Independent researchers have sparked something of a brouhaha this month over access to clinical study reports – the industry’s raw clinical data, some of which forms the basis of submissions to regulators for marketing authorisation review.
Peter Doshi and colleagues highlighted the challenges they faced in obtaining full clinical study reports for oseltamivir (Tamiflu, Roche) in the context of updating a Cochrane review of neuraminidase inhibitors – first in a PLoS Medicine article appearing this month and again in an opinion piece in the New York Times.
They allege that the data they have managed to obtain (through a combination of Freedom of Information requests to European regulators, leaked sources and incomplete manufacturer-provided reports) point toward “reporting biases” and “fundamental problems in trial design,” which they conclude undermine the original effectiveness claims for oseltamivir – which were based on a 2003 manufacturer-supported meta-analysis.
Doshi and colleagues are strong advocates for full public release of clinical study reports, arguing that independent re-analysis of data is in the public interest and that it has the potential to advance medical knowledge.
Regulators Weigh into the Debate
Also writing in PLoS Medicine, regulators from the European Medicines Agency (EMA), the Dutch Medicines Evaluation Board, the Agence Française de Sécurité Sanitaire des Produits de Santé in France and the Medicines and Healthcare products Regulatory Agency in the UK have similarly backed the publication of full clinical trial data – but with a number of conditions. The regulators point out that before such a move can be made, it is critical to:
- Develop and agree to standards for data protection and patient confidentiality
- Adopt quality standards for independent researchers (they point out that these stakeholders are not immune to conflicts of interest)
- Establish guidelines for data sharing
This stance is not entirely unexpected given past announcements by the EMA that it intends to move toward disclosure of clinical study reports supplied to it by industry once marketing authorisation decisions are made – not to mention its responsiveness to Freedom of Information requests, as relied upon by Doshi and colleagues. This is in contrast to the US FDA, which has maintained that this type of data does not fall under the country’s Freedom of Information Act.
Putting it in (a Rapidly Evolving) Context
The issues sparked by these recent publications will inevitably lead to a considerable amount of debate, and one of the first things to spring to my mind was how the implementation of the 2010 pharmacovigilance legislation will impact that debate – at least from a European perspective.
This package of measures will, inter alia, allow competent authorities to impose the obligation to conduct post-authorisation safety studies (PASSs) and post-authorisation efficacy studies (PAESs). From my reading of the Good Pharmacovigilance Practice module on PASSs, I cannot see that disclosure of full clinical data will be required in this context. Meanwhile, scientific guidelines for PAESs and a public consultation on that document are forthcoming sometime later this year.
One strength of the legislation is that competent authorities may request PASSs or PAESs at any time after the granting of a marketing authorisation – meaning that, in cases where questions over safety/efficacy begin to arise, PASSs/PAESs can be utilised in a proactive manner.
Diverging agency viewpoints
Certainly the diverging views of different agencies on the effectiveness of oseltamivir would be one such flag for initiating a PAES (the FDA was at odds with its own country’s Department of Health and Human Services and Advisory Committee on Immunization Practices, which were joined in their viewpoint by Australia and the EMA), although presumably this discrepancy in decision-making is one reason why Doshi and colleagues advocate independent re-analysis in the first place, regardless of the strides regulators are making toward requiring more comprehensive information.
On that note, the EMA is coordinating the Pharmacoepidemiological Research on Outcomes of Therapeutics by a European Consortium (PROTECT) initiative, which will investigate innovative ways for integrating data on the benefits and risks of medicines from different sources and derived from different methodologies (trials, spontaneous reporting, observational studies, etc.)
A greater focus on relative efficacy / effectiveness assessment
Finally, there is the ever-greater focus on relative efficacy and effectiveness assessment in the context of P&R decision-making, meaning that, again as in cases like for oseltamivir, there may be greater scrutiny of claims the first time round, at least where there are comparators on the market.
Although these examples don’t touch on the separate issue of the merit of open access to all raw clinical data, they do indicate the complexity that arises when debates take place in a rapidly evolving environment – and the healthcare landscape seems to be an increasingly fast-moving target these days.