Speaking of clinical trial data transparency – here’s an excellent analysis from the August edition of the Burrill Report ...
Fight for Transparency Heats Up: Industry turns to courts to stop policy in EU, offers own approach
Daniel S. Levine
Tom Jefferson had nagging concerns about his review for the Cochrane Collaboration of clinical studies of the flu drug Tamiflu, but they crystallized in 2009 when a Japanese pediatrician raised a question about the work via email. The doctor wanted to know why the review included eight unpublished trials that Jefferson and his colleagues hadn’t seen, and had only been included as summaries in another study funded by the drug’s producer Roche?
The solution seemed simple enough. Jefferson, an epidemiologist based in Rome who was conducting the updated analysis for the Cochrane Collaboration, an independent group that provides guidance to healthcare professionals on the use of drugs, would reach out to the authors of the studies and get them. But the authors either told him they didn’t have the data, said they had never seen the data, or ignored him completely. When he asked Roche for the studies, it said it would provide them, but insisted at first that he sign a confidentiality agreement, a condition that he found unacceptable.
The controversy came against a backdrop of worldwide fear of a global pandemic of bird flu that sent the World Health Organization, the Centers for Disease Control and Prevention, and governments around the world spending billions of dollars to stockpile Tamiflu. Despite the spending, Jefferson says there were questions about whether the drug did anything to prevent transmission of flu, minimize complications, or whether the health benefits the drug provided justified the risks of using it. Roche, in a statement posted on its web site and updated at the end of February 2013, said that it had disagreed with the analysis plan Cochrane Collaboration shared with it because it was at odds with how Tamiflu has been reviewed and approved by regulatory authorities in more than 80 countries. The company said it stood by the “robustness and integrity” of its data supporting the safety and efficacy of the drug.
“The fight for the last four years hasn’t been so much for clinical study reports,” Jefferson says, “but to have clinical study reports without having to sign confidentiality agreements, funny handshakes, rolling up our trouser leg, and so on.”
The four-year battle to obtain clinical trials data—he believes that there have been 123 studies (74 of which Roche sponsored)—has made Jefferson a flag bearer in an international fight over clinical trials data transparency. The controversy, which is by no means limited to Tamiflu and Roche, is now reaching a critical point. A legal battle is underway in Europe over a European Medicines Agency transparency policy expected to take effect at the start of 2014 that would make public clinical trial data once a drug is approved. At the same time, in the United States, the Institute of Medicine is at work on a consensus study on the issue and the U.S. Food and Drug Administration is seeking comment on a clinical data transparency policy of its own.
Much is at stake as the policymakers embrace comparative effectiveness and the advent of Big Data provides a new means to ferret out untapped information hidden within clinical study reports—the detailed narrative reports at the individual patient level that can run to tens of thousands of pages in a single trial.
Proponents of clinical data transparency argue that it will provide doctors with greater insight into the safety and efficacy of the drugs they prescribe, improve care, cut waste, and minimize scientific misconduct and fraud. Peter Doshi, a post doctoral comparative effectiveness researcher at Johns Hopkins University, who is also working on the Tamiflu review for the Cochrane Collaboration, points to a list of blockbuster drugs that were marketed only to have significant health risks come to light through academicians analyzing clinical trials data. “The whole system has depended on trust in medicine, trust in people to report properly, trust in people to have enough information to credibly analyze trials,” he says. “We know enough not to be so trusting anymore.”
Industry, while not of one mind, says that in principle it doesn’t oppose transparency, but expresses concerns about how it is done. These concerns include the threat to confidential commercial information, the challenge to the authority of regulators, the risk to patient privacy, and the opening of the data to inappropriate uses that could lead to the publication of misleading results and public health scares.
Lawsuits in Europe
In May, the General Court of the European Union prohibited the European Medicines Agency from releasing data from two AbbVie and Intermune trials in an interim ruling, part of a challenge by the two drugmakers to the agency’s decision to grant access to information the companies provided as part of their market approval applications. The challenge is the first to be made to the EMA’s three-year old access-to-documents policy.
Since November 2010, the Agency has released more than 1.9 million pages in response to such requests. The legal battle involves documents relating to AbbVie’s rheumatoid arthritis drug Humira and InterMune’s Esbriet, a treatment for idiopathic pulmonary fibrosis, an unexplained chronic and progressive scarring of the lungs. In both cases, competitors of the companies were seeking the data.
“Biopharmaceutical companies support responsible data sharing that protects patient privacy, maintains the integrity of the regulatory review process, and preserves incentives for biomedical research,” said Matt Bennett, senior vice president for the Pharmaceutical Research and Manufacturers of America in a statement at the time. “Unfortunately, the EMA’s current and proposed policies fail to respect these principles.”
The ruling comes amid increasing pressure from medical journals and patient advocates on pharmaceutical companies to provide complete transparency and make public all of their clinical trials data. It also comes as the EMA readies implementation of a new policy to proactively publish data from clinical trials supporting the approval of new drugs once a decision has been made. The EMA said it would continue the process of drafting its policy on publication of clinical trials data.
Ben Goldacre, author of “Bad Pharma” and co-founder of the AllTrials campaign, which seeks publication of all results from all clinical trials, called the ruling “a disgrace.” “There is no justification for withholding information about the methods and results of clinical trials from doctors, researchers, payers and patients, who need all the information on a medicine to make truly informed decisions,” he says.
The EMA says it welcomes the opportunity for legal clarification of the concept of commercially confidential information, but expressed “regret” over the decision to grant interim relief to AbbVie and InterMune.
FDA, IOM, weigh transparency
In the United States, although efforts to bring about transparency have moved slower, they are advancing. In October 2012, the Institute of Medicine held a two-day workshop to explore the benefits of sharing clinical research data, the issues surrounding it, and how best to do so. The IOM is now working on a consensus study that some hope could provide a roadmap for the development of a clinical trials data transparency policy in the United States.
A group of leading pharmaceutical and biotechnology companies has been involved in the IOM process and issued a press release hailing the effort. “This industry group is strongly in support of enhanced access by third party research to clinical trial data generated by industry and academia in a manner that ensures that patient confidentiality is preserved, scientific integrity is maintained, and intellectual property rights and confidential company information are protected,” they said in a joint statement.
The Food and Drug Administration is seeking public comments through August 5 on a policy to make available de-identified and masked data derived from medical product applications. But advocates of transparency say such data will be of limited utility because the policy under consideration would de-identify it and remove the data’s link to a specific product, study, or application.
Peter Pitts, president of the Center for Medicine in the Public Interest and chairman of a conference on clinical trials data transparency held by FDANews July 23-24, says it’s “a non-arguable fact that FDA is at the nexus of vast amounts of clinical data that if it was appropriately shared it would be of tremendous value.” He says that sharing information, among other things, could help companies fail faster, which would save a lot of money and allow for reinvestment of diminishing resources. But such an effort, he says would require information technology resources the agency is without, decisions on what is and isn’t redacted, discussions about intellectual property rights and commercial confidentiality, and funding for it all.
“We’ve come to the point now, relative to clinical trials data, that transparency is a good thing. The question now becomes, ‘How do you accomplish that and what exactly does transparency mean? And how quickly does it become transparent? And how much of it becomes transparent and how much of it remains confidential? Is it transparent through corporate entities, or is it transparent through a consortium of corporations? Is it transparent through a government organization? And who ultimately decides? Can people opt out? And if people opt out, can you really have transparency?’” says Pitts. “They are all very tough questions.”
(Note: For more on the FDA News conference, see Cry “havoc,” and let slip the dogs of data transparency.
Progress, but no victory
Four years after Jefferson began his efforts to access clinical study reports from Roche and GlaxoSmithKline, which produces the Tamiflu competitor Relenza, both of which are in the class of drugs known as neuraminidase inhibitors, it is Roche and GSK that arguably have the clearest defined and most advanced efforts to provide clinical trials data to outsiders.
Roche has complied with the original request the Cochrane reviewers made in 2009, although the researchers continue to pursue access to additional trials data of which they subsequently learned. The Cochrane Collaboration expects at the start of 2014 to publish a new review of neuraminidase inhibitors based on 23 clinical studies on Relenza and up to 20 studies on Tamiflu.
In February, GSK announced it had signed on in support of the AllTrials campaign and has vowed to publish clinical trials data of all approved drugs dating back to 2000, the date the company was formed through the merger of Glaxo Wellcome and SmithKline Beecham.
“We are committed to being transparent with our clinical trial data to help advance scientific understanding and inform medical judgment,” Patrick Vallance, President, Pharmaceuticals R&D, GlaxoSmithKline, said in announcing support for the campaign. “Our commitment also acknowledges the very great contribution made by the individuals who participate in clinical research. All those involved in the conduct and publication of clinical research, whether healthcare companies like GSK, academia or research organizations, have a role to play in ensuring that the data they generate are made publicly available to help bring patient benefit.”
In response to questions from The Burrill Report, GSK said it believes a broader solution for providing access to trial data from across the research community needs to be developed and it has been in discussion with trials sponsors from industry, academia, and research charities working in partnership to create one. GSK says one solution would be the establishment of an independent data custodian to which research sponsors would deposit anonimized data after a project has been completed and clinical studies are published. Researchers could then submit scientific proposals and analysis plans to independent custodian to request access.
The Pharmaceutical Research and Manufacturers of America and the European Federation of Pharmaceutical Industries and Associations at the end of July jointly issued a list of principles for clinical trial data sharing. They call for a voluntary plan that would make data available to only to “qualified” researchers who sign non-disclosure agreements. They say they would only share data for which they have the informed consent of study participants to do so.
(Note: The full principles can be found on the PhRMA website. PhRMA said implementation of the commitments in the principles will begin on January 1, 2014.)
Companies will start reviewing requests for data from researchers at that point, but Castellani said companies will need to take into account patients’ informed consent. He said it will not be possible to share data from many earlier trials because patients participating in them did not consent to such release.)
But advocates worry that the victories in studies extracted from pharmaceutical companies to date may be short lived, in large part due to the legal battle in Europe over the EMA’s transparency policy. “Societal expectations have dramatically moved in a direction of greater transparency, but there remain serious risks to the process that could potentially derail a lot of the progress that has been made,” says Johns Hopkin’s Doshi. “The progress we’ve seen with Tamiflu may have been a one-off victory for what otherwise may be a return of data secrecy. Many people think the goals of transparency have been achieved, but they have not.”