Regional Differences in Medical Device Development

The regulatory expectations for data necessary for pre-market product approval in the EU are very different to those in the USA, especially in terms of the clinical data required.  This has introduced significant differences in the time to market in the USA compared with the EU, particularly in the case of high-risk devices (e.g. Class III and implantable Class IIb).  As a result, some US physicians have experienced one to three year delays in gaining access to certain important technical innovations such as drug eluting stents or percutaneous heart valves, compared to their European colleagues(1). This issue was the subject of an earlier article by Sarah Sorrel(2), which explained why these timelines can differ so extensively and provided an in depth examination of the clinical data requirements for market approval in the two systems. The present article revisits the issue and speculates on the underlying rationale for the regulatory differences.  More particularly, it brings the subject up to date by analyzing the recent changes in the European Medical Device Directives(3) to investigate whether the gap is widening or narrowing.

Medical devices are regulated in the European Union by three EU Directives(4). The main directive, which covers the vast majority of medical devices from surgical gloves to life sustaining implantable devices such as heart valves, is the Medical Devices Directive (MDD). The MDD is supplemented by an earlier directive on active implantable medical devices and a later directive on in vitro medical devices. These three EC directives are not directly binding on manufacturers, but have been “transposed” into the national laws of each of the 31 Member States that comprise the European Economic Area (EEA). The result is a legislative framework within which many slight regional differences are evident, as dictated by national medical device laws.

The Medical Device Directives are what are known as “New Approach” directives. The underlying principle of the New Approach is that, in order to permit the dismantling of barriers to the free movement of goods and services across national boundaries within the EU, it is necessary to introduce commonly agreed requirements that guarantee adequate safety and appropriate performance for those goods or services. The bulk of responsibility for this guarantee lies, not with government bodies, but with whoever places the product on the market. Examples of other products regulated by New Approach directives include pressure vessels, toys, and personal protective equipment. The basis of market approval in Europe, for a wide range of products, is thus the declaration by the manufacturer, that the product meets certain relevant requirements.  Products that are so certified can carry CE-marking and be placed on the market and circulate freely throughout the EEA.

Each of the Medical Device Directives therefore contains a list of “Essential Requirements” for safety and performance that must be met by any product falling within its scope. Essential Requirements are very general in nature, setting out the basic principles by which conformity can be assessed.  It is left to technical standards, which are consensus documents largely drawn up by industry, to set out detailed guidance on how Essential Requirements can be met in particular circumstances.  European standards that have been identified by citation in the Official Journal of the European Union (so called “harmonized standards”) carry the presumption of conformity with relevant Essential Requirements.   Demonstrating compliance with harmonized standards is not obligatory, but is usually the most convenient and effective route to demonstrating conformity with Essential Requirements.

The role of independent certification organizations (Notified Bodies) in the process and the possible routes to certification (e.g. requirements for quality systems or design dossier review) are critical to the process and depend on the risk category of the product. For low risk devices (Class I) such as a tongue depressor or colostomy bag, the manufacturer can self-declare conformity with the Essential Requirements. For medium to high-risk devices (Class IIa, IIb, and III devices and Class I devices with a measuring function or supplied sterile), the manufacturer must call on a Notified Body to assess conformity. The manufacturer has a choice of methods available for the conformity assessment of the device and/or manufacturing system. The end result is a certificate of conformity that enables the manufacturer to apply “CE marking” to the product.

The use of the New Approach represents a major difference from the arrangements for market approval for pharmaceutical products, for which the assessment of conformity with the regulations is conducted by a government agency, i.e., a Member State Competent Authority for drugs or the European Medicines Agency (EMEA). Notified Bodies are designated, monitored, and audited by the relevant Member States via the national Competent Authorities. There are around 80 Notified Bodies within the EEA. A company is free to choose any Notified Body designated to cover the particular class of device under review. This move away from government control towards self-regulation is justified by the generally lower level of risk associated with medical devices and the less specialized nature of the assessment normally required.  This is in line with one of the fundamental tenets of the European regulatory system, that of proportionality.  The more relaxed government control of market entry for medical devices is, however, balanced by a more hands-on approach by Competent Authorities in the postmarket phase. 

What clinical data is required by the EU medical device regulation and how does this differ from US FDA requirements?  There is no one answer to the first question; it all depends on the circumstances. The recent changes to the Directives3 have emphasized the importance of the clinical risk–benefit assessment (“clinical evaluation”) in CE-marking.  The need for, and nature of, clinical data arises from the requirement to demonstrate that a device is safe, that it performs as intended by the manufacturer, and that any residual risks are acceptable when weighed against the benefits of the device. The term “clinical data” within the meaning of the EC Directives is a broad concept.  Now defined as “the safety and/or performance data generated from the clinical use of a device,” it includes everything from pre-market clinical trials in human subjects, through postmarket clinical follow-up, to literature reviews analyzing the results of use of a product in comparison with other devices. The revised MDD requires that the clinical data used for conformity assessment must include data obtained from any combination of the following options:

• clinical investigation(s) with the device in question;

• clinical investigation(s), or other studies reported in the scientific literature, with a similar device for which equivalence to the device in question can be demonstrated;

• published and/or unpublished data on clinical experience with either the device in question or with a similar device for which equivalence to the device in question can be demonstrated.

The latter option, also referred to as the “literature route” has been commonly used by manufacturers for the CE marking of low to medium risk devices (Class I, IIa, and IIb) for which safety and performance can be adequately demonstrated by a combination of non-clinical data (i.e., bench testing and animal testing) and clinical data that already exists on the device (published or unpublished) or by analogy with published data generated in respect of an equivalent device.  It is now necessary, in such circumstances, to document a rationale to indicate why it is acceptable to use the literature route for a particular device.  Clinical evaluation based on a review of scientific literature remains possible only if:

• there is a demonstration of equivalence between the subject device and the device to which the data relate; 

• the data adequately demonstrate compliance with the relevant Essential Requirements;

• the data address the safety, performance, design characteristics, and intended purpose of the device.

The Global Harmonization Task Force (GHTF) has published useful guidance on data generated through literature searching(5).  A European Commission document contains guidance on the relevance of data obtained from comparator devices(6). The demonstration of equivalence is a key requirement for options 2 and 3 above.

The first option, i.e. the need to furnish clinical data from prospective clinical investigations with the device being developed, may be particularly necessary in the following cases(7):

• where there are significant residual risks for which further information is needed to clarify the risk estimate;

• where components, features and/or method of action are previously unknown;

• where an existing device is modified in such a way that the clinical performance and/or safety may be significantly affected;

• where a device uses new materials for which there is no prior clinical experience;

• where an established device is proposed for a new indication.

In practice, there is still a broad correlation between this list and Class III devices and implants, but the lack of a prospective clinical investigation must be “duly justified” for all classes and types of device.  Even when it is clear that the clinical data should come from a prospective clinical investigation, specific guidance on the criteria to be used is limited. Whilst some guidance is available in the ISO/ European standard on clinical investigations(8), this mainly concerns clinical trial methodology in general. Product-specific guidance for clinical evaluation is lacking for most devices, but a few product-based standards contain relevant requirements. It is not mandatory to follow the requirements of a standard but, where these exist, a rationale should be provided for any deviation from the methods set out in them. In the absence of specific requirements, the manufacturer must decide which data are sufficient for CE marking, i.e., number of subjects, type of study design, primary and secondary endpoints, type and schedule of assessments, minimum patient follow-up period, etc. The Notified Body may be consulted prior to initiating the clinical trial to verify that the assessor accepts that the protocol is likely to yield adequate data for CE marking.

As a basic principle, the objective of a CE marking trial is to demonstrate safety and performance. Therefore, the majority of these trials are non-randomized, single arm studies involving less than 100 patients for which the primary objective is to demonstrate safety. In our experience, CE marking protocols rarely include a study hypothesis and a statistical calculation of sample size even though this is a requirement of the European Standard on clinical investigations of medical devices8. Furthermore, since Notified Bodies tend to function in a closed manner, providing little visibility on criteria required for approval, this situation allows for variations in clinical data requirements despite the fact that guidance documents have been mandated by the Commission to ensure a “level playing field.”

The US approval process for medical devices is very different, especially in terms of the scope and size of clinical trials required for high risk devices. This is a consequence of the fundamental difference between European and US regulations, i.e. the difference between performance and efficacy. To receive approval to market a device in the EU, the manufacturer must demonstrate that the device is safe and that it performs in a manner consistent with the manufacturer’s intended use. To receive approval to market a high-risk device (Class III and some Class II) in the United States, the manufacturer must demonstrate the device to be effective as well as safe. This typically requires a prospective, randomized controlled, adequately powered clinical trial involving hundreds of patients.  It is becoming increasingly cost-effective to carry out such prospective studies in Europe, using FDA pre-market protocols in postmarket clinical studies with CE-marked products.  Such studies can also be used to generate the sort of data that has become indispensable for reimbursement purposes.

This significant trans-Atlantic difference can be illustrated by the example of distal protection systems used in interventional cardiology. The first such system to be developed (GuardWire–Percusurge, Inc., later acquired by Medtronic) was a specialized coronary guide wire with an expandable balloon at its tip. During a coronary angioplasty or stenting procedure, the device is inserted via femoral arterial access and the balloon is inflated distal to the lesion. After the balloon is expanded, the lesion is treated. The purpose of the device is to trap any dislodged arterial debris and thus prevent emboli. After the lesion is treated, any debris is evacuated by aspiration and the GuardWire is removed. In the EU, the GuardWire device was awarded CE marking by demonstrating safety and performance, i.e., ability to aspirate material during the stenting procedure, in a 22 patient single arm study(9). In the USA, this device was designated Class II and to demonstrate safety and effectiveness defined as the ability to reduce complications associated with stenting of saphenous vein grafts, the FDA required an 800-patient multicenter randomized trial comparing distal protection to the standard of care prevailing at that time (no protection)(10).

As a result, the GuardWire was available in the EU at least two years before physicians in the USA had access to the technology. One may argue that this early access must be weighed against the absence of proven efficacy.  On the other hand, once a technology has been shown to work and be safe, is it ethical to withhold it from patients?

Implications
An evaluation of efficacy, based on carefully designed and conducted clinical trials, is essential to any legitimate medical judgement on the appropriate use of a medium to high-risk medical device. Whether the clinical data allowing such judgements are collected before or after marketing approval is granted, remains one of the most significant differences between the regulatory processes in the USA and the EU. However, this regulatory gap has now been narrowed by the emphasis, in the revision to the Medical Device Directives, that market approval in the EU must be based on a risk-benefit assessment based on clinical data.

In practice, this emphasis has very little significance for high risk (Class III) devices.  There has always been an expectation, albeit arising from presumption rather than the wording of the Directives, that clinical investigations are necessary for high-risk devices but not for lower risk devices.  Even now, it should usually be possible to provide a robust rationale to justify conformity assessment of Class I and IIa devices without prospective clinical data.  The real difference is likely to be felt with Class IIb devices, for which the literature route has been almost totally eliminated

What has been perceived by some as a shortcoming in the EU regulatory requirements is further exacerbated by the fact that 1) “real” clinical data in the form of results from prospective clinical investigation is still not mandatory for medical devices, even those that are relatively “high-risk” and 2) Notified Bodies are not on a level playing field in terms of their criteria for approval.

The amendment of the Medical Devices Directive and the Active Implantable Medical Devices Directive has strengthened the language concerning clinical data requirements. However, as predicted in the previous article(2), this has not fundamentally altered the spirit of the requirements which remain the demonstration by the manufacturer of safety, performance, and acceptable risk–benefit ratio. On the other hand, Notified Bodies are increasingly requiring effective postmarket surveillance and commonly expect to see post market surveillance plans, which include postmarket clinical follow-up(11) for certain types of devices.  The increased scrutiny of clinical evaluation should result in a more uniform approach across Europe and, ultimately, increased public confidence in the European regulatory system.

Jeremy Tinkler, is Director of Regulatory Affairs, MedPass International SAS, 95bis Boulevard Péreire, 75017 Paris, France, +33 (0)1 4212 8330, email: [email protected]

References

1. Kaplan AV, Baim DS, Smith JJ, Feigal DA, Simons M, Jefferys D, Fogarty TJ, Kuntz RE, Leon MB Medical device development : from prototype to regulatory approval.

Circulation

. 2004 ; 109 : 3068-3072 2. Sorrel S; Medical Device Development: U.S. and EU Differences; Applied Clinical Trials, August 2006. 3. Council Directive 2007/47/EC of 5 September 2007 amending Council Directive 90/385/EEC on the approximation of the laws of the Member States relating to active implantable medical devices, Council Directive 93/42/EEC concerning medical devices and Directive 98/8/EC concerning the placing of biocidal products on the market. 4. Council Directive 90/385/EEC of 20 June 1990 on the approximation of the laws of the Member States relating to active implantable medical equipment; Council Directive 93/42/EEC of 14 June 1993 concerning medical devices; Council Directive 98/79/EC of 27 October 1998 on in vitro diagnostic medical devices. 5. Global Harmonisation Task Force (GHTF); Clinical Evaluation; Document SG5/N2R8; 28 June, 2007. 6. MEDDEV 2.7.1: European Commission – Guidelines on Medical Devices – Evaluation of Clinical Data: A guide for manufacturers and notified bodies. 7. Co-ordination of Notified Bodies Medical Devices (NB-MED) on Council Directives 90/385/EEC and 93/42/EEC, Recommendation NB-MED/2.7/Rec 1, Guidance on Clinicals, Rev. Nr.2, 20 April 1998. 8. EN ISO 14155 Clinical investigations of medical devices for human subjects, Part1: General requirements; Part 2: Clinical investigation plans. 9. Webb JG, Carere RG, Virmani R, et al. Retrieval and analysis of particulate debris following saphenous vein graft intervention.

J Am Coll Cardiol

. 1999; 34: 468-475 10. Baim DS, Wahr D, George B, et al. Randomized trial of a distal embolic protection device during percutaneous intervention of saphenous vein aorto-coronary bypass grafts.

Circulation

. 2002; 105: 1285-1290 11. MEDDEV 2.12.2: European Commission – Guidelines on Post Market Clinical Follow-up (PMCF).