In the spirit of open collaboration, crowdsourcing and multidisciplinary approaches to disease, Project Data Sphere will make
available, for the first time, clinical trial data sets from pharma and academia, to essentially anyone who registers at http://ProjectDataSphere.org/. The project is housed under the auspices of the Life Sciences Consortium (LSC), itself an arm of George H. Bush's CEO Roundtable
The concept is purposefully simple: biopharmaceutical companies and academic groups provide data sets from past clinical trials,
and anyone with an internet connection can go to the website, register, and download the amassed data for their own research.
SAS, the business analytics company, is providing the web platform and functionality for Project Data Sphere, pro bono.
"Provided you're not on the criminal register, you'll have access," says Charles Hugh-Jones, a member of the LSC, and chief
medical officer at Sanofi US. Access is not dependent on a valid research proposal or academic credential, because "we can't
judge whether someone's got the right research proposal [for fighting cancer], and I would challenge any one committee to
have broad enough knowledge to say where data could go in the hands of different groups of people."
At launch in late June or early July, Project Data Sphere will likely contain five or six data sets, including data sets provided
by Sanofi. Hugh-Jones says four of the top 10 pharmas are currently preparing data contributions. All of the data will come
from cancer trials; data sets will include protocols and case report forms, but will be limited to comparator arm data, at
first. The decision to use only comparator data, not active trial arm data, is "really about getting over certain hurdles,"
says Hugh-Jones. "People have to understand how to de-identify data, they have to get used to some of the IP issues, and legal
issues related to sharing data." Even so, a collection of comparator arm data open for independent research is unprecedented.
"If I could get 100 comparator arm data sets in one area of cancer, or among 10 rare diseases, that's a terrific database"
for modeling disease, says Hugh-Jones. Comparator arm data "feels like a safe space to learn how to work together," says Amy
Abernathy, director of the Duke Center for Learning Health Care. Duke is a participating member of the LSC.
Project Data Sphere will make clinical trial data sets available to essentially anyone.
Data providers will retain ownership and existing IP, but people using the data can build on it, to produce their own intellectual
property. In the second phase of the project, SAS will provide a built-in analytics capability and forum, so data downloaders
can run analytics without leaving the Project Data Sphere website, and talk to each other. Hugh-Jones hopes to eventually
start having data user conferences, to promote further collaboration.
Finally, Project Data Sphere presents a positive PR play for pharma and academia. Organizations gifting data are prioritizing
the discovery of new cancer treatments over business concerns. For academia, it's a shield against the swords clamoring for
the release of data generated via tax dollar-funded NIH grants.
Hugh-Jones is confident that others will contribute data after the initial launch. During a panel on collaboration and transparency
at the 10th Annual SAS Health Care & Life Sciences Executive Conference last month, a SAS employee in the audience suggested
fomenting an Occupy Clinical Trials movement to advocate for greater data transparency. Indeed, it will take many loud voices
and a sustained effort to liberate clinical trial data from life sciences organizations. But cancer, directly or indirectly,
affects everyone's lives. While incremental progress has been made, people with cancer still have disturbing similar outcomes—in
terms of disease progression and death—as they did 30 years ago. Project Data Sphere is a necessary, all-hands-on-deck approach
to tackling the disease.
Ben Comer is Pharm Exec's Senior Editor. He can be reached at firstname.lastname@example.org