To date, manufacturers have had three choices when completing a clinical evaluation report (CER): 1) performing a clinical study, 2) performing a clinical literature review, or 3) a combination of performing a clinical study and literature review. However, with the recent, complete re-write of MEDDEV 2.7/1 (Rev 4), it is now more difficult to use the “literature review only” path. Revision 4 additionally introduces strict requirements for demonstrating equivalence between your device and another. As such, it has become necessary for many companies to update their CERs in order to come into compliance with this guidance document. Changes and clarifications to MEDDEV 2.7/1 include:
CER updates
The frequency of updating a CER must now be defined and justified. There are numerous issues to consider while making this justification: human factors, usability, risk level, degree of innovation, and level of confidence are examples. However, devices with significant risks (e.g., implants) now require at least an annual update to the CER. For devices with non-significant risks; and where the device is well established (e.g., a long clinical history), 2-5 years is the possible frequency range. An update frequency of longer than 5 years is not permissible. Additionally, clinical evaluation throughout the device lifecycle is a key aspect in Revision 4: when new information from post market surveillance (PMS) affects the CER, it is expected that post-market data will be fed into the document in a timely manner.
It is likely that the increased frequency with which evaluations will need to be updated will lead to staffing issues for many medical device manufacturers.
Literature searches
In order to establish current knowledge/state of the art and to obtain data for the device in question, a search strategy must be thorough and objective. As always, Pubmed/Medline is a good starting point, but an EMBASE (Excerpta Medica dataBASE) search may need to be included in order to ensure appropriate coverage of EU journals.
Qualifications of clinical evaluators
Requirements regarding the qualification of the evaluators of clinical data have increased. Competence of Notified Body (NB) staff is also under increased scrutiny. Evaluators should have an extensive knowledge base in subjects such as clinical study design, biostatistics, information management, regulatory requirements, and medical writing. Evaluators also need knowledge specific to the device, its technology, its application, and must now make a declaration of interest. As such, evaluators must have an advanced degree and 5 years of experience, or 10 years of experience if they do not have an advanced degree. These qualifications must be documented, both in your procedures for, and in the actual clinical evaluation submission.
Demonstration of equivalence
In the past, it was appropriate to list several devices similar to the device in question and include those devices in the clinical literature search. Now, equivalence can only be based on a single device. This means all three general criteria (clinical, technical, biological) must be fulfilled by a single device. Equivalent devices being compared should be of the same material and should be for the same intended use and clinical indication as the subject device. If non-CE marked devices are claimed equivalent, differences in patient population or clinical practice between the jurisdictions where the product is approved and the EU must be justified. Revision 4 also requires the NB to investigate the manufacturer’s access to credible information to substantiate the validity for device claimed equivalent to the subject device. This shift will require a manufacturer to have a contract in place allowing access to data for competitor devices to which equivalence is claimed.
Summary of Highlights:
- Rev 4 is a complete re-write of MEDDEV 2.7/1.
- The increased frequency with which evaluations will need to be updated may lead to personnel issues for device manufacturers.
- A quality system update is necessary to define and justify changes to clinical evaluation procedures.
- It is harder to demonstrate equivalence now – basically, one must own the technical file for the equivalent device.
- Requirements for the expertise of the evaluators are clearer and have increased and expectations of the NB are greater now: hence more in-depth reviews.
- Medical writers may need to include an EMBASE search.
- Post-market clinical follow-up investigations may be needed more often. Clinical studies must be sufficiently large to have a reasonable probability of seeing undesirable side-effects. This will probably lead to an increase in the number of first in human and pivotal studies but also with regard to the number of subjects required.