EU MDR Article 61 and equivalence – is there a new way forward?

About this article: Could there be more scope for use of equivalence data under EU MDR than you may have thought? Dr. Amie Smirthwaite, Sr. VP of Global Intelligence & Innovation at RQM+, EU MDR expert, and member of the working groups that authored MDCG 2020-5 and 2020-6, has put a lot of thought into MDR Article 61 and the associated requirements around equivalence and sufficient clinical evidence. From the early days of implementing the MDR at BSI until now, five years later, Amie has found the wording of this article to be puzzling, creating unnecessary challenges for both notified bodies and manufacturers. Amie and the RQM+ team hope that the position taken in this blog will be food for thought for the European Commission, competent authorities and notified bodies, allowing a bit more flexibility for certifying devices that offer a real benefit to patients.

It is somewhat embarrassing to admit, but I have been thinking about Article 61 ever since the MDR was published. In particular, I’ve been drawn to revisit Articles 61(4-6) several times. Not due to their literary merit, but because of the oddness of the wording – there has been something that has never quite sat right with me.

For example, if the intent of Articles 61(4) and (5) is, as many assumed, to make the use of equivalence for Class III and implantable devices exceptional, why are the clauses framed in terms of the need to undertake a clinical investigation? Why don’t they come straight out and say something to the effect of, “Equivalence data shall only be used in the following circumstances: …”? Why is the need to undertake clinical investigations for Class III and implantable devices linked to equivalence at all? Why not simply say that clinical investigations are required unless there is “sufficient clinical evidence”, with the additional requirements for use of equivalence data as a stand-alone clause? This would be much clearer than the current wording. And, finally, if all Article 61(4-6) roads lead to the Rome of “sufficient clinical data” being required to avoid the need for a clinical investigation, how do the requirements of Clauses 61(4), (5) and (6) differ in practice?

Working groups were assembled, involving representatives of Notified Bodies, Competent Authorities, medical professional societies, and industry associations. We debated the issues and together, for well over three years, tried to understand these clauses, their intent and how they should be applied, leading to the publication of MDCG 2020-6 and several other associated guidance documents.

And now, more than five years later, I’ve had a lightbulb moment and finally worked out what’s been bothering me all these years. And what I think is: use of equivalence is not quite as restrictive as we all thought. In particular, I think most of us have been misinterpreting Article 61(6).

First, a bit of background into how I think we got distracted from the actual intended meaning:

Articles 61(4-6) of the MDR are clearly an expansion of Clause 1.1a of Annex X of the MDD, which stated: “In the case of implantable devices and devices in Class III clinical investigations shall be performed unless it is duly justified to rely on existing clinical data”. This clause in the MDD had been troubling for regulators and manufacturers alike, as there was no genuine consensus on what a valid justification to rely on existing clinical data should look like. Some felt that if a device had been on the market for many years with no significant vigilance events or complaint trends, then it could be duly justified to rely on existing clinical data, even if there were no data whatsoever. Indeed, this was probably the prevailing opinion until the mid-2000s; after this point, cumulative medical device failures started to make regulators re-evaluate their assumptions about the implicit safety of medical devices based on lack of vigilance, even those with a long clinical history. In this context, the intent of Articles 61(4-6) can be seen as providing further clarification of when it might be duly justified to rely on existing clinical data and thereby avoid the need for a pre-market clinical investigation. And what Articles 61(4-6) tell us is that a premarket clinical investigation is not required in the following specific cases:

1. For devices not previously certified under the Directives:

a. when there is sufficient clinical data from a demonstrated equivalent marketed by the same manufacturer (Article 61(4))
b. when there is sufficient clinical data from a demonstrated equivalent of a device marketed by another manufacturer, but only if there is a contract in place between the two manufacturers allowing the manufacturer claiming equivalence full access on an ongoing basis to the technical documentation of the equivalent device (Article 61(5))

2. For devices that were previously certified by the manufacturer under the Directives:

• where the clinical evaluation is based on “sufficient clinical data”, and the device complies with the relevant product-specific CS if applicable (Article 61(6a))

3. For “sutures, staples, dental fillings, dental braces, tooth crowns, screws, wedges, plates, wires, pins, clips or connectors”, regardless of whether or not these devices had been previously certified by the manufacturer under the Directives:

• where the clinical evaluation is based on “sufficient clinical data”, and the device complies with the relevant product-specific CS if applicable (Article 61(6b)).

In all of the above cases, there must be an appropriate PMCF plan (more on this later, as it may not be immediately obvious to all that this also applies to the Article 61(6) devices).

As can be seen in points 2 and 3 above, nothing in Articles 61(6a) or 61(6b) requires a contract between manufacturers as described in Article 61(5), nor do these clauses prohibit a manufacturer from claiming equivalence with another manufacturer’s device. All that is required, in common with every other device seeking certification under the MDR, is that the data satisfies the definition of “clinical data” as described in Article 2(48), and that the requirements for demonstration of equivalence as described in Annex XIV Clause 3 [1] , are met. In fact, Article 61(4) points the reader directly to Article 61(6), effectively bypassing Article 61(5) for legacy and WET devices. Why then, did so many of us (including myself, as a frequent contributor to MDCG guidance documents and Team NB training in clinical evaluation) misinterpret this?

Part of the problem is that the Regulation has legal language that can introduce subtle nuances of intent that are not always immediately obvious to people trained as scientists, engineers, or doctors rather than as lawyers. One example of this is the phrase “without prejudice to paragraph 4…” in Article 61(10). Notified Body experts were divided as to what this meant, with half believing it meant you could ignore paragraph 4, and the other half believing it meant you couldn’t ignore paragraph 4 (Commission lawyers later confirmed that the latter interpretation was the correct one).

But, apart from the occasional opacity of the legal language in the MDR, there were a few other factors which I believe confused the issue:

1. The requirement for a contract between manufacturers was met with surprise and a fair amount of skepticism at the time. “Which manufacturers,” some wondered, “are going to be willing to allow another manufacturer access to their commercially sensitive data? [2] What they’re really saying is that we can’t use equivalence anymore.” i.e., that clause was viewed by many as an attempt to effectively eliminate equivalence with another manufacturer’s device as a viable route to CE certification for Class III and implantable devices. This then led to a short-form interpretation of Articles 61(4-6) as, “you can’t use equivalence for any class III or implantable device unless it’s your own device, or unless you have a contract in place if it’s another manufacturer’s device”. Which, in hindsight, is clearly not what Article 61 says.

2. Every clause in Articles 61(4-6) requires, one way or another, that there be sufficient clinical data to demonstrate safety and performance (with clinical benefit being embedded in performance). The repetition of this requirement, in each of Articles 61(4), (5), (6a) and 6(b), focused most people’s attention to the question of “what is sufficient clinical data?”. This is a valid and valuable question to address, but it was conflated with the exasperation some felt at what they considered an unfeasible requirement for a contract that would seldom if ever be granted; therefore the perception that the “sufficient clinical data” referred to in Article 61(6a) and 61(6b) could not include equivalence data (unless it met the additional criteria described in Articles 61(4) or (5)) began to consolidate.

3. Under the Directives, application of MedDev 2.12/2 indicated that manufacturers of devices certified on the basis of equivalence should undertake appropriate PMCF studies, to ensure that they had data on their own devices when it came time for certificate renewal – i.e., the intent was that manufacturers should not normally rely on equivalence indefinitely, but should have appropriate PMCF in place to gather data to confirm conclusions based on equivalence. Therefore, when manufacturers presented one scenario after another to Commission working groups regarding legacy devices that would not meet the bar for “sufficient clinical data” if they could not use existing equivalence routes, the response was typically, “you’ve had years to collect this data, if you don’t already have it then collect it via PMCF under your Directive certificates before the end of the transition period”. And while this may be a fair response for many Class III and implantable devices, it again shifted the discussion away from, “what is the Regulation actually asking for?” to something more like, “are manufacturers really going to have to undertake new clinical investigations for devices that have been on the market for decades, including low risk and standard of care devices, such as sutures?”. This is one reason MDCG 2020-6 devotes considerable attention to requirements for these low-risk devices, rather than clarifying the intent of Article 61(6).

To be clear, I do not think the intent of Article 61(6) is to allow devices to be placed on the market where either:

a) the manufacturer repeatedly failed to meet PMCF commitments agreed to under the Directives; or

b) it is not possible to conclude with a high degree of confidence that the safety, performance and clinical benefit of a device are consistent with the current state of the art, based on existing evidence.

But I do think it means that use of another manufacturer’s equivalence data for the legacy devices described in Article 61(6a) and the “WET” devices described in Article 61(6b) is allowable without a contract. Circumstances in which it might be justifiable to rely on such equivalence data include cases where:

• the devices were initially certified under the Directives after the publication of the Regulation and may not have had enough time to generate sufficient clinical data via PMCF on their own devices to provide comprehensive coverage of every device variant, combination, treatment indication, patient population, end stage and severity of disease, etc. to meet the requirements of the Regulation. Additionally, the pandemic has had a significant impact on elective surgery, and many such manufacturers will not have sufficient enrolment or data generated in time for their MDR submission deadlines.
• the Notified Body accepted PMCF plans limited to key areas of residual risk (for example, in the case of low-risk standard of care devices such as sutures or bone screws). As for bullet 1 above, such studies may not provide the comprehensive data coverage required for confirmation of conformity under the Regulation.
• the Notified Body accepted a “no PMCF” rationale, based on a lack of device risk or novelty. Even if a manufacturer subsequently recognised a need to address this gap following publication of the MDR, the time required to design and initiate such studies, combined with the impact of COVID, might again have made comprehensive data collection unfeasible.
• the technical documentation may not have been sampled for a given device, and therefore justifications to “rely on existing clinical data” may not have been challenged or recognised as insufficient by the manufacturer.

In all such cases it would be expected that:

• the equivalence criteria of Annex XIV are met. [3]
• the clinical evaluation provides “sufficient clinical data” as defined in Article 2(48) and subsequently clarified in MDCG 2020-6, to demonstrate conformity and provide a high degree of confidence in the safety, performance, and clinical benefit of the device in relation to the state of the art and in relation to its stated intended purpose.
• if PMCF studies were considered necessary under the Directives to confirm conclusions reached on the basis of equivalence data, that these have either been completed according to the plan accepted by the Notified Body, or that there are genuine and valid reasons for these studies not generating the required clinical evidence in time for inclusion of the relevant clinical data in the manufacturer’s MDR application, that do not adversely impact the previous conclusions on the safety, performance and clinical benefit of the device.
• PMCF studies appropriate to residual risk are planned in accordance with Annex XIV Part B and included in the technical documentation submitted for MDR application.

A bit more on that last bullet, regarding PMCF: Article 61(6) says nothing about PMCF studies, so it may seem that the MDR does not require it for these devices. However, Article 61(4) says:

“In addition, clinical investigations need not be performed in the cases referred to in paragraph 6”

and Article 61(6) says:

“The requirement to perform clinical investigations pursuant to paragraph 4 shall not apply…”.

i.e., the devices in Article 61(6) are only exempted from the requirement to perform clinical investigations (providing they have sufficient clinical data, whether from their own devices or demonstrated equivalent devices); they are not exempted from the requirement for appropriate PMCF which is also included in Article 61(4). On the other hand, the definition of PMCF under the Regulation is much broader than it was under the Directives, and it is assumed that an appropriate PMCF plan would be proportionate to residual risk and potential “unknown unknowns”.

One final point: although I believe this interpretation is the correct one, and not merely a semantic argument, it will not benefit manufacturers to adopt this position if the Notified Bodies and Competent Authorities do not agree. I have put this question to relevant regulatory leaders. Since implementation would require consensus across Notified Bodies, Competent Authorities and other regulators and stakeholders, they will require some time to consider and respond. Watch this space.

[1] Annex XIV clause 3 requires “sufficient access” to data, but not a contract. The difference in wording between Article 61(5) and Annex XIV Clause 3 allows for other verifiable sources to be used, including data available in the public domain or acquired by comparative testing, proportionate to the risks posed by potential unknowns
[2] one potential scenario would be for virtual manufacturers, although the requirement is for “full” access, which might be more than typically granted to a virtual manufacturer
[3] Use of the equivalence table provided as an Annex to MDCG 2020-5 is strongly recommended

How RQM+ Can Help:

RQM+ has the world’s leading experts and unrivaled collective knowledge on the EU MDR. With former notified body leaders who significantly contributed to committees and guidance documents supporting the MDR, and a team of SMEs who have led MDR implementations and compiled hundreds of technical documentation files, we’ll have a solution to every challenge.

Subscribe to our blog to keep up to date on our journey with the interpretation of Article 61(6).

To easily review the exact language from the sections of the MDR referenced in this article, download the RQM+ EU MDR Filtering Tool and filter lines of the MDR by topic, chapter, section, article, and search term.

About the Author:

Dr. Amie Smirthwaite joined RQM+ in March of 2020. A clinical and regulatory affairs expert, Amie has over 25 years of postdoctoral experience in medical devices, spanning new product development, quality and regulatory systems, and clinical data evaluation. She is leading RQM+ Intelligence and Innovation team and brings a wealth of knowledge and experience having worked for medical device companies, academic institutions, and BSI.

As former Global Head of Clinical Compliance at BSI (12 years, through March 2020), Amie developed BSI’s clinical oversight team and clinical operational processes and led clinical aspects of BSI’s successful MDR designation. Prior to taking responsibility for BSI’s clinical team, Amie was Head of External Training at BSI, and developed a suite of training courses for manufacturers including Clinical Evaluation, Risk Management, Technical Documentation, Biological Safety, ISO 13485.

Amie has been a recent contributor to European Commission Clinical Investigations and Evaluations Expert Group, including subgroups which authored MDCG 2019-9, MDCG 2020-5, 6, 7, 8, and 13 (SSCP, Clinical Evaluation, Sufficient Clinical Evidence, PMCF, and more), and the to-be-released guidance on PSUR, and contributed to development of the requirements for clinical in MedDev 2.7/1 rev 4, which are now reflected in the MDR. Amie is a current contributor to the ISO technical committees 194 (Biological and clinical evaluation of medical devices), 150 (Implants for surgery) and 210 (Risk management) and is acting as interim convenor for a subcommittee that is drafting a new ISO standard for clinical evaluation of medical devices. She is also a member of the advisory boards for CORE-MD (European Union Horizon 2020 project, core-md.eu) and NORE (Network of Orthopaedic Registries of Europe, nore.efort.org).

Prior to her notified body experience, Amie held roles in academic research and industry product development of orthopaedic and cardiovascular devices, leading to multiple publications and three patents.