Publishing clinical trial data: The who, what, when, where, and why

by Joan Rachlin, JD, MPH, Executive Director

Among the many topics that will be presented and discussed at the 2013 Advancing Ethical Research Conference is the ubiquitous issue of data transparency. Some of the most hotly debated questions making the research ethics rounds include who should interpret, distribute, review, and receive data, and with good reason. From WikiLeaks to National Security Agency (NSA) whistleblowers, and all other manner and form of data breaches, the problems of data privacy, data security, data accuracy, and data availability are ones that won’t go away any time soon.

I recently read a piece in The New York Times about Peter Doshi, PhD, a 32-year-old post-doctoral fellow at Johns Hopkins University, who has been pressing pharmaceutical companies to make their clinical trial data public. Dr. Doshi contends that the financial conflicts of interest that permeate considerable segments of biomedical research, as well as the publication bias of journals, make public disclosure necessary so that consumers and patients can participate more fully in making decisions regarding their care. It is well-documented that only half of all clinical trials are published, and that negative results, which are at least as instructive as positive ones, rarely see the light of day.

Although one can hardly imagine a more David-and-Goliath-like match, Dr. Doshi and his allies have made some inroads, most notably in helping to persuade GlaxoSmithKline to agree to share detailed data from all global clinical trials conducted during the past thirteen years, though nothing has yet been released.

Dr. Doshi and his colleagues’ quest for clinical trial transparency began while they were trying to determine whether or not Tamiflu, manufactured by Roche, was an effective treatment for flu; this pursuit proved difficult because the bulk of the data on which claims about Tamiflu’s efficacy had been based was unavailable. Only two out of ten Tamiflu trials, it turned out, had been fully published, and when scientists at the Cochrane Collaboration, a well-respected network of independent researchers working with Dr. Doshi, tried to track down the other eight, they hit repeated road blocks. The Cochrane Collaboration determined that, as a result of the incomplete picture Roche presented about Tamiflu, they could not endorse its effectiveness. Only then did Roche begin to open its file cabinets, but the issue of Tamiflu’s value in reducing flu symptoms has still not been resolved. However, Roche has joined GlaxoSmithKline in agreeing to share detailed data with outside researchers.

This breakthrough, especially when combined with the European Medicines Agency’s proposal to require wider access to trial, has fueled public demand for information that has formerly been entirely within the province of researchers, clinicians, and journal editors. The Pharmaceutical Research and Manufacturers of America, as well as most drug companies—excluding GlaxoSmithKline and Roche—have opposed this push. Some argue that American consumers already have access to raw data via the federal clearinghouse, but there are extensive limitations placed on what kinds of information can be released through that medium.

The New York Times article ends with a powerful and provocative summary of Dr. Doshi’s current message to the drug companies: “publish your data, or we’ll do it for you.”

And there are many other cases fueling the call for requiring data publication. On June 17, for example, two independent reports were issued, indicating that Medtronic’s spinal surgery product, Infuse, carried with it serious risks, including cancer and male infertility, and that there was no demonstrated improvement over already-existing products for the same purpose. The reports, prepared under the auspices of Yale University with funding from Medtronic, reviewed 17 studies. The news reports surrounding these revelations are rife with allegations that negative data was suppressed, that conflicts of interest abounded, and that Medtronics promoted off-label use.

These issues around data transparency and review, among so many others that are rapidly changing the research landscape, will be front and center at the 2013 AER Conference, which will be held November 7-9 in Boston, MA. I hope you will join us there for what promises to be an educational odyssey like none other.