Monday, May 13, 2013

Bringing trial data out of the shadows

Bringing trial data out of the shadows

All sectors have their own mood music, unobtrusive much of the time, but occasionally brought to startling effect into the foreground, dominating all else. The pharmaceutical industry is no different, and at present there is one insistent theme: transparency.

And if there is one area of this debate behind which chords are starting to swell loudly it is the availability of clinical trial data. Unflattering or disappointing trial outcomes have been routinely unpublished, pharma's critics claim, which means patients taking part in trials can be exposed to ineffective or even dangerous products because previous negative research is unavailable.

Clash of opinions
For its part, pharma denies there is a problem. At present, companies operating in the UK, for example, are expected – though not obliged - to register current and future trials within 21 days of enrolling the first patient, with results published within one year of marketing authorisation.

This does not go far enough, according to some. The European Commission, for instance, is in the process of revising the Clinical Trials Directive (2001/20/EC), and putative changes would require a results summary to be put on to the EU's clinical trial database within a year of the end of the relevant trial. The lead legislator on the revision process is Labour MEP Glenis Willmott, who wants a pan-European database, with full publication of trials - and financial penalties for firms which delay. 

“Too many results from clinical trials are misleading, biased or missing,” she said. “It is time that all pharmaceutical companies and researchers made the full results of studies on new and existing drugs publicly available.” 

Others agree with her. In January this year, the AllTrials initiative was launched by the British Medical Journal (BMJ), the James Lind Initiative, the Centre for Evidence-Based Medicine and Ben Goldacre, author of industry critique Bad Pharma

The campaign, backed by heavyweights such as the Wellcome Trust and the National Institute for Health and Care Excellence (NICE), wants governments, regulators and research bodies to ensure the methods and results of all trials, past and present, are published in full. Making available clinical study reports (CSRs) – the formal documents detailing the design, methods and results of trials, which form the basis of submissions to regulators – will improve transparency and aid research, AllTrials says. Summaries alone are not enough.

The idea is that researchers would benefit because they have access to everything relevant that has gone before their project, perhaps allowing them to spot pitfalls or find shortcuts and leading to more rigorous science. The chief beneficiaries would be patients since they will get better medicines, brought to market more quickly and cheaply because research is not being duplicated, leading to better health and wellbeing outcomes. 

Spirit of the age
So why is all this happening now? In the last couple of years a number of cases have created an atmosphere in which pharma is open to accusations of needing to be more clear about how it operates, the most high profile of which culminated in a $3bn fine imposed on GlaxoSmithKline (GSK) by the US government for fraud. 

There is also a philosophical level to the drive for openness, with transparency mirroring the mood of the times, since the ubiquity of the internet means accessing and publishing data have never been easier. In the UK, it is perhaps not stretching a point to suggest that health issues, which are no fault of pharma, such as the Francis Report's findings of abominable mistreatment of patients at Mid Staffordshire NHS Foundation Trust, help add to a feeling that anything less than full disclosure is a bad thing.

Whatever the motivation, Goldacre says that at present only about half of all trial data is ever published – a figure the Association of the British Pharmaceutical Industry (ABPI) and the Pharmaceutical Research and Manufacturers of America (PhRMA) dispute, suggesting the figure is closer to 80 per cent in Europe and the US.

Even so, it is clear that not all trial data is out there. “Biased under-reporting of data is real,” says Dr Fiona Godlee, editor-in-chief of the BMJ. “Data suppression is fraud. I fear we make too much of the complexity issues. Lots of good things are happening in pharma but the industry has shown it can't be trusted.”

Data disputes
Anyone searching for reasons for the current focus on issues of transparency could do worse than look at two things: one, the publication last year of Bad Pharma; and two, Roche's refusal to lay on the table all data for its influenza drug Tamiflu. The latter is a particular bugbear for the Cochrane Collaboration, which seeks to provide third party, evidence-based analyses of drugs and went public in challenging Roche. Tamiflu was stockpiled at great cost by western governments – the bill for the NHS alone was £500m - to guard against a flu pandemic, but concerns emerged that the drug was no better than placebo after summary results and CSRs appeared to suggest different things.

That debate rumbles on [the most recent development being Roche's announcement it intends to release Tamiflu data to Cochrane], but meanwhile the ABPI has argued that it has been looking at transparency for far longer than AllTrials, which it has also dismissed as a “PR-driven initiative”. ABPI chief executive Stephen Whitehead pointedly said: “We were on this journey before Bad Pharma was a twinkle in Ben Goldacre's eyes.”

It's time pharma made results of studies on new and existing drugs publically available

The ABPI line is that summary trial evidence should be made available for both new and existing medicines and it insists mechanisms for doing so are still being finalised. In the US the website is the vehicle by which this currently happens, but PhRMA has been scathing in its response to the debate, accusing critics of cherry-picking examples of poor practice and making damaging allegations. 

“The demands by Dr Goldacre and the BMJ to release patient-level clinical trial data are irresponsible with potentially harmful consequences for future medicine development,” said PhRMA executive vice president Josie Martin. “The recommendations would jeopardise patient privacy and could serve as a deterrent to individuals considering participation in trials.”

Leading the pack
Be that as it may, one pharma company has decided to take the plunge. GSK – putting the recent past behind it - has signed up to AllTrials. As yet the only pharma company to do so, GSK is pledging to publish trial data and CSRs dating back to the company's formation in 2000 for all its medicines, once they have been approved or discontinued from development and the results have been published. No small job, GSK is giving priority to CSRs for its most commonly-prescribed drugs.

All companies, academics and researchers “have a role to play in ensuring that the data they generate are made publicly available to help bring patient benefit,” explained Patrick Vallance, GSK's president, pharmaceuticals R&D. It is fair to say that this puts GSK some way ahead of the pack but Godlee has suggested other firms need to think hard about their own positions.

“GSK is getting all the credit, which doesn't seem fair,” she said. “Other pharma companies can claim that high ground. A call for transparency does not threaten the pharma industry.”

Susie Hackett, director of communications at Roche UK, has already admitted that the company was tardy in taking hold of the Tamiflu situation. “Roche is listening and recognises calls for our industry to be more transparent,” she said. “We could have been more active and engaged in the debate.”

Roche's compromise on Tamiflu data sees the company proposing to allow access to anonymised patient-level data on a case-by-case basis only. Requests would be evaluated by the company and an independent body, after the completion of regulatory reviews in the US and EU. Critics have been underwhelmed.

Call and response
So how would releasing more data actually work? Three problems have got pharma particularly steamed up: how to protect patient anonymity, how to guard intellectual property (IP) and how to ensure commercial confidentiality.

The European Medicines Agency (EMA) is working on how it will release clinical trial data from 2014, and reports from its working parties on this are expected shortly. One thing Europe's pharma trade body, the European Federation of Pharmaceutical Industries and Associations (EFPIA), is sure of is that it is dead against “indiscriminate” transparency. “We believe it is both feasible and necessary to balance the needs of personal privacy and commercial confidentiality, while providing the level of transparency needed to support public health decision-making,” the organisation said.

EFPIA board member Roch Doliveux, who is also chief executive of biopharma firm UCB, voiced the concerns of many in the sector. “I support the notion of transparency, but I don't support the notion of voyeurism,” he said. “There has to be complete transparency with regulatory authorities and every expert authority in the field. But the idea of posting on the net even patient level data – that is overload of information and it leads to misinformation and it is really, really bad for patients.”

The political and commercial winds are blowing in the direction of greater transparency

Goldacre is impatient with this attitude, saying that patients can be trusted to make up their own mind. “People aren't stupid,” he said. “It is okay [for pharma] to share complexities and shortcomings.”

Protecting IP is still a major worry for firms though. “Transparency in the competitive phase of trials would be destructive for research,” suggested Doliveux.

The resources that it will require to publish all this information is another big issue. For example, GSK will redact patient data in the CSRs and their appendices to ensure confidentiality, but has admitted it will take a dedicated team “a number of years” to do this. 

There are also legitimate caveats over the issue of publishing historical data – which could perhaps lead to retrospective class actions against companies shown to have been in some way complicit. 

Building momentum  
Yet for all this, the mood music is getting steadily louder. Much of pharma has been hostile to Goldacre, accusing him of aggressive self-promotion, of simplifying complex issues and even of scaremongering. However, as Cochrane founder Sir Iain Chalmers - a veteran of the battle to get pharma to publish more of its data - said recently: “Thirty years of having a gentlemanly discussion have done bugger all”. 

What Bad Pharma and AllTrials have undoubtedly done is to turn up the volume. The discussions may still be on the gentlemanly side, but publicity has given them greater impetus and exposure. The political and commercial winds seem to be blowing gently in the direction of greater transparency – although whether this movement is, as Goldacre believes, an unstoppable “juggernaut” remains to be seen.

No comments: