How safe are new medical devices?

Abstract

In this article, I identify the peculiar challenges of current regulation in the UK to assess the safety of new medical devices. Not only is there a limited role for the regulatory authority in assessing their safety, but also no clinical investigation might be needed before many new devices can be marketed for use in populations across the European Union. As a lay member of a committee flagged to review research involving medical devices, I describe some of the difficulties we face in this regard as well as recent developments designed to raise regulatory standards.

Introduction

In the UK, both medicines and medical devices are regulated by the Medicines and Healthcare products Regulatory Agency (MHRA) (which in turn implements separate European Directives). However, there is more dissimilarity between the regulations than commonalities. This article aims to highlight these differences and to raise unanswered questions regarding safety assessment of new medical devices in the UK. Before discussing these differences, I will first seek to define a medical device, indicating the various types (there are three) and classes (four).

Device regulations in the UK

The main UK legislation governing medical devices is Medical Devices Regulations 2002 (S.I. 2002/618). This has been amended three times (S.I. 2003/ 1697, S.I. 2007/ 400 and S.I. 2008/ 2936) and brings the following European Directives into UK law: the Active Implantable Medical Device Directive (90/385/EEC), which covers powered devices which are implanted into the human body; the In Vitro Diagnostic Devices Directive (98/79/EC), which regulates testing kits and so forth; and the Medical Device Directive (93/43/EEC), which covers all other devices and thus the widest range of devices from bandages and tongue depressors to prostheses.

Each Directive can be considered to deal with a different type of device, but in general devices are defined as in Article 1 of Directive 93/43:

2(a) any instrument, apparatus, appliance, software, material or other article, whether used alone or in combination, including the software … intended by the manufacturer to be used for human beings for the purpose of:

diagnosis, prevention, monitoring, treatment or alleviation of disease;

diagnosis, monitoring, treatment, alleviation of or compensation for an injury or handicap;

investigation, replacement or modification of the anatomy or of a physiological process;

control of conception;

and which does not achieve its principal intended action in or on the human body by pharmacological, immunological or metabolic means, but which may be assisted in its function by such means …

Medical devices fall into one of four classes. Class I devices neither enter nor interact with the body and, if the device is both non-sterile and non-measuring, it may be subject to a self-declaration by the manufacturer that the device conforms to the requirements (in such cases there will be no unique registration number otherwise associated with the CE marking). Class IIa and IIb devices are those that are invasive or non-active implantable or interact with the body; and Class III devices (which include all active implantable devices) are those that affect the functions of vital organs. If there is any uncertainty over classification, the rules regarding the stricter class are taken to be applicable.

Role of the regulatory authority

If there is any confusion about whether a drug-device combination should be regarded for licensing purposes as a drug or a device, then in practice the MHRA will decide the matter. The MHRA, which is the Competent Authority for both medicines and medical devices, does not, however, approve medical devices for the market as it does with medicines. Instead the MHRA approves ‘notified bodies’, which in turn assess and certify a device’s conformity to the ‘essential requirements’ necessary for approval for the relevant class of the appropriate type of device. These essential requirements relate to the safety of the device and are expressed in terms of some half-dozen or so broad technical principles to do with performance criteria.

These notified bodies are private certification organizations of which there are six in the UK and approximately 75 in the EU. They provide a public regulatory function by reviewing submitted data and inspecting any manufacturing processes involved. Each notified body may approve those classes of medical device for which it has MHRA authorization. Approval entitles the manufacturer to a certificate of conformity for the device – the CE mark – which in turn enables the device to be placed on the market throughout the European Economic Area.

Proof of the efficacy of a device is not regarded before a new medical device gains a licence for the EU market. So long as a notified body can be satisfied that the device is fit for purpose it will award the mark.

Clinical investigation of safety

To get approval for a new medical device – even a Class III device – it may not be necessary for any clinical investigations to be performed and thus there will be no role for an ethics committee. In order to demonstrate a new medical device’s conformity to the essential requirements appropriate for the device it may be possible merely to rely on a review of the scientific literature, and this is permissible (see Annex X of Directive 93/43) where the new device is equivalent in terms of product construction to an existing device and where the operator adopts the same procedures as apply to an existing device at the same anatomical site(s) and in the same type of patient. This is also the situation in the USA, where the Food and Drugs Administration (FDA) will accept the demonstration of ‘substantial equivalency’ without a new clinical investigation – the 510(k) route.

Where equivalency cannot be demonstrated though, the evidence of more extensive testing to demonstrate both the new device’s safety and – in a departure from European practices – its effectiveness for intended use will be required (Fraser et al., 2011). When a literature review would not be adequate and the higher the class, the higher the risk involved, the more likely it is that human studies will be needed to support an application for approval to establish safety.

Clinical investigation is needed for a new medicine to gain a licence for market, and such trials must legally first be reviewed by a recognized ethics committee, but some new medical devices can wholly escape the ethics review process and still attain full market access.

Ethics review of clinical investigations

When a clinical investigation is required, there are currently some 14 NHS Research Ethics Committees (RECs) which are ‘flagged’ to review research involving medical device applications within the UK National Research Ethics Service (NRES) portfolio, and whilst technically flagging is only an administrative mechanism for developing committee expertise and any NHS REC is legally competent to assess such research, both NRES and the MHRA prefer to steer applicants towards flagged committees.

An NHS ethics review can still, however, be avoided in a number of ways. Firstly, investigations can be conducted on non-NHS patients (and perhaps using a non-NHS committee). For example, healthy volunteers could be used to test a new sticking plaster or contact lens. However, any physicians would still have to adhere to the rules of the General Medical Council (GMC), which requires an ethics review of all research, whether in the public or private sectors.

If NHS patients are recruited, it is still possible to circumvent the ethics committee where the manufacturer (who may be a single clinician) fashions a device specifically for one of his or her own patients (a ‘custom-made device’). If the clinician does this sort of thing every ‘now and again’ for particular patients, and gets good results, it may occur to the clinician that s/he ought to make the device more widely available. The experience with the ‘custom-made’ devices effectively provides ‘proof of concept’ evidence, and a retrospective analysis of the data from these ‘natural experiments’ could conceivably be used to support certification. If the manufacturer then wishes to broaden the use of the device, he or she might teach colleagues in the same hospital trust (i.e. the same legal entity) to use the device on their patients, in which case NHS Research & Development permission may be needed and this may, but need not, trigger NHS REC approval – it will depend on whether those involved regard the wider use as research.

When the MHRA is involved, the REC can be reassured that the safety of the investigation will have been independently assessed by experts appointed by that body as to the research’s scientific validity and the safety aspects of the study (see, for example, the Integrated Research Application System’s (v.3.3) Part B question 14-6, ‘Has the IMP been the subject of scientific advice related to this clinical trial?... From a M[ember] S[tate] competent authority?”). However, when the MHRA is not involved, the REC may need to look very carefully at the safety of the proposed investigation itself.

Yet the REC will almost certainly have to recognize that it is not constituted to perform such a task. Its ‘expert’ members may know nothing about the morbidity within which the device attempts to seek a role. It may have to assure itself that the proposal has been subject to appropriate, independent scientific appraisal and scrutiny of safety, perhaps commissioning an independent expert to verify these aspects of the study. It is for the REC to satisfy itself of the adequacy of arrangements for the device’s safety, the suitability of its design and its material composition (which will probably necessitate a review of the pre-clinical data and perhaps the arrangements for sterilization and training). This is not a job for amateurs.

The first clinical use for a device can be after CE-marking, perhaps when the device appears for ethics review as a ‘post-market review’. These studies will not necessarily be clinical trials at all. Instead, there may be what is termed an ‘investigation’ because there need be no comparator or any complex statistics, and the purpose of the data is merely to demonstrate that the device can do what is to be claimed for it.

Environmental concerns

In another indication of just how fast-changing the medical device review environment can be, a potential new area for ethics review of medical devices is suggested by the Restriction of Use of Certain Hazardous Substances in Electrical and Electronic Equipment Directive. This Directive (2002/95/EC) came into force on 1 July 2006 and placed obligations on the producers of equipment to reduce the negative environmental impacts of that equipment. The Directive, however, exempted producers of medical devices from the obligation due to concerns that the producers had especially complex needs to ensure patient safety which trumped immediate environmental concerns. However, the resultant moratorium on this exemption looks likely to be removed in the near future (Townsend and Cofré, 2011).

Ethics committees may wish to anticipate the legislation by asking about environmental considerations at the earliest possibility, as it may be too late to alter a device’s materials once the device has been granted CE marking. If more environmentally friendly considerations can be brought to the attention of researchers and developers, then it is possible that any clinical investigation will not have to be repeated merely to gather data to support the development of a medical device solely because of a predictable need to change its material composition.

Rules for clinical trials of devices

The situation as described above was much the same in terms of IMPs before the International Committee on Harmonisation – Good Clinical Practice (GCP) standard was established, and not surprisingly there is now a move to develop a similar standard for medical devices to GCP so that what is required to meet the regulatory standards in the different world markets can be met by adopting a harmonised approach. This is the aim of BS EN ISO 14155, the latest version of which was published in March 2011. As this standard develops and gains acceptance it will require the use of clinical investigations in situations where they can currently be avoided. This should ensure the MHRA’s involvement, and as such may ease the science/safety concerns from the REC. But unless the custom-made route is to be tightened up too, then any scrutiny of such devices could still elude the RECs or is likely to remain a potentially difficult area requiring expertise uncommon on a REC.