EMA’s evolving stance on external controls: Key takeaways and preparation strategies for sponsors

This blog is part of The Regulatory Navigator series, where we explore the evolving regulatory landscape with actionable insight from Parexel's experts, sharing their experience to maximize success for clinical development and patient access.

External control arms (ECAs) continue to emerge as a valuable tool for evidence generation, offering innovative approaches to clinical research and addressing ethical and practical challenges that can arise with traditional randomized controlled trials (RCTs), as discussed in our previous article. Trials with ECAs leverage data from sources outside the investigational study, including real-world cohorts, historical studies, and non-interventional research to provide meaningful context for interpreting the effects of new therapies. 

While EMA has historically taken a cautious stance toward ECAs¹, the agency convened a workshop on November 3, 2025, on the use of external controls for evidence generation in regulatory decision-making.²  

This workshop follows the publication of Concept Paper³ outlining plans for a Reflection Paper on ECAs⁴, expected in 2026, with the aim of providing a structured framework for consistency, transparency, and methodological rigor. These recent initiatives indicate a gradual shift toward broader consideration of ECAs in specific contexts. 

The acceptance of external controls in the centralized marketing authorization applications (MAA), such as those for Abecma (idecabtagene vicleucel)⁵ and several other products for the treatment of hereditary transthyretin (ATTRv) amyloidosis^6,7,8, demonstrate that under the right conditions ECAs approaches can play a meaningful role in regulatory decision-making. 

The value of ECAs  

EMA’s evolving position has important implications for sponsors. ECAs can accelerate timelines and reduce patient burden when randomization is impractical or unethical, which is often the case in rare diseases, advanced therapies, and high-unmet-need oncology indications. They also provide a way to contextualize time-to-event endpoints in single-arm trials, strengthening benefit–risk narratives for pathways such as conditional approval, PRIME, and orphan maintenance. 

During the workshop, participants reinforced that RCTs remain the preferred approach but also acknowledged that external controls may be considered when strong treatment effects make meaningful comparisons possible. Using external data to compensate for inconclusive trial results, however, is generally viewed as inappropriate. ECAs are not a shortcut, but rather a complementary tool that requires careful design and implementation. 

What sponsors need to know 

The forthcoming Reflection Paper⁴ will likely address definitions, minimal requirements, and operational considerations for ECAs, with topics such as bias mitigation, estimands, target trial emulation, and data quality standards featured prominently in the workshop discussions. EMA is signaling that sponsors will need to demonstrate methodological robustness, including pre-specification of ECA design, alignment of endpoints, and transparent statistical analysis plans. 

• Target trial emulation emerged as a cornerstone concept. Sponsors should be prepared to describe the hypothetical randomized trial they aim to emulate, including eligibility criteria, treatment strategies, outcomes, and follow-up. Correctly defining time zero – the point at which follow-up begins – is essential to avoid biases such as immortal time or calendar-time effects. 
• Endpoint alignment is also critical: Differences between real-world progression-free survival (PFS) and trial-based PFS can introduce substantial bias if not addressed through calibration or sensitivity analyses. 
• Data quality is another essential element. EMA’s Data Quality Framework provides a structured approach to assess relevance, reliability, completeness, and timeliness.⁹ Sponsors should anticipate rigorous scrutiny of how external data sources are selected and qualified. Feasibility assessments, including identification of major and minor limitations, will be a key to building confidence in the evidence package. 

Proven ECAs approaches from real-world lessons  

Case studies discussed during the EMA workshop illustrate both the opportunities and challenges of ECAs: 

• In Abecma’s marketing authorization application⁵, EMA accepted a global non-interventional study as an external comparator, acknowledging clinically meaningful benefit despite limitations such as missing data and covariate imbalances.  
• Another example that illustrates EMA’s willingness to consider external comparative evidence, including systematic literature review and real-world cohorts, to contextualize outcomes from single-arm trials is Breyanzi (lisocabtagene maraleucel) case.¹⁰ The approach relied on pre-specified covariates and methodological rigor to minimize bias, rather than serving as a formal pivotal comparator.  
• Libmeldy (atidarsagene autotemcel) showed that in ultra-rare diseases, external controls can provide compelling evidence when effect sizes are large and clinically unambiguous.¹¹ 

These examples highlight a consistent theme – success with external controls depends on early planning, transparency, and methodological rigor. Approaches that are developed after the fact, use poorly aligned endpoints, or lack adequate bias mitigation remain high-risk strategies that can introduce significant uncertainty and are unlikely to inspire confidence during regulatory review. 

Conclusion and next steps  

EMA’s evolving stance on external controls represents an important opportunity for sponsors, but it comes with clear expectations. ECAs are not a substitute for randomized trials, they are a complementary contextualizing approach that requires careful design and execution. Sponsors who act now by embedding ECAs into trial design, qualifying data sources, and applying rigorous analytical methods will be well positioned as guidance formalizes in 2026. 

Consider integrating ECAs into development plans ahead of the Reflection Paper⁴ finalization, where appropriate for your specific context. This approach may include (1) designing experimental arms that can be emulated; (2) qualifying data sources using EMA’s Data Quality Framework⁹; and (3) implementing robust bias mitigation strategies.  

For implementation, a dedicated protocol and statistical analysis plan (SAP) should be prepared for the ECAs. Prespecified analyses should detail the statistical approach for addressing confounding, as well as sensitivity analyses to assess robustness against potential sources of bias, including unobserved confounding and missing data. Early engagement with the CHMP through Scientific Advice is essential to align expectations and reduce risk. 

Parexel brings practical experience in navigating these opportunities and challenges, providing solutions especially where there is yet-to-be defined regulatory clarity. We help sponsors move beyond theory by combining regulatory strategy, advanced biostatistics, and real-world data expertise to create evidence packages that are both credible and defensible, and that withstand regulatory scrutiny. 

If you are planning a single-arm trial or seeking to accelerate evidence generation for a marketing authorization application, reach out to us. Our integrated approach ensures that external controls are not only considered but implemented in a way that meets EMA’s evolving standards. 

Start a conversation

References 

1. Wang et al., Current perspectives for external control arms in oncology clinical trials: Analysis of EMA approvals 2016–2021. J Cancer Policy 2023 Mar:35:100403 doi: 10.1016/j.jcpo.2023.100403. 
2. EMA Workshop on the use of external controls for evidence generation in regulatory decision-making, November 2025 
3. EMA Draft concept paper on the development of a reflection paper on the use of external controls for evidence generation in regulatory decision-making, July 2025 
4. EMA Development of a reflection paper on the use of external controls for evidence generation in regulatory decision-making - Scientific guideline, July 2025 
5. EMA CHMP Abecma EPAR, EMA/409800/20212021, June 2021  
6. EMA CHMP Attrogy EPAR, EMA/161582/2025, April 2025 
7. EMA CHMP Wainzua EPAR, EMA/CHMP/515905/2024, January 2025 
8. EMA CHMP Amvuttra EPAR, EMA/CHMP/689555/2022, July 2022 
9. EMA Data quality framework for medicines regulation, December 2023 
10. EMA CHMP Breyanzi EPAR, EMA/134759/2022, January 2022 
11. EMA CHMP Libmeldy EPAR, EMA/584450/2020, October 2020 

Return to Insights Center