Science and Technology CommitteeWritten evidence submitted from the Editor and Deputy Editor of the British Medical Journal

The BMJ (British Medical Journal) appreciates the opportunity to contribute to this inquiry. We would be pleased to also provide oral evidence if necessary, and we look forward to the Committee’s conclusions.

Here is our response to the Committee’s questions:

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(i) Not entirely, although the revisions will greatly improve the regulation of clinical trials (Proposal for a Regulation of the European Parliament and of the Council on clinical trials on medicinal products for human use, and repealing Directive 2001/20/EC. European Commission. 2012).1 We agree with Dr Peter Gøtzsche of the Nordic Cochrane Centre, whose recent peer reviewed article in the BMJ said “The new European Union Regulation on Clinical Trials ... aims to simplify the process for application and approval of trials and make it more uniform throughout the EU. It also includes a lighter regime for low risk trials—for example, those using licensed medicines. It contains much good sense, but there are still deficiencies in providing access to information and protection to patients.”2

1(ii) How the EU regulation will improve applications and approvals for drug trials:

It provides for a single EU portal for applications, set up and maintained by the European Commission.

It allows lighter regulation for low risk trials.

All application information will be publicly accessible unless confidentiality is justified to protect personal data or commercially confidential information.

Summary results must be reported to the single EU database (which will be controlled by the European Commission) within one year of the end of the trial.

Each trial protocol should include a description of the publication policy for results.

All information given to trial participants, including the form for informed consent, must be included in each application.

1(iii) The BMJ endorses Gøtzsche’s evidence based recommendations for further amendments to the proposed EU Regulation on Clinical Trials, as summarised here:3

Citizens’ right to know should override commercial confidentiality. The EU database will contain no personal data on trial participants, and the European ombudsman has declared that there is no commercially confidential information in trial protocols or clinical study reports. Patients volunteer for research to benefit society and future patients, not to benefit industry.

Results and data should be provided within one year after trial completion, with no exceptions. The current proposal for the EU Regulation allows for postponement for substantiated “scientific reasons”.

Violations of the one year deadline should be punished.

A public audit process should be established.

Clinical study reports, raw anonymised patient level data, and statistical codes should be published on the portal, not simply summaries of results.

Trial protocols should be easily accessible and all amendments should be dated and submitted to the EU portal.

The protocol should contain the full statistical analysis plan and case report forms.

The scientific and ethical justification for a trial should be based on a systematic review of similar trials, whether registered or not. Many old, unregistered trials are highly relevant for evaluating the scientific and ethical justifications for new trials.

Trial populations should be similar to the populations expected to use the drug.

Certificates of analysis of both active drugs and any placebos should be submitted together with visual records (images): even in “double blind” trials active drugs and placebos sometimes differ in texture, colour, and size and the study is not truly blinded.

The consent form for trial participants should state that all results and anonymised trial data will be made publicly available within a year after the end of the trial.

The clinical trial master file should be stored indefinitely (not just for the proposed five year period), in preserved electronic formats: data may be essential for interpreting trial results or for litigation at any time in the future.

All serious adverse events—including those occurring in trials conducted outside the EU—should be reported without delay. The current proposal for the EU Regulation requires sponsors to report adverse events that affect the benefit-risk balance, but only if these events are unexpected and only from trials in the EU.

Patients should be followed up closely for some time after they come off a trial drug. Such follow up is not currently provided for in the proposal.

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(i) The HRA is relatively new and is finding its feet. However, we are encouraged by the statements made by Janet Wisely of the HRA’s National Research Ethics Service about how she plans to monitor compliance on registration and reporting of the results of trials.4

2(ii) We would recommend that the NRES and HRA take a firm line against trialists and sponsors who fail in their responsibilities to publish the results of trials, and that any failures to comply should be sanctioned by withholding ethical approval on future trials until results are fully reported.

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

3(i) What evidence is there? As the BMJ states hidden clinical trial data are systematically undermining doctors’ ability to prescribe treatment with confidence.5 Many widely used drugs across all fields of medicine have been represented as safer and more effective than they are, endangering people’s lives and wasting public money. It is well documented that researchers and companies often withhold clinical trial results from doctors and patients. Half of all trials are never published (Song F, Parekh S, Hooper L, Loke Y K, Ryder J, Sutton A J, Hing C, Kwok , Pang C, Harvey I. Dissemination and publication of research findings: an updated review of related biases. Health Technology Assessment 2010; Vol. 14: No. 8). In many cases—such as those of oseltamivir (Tamiflu) and rimonabant—direct requests for information about trials have been refused. The US government’s requirements for timely reporting of clinical trial results at its register have been ignored by the authors and sponsors of four out of five eligible trials.6 Numerous studies have shown strong evidence of publication bias, where unfavourable or negative results are not published.

3(ii) The BMJ has shown, through publishing research, investigative journalism, and commentaries, that current knowledge about the effectiveness and safety of specific medicines and medical devices is seriously incomplete. Doctors cannot always make fully informed and accurate decisions about which tests and treatments to offer their patients. Well documented cases in which hidden clinical trial data have had or may have serious consequences for human health include the cases of these drugs:

paroxetine (GSK) (Lenzer J. Manufacturer admits increase in suicidal behaviour in patients taking paroxetine. BMJ 2006; 332 doi: http://dx.doi.org/10.1136/bmj.332.7551.1175);

oseltamivir (Roche) (Doshi, P. Neuraminidase inhibitors—the story behind the Cochrane review. BMJ 2009;339:b5164 doi: http://dx.doi.org/10.1136/bmj.b5164);

reboxetine (Pfizer) (Wieseler B, McGauran N, Kaiser T. Finding studies on reboxetine: a tale of hide and seek. BMJ 2010;341:c4942 doi: http://dx.doi.org/10.1136/bmj.c4942)

rosiglitazone (GSK) (Cohen D. Rosiglitazone: what went wrong? BMJ 2010; 341 doi: http://dx.doi.org/10.1136/bmj.c4848); and

rimonabant (Sanofi-Aventis) and orlistat (Roche) (Gøtzsche PC, Jørgensen AW. Opening up data at the European Medicines Agency. BMJ2011;342:d2686 doi: http://dx.doi.org/10.1136/bmj.d2686).

3(iii) What impact does withholding clinical trial data have on public health?

This is not simply an academic matter. Missing data about the risks of medical interventions in trials can skew the evidence base (body of knowledge) that drives medical practice and policy, can harm patients by exposing them to drugs and devices that may be less effective and safe than we currently think, and can lead to futile priority setting and expense by health systems. Moreover, researchers or others who deliberately conceal trial results have breached their ethical duty to trial participants. A BMJ editorial last year said: “Concealment of data should be regarded as the serious ethical breach that it is, and clinical researchers who fail to disclose data should be subject to disciplinary action by professional organisations. This may achieve quicker results than legislation in individual countries, although this is also desirable.”7

Fiona Godlee, editor in chief of the BMJ, was co-author of the briefing note at www.alltrials.net on missing trial data.

3(iv) One key example is that of the neuraminidase inhibitors oseltamivir (Tamiflu). In 2009, during the swine flu (H1N1 influenza) pandemic, the BMJ published an updated Cochrane review on neuraminidase inhibitors in adults with influenza (Jefferson T, Jones M, Doshi P, Del Mar C. Neuraminidase inhibitors for preventing and treating influenza in healthy adults: systematic review and meta-analysis. BMJ 2009;339:b5106).The review and a linked investigation undertaken jointly by the BMJ and Channel 4 News cast doubt not only on the effectiveness and safety of oseltamivir (Tamiflu), but also on the system by which drugs are evaluated, regulated, and promoted, which is giving doctors, patients, and the public a false sense of security (Cohen D. Complications: tracking down the data on oseltamivir. BMJ2009;339:b5387). Jefferson et al, concluded that they had no confidence in claims that oseltamivir reduces the risk of complications and hospital admission in people with influenza. In doing so they reached a similar conclusion to the Food and Drug Administration in the United States and a health technology assessment performed for the UK’s National Institute for Health and Clinical Excellence (NICE), which both found insufficient evidence on complications. Yet claims that oseltamivir reduces complications have been a key justification for promoting the drug’s widespread use. Governments around the world have spent billions of pounds on a drug that the scientific community has found itself unable to judge. The BMJ has made public key correspondence with the drug’s manufacturer, Roche, and with international organisations that recommend or regulate drugs.8 Currently:

The World Health Organization (WHO) recommends Tamiflu, but has not vetted the Tamiflu data.

The European Medicines Agency (EMA) approved Tamiflu, but did not review the full Tamiflu dataset.

The US Centers for Disease Control and Prevention (CDC) and the European Centre for Disease Control and Prevention (ECDC) encourage the use and stockpiling of Tamiflu for influenza epidemics, but have not vetted the Tamiflu data.

The majority of Roche’s Phase III treatment trials of oseltamivir remain unpublished over a decade after completion.

In Dec 2009 Roche publicly promised independent scientists access to “full study reports” for selected Tamiflu trials but, to date, the company has not made even one full report available.

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

4(i) Governments: Voluntary agreements have been shown to be ineffective in achieving transparency on the results of clinical trials. Calls for greater transparency have been made for the past 20 years with little visible effect. Where individuals have made important strides, as for example 20 years ago at GSK, a change of leadership at an organisation can reverse any temporary gains. Legislation is the only certain way of mandating publication of trial results. This should be enacted at the European level. Experience at the FDA shows that legislation is not in itself enough. Rigorous audit and sanctions for trialists and sponsors who fail to comply will also be needed.

4(ii) Regulators: The European Medicines Agency (EMA) announced in November 2012 that, from 1 January 2014, it would proactively publish all clinical study reports in order to allow reanalysis of clinical trial data by stakeholders. The BMJ sees this as a major and essential step towards improving the evidence base for medicine and healthcare. At the moment, the EMA’s policy is the only ray of light on a dark horizon. Working through the detail of how it will work, and fighting off attempts to block the policy, will require steadfast support from all parties interested in the public good.

4(iii) Ethics committees: As mentioned above, ethics committees have an important role to play but have not so far seemed able to live up to their responsibilities. It should be a pre-requisite of approval that trials will be registered and the results published in a timely manner. Ethics committees need to be charged with auditing compliance and should be required to report to the HRA any trials that fail in this regard. Approval of further trials by the same trialists or sponsors should then be withheld until the results of completed trials are reported in full.

4(iv) Study sponsors and funders: It should be their legal responsibility to register and report the results of all trials they have sponsored. They should withhold some of the funds until the results have been published, as the UK HTA does to good effect.

4(v) Investigators: It is or should be ultimately the principle investigators responsibility to ensure that the trial is registered and the results made public in a timely manner. The extent to which this might be enshrined in legislation if obviously a matter for consideration.

4(vi) Journals: Journals can and should act as advocates for transparency and integrity in medical science. Through its open data initiative the BMJ aims to achieve appropriate and necessary independent scrutiny of data from clinical trials (http://www.bmj.com/open-data). Working with others, we seek to highlight the problems caused by lack of access to data and to call for and support ways to release the data. Journals can also apply pressure directly on trialists and sponsors who still value publication in a journal for academic and marketing purposes. To this end, the BMJ announced in January 2013 that we will no longer publish any trial of drugs or devices where the authors do not commit to making the relevant anonymised patient level data available upon reasonable request. Whether or not a request is reasonable will be adjudicated by readers of the BMJ through our rapid response system. Those making the request will be asked to post an account of what they wish to do with the data, and the trial’s authors will be asked to respond, giving details for any refusal.

4(vii) Campaigners: The BMJ is a joint founder of the current AllTrials campaign.9 The campaign calls on governments, regulators, and research bodies to ensure that the aims and methodological details of all clinical trials, past and present, are publicly registered and their full methods and results reported. By 19 February 2013, just over a month after launch, the AllTrials online petition had more than 30,800 signatories including many international organisations concerned with medical research and treatments and more than 80 patients’ organisations.10

4(viii) If all of the above fails, pharmaceutical and medical device companies should be prevented from evaluating their own products. A central fund could be established, as exists in Italy, into which companies would pay if they wanted their drugs to be licensed. The fund would be used to support independent phase 3 and 4 trials comparing new treatments with existing treatments. For more on how such a system might work, read Garattini S, Chalmers I. BMJ 2009;338:b1025. BMJ 20 http://www.bmj.com/content/338/bmj.b1025

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

It is not clear that any country has cracked the problem of hidden clinical trial data. (Major discrepancies in how regulators in different jurisdictions interpret the clinical trial data available to them is itself a sign of how mad the current situation is—see Doshi P. Neuraminidase inhibitors: the story behind the Cochrane review. BMJ 2009;339:b5164). The USA has taken the legislative route but has so far failed to enforce the law adequately. Other countries have retained a voluntary approach but with variable and mainly little effect. This is why the EMA’s recent announcement has been greeted with such enthusiasm, and why it is essential that we all give the EMA as much support as possible in working through the difficult issues it will face in delivering on its promise. Italy has shown innovative thinking in setting up a fund for supporting independent head to head trials, drawn from a proportion of the marketing revenues of pharmaceutical companies. This is an approach that other countries should consider.

Submitted by:

Dr Fiona Godlee, Editor in Chief, BMJ, and Dr Trish Groves, Deputy Editor, BMJ.

Dr Fiona Godlee has been Editor in Chief of the BMJ since 2005. She qualified as a doctor in 1985, trained as a general physician in Cambridge and London, and is a Fellow of the Royal College of Physicians. Since joining the BMJ (British Medical Journal) in 1990 she has written on a broad range of issues, including the impact of environmental degradation on health, the future of the World Health Organisation, the ethics of academic publication, the problems of editorial peer review, and the need for greater openness in medicine and research. In 1994 she spent a year at Harvard University as a Harkness Fellow evaluating efforts to bridge the gap between medical research and practice. On returning to the UK, she led the development of BMJ Clinical Evidence, which evaluates the best available evidence on the benefits and harms of treatments and is now provided worldwide to over a million clinicians in nine languages. In 2000 she moved to Current Science Group to establish the open access online publisher BioMed Central as Editorial Director for Medicine. In 2003 she returned to the BMJ Group to head up its new Knowledge division. She has served as President of the World Association of Medical Editors (WAME) and Chair of the Committee on Publication Ethics (COPE) and is co-editor of Peer Review in Health Sciences.

Dr Trish Groves is Deputy Editor, BMJ and Editor-in-Chief, BMJ Open. Trish qualified in medicine and psychiatry before moving to the BMJ in 1989. Trish leads the team that peer reviews original research and articles on research methods, and is responsible at both BMJ and BMJ Open for policies on open access and sharing of raw research data from studies. She has been a member of the council of the Committee on Publication Ethics and of international groups including those developing guidelines on transparent reporting of clinical trials and trial protocols (CONSORT2010, SPIRIT); those working with the European Medical Research Councils and European Science Foundation on the effectiveness of medical; and the IDEAL collaboration that is developing stronger surgical research methods and is working with the US FDA devices division on improving regulatory pathways.

The BMJ: The BMJ (British Medical Journal) is an international peer reviewed medical journal published online at bmj.com with more than 14.5 million unique users internationally (1.6 million unique users a month). It also appears in weekly print and iPad editions. The print BMJ’s weekly circulation is 122,000, of which 10,000 copies are distributed outside Britain. International editions reach another 55,000 readers. The BMJ has been published without interruption since 1840. The journal’s mission is to lead the debate on health and to engage, inform, and stimulate doctors, researchers, and other health professionals in ways that will improve outcomes for patients. We aim to help doctors worldwide to make better decisions.

Conflicts of interest: Both authors of this statement are full time editors employed by the BMJ. A small proportion of our remuneration is affected by the performance of the journal and the publishing group, which could in a very small way be boosted by the success of the BMJ’s open data campaign. Both authors are long standing advocates for transparency in research and for open access to peer reviewed research.

February 2013

1 http://ec.europa.eu/health/files/clinicaltrials/2012_07/proposal/2012_07_proposal

2 Gøtzsche P C. Deficiencies in proposed new EU regulation of clinical trials. BMJ 2013;345:e8522 doi: http://dx.doi.org/10.1136/bmj.e8522

3 All of the points come from this detailed article: Gøtzsche PC. Deficiencies in proposed new EU regulation of clinical trials. BMJ 2013;345:e8522 doi: http://dx.doi.org/10.1136/bmj.e8522.

4 http://www.bmj.com/content/345/bmj.e7304/rr/613776

5 http://www.bmj.com/open-data

6 http://www.clinicaltrials.gov/

7 Lehman R, Loder E. Missing clinical trial data. BMJ 2013;343:d8158 doi: http://dx.doi.org/10.1136/bmj.d8158

8 http://www.bmj.com/tamiflu

9 http://www.alltrials.net/

10 http://www.alltrials.net/supporters/

Prepared 16th September 2013