A coalition to promote and implement data sharing in cancer by facilitating data ‘liquidity,’ first proposed in February at an Institute of Medicine (IOM) workshop, is taking shape with the formation of a steering committee and action plans that include a demonstration project.

For several years, data sharing has been on the agenda of oncology meetings and organizations, spurred by the growing volume of genomic and clinical data that reside in separate institutional databases. Proponents cite many benefits to sharing: reducing duplication and costs, informing clinical care, fueling research, and the overarching goal, getting new and effective treatments to patients more quickly.

Biggest Bottleneck

Three IOM meetings over the past few years have addressed the issue. At a recent meeting on personalized cancer care in Washington, DC, keynote speaker John Mendelsohn, MD, of The University of Texas MD Anderson Cancer Center, Houston, said that for personalized cancer care, “the biggest bottleneck is information management.” Massive amounts of data need to be analyzed and stored, and that is made more difficult by “data balkanization.”

Now, the champions of data sharing say the time may be ripe for significant movement forward. And the new coalition, they say, could be a catalyst for achieving not just data sharing, but data liquidity—the rapid, seamless, secure exchange of useful, standards-based information among authorized individual and institutional senders and recipients.

“We were happily surprised by the amount of engagement at the IOM Workshop [on Informatics Needs and Challenges in Cancer Research],” said Amy Abernethy, MD, of Duke University, Durham, North Carolina, one of the authors of the proposal and a member of the new steering committee. “There seemed to be a real groundswell of interest.”

Steering Committee

Encouraged by that response, the authors, led by Marcia Kean, MBA of Feinstein Kean Healthcare, quickly formed a steering committee for the group, which they named the Data Liquidity Coalition. Steering committee members include a range of experts and stakeholders. The cancer centers, for instance, are represented by William Dalton, MD, PhD, from H. Lee Moffitt Cancer Center, Tampa, Florida, and Lawrence Shulman, MD, of Dana-Farber Cancer Institute, Boston, as well as Dr. Abernethy from Duke; the pharmaceutical industry by Charles Hugh-Jones, MD, and Joel Beetsch, MD, both from Sanofi; patient groups by Andrea Ferris, the President of LUNGevity; and the IT industry by Kris Joshi, PhD, of Oracle and William Tulskie of Healthcare IT, Inc.

Ms. Kean is the Chairman of Strategic Initiatives at Feinstein Kean Healthcare, a strategy and communications firm that assists life sciences innovators. Other members include Kathleen Foley, PhD, of Thomson Reuters Healthcare; George Poste, DVM, PhD, of Arizona State University; Tom Kean, CEO of C-Change; and Brad Pollock, PhD, from the Epidemiology and Biostatistics Department of The University of Texas Health Science Center at San Antonio.

One of the committee’s first acts was to lay out five activity tracks (see sidebar), including demonstration projects. Some members are now working on a proposal for the first demo and beginning to look for funding. Coalition participants will provide some resources, such as in-kind donations of services, but they are also talking to potential funders about grants, Ms. Kean said.

The demonstration project will need to show that a data-sharing project can make data accessible and usable by the whole research community, serving as proof of that broad principle. More specifically, it will also test and demonstrate how the coalition will operate.

An Honest Broker

Envisioned as an “honest broker,” the coalition will not undertake data-sharing projects directly. Instead, it will work with cancer organizations needing data exchange to develop specifications, standards, and software. The actual IT work will be done by vendors or the open-source community, but the coalition will help craft the request for proposal, which would require that the system developed be kept open at the interfaces so that others could connect to it. The coalition will share the framework with all members. It could also maintain the common infrastructure so that over time, Ms. Kean said, the data-exchange capabilities could constitute a national biomedical resource.

For the demonstration projects—and the coalition itself—to succeed, it will need to deal with more than technical issues. “Data liquidity is about more than IT,” Ms. Kean said. “Potentially it involves policy, advocacy, culture.… It’s what people are calling a systems problem.”

The other four activity tracks are designed to address such problems. Industry competition and intellectual property issues are the most obvious obstacles, but other entrenched practices also work against data sharing. The academic hierarchy, for example, rewards individual researcher achievement and publication rather than collaboration. And health-care providers and patient groups have economic incentives to guard the data gained from their own patient population.

Life Sciences Consortium

The question remains whether such systems will change. Steering committee member Charles Hugh-Jones, MD, of Sanofi, thinks the time is ripe. “There’s a confluence of several factors,” he said, including the technology to make data sharing possible and the growing realization that it could lead to faster, cheaper, and better drugs.

Dr. Hugh-Jones is involved in another collaborative effort, the CEO Roundtable on Cancer’s Life Sciences Consortium, a task force of Roundtable companies and academic organizations involved in the life sciences. The Consortium is working on a project called DataSphere (formerly MetaPharm), in which research groups will put their clinical trial data online in a format accessible to any bone fide research organization, Dr. Hugh-Jones said.

To that end, Sanofi-Aventis is releasing two recent large phase III datasets together with case report forms and data descriptors. Other Consortium members, academic and commercial, are working to identify oncology datasets to contribute. Work on the technical infrastructure by the SAS Corporation, a Life Sciences Consortium member, is underway, and the first data should be available by the end of this year.

Progress on DataSphere may be one more sign that the benefits of data sharing are increasingly seen to outweigh the risks.

“Somewhere along the line, all will benefit by sharing data,” said Ms. Kean. This has been very hard to convey in the biomedical community, she added, but interest is clearly growing. “I think we’re approaching a tipping point.” ■

Disclosure: Dr. Abernethy has received research funding from the National Institutes of Health, Agency for Healthcare Research and Quality, Robert Wood Johnson Foundation, Pfizer, Helsinn, Amgen, Kanglaite, Alexion, Biovex, DARA, and MiCo, has served as a consultant for Novartis and Pfizer, and is on the corporate board of directors for Advoset and Orange Leaf Associates, LLC. Ms. Kean, Dr. Hugh-Jones, and Dr. Mendelsohn reported no potential conflicts of interest.