Article 15 of eu clinical trials directive 2001/20/ec

Therefore, all interested parties, and especially economic operators, are reminded of the legal situation applicable after the end of the transition period Part A below. This notice also explains certain relevant separation provisions of the Withdrawal Agreement Part B below , as well as the rules applicable to Northern Ireland after the end of the transition period Part C below.

The UK government has so far ruled out such an extension. Nor does a free trade agreement remove customs formalities and controls, including those concerning the origin of goods and their input, as well as prohibitions and restrictions for imports and exports. This authorisation is also required if only part of the manufacturing e. Article 3 2 a is a typical textual formula that encourages aggregate risk analysis. It offers little protection to an IA.

It allows the risks of research to be offset by anticipated benefits of clinical care. No mention is made of clinical equipoise, nor it is stated that the risks should be necessary, proportionate or minimised.

Article 5 i is problematic in other ways. In particular, it is not clear how it can apply in circumstances of clinical equipoise. Construed lit- erally, it sets a risk:anticipated benefit threshold that requires an expec- tation that patients will benefit from the test-drug to an extent that outweighs the risks of not receiving the test-drug.

Alternatively, there must be evidence to expect that the test-drug involves no risk which is very rare. Since IAs not given the test-drug receive standard available treatment, the literal reading means there must be an expectation that the test-drug is better than standard available treatment or is risk free. This is highly problematic if physician-researchers observe clinical equi- poise.

In such situations, there is genuine uncertainty amongst the profession as to which treatment—test-drug or standard available treat- ment—is better. Although a good trial is expected to be conclusive meaning one arm of the trial is expected to have a better outcome , there is substantial uncertainty as to which of the two groups this will be.

It is difficult to know why the principle of clinical equipoise was over- looked in this way. One explanation is that those drafting the Directive were strongly influenced by the pharmaceutical industry which, prefer- ring the ease of placebo-controlled trials, downplayed the importance of The problems and irrationalities of aggregate risk analysis were explained above: Section II-A.

For instance, an individual investi- gator could honestly claim to observe the principle of clinical equipoise it is measured against the views of medical community whilst simultaneously expecting that the test- drug will be beneficial. This may absolve a few investigators from legal liability, but it will not absolve the majority since, overall, the profession is uncertain whether the test-drug is better and it does not resolve the difficulties of legal interpretation for research ethics committees or Member State authorities.

The latter should adopt the objective stance of a reasonable hypothetical physican-researcher. The Helsinki Declaration was also going through a period of turmoil at the time.

A form of words to describe acceptable risks compatible with clinical equipoise did not emerge until the work of NBAC, Weijer and the framers of the Additional Protocol to the Oviedo Convention gained currency.

Member States must now handle the deficiencies of Articles 3 2 a and 5 i as best they can. One option is for investigators, Member States and ethics committees to ignore the principle of clinical equipoise when assessing the lawfulness of research with IAs. It need not be certain or likely in the eyes of the medical community; just possible. A chance of roughly 25 per cent might suffice. To reject the concept of equipoise would also be inconsistent with the Declaration of Helsinki potentially causing problems when researchers reach the point of publication and the Additional Protocol to the Oviedo Convention which is binding in many Member States.

Another approach would be to alter the frame of risk analysis so that rather than compare the test-drug against standard available treatment, one compares it with no treatment. Clarifica- tory words italicised Although this entails some stretching of language, we think not to an illegitimate extent. The words of clarification simply ensure that the provision coheres with clinical equipoise and component risk analysis.

The Helsinki Declaration endorses clinical equipoise and is referred to in the recitals of the Directive. Although legal instruments this side of the Atlantic have yet to insist upon component analysis, it was rec- ommended by NBAC and the logic of the argument speaks for itself.

The risk may range from trivial risks of chart review and additional blood sampling almost an invariable requirement in any drug trial to for example the radiation burden associated with serial X-ray or CT scan- ning to monitor the efficacy of a novel drug for treating cerebral should be offered the opportunity to participate in the trial at least they receive an unproven drug rather than unaffordable therapy , or whether this unreasonably exploits their economic vulnerability.

That said, we intend it to mean the same thing as minimal risk. A persuasive case for allowing minimal risk also emerges when one studies international ethical standards.

It agreed that research should be permitted where it poses no more than minimal risk and further recommended that research posing more than minimal risk should be allowed in excep- tional circumstances. Mayer et al. It proposed stricter scrutiny where the research involved risks more than minimal. See especially: 45 CFR Research Involving Incapacitated Adults equivalent to minimal risk, if the ethical arguments we outline in Section IV were observed. The only disagreement amongst them is whether anything more than minimal should be permitted.

The language used to define minimal risk has tended to vary, but this is not due to substantial disagreement about its merits or scope. It is less clear whether they allow bronchoscopy, spinal taps or cardiac puncture. Karlawish and J. A residual criticism is that the standard is too vague: NBAC , op.

McRae, Weijer , op. NBAC suggests that the general population be used as a benchmark because the risks faced in their daily lives are usually lower than those confronted by patients.

NBAC added a caveat that if the study nonetheless poses higher risk for any prospective participants then the ethics committee should insist on additional safeguards. Perhaps the European legal framework should allow research risks slightly above minimal provided such research is subject to more rigorous scrutiny. Further discus- sion of the ethical legitimacy of such a framework would be required. For example, in severe traumatic brain injury, is death a more serious risk for a patient than staying alive in a persistent vegetative state or with minimal consciousness?

Or in stroke research is one to regard having aphasia as more harmful than a paralysed arm? While the subjectivity and incommensurability of risk is not caused by the Directive, it is an issue that must nevertheless be tackled by Member States when implementing governance frame- works. Component analysis helps ameliorate the difficulties for ethics committees and competent authorities. The non-therapeutic com- ponents can, in most circumstances, be assessed straightforwardly by an ethics committee.

The risks associated with therapeutic components of the protocol need not be measured precisely. The two questions are: a does the medical profession have a bona fide belief that the net balance of risks and expected benefits in each trial arm are equivalent, meaning the research subjects will not be denied clinical care at least as good as standard treatment?

Permitted Investigations Notwithstanding sensible risk thresholds, further protections are needed to protect IAs from unfair studies. For example, unscrupulous research- ers might seek to enrol IAs in research projects because they find them more compliant, easier to enrol e. Amended Helsinki Declaration Article It states that research is prohibited unless it is essential to validate data [already] obtained in clinical trials on persons able to give informed consent or by other research methods.

The unusual phrasing is helpful because it prevents research until appro- priate preparatory studies with animals and other adults have been com- pleted. This issue is not explicitly observed by the Helsinki Declaration although ethics committees usually require it.

It states that research is prohibited unless it relates directly to a life-threatening or debilitating clinical condition from which the incapacitated adult suffers[.

The phrasing would have been better if it had simply used the familiar wording that research involving IAs is prohibited if it can be performed as effectively on competent persons. Interpreted narrowly it prohibits research into the general care of IAs that falls short of direct clinical treatment.

A relevant example is research into the rate or volume of ventilatory support in those being treated during or after cardiac arrest, the importance of which was highlighted in recent observational and experimental It is generally unproblematic to read it more narrowly—for example that persons suf- fering from a stroke must not be involved in research unless it relates to treatments for stroke or knowledge about strokes —because physicians rarely want to involve IAs in other research.

However, situations can be envisaged which demonstrate that the con- dition is not helpful. For example, a physician-researcher might propose a large-scale study to investigate the impact of socio-economic factors on the quality of proxy decision-making. He might propose to enrol a wide variety of IAs—some suffering stroke, cardiac arrest, sepsis, etc.

The issue here is not to ensure that the patient-sub- jects suffer from the same condition, but that they have not been chosen because they are easy to exploit. There should be an independent, demonstrably valid reason for including them. A further confusion to which this language may lead is an insistence that research relate to a condition which is caused by mental incapacity.

This is not what was intended and, to our knowledge, Member States have thus far avoided this reading. However, confusion arose in England during the passage of the Mental Capacity Act , and it is important that similar mistakes are not made by ethics committees. Some debilitating clinical conditions result in mental incapacity as a consequence of non-neurological disease or essential therapy. For example, patients with serious infections often develop severe respirat- ory failure.

In such patients, the severe systemic illness affects capacity, rather than the other way round. Thus, policies requiring the mental incapacity to be a cause of the condition studied would be unworkable. Another example arises when patients have artificial ventilation deliv- ered through a tube in the windpipe. This is usually very uncomfortable and patients are given strong painkillers and sedatives to make them more comfortable.

The condition studied arises independently of the incapacity, so a policy insisting on a causal relation would be problematic. Care must also be taken to avoid reading Article 5 e in a way that prevents research that seeks to study possible complications of incapaci- tating disease. Airway reflexes are compromised in patients with a depressed level of consciousness, resulting in an increase in the incidence of aspiration pneumonia.

Research aimed at preventing such pneumonia is clearly in the public interest and can only be studied effectively with the involvement of incapacitated patients, some of whom may not have developed pneumonia. A policy that insists that the patient belong to a community of persons suffering the condition would be problematic. Proxy Consent One of the most familiar legal and ethical standards for research with IAs is the requirement to seek consent for research from a person acting as the representative of the IA.

The CTD reflects this in Article 5 a. It poses three problems. Most significantly, the Directive fails to recognise any exceptions to the requirement for prior consent, which has caused severe problems for research into emergency and critical T. Aufderheide et al. Menon and K. We examine this issue below. A second problem is that those drafting the Directive shied away from defining who should be recognised as the legal representative of an IA.

A comparison is provided by Lemaire et al. In the Netherlands, a spouse or life companion may act as proxy. England, Wales and North- ern Ireland have adopted a much broader and highly unusual policy. The important caveat is that the doctor or NHS nominee must not be connected with the conduct of the trial. This means they must not be an investigator, a sponsor, a person engaged by, or acting in concert with, the sponsor or trial manager or under the direction or control of the sponsor, inves- tigator or collaborator.

Although some pluralism is justified, it pre- sents three sets of problems. In the first place, the various definitions impact on international trials making them cumbersome and unwieldy. In the main, it is simply not feasible for researchers to instigate legal proceed- ings before commencing research.

This is relevant not only to emergency research discussed below but also to research on diseases such as dementia and stroke. Third, it allows Members States to define legal representatives in an unethically broad way, irrespective of whether the representatives are appropriately qualified or subject to potential conflicts of interest.

Recent experience in the US after US federal regulations left the definition to individual States suggests these issues could be quite divisive. It ought to have focused on the purpose of seeking proxy-consent and defined legal representatives as including a current partner, close Lemaire et al. Following the Kantian ideal described below, the proxy should give some thought to principled autonomy, including obligations to reason rationally and not ignore the importance of research for other vulnerable IAs.

The proxy should be present for the duration of the research, ready to revoke consent if the burdens become too great in the same way that a competent adult might remove themselves from a trial. That said, the proxy should be regarded as holding a power of veto rather than being the primary risk assessor. If there had been some serious deliberation about the role of proxy- consent, consensus might have been possible in the CTD. We recognise that it is now unlikely that Member State governments will tackle this task.

The most practical course is thus to encourage countries with narrow definitions to appreciate the value of a definition of legal repre- sentative that does not require court involvement. Furthermore, countries following the English definition should allow doctors to give proxy-consent only in exceptional circumstances.

Harmonisation might in this way be indirectly achieved. A third problem stemming from Article 5 a concerns the revocation of consent.

Article 5 a states that consent to enrol an IA in a clinical trial may be revoked at any time. In some senses this is clear and pro- vides important safeguards. The proxy may order researchers to cease administering a test drug or placebo. They may also prospectively decide that no additional tissue or data be collected.

What is less clear A personal representative is more likely to know whether the IA has special idiosyncra- sies that are likely to affect the way they experience the research. For example, the mag- nitude of pain and anxiety may be much greater in an intellectually disabled person with a fear of needles than a year-old diabetic.

They may know that the IA has a duodenal ulcer, meaning an aspirin may unusually present more than minimal risk: Karlawish and Hall, op. Research Involving Incapacitated Adults is whether the legal representative has the power to order that the tissue and data collected up to that point be destroyed or not be used for research. For example, might the legal representative order that data about the patient recorded in tables and databases be erased or blocked? A power of this kind could create practical difficulties for researchers.

Some have also argued that it would introduce serious bias in research with critically ill patients on the assumption that survivors will be more likely than non-survivors to order the destruction of data. However, such arguments will be less persuasive where the requirements for proxy-consent have been waived or deferred see Section III-D , the data is sensitive, the analysis is likely to cause harm or substantial dis- tress, it is possible to separate the data collected before withdrawal and the risk of bias is small.

Research without Proxy Consent e. Results of clinical trials conducted only in the United Kingdom and results of multi-country trials where the United Kingdom was the only EU Member state where the clinical trial was conducted have to be submitted to EudraCT, also after the end of the transition period, if this is required for non-EU studies i. Article 41 1 of the Withdrawal Agreement provides that an existing and individually identifiable good lawfully placed on the market in the EU or the United Kingdom before the end of the transition period may be further made available on the market of the EU or of the United Kingdom and circulate between these two markets until it reaches its end- user.

The economic operator relying on that provision bears the burden of proof of demonstrating on the basis of any relevant document that the good was placed on the market in the EU or the United Kingdom before the end of the transition period. These pages will be updated with further information, where necessary. Further, France goes beyond the requirements of the Directive by requiring a final report to be drafted during the year following the termination of the trial.

Each of the member states' legislation reviewed also provides an obligation to inform the competent authority and the ethics committee of the end of the trial within 90 days. This deadline, however, is reduced to 15 days in the case of the early termination of the research.

As indicated in the Directive, the sponsor of the trial is in charge of informing both the ethics committee and the competent authority, and this was adopted in the domestic laws, including in Spain where the sponsor alone is in charge of notifying the end of the trial. Domestic regulations are similar in that they require the sponsor to file a new application if there are substantial amendments to the protocol, to receive a new favourable opinion and new approval. However, under Belgian legislation, the investigator is in charge of notifying the ethics committee of the reasons and content of the substantial modifications, and the sponsor is in charge of informing the competent authority.

The Spanish implementing legislation indicates, without providing further details, that the sponsor and the investigator must notify the modifications to the competent authority Spanish Drugs Agency and the ethics committee. Surveillance procedures of investigational medicinal products The Directive requirement that the sponsor must notify the competent authority and ethics committee of any suspected serious unexpected adverse reaction has been adopted in the various implementing laws of the member states reviewed.

This notification must be made immediately and, at the latest, within seven days of the date on which the sponsor became aware of effects which caused the death or endangered the life of a person, or within 15 days in other cases. However, there are significant differences, mainly between French law and EU law. For example, the Directive provides that "For reported deaths of a subject, the investigator shall supply the sponsor and the Ethics Committee with any additional information requested" Article French law does not provide an option for the investigator to communicate this type of information directly to the ethics committee.

The only possibility available to the French ethics committee in this context is to ensure "that participants in the research have been informed of the adverse reactions and have confirmed their consent".

French law differs as well with the half-yearly declaration of serious unexpected adverse reactions which the sponsor must send to the ethics committee and AFSSAPS, and which must contain not only the list of adverse reactions but also a summary description. Finnish law also differs from the Directive in that it requires information from the sponsor, not on the serious side effects, but on the evaluation of the safety of the medicinal product which is the object of the clinical trial.

Further, the Directive requires the sponsor to inform the competent authorities as soon as possible of any urgent safety measure taken to protect the participants against an immediate danger. On this point, English law requires the sponsor to inform the competent authority as well as the ethics committee within a maximum of three days of the measure, however, for example, the Belgian and German laws require notification to be made immediately.

Liability and insurance of the sponsor Many other differences relate to the regime governing the liability of the sponsor or the insurance which each sponsor should, in principle, take out.

The Directive does not actually deal with any liability regime. It provides that it is "without prejudice to the civil and criminal liability of the sponsor or the investigator" Article 19 , and only states that "any insurance or indemnity to cover the liability of the investigator and sponsor" should be taken into account by the sponsor and the committee Article 6. Despite this lack of any formal requirement, most member states have provided rules governing liability and insurance relating to trials.

French, German, Belgian and Spanish law expressly set up a liability regime that is favourable to victims and borne by sponsors, as opposed to English and Finnish regulations which have not provided anything. Under Spanish, French and Belgian law, it is even mandatory for the sponsor to take out insurance.

Further, in Spain, the sponsor is jointly and severally liable with the investigator, where the sponsor has not taken out any insurance or where there is insufficient insurance to cover injury. It is interesting to note that the Directive has created the concept of a "legal representative of the sponsor", obliging sponsors which are not established in the EU to appoint a legal representative within the EU.

This requirement has been adopted in all the domestic laws under review, in particular in French law, which provides that the sponsor "or his legal representative must be established in the European Community".

However, it seems that the legal representative under French law is just a spokesperson for administrative matters with the authorities.

This concept is referred to in the procedures for filing research applications but it is not used, for example, in the provisions relating to the sponsor's liability. Even so, the Directive clearly connects the concept of the legal representative with liability, which therefore goes further than just a spokesman for administrative matters and also causes the legal representative to be at greater risk.

Operators will therefore have to be especially careful in drafting their contracts, and it would be prudent to rigorously define this concept. The domestic laws contain three other provisions which are not contained in the Directive:. Insurance amounts. Under French law the minimum insurance amounts are:. The second difference is the limitation of the appeal periods against insurers. According to French law, insurance only covers loss for claims which have been notified within a period of time determined by the insurer, but which cannot be less than ten years.