Science and Technology CommitteeJoint written evidence submitted by The Institute of Cancer Research, London, and The Royal Marsden NHS Foundation Trust

1. Do the European Commission’s proposed revisions to the Clinical Trials Directive address the main barriers to conducting clinical trials in the UK and EU?

The EU Clinical Trials Directive has increased the regulatory burden on clinical trials, making them slower to set up and more expensive to run. The drafted revisions do attempt to ease the regulatory burden in some respects, but don’t go far enough and aren’t clear enough. In particular, the revisions make only a crude attempt at distinguishing between the different levels of risk that trials present, and so fail to properly reshape the directive’s flawed, one-size-fits-all approach to trial regulation.

Our specific comments are as follows:

Scope The scope of the directive is not effectively clarified. The definitions provided (clinical study versus clinical trial versus non‐interventional study) are confusing, which risks them being applied inconsistently across the EU and could even widen the scope of regulation.

Risk The draft introduces the concept of “low interventional” trials (trials using drugs within their licensed indications or where their use is considered standard practice), which could be eligible for lighter-touch regulation. All other drug trials are considered to pose the same level of risk—so use of a well-established drug in a slightly different patient group from the one it is licensed for is treated the same as use of a novel therapeutic supported by little safety data. But for regulation to be proportionate, it needs to take into account other factors influencing risk including the safety profile of the drug and the patient population involved. In the UK, the MHRA and academic community have already established a three-tiered system adapting safety reporting and other elements of trial conduct to the level of risk, and we would like to see this approach refined and adopted by the EU.

Streamlining authorisations and approvals: The draft describes a streamlined process for authorising clinical trials and approving modifications through a single application via an EU portal and a single decision within each member state. But it is unclear how this will work in reality because of a lack of detail. Establishing such a system is likely to be costly and time-consuming for regulators and sponsors, and the lack of detail could result in divergent implementation. Ensuring a robust and reliable IT system which is compatible with national infrastructures will also be a major challenge.

Safety reporting: The draft does little to reduce the burdensome and duplicative safety reporting requirements for sponsors. Instead the reporting burden is increased with sponsors required to report all serious adverse events (rather than just those that are unexpected and related) to participating investigators at the end of the trial, and to report all serious adverse reactions to the marketing authorisation holder. This will be onerous and difficult to comply with, particularly where combinations of therapies are being evaluated or generic drugs are used. Nor do the revisions allow for a risk-adapted approach to safety reporting.

Sponsorship and indemnity: The revisions make a number of proposals which we feel should make it easier to collaborate internationally. We welcome the formal introduction of co‐sponsorship, a model employed by non‐commercial organisations in the UK but not widely recognised by other member states. The draft also clarifies the role of the EU contact person for sponsors established outside the EU and introduces a national indemnity mechanism across member states.

2. What is the role of the Health Research Authority (HRA) in relation to clinical trials and how effective has it been to date?

The HRA only came into being on 1 December 2011 and it is too early to tell how well it is doing, although it does face some significant constraints. It has a separate set of responsibilities from the MHRA, whereas there had been hopes that those responsibilities would be brought together within a single organisation, with a clear remit for global R&D. The HRA also has responsibility only for England and Wales, and in Scotland approval mechanism for trials remain different. On occasion that has acted as a barrier to collaborations between researchers in different parts of the UK.

3. What evidence is there that pharmaceutical companies withhold clinical trial data and what impact does this have on public health?

It is difficult to provide clear evidence that pharmaceutical companies are burying data. However, some of our senior clinicians have expressed concerns over instances where publication of data has been delayed, and there is a more general perception that if a pharma-sponsored trial has produced negative results, there is less impetus to see it published. These concerns relate in particular to formal, peer-reviewed publication in a journal—trial data is normally presented in an abbreviated form at conferences where there will be public acknowledgment and discussion of the results.

Trials are sometimes stopped early because the benefits of a treatment appear to exceed pre-defined criteria. In these cases, sponsors should be encouraged to publish longer-term data in full and in a timely fashion. There are agreed procedures for deciding when to stop a trial early, but even so early analyses can be dominated by outcomes in a particularly sensitive sub-population, and striking early benefits may be attenuated once more mature results are available. If only the initial analysis of a trial is published, doctors and patients may gain a misleading picture of a drug’s benefits and side-effects.

Assessment of safety data is critical in determining the overall risk/benefit ratio of any treatment and lack of access to full trial data could have a major adverse impact on public health. If trial results finding little benefit for a drug or concerning levels of side-effects are not published, there is potential for patients to receive ineffective or even dangerous treatments.

4. How could the occurrence and results of clinical trials be made more open to scrutiny? Who should be responsible?

We would support some form of action to ensure trial results are published, in order to avoid deliberate or accidental publication bias. It will be important for regulators to monitor the expected completion dates of clinical trials, and any discrepancy between the number they have registered and the number which end up publishing their results. Ensuring registration on a trial registry such as clinical makes available the status of the trial and would normally provide an approximate timeframe for the presentation of initial or final trial data. Use of databases like this can increase transparency in access to data. A legal requirement to publish clinical trial data within a certain timeframe might also reduce the potential for delayed publication of unfavourable results.

Currently, even a full trial publication may only present selected safety data. Any potentially serious adverse event that could be related to the study drug should normally be flagged, but there remains the potential for under-reporting. Trial sponsors are expected to provide study data in response to requests from researchers conducting meta-analysis of trials, but there can be delays in doing so, sometimes for technical reasons such as the failure of the original consent forms to cover secondary research. Legislative pressure to make all safety data available could help ensure that unfavourable safety data is not withheld. One possible way to prevent under-reporting of adverse events would be for journal publications to provide the entire safety listing as a supplementary online file, for open public scrutiny. However, moves to provide open access to raw trial data do need to be balanced against the requirement to avoid any breach of patient confidentiality.

5. Can lessons about transparency and disclosure of clinical data be learned from other countries?

Clinical trials are conducted and published in an international arena, and in general concerns about access to trial data relate not to UK law but to the approach internationally. Some countries may place a greater emphasis than others on transparency of data as opposed to patient confidentiality, but if anything the UK is more open to data access than most.

But the UK’s regulatory authorities can play an important role. It should be possible to demand publication within a certain time limit for trials that have authorisation and there could be closer scrutiny of the completion rate of studies, which some analysis suggests is low. Regulatory authorities do continue to look at data on effectiveness and side-effects even after a drug has a licence, and must continue to be prepared withdraw a drug’s licence if concerns arise about its risk/benefit ratio.

February 2013

Prepared 16th September 2013