Clinical Trial Data Sharing Off to a Slow Start

Clinical Trial Data Sharing Off to a Slow Start

A Rutgers paper in The New England Journal of Medicine explores why long-awaited access to drug trial data is underused

Clinical data-sharing is touted as key to accelerating medical discoveries, but a two-year-old sharing program is surprisingly underused.
Photo: Shutterstock
For many years, doctors, scientists and researchers have urged that clinical drug trial data be shared to accelerate medical advances in treating multiple diseases. But two years after free patient data became available in a major data-sharing project, the biggest surprise is how little it is being used.

“Given the risks patients are taking and given the data being collected, there’s an ethical imperative to make maximum use of it,” said Brian L. Strom, chancellor of Rutgers Biomedical and Health Sciences. “That’s why it’s frustrating that more people are not using it and we don’t fully know why.”

Strom is lead author of a paper published today in the New England Journal of Medicine that explores the unexpectedly low usage of the free data. Since GlaxoSmithKline began making anonymous patient data available in May 2013, 177 research proposals were submitted and access granted to 144 by an independent review committee headed by Strom. No applications were denied; 33 were withdrawn after requests for more information.

GlaxoSmithKline funded millions of dollars to launch the data-sharing system and make data available from 1,500 of its drug trials. The clinicalstudydatarequest.com website was set up in 2014 and data from other pharmaceutical companies was added, bringing the total to 2,800 trials from 13 companies, including Astellas, Bayer, Lilly, Novartis, Roche and Sanofi.

Brian L. Strom
Brian L. Strom, chancellor of Rutgers Biomedical and Health Sciences, is the lead author on a paper exploring why use of clinical drug trial data is unexpectedly low.
To date, only one research paper using the data has been published. To find out why the data is so underused, Strom and co-authors surveyed the 144 participating investigators, but only 24 responded. More than half of them said they were still analyzing data.

From the responses, the authors gleaned that two years may be too little time to judge the data-sharing system, which researchers can apply to access via the website. Some survey respondents said it is difficult to work with the data behind a secure network firewall and would like a simpler way to share data.

“We may need to make it easier for people to use the data, but we don’t know if that is the issue,” Strom said. “Before we invest an enormous amount of money to expand data-sharing to include academic data as well, we should learn from our experiences in order to do it right.”

One positive outcome: Concerns that the data would mostly be used by researchers looking to disprove original findings and in lawsuits against pharmaceutical companies were not borne out, Strom noted.

Management of the Clinical Study Data Request system was turned over to The Wellcome Trust, a global charitable foundation that supports scientists and researchers, at the end of 2015. It is one of a few clinical data-sharing programs.

Strom suggested a more extensive survey of participants may better reveal why researchers are not accessing data that so many agree could help speed medical innovations.

“We can’t assume that just making the data available is going to lead to enormous use of it,” Strom said. “This suggests we should take a timeout and look at how to do it right.”

Co-authors of the study, “Data Sharing: Is the Juice Worth the Squeeze,” are Marc E. Buyse, Hasselt University; John Hughes, The United Kingdom Clinical Research Collaboration Board; and Bartha M. Knoppers, McGill University.


Media contact: Dory Devlin, dory.devlin@rutgers.edu or 973-972-7276