The introduction of MDR has significantly increased the importance of clinical evaluation for medical devices. And whilst several guidelines exist on clinical evaluation under MDR, many manufacturers are often still unsure of how to interpret the rules for providing sufficient clinical evidence for their medical devices. In particular, Article 61(10) of the MDR which addresses clinical evaluation based on non-clinical testing methods alone, is becoming increasingly relevant, presenting challenges for manufacturers and notified bodies (NBs).

As such, medical device manufacturers commonly need to answer this important question:

“Which kind of clinical or non-clinical data are most adequate to demonstrate that my medical device conforms with all applicable general safety and performance requirements (GSPRs)?”

Determining this requires profound analysis of the individual device characteristics, and the current state-of-the-art (SOTA) in the respective medical field. However, generally in regulatory and clinical affairs there are many grey zones, and the definition of data, applications, and features can often be read in several ways.

Brought to you in collaboration with our esteemed Clinical Affairs partner, Dr med. Holger Anhalt – Founder and CEO of Medical Mind RCS GmbH, the article below outlines the different regulatory and clinical strategies related to this topic, and offers some practical advice based on our extensive clinical experience in the medical device and pharmaceutical industry.

Read on to find out more about:

• Identifying the benchmarks for pivotal clinical data
• Clinical vs. non-clinical data provision under MDR | Article 61(1) vs. Article 61(10)
• What to include in your clinical evaluation plan (CEP) and clinical evaluation report (CER) regarding clinical evidence

Identifying the benchmarks for pivotal clinical data

Among other aspects, the most important objective of clinical evaluation is to confirm the safety and performance of a medical device.

To demonstrate conformity to all applicable GSPRs, the MDR requires manufacturers to determine particular benchmarks for the pivotal clinical data. Within the medical device regulatory environment benchmarks are understood as acceptance criteria. These benchmarks should ideally be quantitative. This should take place at the beginning of the clinical evaluation process, be documented in the clinical evaluation plan (CEP), and is part of the whole clinical development process.

Although already required under the MDD, this requirement represents one of the most significant changes for clinical evaluations under the MDR, as it was previously not widely imposed by the notified bodies. Under the MDR this must now be implemented, and as a result requires substantial changes in the whole clinical evaluation process and document structure. In particular, the CEPs have become much more burdensome with extensive medical backgrounds and state-of-the-art (SOTA), as the benchmarks must be derived considering the gold standard, treatment alternatives, and other devices and drugs currently available on the market.

To derive the benchmarks, all claims that need to be supported by (pre)clinical data must be identified. Then the most adequate types of pivotal data and their acceptance criteria can be defined. These data might be of clinical or preclinical origin.

For the correct proceeding, it is crucial to define the kind of benefit that is related to the intended use of a medical device. In this context, it can be challenging to understand the different terms related to clinical performance. Therefore, it is important to note that “clinical performance”, “clinical benefit”, and “clinical evidence” are not synonymous.

Article 2 of the MDR defines these terms, which are relevant for deciding how to address the GSPRs:

• Clinical performance means the ability of a device, resulting from any direct or indirect medical effects which stem from its technical or functional characteristics, including diagnostic characteristics, to achieve its intended purpose as claimed by the manufacturer, thereby leading to a clinical benefit for patients, when used as intended by the manufacturer
• Clinical benefit means the positive impact of a device on the health of an individual, expressed in terms of a meaningful, measurable, patient-relevant clinical outcome(s), including outcome(s) related to diagnosis, or a positive impact on patient management or public health
• Clinical evidence means clinical data and clinical evaluation results pertaining to a device of a sufficient amount and quality to allow a qualified assessment of whether the device is safe, and achieves the intended clinical benefit(s) when used as intended by the manufacturer

This implies that the clinical performance of a medical device represents the intended function within a medical procedure or treatment concept, but it is not automatically a direct clinical feature. It can also be a preclinical or technical function (e.g., a diagnostic measuring tool, or a special laboratory fridge that preserves clinical samples). In all such cases, manufacturers must determine the best suited data sources and benchmarks to confirm the device performance and safety.

Clinical vs. non-Clinical data provision

Direct & indirect clinical data

Although the terms “direct” and “indirect” clinical data are not used in the MDR, and only mentioned in MDCG 2020-6 related to legacy devices which were already marketed under the MDD, they represent a good concept to define which kind of data are needed for successful CE-marking.

Generally, clinical data should support direct benefits whereas indirect clinical benefits may be demonstrated by non-clinical data, such as pre-clinical tests, bench testing, or usability data based on formative and summative studies.

CE-marking based on non-clinical data alone – Article 61(10)

Although Article 61(1) in the MDR states that “Confirmation of conformity with relevant general safety and performance requirements …, shall be based on clinical data providing sufficient clinical evidence”, according to Article 61(10), demonstration of conformity with GSPRs that is based on the results of non-clinical testing methods alone can be adequate, if duly justified.

However, adequate justification for any such exception shall be given based on:

• The results of the manufacturer’s risk management
• Consideration of the specifics of the interaction between the device and the human body
• The intended clinical performance
• The claims of the manufacturer
• The risk class of the device (i.e. Class III and implantable devices are not in scope)

Key aspects regarding Article 61(10) include the following:

• If the manufacturer claims a direct clinical benefit, Article 61(10) is no longer feasible and clinical data must be provided
• The Article applies to devices that do not have ‘direct’ interaction with the human body
• But the Article does not exclude devices that have direct contact with the human body
• Duration of body contact, design, particular risks, the grade of innovation and the role in the overall clinical procedure must be critically considered. For example, a simple surgical instrument that is critical to the success of the implant may not be applicable, whereas basic surgical instruments such as scissors, forceps, and standard scalpels can be feasible for non-clinical testing methods alone

Other typical examples of medical devices to which Article 61(10) would apply, include devices with a low risk profile, have little or no patient contact, and/or have either indirect or no clinical claims, such as lab fridges, disinfectants, or simple measuring tools.

It is important to note that “indirect clinical benefit” is not a synonym for applicability of Article 61(10). In such cases, it can be more feasible to establish the clinical benefit with clinical data, e.g., from the overall procedure, and/or data from other devices. For example, the clinical evidence of a cardiological guiding catheter can be established by the clinical data of the accompanying stent system, although its clinical benefit is indirect (facilitation of the delivery of the stent to the coronary artery).

Article 61(10) does not relieve the manufacturer of the duty to prove clinical performance claims with adequate measurable (non-clinical) data. The following data sources are considered adequate under Article 61(10):

• Compliance to standards
• Preclinical verification and validation activities
• Bench tests
• Simulated use tests
• Usability testing
• Data from clinical experience (PMS, PMCF, safety database evaluations)
• Supportive clinical data from scientific literature searches or other sources

Bear in mind, that even if pre-clinical data are considered most feasible to establish the clinical evidence under Article 61(10), additional clinical data can (and should) be provided as supportive data, e.g., literature data and/or PMS data from comparable devices and/or safety database evaluations to further confirm the performance and overall safety of the device under evaluation.

MDCG 2020-13 provides further useful guidance for manufacturers to help determine if Article 61(10) is applicable for their medical device, and to establish a justification accordingly.

Special Scenarios

Medical devices with a supportive function

Importantly, clinical performance does not automatically provide a direct positive impact (clinical benefit) on the health of a patient. For example, the device may only have a supportive function, and a clinical benefit can only be achieved in conjunction with another device within the same clinical procedure – such as a guide wire that facilitates the delivery of a cardiac stent system in the correct position, or a PTCA balloon catheter that expands the stent.

The clinical performance in these examples would be indirect but related to the performance of the stent, which ultimately provides the direct clinical benefit for the patient. In this case clinical data of the outcome (e.g. procedural success, survival, or rehospitalization rate, etc.) of the complete procedure (stent implantation) would be adequate to prove the safety and performance of the PTCA catheter or guide wire.

Medical devices with a measuring function

Often, accuracy is an adequate parameter to confirm the clinical performance, especially in devices with a measuring function. In such cases it needs to be decided whether the accuracy should be tested in a clinical setting or if bench tests would be sufficient. This needs to be evaluated on a case-by-case basis.

If, for example, the performance of a fever thermometer or blood pressure meter is tested in a certain population, the accuracy of the measurement device might be evaluated, but not necessarily the effect of a therapy or any clinical outcome for the patients. In this case the accuracy of the measurement is an appropriate parameter to address the clinical evaluation of the device as no clinical endpoint (improvement of patient’s health) is apparent.

Here, in accordance with Art. 61(10) preclinical bench testing, considering the risk profile, established technology, and intended performance, would probably be suitable to provide adequate clinical evidence. Normally, bench tests should be further supported by usability testing according to IEC 62366, which is an important data source in cases where direct clinical data is inadequate.

In this context it is important to note that all scientific data which are directly derived from human patients (including laboratory tests) are clinical data by definition, but not all clinical data automatically address clinical endpoints, for example accuracy data of devices. Usability data are normally derived from healthy volunteers and are traditionally seen as non-clinical data, comparable with data from bench testing.

What to include in your Clinical Evaluation Plan regarding clinical evidence

The following steps should be conducted and described in your CEP and considered in the clinical development plan:

Identify your clinical claims

An exhaustive identification of all relevant clinical claims which are based on the intended use must be conducted. This includes an alignment and consistency check of the device homepage, risk management, IFU, CEP, and CER.

It is worth noting here, that often small differences in the formulation of a clinical claim or the intended performance can have a dramatic impact on the regulatory route, which highlights the importance of early identification as part of your CEP.

Describe SOTA

Besides a detailed description of all indications, user groups, and technical/clinical features of the device under evaluation, a systematic literature search should be performed to describe the current state-of-the-art (SOTA).

This should describe the disease, typical user groups and their current treatment approaches, and alternative treatment methods, including current success rates and side effects. It should also include current guidelines, publications e.g., related to competitor device technical standards, a list of similar (competitor) devices, and all other relevant aspects.

Decide on your clinical route

The next step involves deciding whether demonstration of conformity based on non-clinical data alone according to Article 61(10) is applicable, or if the clinical route is more feasible according to Article 61(1).

Identify adequate acceptance criteria & their values (benchmarks)

This involves identifying benchmarks for both safety and performance. Then the most adequate types of pivotal data can be defined. These might be of clinical and/or preclinical origin. Based on the information gathered in the SOTA, applicable success criteria should be defined, such as clinical endpoints for safety and performance or non-clinical criteria (especially in the case of Article 61(10)) from bench testing, usability testing, and/or conformity with standards. Therefore, the CEP should contain a detailed section on applicable standards and their success criteria.

It is not sufficient to just name certain criteria. Most importantly, clear quantifiable benchmarks, such as clinical success rates, percentage of side effects, or concrete values of technical dimensions (e.g., time, voltage, temperature) which must be met by the medical device under evaluation, should be listed. In exceptional cases, qualitative data can also be adequate – however, this requires justification.

Moreover, Article 61(10) does not free the manufacturer from the aforementioned steps. Even for devices where clinical test methods are not deemed appropriate, a systematic literature search and safety database review should be performed to describe the different user groups, their performance, and the most relevant side effects and safety aspects. Measurable (non-clinical) benchmarks must also be defined and profound justification for why clinical data is not adequate must be given.

Importantly, all indications and population groups must be addressed by adequate data. This is often difficult to achieve, e.g., in cases where a device has an unlimited application range, such as surgical tools and robots, diagnostic devices, etc. In such cases, a worst-case scenario or other argumentation line should be given, which justifies why certain clinical endpoints are representative for a particular indication range or user group.

What to include in your Clinical Evaluation Report regarding clinical evidence

When the development process is further advanced, the clinical and/or non-clinical data which were defined in the CEP as being adequate to confirm the GSPRs, need to be outlined in detail and compared with the benchmarks in the accompanying CER. This should include the following steps:

Repeat a short summary of the SOTA

Include the previously defined benchmarks to enable the reviewer to better understand the comparison and critical discussion of the outcomes of the defined clinical and/or non-clinical testing program.

Describe the results of the pre(clinical) testing program in adequate detail

This does not mean that all details of all technical and clinical tests must be shown, but at the least reference should be made to where they can be found. The concrete outcomes of the testing program should be provided, ideally in a table which directly compares the benchmarks from the CEP with the pivotal data pertaining to the device under evaluation.

Critically discuss if the benchmarks have been successfully met 

And if not, justify if certain GSPRs are considered confirmed despite unsuccessful test results, and/or define further planned measures, such as PMCF, additional testing, etc.

Declare that conformity with all relevant GSPRs has been successfully confirmed 

Particularly for requirements 1-8. This includes the requirements on performance, safety, acceptable risk-benefit-ratio, and undesirable side effects, as listed in MEDDEV 2.7/1, rev. 4.

In conclusion, determining the adequate clinical evidence to ensure your medical device complies with the GSPRs under MDR requires the identification of all relevant (pre)clinical claims, and definition of appropriate acceptance criteria and their values (benchmarks) based on comprehensive analysis of device characteristics and the current state-of-the-art.

Whilst the definition of data, applications, and features can be read in various ways, by understanding the relevant regulatory benchmarks, generating the appropriate pivotal data, and including the aspects outlined above in your CEP and CER, you will be able to effectively curate adequate clinical evidence and ultimately achieve compliance for successful CE-marking.

Should you have a clinical evaluation challenge, feel free to get in touch – our Clinical experts are ready and happy to help. And for further reading on clinical affairs for medical devices, see here.