So naturally my interest was seized by an article in the local newspaper, about the problems a nearby children's cancer charity was having with its research programme. There were dire warnings that they would have to stop their research or even close down, because they could not afford all this new bureaucracy. Not being one to hold back when there's a chance of controversy, I bashed out a letter to the paper asking whether patients should expect lower data quality standards for drugs developed by universities, than they quite rightly expect from pharmaceutical companies. The doctor who heads the charity eventually replied, basically saying that these trials were absolutely necessary and we should get them done any way we can.

This raises a number of questions. Firstly, now that the Directive and its follow, up the Good Clinical Practice Directive, have both been enacted by member states in national legislation, has it been the death knell for academic clinical research? The answer of course is, no, it hasn't. True, changes were made to smooth out some of the problems impacting on non-commercial trials, but the basic principles of accountability and good practice remain. So now we have the level playing field the Euro-legislators wanted - or have we? My direct experience of working with the legislation is that it's far from level. In fact, working as I do on multinational trials, it is the differences between the EU countries that are most striking. Let's look at ethics committees first.

Ethics committees

The EU Clinical Trials Directive established clear terms of reference for ethics committees. Members now have to be formally trained, and there are strict time limits for responding to submissions. This you might think should ensure common standards.

But at present I am involved with a global phase III programme with sites in North and South America, Asia and seven EU member states. Suddenly, one committee has uniquely issued an outright rejection, on study design grounds. How can this possibly make any sense? We have around 70 investigators, all with their own local ethics committees, and just one committee claims to have special scientific insight that nobody else possesses. Note also that this is non-negotiable - they are saying they are right and all the other committees are wrong.

So it is clear that we are a long way from consistency in the way ethics committees operate. In the UK, efforts have been made to approach consistency, starting with the two-level approval of multi-centre trials. The key feature of this was that the upper level of review, by the Multicentre Research Ethics Committee or MREC, addressed patient protection and scientific issues.

Only one MREC submission was needed, and after that an approved protocol went to local committees or LRECs. These could only consider local issues, in particular whether the investigator had the expertise and facilities to carry out the study. They were not allowed to reject a protocol on scientific grounds. This system worked reasonably well (although not all readers will agree), but on 1 April this year it was rolled into the National Research Ethics Service, part of the National Patient Safety Agency (NPSA). This is still a two-tier system, with the same aims, but where previously the choice of MREC was based on geographical criteria, now any ethics committee 'recognised' by NPSA can receive submissions, via a centralised process. The recognised committee does not issue approvals as such, it expresses a 'favourable opinion', after which local capability is assessed by a 'site specific assessment' (SSA) if there is more than one investigator site. This can be carried out by a local ethics committee or by the institution's (e.g. hospital or university) research and development committee. Indeed, approval by the local R&D committee, which has become mandatory at most sites over recent years, has now been combined with ethics submissions in the centralised process at the NPSA's National Research Ethics Service.

European variety

This quick review of the UK process illustrates my point: none of this detail is in the EU legislation. It doesn't even say who can serve on an ethics committee. We are accustomed in the UK to seeing at least one lay member listed but, for example in Italy, I know committees who have none. Members there are nearly all academics. Neither is there any mention of how many members there should be, or who can apply to a committee (in the UK it must be the investigator), or whether there should be a two level system. It was always the EU's intention that member states should fill in these gaps with their own laws in this way, but the inevitable result is huge variation across Europe. Can we be surprised therefore if some committees like to flex their muscles?

Of course, ethics committees are just one of the hurdles we have to jump to get a study started, and there is a complex web of interfaces to manage. In all member states, for drugs not yet marketed, we have to apply for regulatory approval, generally termed 'competent authority' (CA) approval. The CA normally also handles import licences for study drugs manufactured outside - which will be the case for all (or nearly all) countries in most multinational studies. All clear so far. But then there's the need for insurance, which most member states regard as something for the ethics committees. This is because insurance is needed against damage to patients participating in the trial, on a 'no fault' basis (so that patients don't have to prove negligence). Despite being an EU member state, Lithuania I have found is different, because the CA there wants the insurance policy submitted before it will approve the study, unlike other member states in my experience. This has a major impact on project planning, because there is no possibility of overlapping CA approval and insurance tasks - the latter has to come before the former. It might make you think twice about running studies in that country.

But please don't take any of this as a manual for how to do trials in Europe, because local practices are constantly changing. By the time you read this they could well be different. I had some experience last year with a major CRO, which had made great efforts to get planning templates in place for all countries. These set out all the tasks required, with estimates of time scale for each task. This is what I recommend when training people in how to plan clinical trial projects, so they were doing things right. Project management best practice says that, once an initial plan has been set up, it needs to be circulated to the project team for review. I remember that the Italian team member rejected the plan, saying that the time scales for approvals were all wrong and were now much longer. For a CRO this can be disastrous, if the contract has been agreed on the basis of outdated templates. It would also be bad news for a sponsor, especially one which has investors to keep happy.

The UK picture

I want to return again to the subject of UK legislation. This is because it appears to be the most rigorous EU member states when it comes to national implementations of regulations (see box overleaf). For example, both ICH guidelines and the EU Directives require sponsors to report serious non-compliance of Good Clinical Practice, but UK law (specifically SI 1928) is much more specific, setting strict time limits for reporting what are termed 'serious breaches'. This extends to sponsors' contractors, whose contracts must include obligations for such reporting, and to specific training. It is not good enough for sponsors to tell the regulator the MHRA that they have delegated most of the work to contractors. They are as responsible for what the contractors do as if they were doing it themselves. Therefore, an MHRA inspector will ask whether the sponsor has audited the contractors, whether the contractors' standard operating procedures (SOPs) include the mandatory functions, and whether the contracts oblige the contractors to comply with the legislation. For international studies this raises issues with contract wording. Is it really practical for a contractor that's US-based, for example, to include in a contract for a global study all the specific obligations required by a multitude of national laws? I am finding so far that some US contractors have very little knowledge of UK legislation, and are hoping to get away with rather vague 'catch-all' wording. I see their problem.

Living and working in the UK as I do, I am most aware of how the MHRA is approaching all this. I mentioned inspections, and since the new legislation came in there seems to have been a big escalation of these. Inspectors are no longer looking first at more prominent sponsors (if they ever did) - sponsors of all sizes and types are now receiving the dreaded letter. Not only that, but all kinds of contractors are being inspected; not just clinical CROs, but specialist providers of technologies and services. So beware - sponsors could have problems if their own inspection goes well but one of their contractors is running a poor ship.

This last point has implications particularly for small companies such as biotechs. These operate almost entirely by contracting out whole programmes. An inspection for a small company is a double-edged sword. On one hand there is less scope to be inspected and it may not be necessary to have a full range of Standard Operating Procedures (SOPs) if contractors' SOPs are used. But be sure that it's not just the SOPs that form the basis of the inspection, it's the legislation. That means the small company has to tick the same number of boxes as the big one, a burden that such companies have complained vociferously about. Some of these companies may be developing drugs with small markets - orphan drugs - and they are finding this level of bureaucracy a disproportionate burden.

I have seen many reports suggesting clinical research is drifting away from the UK. What is driving this is not easy to identify - there must be many factors, including a lack of willingness among NHS trusts to collaborate with industry in creative ways (e.g. in recruiting patients). I think it is too early necessarily to blame the UK legislation for this. UK sponsors that I work with still seem to be placing studies in the UK, and I continue to work with US sponsors who do the same. A stronger engine is the move eastwards from Europe in general, into Asia, which can offer large populations of treatment-naïve patients.

Beyond ethics and regulatory approvals

It goes almost without saying that I could write many pages about how EU member states have interpreted the Directives in their own ways. I have chosen regulatory and ethics approvals largely because they are uppermost in my mind, resulting from very recent experience. There are many other examples of practices which vary horizontally (across national boundaries) and longitudinally (over time). This must have a substantial impact on our ability to complete drug development projects on time. Indeed I have commented in a previous article on the disappointing metrics describing this process. Clearly it is in the public interest to re-examine the implementation of the Directives, with a view to achieving stability and consistency. However my focus is not really arbitrary, and surely considerable benefit would accrue from a common structure for regulatory and ethics approvals. The present variability gives the lie to critics of the EU who claim that national sovereignty is being eroded, and everything is being dictated by Brussels.

Also, it is possible for EU legislation to be challenged, and there is a recent example of that. The 2004 EU Physical Agents Directive would, if fully enacted, have made MRI scans largely impossible. This was because of a misguided attempt to protect the public against magnetic radiation. After an evidence-based campaign by scientists, the EU recently decided to postpone its implementation of this Directive. This shows it is possible to get politicians to listen to sense, and let's hope that they will do so regarding clinical trials.

Returning to the opening theme, the public rightly expects the same high standards for medicines development, whoever is doing the work, and wherever they are doing it. But the public is not best served by processes that unnecessarily delay the availability of those medicines to people who need them.

Box: A whistle-stop tour of European clinical research regulation

Good Clinical Practice first appeared as a set of guidelines issued by the US Food and Drug Administration in 1977.

Various guidelines appeared around the world over the next 20 years or so, issued by national bodies and lacking any legal force.

The first attempt at standardisation of practices was the International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH), dating from 1989.

This is a series of meetings and an organisation in its own right - see www.ich.org - and involving the three major markets of Europe, North America and Japan.

ICH has issued its own guidelines, and these form the basis of the EU's two directives, The Clinical Trials Directive and the Good Clinical Practice (GCP) Directive.

All EU member states were obliged to enact their own national legislation, and for the UK these are Statutory Instruments 1031 (1 May 2004), 1928 (29 August 2006), and 2984 (12 December 2006).

These together amended the 1968 Medicines Act. SIs 1031 and 1928 contain the major part of the new legislation. Until the EU Directives appeared, GCP was only a standard - the Directives for the first time gave it legal force.

Les Rose is a freelance writer and clinical science consultant with Pharmavision Consulting. For more information e-mail: lesrose@ntlworld.com