Under the EU In Vitro Diagnostic Medical Devices Regulation 2017/746 (IVDR), every IVD must have a Performance Evaluation Report (PER). PERs consist of three pillars: scientific validity, analytical performance and clinical performance.
Collating data to satisfactorily address each pillar must be a priority for IVD manufacturers. With notified bodies under severe pressure and resource limitations, they will not engage with manufacturers who cannot demonstrate that they have processes for IVDR compliance in place.
As manufacturers were not previously required to provide a formal PER under the IVDD, some are unsure how to prepare compliant PERs. Under the IVDD, manufacturers could refer to guidance from the Global Harmonization Task Force (GHTF) and relevant ISO standards.
GHTF guidance can still provide initial support for the three key strands of the PER. However, there is a need for IVDR-specific guidance to give manufacturers a clearer idea of expectations.
In general terms, the process and the end goal of the PER is similar to the Clinical Evaluation Report (CER) under the EU Medical Device Regulation. IVD manufacturers who also manufacture medical devices may have useful experience with CERs that could support PER drafting.
However, they must bear in mind that IVD risks are usually indirect. Rather than the device itself causing harm, the risks lie in the possibility of incorrect diagnosis, and the resulting consequences.
Manufacturers must consider how misdiagnosis could affect patients, and whether it may result in the wrong treatment being prescribed or a lack of treatment altogether.
A risk-based approach
One thing manufacturers can be sure of is that the IVDR expects them to assess benefit-risk on the basis of clinical data, existing literature and continuous assessment of post market data. The depth of this assessment is proportional to device risk.
For instance, for a well-established and standardized Class B device, clinical performance may be demonstrated by peer-reviewed literature and published experience of routine diagnostic testing or even by only one of these sources. On the other hand, for a novel Class D device, clinical performance studies with prospective sample collection are typically required.
For analytical performance, analytical studies are always required. They must address all the applicable parameters in point (a) of Section 9.1 of Annex I of the IVDR, for example sensitivity, specificity, accuracy, LoD, linearity, etc.
CLSI (Clinical and Laboratory Standards Institute) guidelines are typically used to demonstrate analytical performance. Manufacturers should be aware that any aspects of analytical performance that are omitted due to their non-applicability require a justification.
How many can you answer?
- What research is available in relation to the association of the analyte and the clinical condition
or physiological state? Is the scientific literature peer reviewed? Are clinical guidelines from
professional associations or consensus expert opinions available?
- How established is the scientific validity of the analyte and its clinical application? Are other
devices available that measure the same analyte/biomarker?
- Have favourable and unfavourable publications been considered?
- Is the analyte, the technology, intended use and/or target population new? Are proof of
concept studies available? Are clinical performance studies available?
- Have all relevant tests been carried out? (e.g., sensitivity, specificity, interference, LoD,
accuracy, specimen claims)
- Have relevant standards and guidelines been used? (e.g., ISO, CLSI)
- Have certified reference materials or reference methods been used? (e.g., trueness). Is the
- Is the approach used to gather clinical performance appropriate considering aspects such as
novelty, established device or test standardisation?
- When used, has method comparison been performed against the state of the art (e.g., against
another CE marked device)?
- Has clinical performance been demonstrated to prove the device is safe and effective for the
intended use claims?
Clinical performance studies tend to focus on the patient, the clinical condition or physiological state. The depth of clinical evaluation is therefore proportional to the device risk, novelty, intended purpose, and specific claims.
Data can be sourced from a combination of clinical performance studies, scientific peer reviewed literature and/or published experience from routine diagnostic testing.
The IVDR does not explain what is meant by published experience gained by routine diagnostic testing. However, this is this is understood as being data from proficiency testing schemes, external quality assessment (EQA), ring-trials (inter-laboratory comparison), amongst others.
For a well-established device that was self-certified under the IVDD and did not undergo actual clinical performance studies, peer-reviewed literature or published experience may be the only source of clinical performance data available to the manufacturer. This may also be the case for other legacy devices.
Since there is no grandfathering under the IVDR, notified bodies will typically expect to see actual clinical data being generated through clinical performance studies, especially for higher risk devices, even if they had been on the market previously. Manufacturer preparedness is everything and any potential gaps should be addressed before submission to the notified body.
General safety and performance
The PER should also assess all relevant performance data and demonstrate overall conformity with the General Safety & Performance Requirements. Risk management and usability is an integral part of the GSPRs, and so manufacturers must demonstrate that they have considered the final intended user and use environment.
For instance, for near-patient tests, where the test is performed outside the laboratory, it would not be acceptable to demonstrate clinical evidence in the PER solely based on data obtained with professional laboratory users.
An ongoing process
Performance evaluation is not a one-off process. The IVDR requires that performance evaluation is continually updated throughout the lifetime of a device. It is therefore highly recommended that manufacturers implement suitable, standardized processes to make updating as easy as possible. For a deeper dive into this topic, download our white paper "Performance Data and EU IVDR: Improving operational efficiencies through compliance".
If you’re unsure where to start, RQM+ can provide project management and technical leadership, to strengthen your team and processes, and ultimately enabling you to stand on your own when you’re ready. We are continually developing effective PERs for our clients, building in our learnings, as well as the latest advice and best practice. Book a consultation with us - we’re here to help!