Why better data sharing is needed to aid cancer research

Healthcare organizations that have clinical data on cancer trials need to share that information with global researchers to advance investigations into solving the disease.

The call comes from Project Data Sphere, which is making the plea in an article in the New England Journal of Medicine. Data sharing is being hampered by the lack of infrastructure and appropriate methods to manage data exchange, the article says.



Project Data Sphere is an independent not-for-profit initiative of the CEO Roundtable on Cancer’s Life Science Consortium. The roundtable aims to develop and implement initiatives that reduce the risk of cancer, enable early diagnosis, and facilitate the development of and access to the best available treatments.

Better access to all clinical trial information is essential for accelerating progress in cancer research and moving the best treatments to patients, particularly those who have volunteered to participate in clinical trials.

“It’s critical that we find ways to break down siloes and allow researchers direct data access to ensure we maximize the impact of decades of clinical research,” says Monica M. Bertagnolli, MD, the group chair for the Alliance for Clinical Trials in Oncology at Brigham and Women’s Hospital and President-Elect, American Society of Clinical Oncology.

“Responsible sharing of data is something we owe all trial participants so that we may learn from and expand the research they have enabled,” wrote Bertagnolli, the article’s lead author.

One approach for improving data sharing would be Project Data Sphere’s Cancer Research Platform, an open-access approach for holding data. It was launched in April 2014. The Project Data Sphere platform was developed by SAS, a healthcare industry company specializing in data security and analytics. The cancer platform enables registered users to use analytic tools on data contained on the platform.

Using the platform, researchers can download, share, integrate, and analyze anonymized patient-level data. Data providers—pharmaceutical companies and academic institutions—contribute cancer patient de-identified data. Currently the platform includes 72 datasets derived from almost 42,000 patient lives.

Slow progress

“While there is broad agreement on the value of optimizing access to clinical trial data, the lack of infrastructure and consensus on appropriate methods to manage data exchange and ownership have slowed progress,” says Martin J. Murphy, MD, chief executive officer of Project Data Sphere and a co-author of the NEJM article. “Recently published research documents show that big data analysis yields insights that may have an impact on cancer patient care. This manuscript illuminates the need to advance open-access data-sharing from all sectors that house clinical trial data such as industry, academia and government.

“It is now time for all holders of clinical trial data to fulfill the intentions of volunteer clinical trial subjects to derive all the knowledge possible from their clinical trial contributions,” Murphy concludes. “We seek new data-sharing partners to build a global research resource for cancer patient benefit.”