November 29th, 2012
European Medicines Agency Destroys the Dam: Slight Dampness Results
It is odd how a dam can burst with hardly anyone noticing. For decades, the pharmaceutical industry has been conducting trials on its own products that have never been made public, though many of them have been disclosed to licensing agencies such as the FDA and the European Medicines Agency (EMA). These agencies have also received huge volumes of clinical study reports, trial protocols, and individual patient data that have been kept confidential. Until this year, the EMA has been dismissive of any attempts by researchers to gain access to such data. However, in the face of criticism from the European Ombudsman and the European Parliament, the EMA announced in April 2012 that it would be henceforth allowing access to all data submitted to it in applications for drug licensing in Europe, whether those applications were successful or not, and that it would be holding a workshop on data transparency in November to discuss the best methods of carrying out such disclosure proactively.
Many of us rushed to book places at this workshop, and it was so over-subscribed that months went by before we knew whether we had a seat for the meeting, which took place last week on November 22nd, at a time when most of you were getting up to face a day of family togetherness culminating in a turkey dinner.
Some of us were elated to hear the EMA’s Executive Director state at the beginning of the meeting that the EMA’s decision was irrevocable and that we were here to discuss how all data submitted to the EMA could be shared, not if. There was then a panel discussion in which two representatives of the European Federation of Pharmaceutical Industries and Associations expressed considerable disquiet about this decision, repeatedly citing considerations of commercial sensitivity and competitive disadvantage. Other members of the panel countered these objections and invited the pharma representatives to put their case to the public: Do people want trial data to be used for the commercial advantage of drug manufacturers, or for the public and health professionals to gain better information about the harms and benefits of the drugs that they use on millions of people every day?
In the second part of the meeting, similar arguments went back and forth, and there were few respondents who actually addressed the purpose of the meeting by offering concrete suggestions on how to optimize the proactive release of data submitted to the EMA from January 2014 onwards.
So the meeting was lively, momentous even, but not very productive. What most of us — including the panelists — had failed to notice was that the EMA is already releasing huge quantities of data. If you would like to see everything that has been submitted to the EMA relating to a drug already on the European market, simply apply while providing the purpose of your inquiry, and you will be given the whole file.
Yes, you read that right. The dam is already bust, and lies in pieces in Europe. The data that industry is trying so hard to fence in with legal restrictions, careful vetting of applicants, clauses about indemnity against commercial losses, etc., is in large part out there for open use if you have a reasonable case for accessing it. Moreover, the agency is trying to make the process more comprehensive, better structured, and proactive.
There is a clear challenge here to other licensing agencies such as the FDA. It is also a challenge to medical editors, and one that the BMJ has already decided to address by insisting that all authors disclose their full data for studies that it publishes from January 2013 onwards. Other leading journals will need to take note: the EMA decision calls time on the current model by which pharmaceutical companies design, conduct, and report phase-3 trials of their own new products in leading journals, and then buy up large numbers of reprints for promotional purposes. Products will henceforth be judged on the totality of data relating to the human trials in which they have been used — and about time, too.