EU moves to reset medical devices regulation

The European Commission has published its long-awaited proposal for a regulation amending and simplifying the EU Medical Devices Regulation (EU MDR) and the In Vitro Diagnostic Medical Devices Regulation (IVDR). The full proposal is available on the Commission’s website.

While some of the proposed amendments are technical or incremental, others would materially reshape how the EU medical devices framework operates over the next decade. The proposal should therefore be read not as a minor corrective exercise, but as a structural recalibration of a regulatory system that has struggled since the MDR and IVDR entered into force.

According to the Commission, the proposal pursues eight overarching objectives:

• regulatory simplification and proportionality;
• reduction of administrative burden;
• promotion of innovation and availability of devices for special patient groups and unmet needs;
• greater predictability and cost-efficiency in conformity assessment;
• improved coordination within the decentralised EU medical device regulatory system;
• greater use of digitalisation;
• promotion of international regulatory cooperation; and
• improved interaction with other EU legislation, including the AI Act.

The number of individual amendments proposed under these headings is extensive. What follows therefore highlights the most legally and practically significant changes, with a particular focus on their implications for manufacturers, notified bodies and regulators.

This proposal also needs to be viewed in the context of a broader wave of EU legislative reform in life sciences and technology. In recent weeks alone, the EU has advanced simplification measures for the AI Act, reached political agreement on the revised pharmaceutical legislation, and announced a forthcoming EU Biotech Act, expected to be adopted in parallel with the present proposal. Against that backdrop, the Commission’s message is clear: regulatory ambition must now be matched by regulatory operability.

Simplification and proportionality

The Commission proposes a series of measures intended to make the MDR and IVDR more proportionate and workable. Some are genuinely significant.

Most notably, the proposal removes the general five-year validity cap on certificates of conformity. Instead of routine re-certification, notified bodies would “periodically review” certificates throughout their lifetime. Certificates would only be time-limited where this is considered exceptionally necessary, such as where approval is granted subject to conditions, including post-market clinical follow-up (PMCF) obligations.

This change aligns with the Commission’s earlier implementing regulation aimed at standardising notified bodies’ conformity assessment activities, which already sought to narrow the scope of re-certification reviews.² Taken together, these measures should materially reduce cost, uncertainty and regulatory congestion.

Other simplification measures include:

• removal of the minimum qualification requirements for the Person Responsible for Regulatory Compliance (PRRC);
• removal of the requirement for SMEs to maintain a PRRC who is “permanently and continuously available”; and
• introduction of a new category of “well-established technology devices”, subject to lighter conformity assessment requirements, replacing the outdated and overly prescriptive list of specific products previously relied upon.

By contrast, some of the Commission’s proposals are less transformative than advertised. For example, the text suggests a broader range of acceptable clinical and non-clinical data for conformity assessment, including non-peer-reviewed clinical data, data from equivalent devices, and in vitro, ex vivo and in silico testing.

In practice, however, it is not clear that these changes materially expand what is already permitted. The existing framework already allows laboratory testing, simulated use testing, computer modelling and animal models, and the new categories may amount more to repackaging than reform.

Similarly, while the proposal removes the requirement for a manufacturer to hold a contract granting access to the technical documentation of an equivalent device, it retains the requirement to demonstrate “sufficient access” to the underlying data, as well as technical, biological and clinical equivalence. These thresholds remain high and are unlikely to make equivalence claims significantly easier in practice.

Changes to classification rules are also modest. The Commission proposes no amendments to IVDR classification rules, offering no relief from the sweeping up-classification of IVDs under the IVDR. Amendments to MDR Rule 11 on software largely confirm what industry has already concluded: almost all medical device software falls at least within Class IIa. Software used for any medical purpose, even in non-serious situations, will continue to be classified accordingly.

Reduction of administrative burden

The proposal also introduces targeted measures to reduce compliance burdens, particularly in post-market obligations.

Key changes include:

• reduced frequency of Periodic Safety Update Reports (PSURs), depending on device class;
• extension of the vigilance reporting deadline for serious incidents from 15 to 30 days; and
• exemption from prior authorisation for IVD performance studies involving routine blood draws, and exemption from prior notification for companion diagnostic studies using leftover tissue samples.

Of particular interest is the proposal to permit pre-determined change control plans agreed in advance between manufacturers and notified bodies. Under such plans, manufacturers could implement pre-approved changes to devices or quality management systems without seeking fresh approval.

This concept has long been discussed in relation to AI-enabled medical devices and suggests that the Commission is consciously future-proofing the MDR and IVDR against adaptive technologies.

Innovation and access for special patient groups

While simplification and burden reduction dominate much of the proposal, the most ambitious reforms appear under the heading of innovation and access.

First, the Commission proposes to broaden the in-house manufacturing exemption. The prohibition on supply between health institutions would be removed, as would the requirement that no equivalent device exists on the market. Documentation obligations would also be simplified, and laboratories would be permitted to rely on the exemption for in-house IVDs used in clinical trials.

More significantly, the proposal introduces two entirely new regulatory categories: “breakthrough” and “orphan” medical devices and IVDs.

A device would qualify as a breakthrough device where an expert panel considers that it introduces a high degree of novelty and is expected to deliver a significant positive clinical impact for a life-threatening or irreversibly debilitating condition, either by outperforming existing alternatives or addressing an unmet medical need.

An orphan device designation would apply where a device targets a condition affecting no more than 12,000 individuals per year in the EU, and either addresses an absence of alternatives or provides a clinical benefit over existing options.

Designation as a breakthrough or orphan device would trigger priority rolling review by a notified body and entitle the manufacturer to expert panel advice on clinical development strategy and evidence requirements. Crucially, notified bodies would be required to give “due consideration” to that advice and could issue conditional certificates based on limited clinical data, supported by PMCF obligations.

If implemented robustly, this framework could significantly accelerate market access for innovative technologies and underserved patient populations. However, its success will depend heavily on guidance, resourcing and the willingness of expert panels to assert influence over notified bodies.

The proposal also provides that orphan devices previously approved under the old directives would benefit from an indefinite extension of their transitional period, allowing them to remain on the market indefinitely, subject to stability and continued orphan status.

Finally, the Commission proposes regulatory sandboxes at both national and EU level. National sandboxes would permit real-world testing of devices addressing unmet medical needs where MDR/IVDR requirements would otherwise impede development. EU-level sandboxes would be used to test the adequacy of regulatory requirements for emerging technologies, though without real-world testing.

The legal status of data generated in sandboxes remains unclear and will require guidance.

Predictability and cost-efficiency of certification

The proposal also seeks to address long-standing concerns about inconsistency, delay and cost in conformity assessment.

A formal legal basis would be introduced for “structured dialogue” between manufacturers and notified bodies, clarifying the boundary between permissible clarification and prohibited consultancy.

Notified body review would be narrowed for many devices, with only one representative device per group or portfolio subject to full review. Surveillance audits would be reduced to once every two years, remote audits permitted, and unannounced audits restricted to cases “for cause”.

The Commission would also gain power to cap notified body fees for SMEs and manufacturers of orphan devices.

Coordination within a decentralised system

One of the most consequential aspects of the proposal is the significantly expanded role envisaged for the European Medicines Agency (EMA).

Expert panels, already operational but under-utilised, would become central actors in the system, providing scientific, clinical and regulatory advice to the Commission, the Medical Device Coordination Group (MDCG), member states, notified bodies and, in some cases, manufacturers. The EMA would appoint panel members and charge fees for panel advice.

The EMA would also coordinate national competent authorities on issues such as borderline determinations, multi-country clinical studies, derogations and market surveillance.

This includes codifying the “Helsinki Procedure” for borderline and classification disputes. Where authorities cannot agree, an expert panel opinion would be issued, and the referring authority would be required to give it “utmost consideration” when adopting its decision.

A new dispute resolution mechanism would also be introduced for conflicts between manufacturers and notified bodies, operated by national authorities acting as ombudspeople.

Digitalisation

The proposal supports wider use of digital tools, including electronic declarations of conformity, digital technical documentation, digital labelling and expanded use of electronic instructions for use.

Interestingly, the Commission also signals that some Eudamed functions may be delivered via alternative IT systems. While several Eudamed modules will become mandatory in May 2026, others, particularly clinical investigation and performance study modules, remain delayed. This raises the possibility of functional overlap with the Clinical Trials Information System (CTIS) in future.

International cooperation

Two new provisions address international cooperation.

First, the Commission would be mandated to pursue regulatory cooperation through fora such as IMDRF, MDSAP and ISO, including joint inspections and assessments.

Second, the Commission would be empowered to participate in bilateral and multilateral reliance schemes, allowing regulatory decisions from third countries to be used to expedite EU approvals, provided reciprocity is ensured.

At minimum, this could lead to fuller recognition of MDSAP audit reports. More ambitiously, it could open the door to mutual reliance arrangements with jurisdictions such as the UK or Australia. However, the reciprocity requirement effectively excludes unilateral reliance on FDA approvals.

Interaction with other EU legislation

Perhaps the most consequential reform is the proposed treatment of AI-enabled medical devices under the AI Act.

Under the proposal, medical devices and IVDs would be excluded from classification as “high-risk AI systems” and instead regulated exclusively under sector-specific legislation, in line with aviation and automotive AI systems. Notified bodies would not require separate AI Act designation, and manufacturers would avoid overlapping quality management requirements.

Given industry concerns about the AI Act’s restrictions on real-world testing, this exemption is likely to be widely welcomed.

Other changes include a pathway for a single approval covering combined clinical trials involving medicinal products and medical devices or IVDs, aligning with forthcoming amendments under the EU Biotech Act.

Cybersecurity would also become an explicit general safety requirement, with non-safety cybersecurity incidents reportable via Eudamed to EU cybersecurity authorities.

Conclusion

The repeated extension of MDR and IVDR transitional periods was always a temporary fix. It mitigated shortages but could not address the structural weaknesses of the system. The Commission’s own targeted evaluation has now acknowledged those shortcomings.

This proposal represents a serious attempt to stabilise, simplify and future-proof the EU medical devices framework. While the legislative process remains long, and outcomes uncertain, the direction of travel is clear: fewer procedural bottlenecks, greater reliance on expert judgement, and closer alignment with innovation policy.

If adopted in time, the reforms may also spare manufacturers from the worst effects of overlapping AI regulation. For an industry that has endured years of regulatory turbulence, that alone would be no small achievement.