Science and Technology CommitteeWritten evidence submitted by King’s Health Partners’

Question 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?

1. Many of the changes from the CTD proposed in the text are welcomed; in particular, the acceptance of co-sponsorship, the national indemnity scheme and some of the risk adapted modifications. On balance, we welcome the replacement of the CTD with a Regulation as it will remove much of the disparity introduced by Member State interpretation of the requirements of the Directive. However we have practical concerns about how the transition will work which may need to be addressed, in the following areas. Key considerations from KHP:

(i)The Directive and Regulation are to run in parallel for three years. It is unclear how this will work. Sponsors will be keen to know which will take precedence during this period.

(ii)It is unclear how the ICH requirement will work. There is no mention of repealing the GCP Directive (2005/28EC) but the Regulation cites ICH GCP with a cryptic caveat “... provided there is no other specific guidance issued by the Commission and that those guidelines are without prejudice to this Regulation.” (Recitals; Clause 29; p 19)

(iii)It is unclear whether the parallel period starts in 2014 or 2016. In the Explanatory Memorandum 3.13; p11 it states that: “... the regulatory framework at EU level will be complemented by national laws.” It gives an example but a more comprehensive list of where this will apply would be useful. If national laws permit member states to variously tighten and relax the impact of the Regulation we may not be much better off than with the CTD.

(iv)Proposed reporting times—times for declaring start and end of trials are short and potentially onerous.

(v)Safety reporting requirements for Sponsors—there are requirements for Sponsors to make periodic reports to MA holders for IMPs. This will be difficult and resource intensive for non-commercial Sponsors.

(vi)Co-Sponsorship—The proposal to permit co-sponsorship is welcomed. To date most member states have not permitted co-Sponsorship. In the UK non-commercial Sponsors have successfully engaged in co-Sponsorship arrangements which allow sponsor obligations to be distributed between institutions. KHP routinely co-sponsors within the partner institutions and with third parties and welcomes the potential ability to co-Sponsor with institutions in other member states.

(vii)Risk Adapted modifications—The proportional approach to approvals and requirements for low risk trials is welcome. However, we have concerns that compiling the evidence (as currently required) to establish the risk of a given trial may prove onerous for the non-commercial sector. Consideration should be given to simplifying this process as it currently discourages researchers from attempting to have their studies classified as low risk.

(viii)In type A trials, we believe consideration should be given to adopting a system where no adverse event reporting beyond that normally occurring in routine clinical practice should be required—ie investigator sites and not sponsors should be responsible for submitting reports via Member States’ national pharmacovigilance reporting systems (In the UK this is the “Yellow Card” scheme https://yellowcard.mhra.gov.uk/hcp-form/reporterdetails/) and recording the event in the study CRF for the purpose of Data Monitoring Committee (safety committee) review, if one is established. Beyond this, any requirement to report adverse events centrally which do not meet “Yellow Card” reporting requirements would be at the discretion of the Trial Steering Committee/Chief Investigator and would depend on the purpose of the trial. The “Yellow Card” system is already in place and clinicians are fully aware of their responsibilities with regard to “Yellow Card” reporting. Currently, all SAE reports for low risk trials are collected centrally at sponsor offices and are filed there, typically with no SUSAR’s being identified at all, creating a cumbersome paper trail for no discernible purpose simply because the regulations require that the Sponsor, rather than the study site investigator, report SUSAR’s. Although this may be appropriate in Type B and Type C studies, it serves no useful purpose in Type A studies. Allowing studies teams to specify in protocols that “Yellow Card” reporting is all that is required is meaningful to site clinicians and simple to implement without the need for study specific documentation and procedures and central resource to track and monitor such events, which is particularly important in large, multicenter, pragmatic trials in routine clinical practice. However, the current model of specifying in the protocol which events do not need reporting is confusing for investigators and results in over-reporting.

(ix)We support the view of other academic organisations that more emphasis should be placed on the use of DMC’s and the benefits of DMC monitoring rather than collecting centrally individual SAE reports within sponsor organisations, where the IMP used has been on the market for a considerable time. Permission to use Yellow Card reporting should be an option on the Type A notification system, and possibly for some Type B studies, with the MHRA requesting additional reporting only in those trials where there is felt to be a specific concern.

(x)Portal and Database –a single comprehensive database, accessed via a portal would be expected to save administrative effort and to present a solution to many of the publication/transparency considerations. It is to be hoped that a suitable robust platform can be developed and implemented in time for the implementation of the Regulation.

(xi)Indemnity/insurance—the proposal to require member states to set up a national indemnification mechanism is welcome. We believe that it will provide a level of confidence for trial subjects regardless of whether the Sponsor is a commercial or non-commercial organization and will make multistate trials much easier to set up and potentially less costly for the non-commercial sector.

(xii)While obtaining insurance in the UK has not presented difficulties, non-commercial organisations have reported difficulties in obtaining insurance in a number of EU member states (eg France). This has largely arisen in multi-state trials where a participating member state does not accept existing insurance and demands that local insurance is secured. In some instances the barriers of cost and resource in obtaining this have led to the termination of clinical trials. Publishing an annual national/EU report on the incidence of trial related litigation would help balance the often exaggerated concerns of non-commercial sponsors of the risks of trials, given the relative attention given to the very few where serious adverse events occur at rates beyond those of routine care.

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

2. The majority of clinical researchers and R&D staff in KHP have very little awareness of the HRA aside from the fact it exists. One or two members of staff who are involved in national level with professional networks understand that the HRA is an “authority in waiting” it is understood that NRES and several associated functions have moved to fall within HRA remit and that HRA is conducting scoping activities to reshape the NHS R&D/REC interface.

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

3. There is extensive evidence in the public domain to show that pharmaceutical companies withhold clinical trial data. KHP is in no doubt about the veracity of this evidence and is aware of a number of specific pharmaceutical trials which have failed to publish the results in full (approximately 33% in a small subset of studies in a specific disease area reviewed, which is in line with published estimates.1

4. There are two stages of trials where failure to publish has been raised as a concern. The first is where pharmaceutical companies conduct trials which are essentially “invisible” to the public domain and this is likely to be during the early development phase of new drugs, when novel molecules are being tested in healthy volunteers. These are the studies more likely to cite commercial sensitivity and it would be suggested that any willingness to allow such studies to conceal their results should be strictly restricted to the time until the drug is licensed. At that stage, all studies pre-licensing should be published in full and the recent GSK announcement to this effect is to be welcomed (see http://www.gsk.com/media/press-releases/2013/GSK-announces-support-forAll-Trials-campaign-for-clinical-data-transparency.html). Once studies reach the stage of being conducted in NHS patients, it is unlikely they will be “invisible” in the public domain. Academics investigators will have recruited patients to the studies and will make reference to those studies in papers and reviews relating to the disease area. However it is not uncommon for the “results” of such studies to merely take the form of a press release, with limited data regarding the methodology, outcome measures, analysis methods, confidence intervals and so forth. The practice of medicine would be greatly improved by mandating that such data be made available in full, through publication in peer reviewed medical journals or full online reports.

5. The impact on public health is considerable. Unpublished evidence, if available, may prevent unnecessary trials being undertaken and act to obstruct meaningful meta-analysis resulting both in expensive treatments being prescribed when they are not needed and effective treatments being denied to patients even though the evidence exists, simply because it cannot be reviewed properly. KHP feels that swift action should be taken to ensure that pharmaceutical trial data is made available in full for all licensed medicines, regardless of whether the trials conducted relate to the licensed disease area or another.

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

6. Occurrence of trials: The International Committee of Journal Editors (ICJE) statement on trial registration has had a significant and welcome improvement to trial reporting (eg EudraCT/clinicaltrials.gov/ISRCTN) for both medicinal and non-medicinal trials but the issue of unregistered trials being published remains. Without enforcement the voluntary requirement to register trials rests with the sponsor.

7. Results of trials: Historically authors and publishers have refrained from publishing trial data with negative/abandoned/uninteresting results (both academic and pharmaceutical led trials) for very different reasons. More recently, increased awareness has highlighted the importance of publishing neutral/negative results. Further work now needs to ensure that all clinical trials and their primary results (in full) are made available routinely at the end of any trial, not limited to CTIMPs subject to regulations (ie surgical techniques, devices, psychological therapies, educational programmes or non-CTIMP medicinal trials). Ensuring all clinical trials are open to full public scrutiny cannot fall to the Competent Authority (CA) alone since only a subset of trials i.e CTIMPs and some device trials are subject to CA approval (MHRA in the UK). In general, it is the funder of an academic trial who ensures publication, as it is often made a condition of the funding award (eg NIHR HTA programme where it is overwhelmingly successful—see http://www.bmj.com/content/346/bmj.f105#aff-1) and while this is effective it is not the solution. “Own account” or unfunded clinical trials also complete and fail to publish.

8. Responsibility to ensure both registration and publication of all trials: The simplest solution would be for the relevant ethics committee to mandate that trial teams produce evidence that they have registered their trial on a public database prior to their trial starting, since all trials (regardless of intervention) require ethical approval. If this became a routine condition of approval for all trials, it would keep any additional administrative burden to a minimum and would not be overly complex to administer, while still ensuring all trials from PhD projects through to large multinational trials, in all interventions from CTIMPs to educational programs, are registered. It is important that the solution resolves this for all clinical trials and not just a subset. Were it mandatory, to register it would be a relatively simple process to require the investigator to submit a copy of the primary publication (ideally in a peer reviewed journal within one or two years of the “last patient last visit”) to the ethics committee as the final step in the process. Where this is not possible investigators should continue to provide annual progress reports to the ethics committees, explaining delays to analysis/publication of results. By having access to full information about trials and their statuses, ethics committees are well placed to communicate and escalate any concerns to both Sponsors/investigators about publishing delays. In academic trials this would ensure unpublished trials come to the attention of the university administration or NHS R&D office routinely. For CTIMPs (most pharmaceutical trials), submission of the “end of trial” declaration to either ethics or the MHRA could be permitted only after final publication. This would mean annual fees to the MHRA would be payable until publication and would continue indefinitely if the trial fails to publish. Provision could be made for specific studies to be exempt from the need to publish, if this is agreed with the CA, but this should only be for drugs in development. Once licensed, all retrospective trial data should be immediately published, in line with the GSK commitment made on 5th February (see http://www.gsk.com/media/press-releases/2013/GSK-announces-support-forAll-Trials-campaign-for-clinical-data-transparency.html). Importantly, the ethics committees for such trials would consider the trial “open” unless the MHRA/CA confirms the exemption to the requirement for immediate publication. The ethics committees therefore will also have a record of all trials where the results will not be published until the IMP is licensed.

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

9. KHP does not have any detailed information about systems in place in other countries, in order to provide a view

Declaration of Interests

Professor Andrew Pickles, Director, King’s Clinical Trials Unit, Department of Biostatistics, Institure of Psychiatry, King’s College London
No interests to declare

Mrs Jackie Powell, KHP-CTO, Research & Development, King’s Health Partners
No interests to declare

Miss Caroline Murphy, Manager, King’s Clinical Trials Unit, Department of Biostatistics, Institute of Psychiatry, King’s College London
No interests to declare

Contributors

Miss Joanna Kelly, King’s Clinical Trials Unit, Department of Biostatistics, Institute of Psychiatry, KCL

Dr James Galloway, Department of Academic Rheumatology, School of Medicine, KCL

Professor Richard Hughes, Emeritus Professor of Neurology, KCL

Professor Alison Metcalfe, School of Nursing, KCL

February 2013

1 http://www.bmj.com/content/346/bmj.f105#aff-1

Prepared 16th September 2013