856 doi:10.1136/bmj.m4856

Link: https://www.bmj.com/content/372/bmj.m4856 

22. Harvard Dataverse – Structured Template and Reporting Tool for Real World Evidence (STaRT-RWE)

Link: https://dataverse.harvard.edu/dataverse/STaRT-RWE;jsessionid=952c11f48e3021a6ddcd8c9c9822 

RWR Insights | USA – Lessons Learned from FDA Reviews of External Control Arms2023-03-05T12:37:05+00:00

EU | EMA 2025 Vision for the Regulatory Use of RWE – Connecting the Pieces

RWR CONTEXT

In the past 12 months, the EMA has been busy…publishing strategies and frameworks, creating coordination centres, and requesting proposals from training vendors! The question is…how does this all fit together?

Here we provide a a high-level overview of what has happened and what role each of the activities plays in helping to realise the EMA’s 2025 vision for the use of RWE in regulatory decision making.

DECEMBER 2022 – The European Medicines Agency (EMA) has a very ambitious goal for the use of real-world evidence (RWE):

“…by 2025 the use of real-world evidence will have been enabled and the value will have been established across the spectrum of regulatory use cases”

[EMA – A vision for use of real-world evidence in EU medicines regulation (24 November 2021)] [ref 1]

In the past 12 months, the EMA has been busy…publishing strategies and frameworks, creating coordination centres, and requesting proposals from training vendors! The question is…how does this all fit together?  The following image provides a high-level overview of what has happened and what role each of the activities plays in helping to realise the EMA’s 2025 vision for the use of RWE in regulatory decision making…exciting times!

 

DECEMBER 2021 = EMA – Data Standardisation Strategy [ref 2]

      • The European medicines regulatory network’s data standardisation strategy sets out principles to guide the definition, adoption and implementation of international data standards by the network.
      • It aims to:
        • enable quicker uptake of international data standards across the EU
        • improve data quality
        • enable data linkage and data analysis to support medicine regulation

FEBRUARY 2022 = EMA – Data Analysis and Real World Interrogation Network (DARWIN EU®) [ref 3] [ref 4]

      • In February 2022, EMA selected a service provider (Erasmus University Medical Center Rotterdam) to deliver DARWIN EU
      • By 2024, DARWIN EU will be fully operational with involvement of data partners and medicine regulators
      • The vision of DARWIN EU® is to give EMA and national competent authorities in EU Member States access to valid and trustworthy real-world evidence, for example on diseases, patient populations, and the use, safety and effectiveness of medicines, including vaccines, throughout the lifecycle of a medicinal product
      • The role of the Coordination Centre is to develop and manage a network of real-world healthcare data sources across the EU and to conduct scientific studies requested by medicines regulators and, at a later stage, requested by other stakeholders.
      • By supporting decision-making on the development, authorisation and surveillance of medicines, a wide range of stakeholders will benefit, from patients and healthcare professionals to health technology assessment bodies and the pharmaceutical industry. Additionally, DARWIN EU® will provide an invaluable resource to prepare for and respond to future healthcare crises and pandemics.
      • For example, the availability of timely and reliable real-world evidence can lead to innovative medicines becoming more quickly available to patients. Better evidence also supports more informed regulatory decision-making on the safe and effective use by patients of medicines on the market.  

APRIL 2022 = EMA – Big Data Training Tender – Pharmacoepidemiology and Read-world Evidence curriculum 

The Pharmacoepidemiology and Read-world Evidence curriculum (Lot 2) aims at training regulators and scientists for the collection, analysis, interpretation, and use of observational/real-world data for regulatory evaluation and decision-making on medicinal products. Upon completion of the curriculum, training participants should have a good understanding of the pharmacoepidemiological and real-world evidence concepts and methods to develop, analyse and critically review protocols and reports of non-interventional studies.

MAY 2022 = European Commission – European Health Data Space (EHDS) Proposal [ref 5]

      • Proposal for a European Health Data Space published in May 2022
      • Planned to be effective (in force) by 2025
      • Implements a European electronica health record (EHR) exchange format
      • MyHealth@EU = Empowers individuals through increased digital access to and control of their electronic personal health data, nationally and cross-borders, as well as support to their free movement, fostering a genuine single market for electronic health record systems, relevant medical devices and high-risk artificial intelligence (AI) systems (primary use of data)
      • HealthData@EU = Provides a consistent, trustworthy, and efficient set-up for the use of health data for research, innovation, policy-making and regulatory activities (secondary use of data)

JUNE 2022 = EMA – Metadata List Describing Real World Data [ref 6]

A list of metadata describing real-world data sources and studies is available to help pharmaceutical companies and researchers to identify and use such data when investigating the use, safety and effectiveness of medicines.

      • This metadata list will feed into two future EU catalogues on real-world data sources and studies:
        • The catalogue of data sources will cover information on real-world databases, and is due to replace the European Network of Centres for Pharmacoepidemiology and Pharmacovigilance (ENCePP) catalogue in late 2023
        • The catalogue of studies will cover studies performed on the data sources, enhancing and replacing the European Union electronic register of post-authorisation studies (EU PAS Register)

JULY 2022 = EMA – Global Regulators Call for International Collaboration to Integrate Real-World Evidence into Regulatory Decision-Making [ref 7]

      • EMA has endorsed a joint statement calling for international collaboration to enable the generation and use of real-world evidence for regulatory decision-making published today by the International Coalition of Medicines Regulatory Authorities (ICMRA).
      • The use of real-world data and real-world evidence in the development, authorisation and monitoring of medicines to support regulatory decision-making is rapidly increasing. Although real-world evidence can play an important role in bridging knowledge gaps, there are still challenges that need to be addressed, such as heterogeneous data sources across the globe and different levels of quality of the data. Interested parties also need to deal with various processes for data sharing and access.
      • During the COVID-19 pandemic, international medicines regulators and researchers have worked together to establish or reinforce collaboration allowing efficient sharing of data and experience in relation to real-world evidence. They agreed to further such collaboration beyond the pandemic.
      • In their statement, ICMRA members pledge to foster global efforts and further enable the integration of real-world evidence into regulatory decision-making. They identify four focus areas for regulatory cooperation:
        • harmonisation of terminologies for real-world data and real-world evidence
        • regulatory convergence on real-world data and real-world evidence guidance and best practice
        • readiness to address public health challenges and emerging health threats; and
        • transparency

SEPTEMBER 2022 = EMA – A Good Practice Guide for the Use of Real-World Metadata [ref 8] [ref 9]

      • This draft guide aims to help regulators, data holders, researchers, pharmaceutical companies, and other interested stakeholders to use the catalogue of data sources that will replace the currently available ENCePP catalogue.
      • For instance, it provides recommendations on how to identify suitable real-world data sources for studies and describes the required metadata elements.
      • Suggestions for consideration received from industry included:
        • Link and cross-reference the Good Practice Guide with the Data Quality Framework to ensure consistency in the use of concepts (i.e., data quality, data reliability) and terminology
        • Clarify the document’s scope from the geographical (EU vs non-EU) and data source type perspectives
        • Add explanations of the metadata and proposed values for key variables and consider additional metadata such as time lag between the collection and availability of data, more granular detail on laboratory data and inclusion of genomic data sources
        • Consider linking the catalogue to similar initiatives such as EHDEN or EHDS catalogues to harmonise definitions and avoid duplication of effort in providing information to different sources from the data holder point of view

OCTOBER 2022 = EMA – Data Quality Framework for EU Medicine Regulation [ref 10]

      • A draft data quality framework for medicine regulation was available for public consultation until 18 November 2022. 
      • The purpose of this framework was to be applicable for any type of human and veterinary data that might be submitted to medicines regulator. The work will continue next year to apply data quality principles outlined in the framework to specific domains (e.g. real-world, manufacturing etc.) and to ensure alignment with developments coming from the European Health Data Space.
      • This guidance document sets out the criteria for a more consistent and standardised approach to the quality of data used in medicine regulation to support benefit-risk decisions. 
      • RWD quality should be considered in terms of (1) Relevance, (2) Reliability, (3) Extensiveness, (4) Coherence, and (5) Timeliness

NOVEMBER 2022 = EMA – DARWIN EU Data Partners Onboarded [ref 11] [ref 12]

In November 2022, DARWIN EU completed the onboarding of the first set of data partners with access to real-world healthcare data from sources such as hospitals, primary care, health insurance, registries and biobanks.

References

1. EMA – A vision for use of real-world evidence in EU medicines regulation (24 November 2021)

Link: https://www.ema.europa.eu/en/news/vision-use-real-world-evidence-eu-medicines-regulation

2. EMA – Data Standardisation Strategy [17 Dec 2021]

Link: https://www.ema.europa.eu/en/about-us/how-we-work/big-data#data-standardisation-strategy-section  

3. EMA – Initiation of DARWIN EU® Coordination Centre advances integration of real-world evidence into assessment of medicines in the EU

Link: https://www.ema.europa.eu/en/news/initiation-darwin-eur-coordination-centre-advances-integration-real-world-evidence-assessment  

4. EMA – Data Analysis and Real World Interrogation Network (DARWIN EU) 

Link: https://www.ema.europa.eu/en/about-us/how-we-work/big-data/data-analysis-real-world-interrogation-network-darwin-eu   

5. European Commission – Proposal for a regulation – The European Health Data Space (May 2022)

Link: https://health.ec.europa.eu/publications/proposal-regulation-european-health-data-space_en    

6. EMA – Metadata List Describing Real World Data [10 June 2022]

Link: https://www.ema.europa.eu/en/about-us/how-we-work/big-data#metadata-list-describing-real-world-data-section 

7. EMA – Global regulators call for international collaboration to integrate real-world evidence into regulatory decision-making [22 Jul 2022]

Link: https://www.ema.europa.eu/en/news/global-regulators-call-international-collaboration-integrate-real-world-evidence-regulatory-decision  

8. EMA – A Good Practice Guide for the Use of Real-World Metadata [27 Sept 2022]

Link: https://www.ema.europa.eu/en/about-us/how-we-work/big-data#metadata-list-describing-real-world-data-section  

9. EMA Report from the Second bi-annual Big Data Steering Group and industry stakeholders meeting [3 Nov 2022]

Link: https://www.ema.europa.eu/en/events/second-bi-annual-big-data-steering-group-industry-stakeholders-meeting    

10. EMA – Data quality framework for EU medicine regulation [10 Oct 2022]

Link: https://www.ema.europa.eu/en/about-us/how-we-work/big-data    

11. Data Analysis and Real World Interrogation Network (DARWIN EU®) 

Link: https://www.ema.europa.eu/en/about-us/how-we-work/big-data/data-analysis-real-world-interrogation-network-darwin-eu 

12. Data Analysis and Real World Interrogation Network (DARWIN EU®) – Data Partners

Link: https://www.ema.europa.eu/en/about-us/how-we-work/big-data/data-analysis-real-world-interrogation-network-darwin-eu#data-partners-(new)-section  

EU | EMA 2025 Vision for the Regulatory Use of RWE – Connecting the Pieces2023-02-06T12:20:34+00:00
Help

About Us

Contact Us

Our Values

SSO Login

Follow Us

Newsletter

LinkedIn

[…]

Legal Info

Privacy Policy

Terms and Conditions

Copyright 2026 | Real-World Research Limited | All Rights Reserved |

This website uses cookies and third party services. Settings OK Reject

Cookies

We use analytical tools (Google Analytics) to help us improve the user experience. We DO NOT sell or use any of this data for any purpose beyond improving the design and usability of our own website (rwr-regs.com).
Go to Top