EMA ERA Guidance One Year On and What the EU Pharma Reform Means Next

In our previous article, we set out the key changes in the revised EMA guideline on Environmental Risk Assessment (ERA) for human medicines. When it came into effect in September 2024 (EMA, 2024), it provided a clearer, more detailed framework for evaluating the environmental impact of medicinal products.

At the time, the priority was understanding what had changed. A year on, the focus has shifted: how are assessors applying the new expectations, and what does that tell us about where regulation is heading?

Experience across current procedures, supported by EMA stakeholder discussions during the first year of implementation, shows that an ERA is no longer treated as a largely descriptive annex. It is being reviewed more consistently and with real impact on procedure timelines and outcomes.

This is reinforced by the direction of EU pharmaceutical reform (the “Pharma Package”): political agreement was reached in December 2025, and the compromise texts for a new Directive and Regulation were published in March 2026, pointing to stronger, more enforceable and more lifecycle-focused ERA expectations across the system.

Overall, the first year has made one thing clear: ERA is being treated as a critical regulatory workstream. Phase I decision logic is being checked in detail, exposure refinements need traceable supporting evidence, and gaps, particularly for older substances, are leading to requests for additional studies or well-substantiated justification. In this context, the EU Pharma Package is best seen as formalising what is already happening: clearer lifecycle responsibilities and more tangible consequences where ERA submissions are incomplete or not aligned with current expectations.

 

From clarification to enforcement

The revised guideline sets out a structured approach to developing an ERA, combining exposure assessment, hazard identification, and a tiered risk evaluation. At first, many saw it as a clearer statement of what regulators already expected. In practice, it has changed how the ERA is handled during assessment.

Assessors are more willing to challenge submissions that lean on historic assumptions rather than current, case-specific evidence. Citing legacy ERAs, relying on legal-basis arguments, or presenting exposure-driven justifications in isolation is rarely enough. ERA is increasingly expected to read as a complete, guideline-aligned and scientifically robust part of the dossier.

The shift is subtle but important. An ERA is no longer something that can simply be included to “tick the box”. It must now withstand technical scrutiny.

Across procedures, five themes have come up repeatedly in the first year:

1. An ERA is now a core part of regulatory acceptability: a complete, reworked ERA is expected for all applications (including generics), and references to legacy assessments or legal-basis arguments are not accepted as a substitute for a guideline-aligned submission.
2. Phase I decision logic has become a regulatory control point: assessors are auditing the decision tree itself. Errors in applicability, thresholds or interpretation can trigger rework and a fresh look at whether Phase II is needed.
3. Exposure refinements must be evidence-led: refinements (e.g., prevalence-based approaches) are still possible, but need recent, traceable epidemiological sources, with underlying reports available for review. Commercial proxies (e.g. sales data or prescription volumes) are being rejected.
4. Current scientific standards highlight major gaps for older products: common pressure points include adsorption/mobility, chronic aquatic endpoints, sediment and soil compartments, and bioaccumulation, where older datasets often fall short of today’s requirements.
5. Waiver-based strategies rarely work on their own: arguments such as short-term use or low perceived exposure need to sit within a complete, guideline-compliant ERA, rather than being used to replace missing elements.

 

 

Closer scrutiny of decision logic and exposure assumptions

One of the clearest developments over the past year has been the attention now paid to the “working” behind the ERA conclusions, especially in Phase I. The decision tree for determining whether further assessment is needed; assessors are increasingly checking not just that the framework is used, but that it is applied correctly.

Misread thresholds, inconsistent decision points, or unclear reasoning are being challenged more often. In some cases, applicants are asked to rerun the Phase I decision tree and confirm whether Phase II should have been triggered. Phase I has, in effect, become an auditable control point; errors here can undermine downstream arguments.

A similar shift is playing out in exposure assessment. The guideline is explicit that predicted environmental concentrations must be based on scientifically justified assumptions. EMA has confirmed that commercial data, such as sales or prescription volumes, are not appropriate for refining exposure estimates.

In practice, many established approaches are being revisited. Where refinement is used, it needs to be supported by transparent, traceable data, typically epidemiology, that reflects real-world use (often one-year prevalence), with sources available for review. Exposure arguments still matter, but they are expected to be supported by evidence rather than inferred.

 

Higher scientific expectations and the legacy product challenge

The revised guideline also sharpens expectations for environmental fate, ecotoxicology and bioaccumulation, with clearer direction on study design and how endpoints should be derived.

For new substances, these requirements can be addressed prospectively. For older active substances, this exposes a structural problem: many products authorised under earlier frameworks simply do not have datasets that meet today’s standards, particularly for adsorption, chronic aquatic toxicity, and bioaccumulation.

Commonly identified gaps include adsorption/desorption and mobility (including higher-tier expectations aligned with OECD 106), chronic aquatic studies suitable for NOEC/EC10 derivation, sediment and soil compartment assessments, and bioaccumulation/secondary poisoning lines of evidence.

Literature can be used, but it needs to be supported by a structured search and a robust evaluation of data quality. Assessors have also been clear that each application should contain a complete, compliant ERA, even where earlier assessments exist.

In practice, that shifts the conversation from approach to evidence. For many generics and established products, the issue is less whether the ERA narrative is plausible, and more whether there is enough fit-for-purpose data to demonstrate alignment with current expectations.

 

What this looks like in practice

These trends are already showing up in day-to-day regulatory interactions.

Recent regulatory experience shows how these expectations are being applied in practice, particularly for well established products. Recent procedures show that even well established products can face detailed ERA review, particularly where datasets are incomplete or not aligned with the revised framework. The question is not simply whether an ERA is present, but whether it follows the current decision logic and data expectations. That can translate into highly prescriptive requests for an update.

 

Case example: variation procedure

In one anonymised example involving a variation procedure for a widely used active substance, the regulator concluded that the application was not approvable unless outstanding ERA deficiencies were addressed and requested a comprehensive update aligned with the revised guideline.

The request package included:

• Correction of the Phase I decision tree, including reassessment of key decision points
• Revision of decision-tree responses where applicability thresholds had been misinterpreted
• Submission of supporting evidence for prevalence-based PEC refinement, including recent, Member State-specific data sources
• Provision of missing studies and reports covering physico-chemical properties, environmental fate and chronic ecotoxicity endpoints
• Confirmation that endpoints (e.g., NOEC/EC10) were derived in line with current expectations
• Inclusion of a targeted literature review, even where reference to an existing ERA had been made

 

The applicant’s original position, emphasising short duration of use, low prevalence, no increased environmental burden compared to existing products, and low perceived ecological risk, was not considered sufficient to avoid further assessment.

 

A broader trend for legacy and generic products

Importantly, this example reflects a broader trend: even long established products and generic applications are being reassessed against current ERA expectations, and deficiencies can become a rate limiting step in regulatory procedures. It is increasingly being applied to long established active substances, including generics, where the original ERA was incomplete, based on older methods, or never produced to the standard now expected. Arguments that once carried more weight, such as short duration of use, low prevalence, or limited perceived environmental risk, may still be considered, but they are less likely to succeed on their own and need to be set within a fully guideline-compliant ERA supported by appropriate evidence.

 

Escalation during procedures

Another feature is how quickly ERA questions can escalate during a procedure. What starts as a clarification request can become a requirement to update the assessment or generate additional data. In some cases, the ERA issues become critical path items, holding up progress until they are resolved.

 

A shift in regulatory mindset

Taken together, these points point to a clear change in how the ERA is treated. It is no longer anchored to a product’s history, and established use is no longer assumed to mean low regulatory risk. Instead, ERA is increasingly judged against current scientific standards and regulatory expectations, regardless of how long the substance has been on the market.

 

Looking ahead: what the EU Pharma Package could mean for ERA

A longer-term shift, not a temporary change

The revised EMA guideline has already changed how ERAs are assessed. The EU’s wider pharmaceutical reform agenda (the “Pharma Package”) (European Commission, 2023) suggests this is not a temporary uplift but a longer term direction of travel. A political agreement was reached in December 2025, and the compromise texts for a new Directive and Regulation were published in March 2026. Formal adoption is expected in 2026, with transition periods before the new rules become fully applicable.

At its core, the Pharma Package places the ERA within a broader One Health framing, recognising links between environmental exposure, ecosystem impacts and public health outcomes, including antimicrobial resistance. This is more than a change in wording. It positions the ERA not just as a dossier component, but as part of a wider public health approach, with implications beyond individual products.

 

From one-off submission to lifecycle responsibility

A major change is the move away from an ERA as a one‑off, point‑in‑time submission. Instead, it becomes an ongoing responsibility: marketing authorisation holders would be expected to update the ERA when new information could change its conclusions. That could include monitoring results, new ecotoxicology findings, or revised exposure information driven by changes in use or scientific understanding.

In practice, the elements of an ERA can already be revisited through certain post‑authorisation activities (for example, Type II variations where supporting data are resubmitted or reassessed, even if the overall conclusion does not change). The difference in this context is the direction of travel towards a more explicit, legally anchored lifecycle duty: the expectation would be to review and update the ERA when relevant new environmental information emerges, not only when a product change triggers a regulatory submission.

This direction of travel is further reinforced by the EU pharmaceutical reform (‘Pharma Package’), which broadens the ERA’s role beyond a static assessment of exposure following patient use. In particular, it introduces explicit requirements for certain product classes, such as antimicrobials, where the ERA is expected to address the risk of selection for antimicrobial resistance in the environment across the full lifecycle, including manufacturing supply chains inside and outside the EU, as well as use and disposal by healthcare professionals and patients.

In this context, the reform also points to alignment with established international standards, including the use of antibiotic‑specific PNECs where relevant for assessing resistance selection. While this shift is most clearly articulated for antimicrobials, it also signals a broader move to integrate mechanism‑driven risks within a broader environmental and public health framework. Endocrine‑active substances, already recognised as triggers for more detailed, mechanism‑based assessment under the EMA guideline, sit naturally within this expanded perspective, where environmental effects are considered not solely in compartmental ecotoxicological terms but in the context of wider biological and population‑level impacts.

Operationally, that makes the ERA feel closer to a lifecycle activity, more akin to pharmacovigilance than to traditional nonclinical documentation. It also mirrors what many applicants are already experiencing under the revised guideline, where assessors ask for assumptions to be revisited, decision logic to be clarified, and datasets to be updated, even for established products.

 

Extending expectations to legacy products

Another significant development is the extension of ERA expectations to legacy products, including medicines authorised before 2005 that were never subject to a formal environmental assessment. Under the Pharma Package proposals, EMA would be tasked with identifying products that may pose environmental concern and requiring the generation of an ERA or updates, mirroring the direction already set by the revised EMA ERA 2024 guideline that took effect in 2024, where complete, guideline-aligned ERAs are increasingly being expected even for well established substances.

 

Stronger links to regulatory decision-making

Perhaps the most consequential change is how the ERA is tied to regulatory decision-making. Today, an ERA is required, but deficiencies do not always translate into clear procedural consequences. The Pharma Package would change that by setting out more explicit, enforceable outcomes. Where the ERA data are incomplete or weakly substantiated, procedures may be delayed or unable to progress. Approvals may come with post authorisation obligations, such as additional studies or monitoring. In more serious cases, obligations could be imposed on products already on the market, and authorities may take further action where environmental risks are not adequately addressed.

This would be a real shift in how the ERA is treated, it moves closer to the centre of regulatory judgement rather than sitting at the margins of benefit–risk thinking. Importantly, it would not arrive out of nowhere. The tougher review of the ERA within submissions, the specificity of supplementary information requests and the fact that ERA issues can affect procedural timelines all suggest the system is already moving in this direction.

 

Increased transparency and consistency

The Pharma Package also places more weight on transparency. More ERA content in public assessment reports, alongside centralised repositories and substance level assessments, including the proposed use of ERA monographs, would make approaches easier to compare. Over time, that may support data sharing and reduce duplication, but it also raises the bar for consistency, robustness and scientific justification across submissions.

The possibility for the European Commission to define the content and format of ERA monographs, as well as the procedures for their development, updating and use in marketing authorisation applications, points towards a more structured and standardised approach at the substance level. While the detail will depend on how these provisions are implemented, the direction is clearly towards greater harmonisation and reuse of ERA data.

 

ERA in Clinical Trials for GMO-containing products: what’s actually changing?

A key theme in the new Pharma Package is the move towards a One Health approach, recognising the link between human, animal and environmental health. In practice, this is one of the areas where that concept starts to translate into something more tangible, with a clear strengthening of ERA requirements, not just at marketing authorisation, but earlier in development. This is particularly relevant for medicinal products that contain or consist of genetically modified organisms (GMOs).

An ERA is already required for CTA submissions for GMO-containing investigational medicinal products. That’s not new. However, the current system is fragmented. Sponsors are dealing with different Member States, different expectations, and separate submissions to multiple national competent authorities. For multinational trials, this often means running several processes in parallel, adding both complexity and timelines, with limited predictability.

The Pharma Package seeks to address this through a single, centralised ERA submission via CTIS. In simple terms, the sponsor submits one ERA centrally. The CHMP then reviews it and issues a scientific opinion within 45 days of validation. That opinion feeds into the assessment by the reporting Member State and the other Member States involved in the trial. On paper, this should reduce duplication and make multinational studies more manageable from an ERA perspective.

 

Centralisation without lowering expectations

The underlying scientific framework doesn’t really change. The ERA still needs to follow the principles set out in Annex II of Directive 2001/18/EC, and it should preferably be submitted in English. EMA is expected to develop more detailed guidance with Member States, which will be important in terms of how consistent this becomes in practice. One point that is likely to be more visible is the role of risk mitigation measures. The CHMP opinion can include these, and sponsors will need to show that they can be implemented in the context of the trial. So, it’s not just about identifying risk, but also about demonstrating that it is appropriately managed.

This clearly addresses a real pain point, and centralisation should help from a process perspective. But it’s worth being clear about what it doesn’t do. It doesn’t lower the bar scientifically, and it doesn’t remove the need for a well thought-through ERA. If anything, having a single assessment and explicit CHMP opinion may bring more consistency, but highlight where the gaps or weaknesses are. The focus on mitigation and implementation suggests that regulators will expect sponsors to go beyond a purely theoretical assessment. As with much of the Pharma Package, the detail will matter. How this works in practice will depend on the forthcoming delegated acts and guidance.

 

Key takeaways

Key themes from the revised guideline’s first year in operation and what EU pharmaceutical legislation reform (the Pharma Package) reinforces:

• An ERA is being treated as a core acceptability item, not a descriptive annex, with deficiencies that can affect approvability and timelines.
• Phase I decision logic is now a control point: assessors are checking the decision tree itself, and errors can trigger rework and reconsideration of Phase II.
• Exposure refinements must be evidence-led, with traceable epidemiology (often prevalence-based) and sources available for review; commercial proxies (e.g., sales data) are being rejected.
• Legacy and generic products are increasingly being reassessed against today’s standards, exposing recurring gaps in fate, chronic ecotoxicology, sediment/soil compartments and bioaccumulation.
• Waiver-based narratives (short duration, low prevalence, low perceived risk) rarely succeed on their own and need to sit within a complete, guideline-compliant ERA supported by appropriate data.
• Regulatory interactions are becoming more prescriptive, requests increasingly specify the studies, endpoints and derivations expected (e.g., NOEC/EC10 aligned with relevant guidance).
• The Pharma Package would make the ERA more explicitly lifecycle based, with a clearer expectation to review and update assessments when relevant new environmental information emerges, not only when a variation or other product change triggers a submission.
• Proposed reforms broaden scope and consequences: manufacturing and AMR-related environmental risks gain prominence, transparency increases, and procedural outcomes may more explicitly depend on ERA completeness and robustness.

 

How DLRC can help

DLRC supports clients in interpreting the updated ERA expectations, navigating evolving EU requirements, identifying and addressing data gaps, and developing ERA strategies that align with both current practice and anticipated regulatory change. This includes supporting Phase I decision logic, exposure refinement approaches, and the generation or justification of data to meet evolving scientific standards, ensuring that ERA does not become a rate-limiting step in regulatory procedures.

DLRC is also able to assist in the transition toward implementation of the new EU pharmaceutical legislative package, helping companies identify risks and opportunities based on their portfolio and providing strategic advice.

Contact our experts by emailing us at hello@dlrcgroup.com