The Dana-Farber Cancer Institute (DFCI) of Boston and the Ontario Institute for Cancer Research (OICR) in Toronto have joined the Collaborative Cancer Cloud, a partnership formed in August 2015 by the Knight Cancer Institute at Oregon Health and Science University (OHSU) of Portland, Oregon, and Intel. By allowing research institutions to securely share genomic, imaging and clinical data on a precision medical analytics platform, the Collaborative Cancer Cloud is positioning itself to make rapid strides in cancer research and lifesaving therapies.

DFCI brings its Profile database, one of the most comprehensive personalized cancer medicine initiatives in the U.S., to the mix. “Through the Profile project, we have created one of the world’s largest databases of genetic abnormalities that drive cancer, with over 15,000 genetic profiles of patients’ tumors, adding about 400 each month to the database,” said Barrett Rollins, M.D., Ph.D, and chief scientific officer of DFCI. The data includes profiles from patients at Boston’s Brigham and Women’s Hospital and Boston Children’s Hospital. 

Similarly, the OICR is the largest repository of cancer genomes in Canada, hosting the Secretariat and Data Coordination Center (DCC) for the International Cancer Genome Consortium (ICGC) and co-hosting the Secretariat for the Global Alliance for Genomics and Health. “To understand the causes of cancer and to develop more effective methods of prevention, detection and treatment, cancer researchers need access to rich molecular and clinical data sets,” said Lincoln Stein, M.D., Ph.D., director of the Informatics and Bio-computing program at OICR and professor of molecular genetics at the University of Toronto.

Both of the joining institutes are energized by the powerful and reliable Intel platform, as well as the expanding cohort offered by the cloud. “We are convinced that the innovative data sharing structure developed by Intel and our academic partners will accelerate the delivery of better treatments for our patients,” said Rollins of DFCI. OICR’s Stein notes that information in rich data sets is “often siloed and unmanageably large, rendering it effectively inaccessible. Projects like the Collaborative Cancer Cloud allow for multiple institutions to pool their data and provide researchers with the computer power needed to work on data remotely.”

The quality of data in the profiles will require iterative development of standards, best practices, and scalable models. According to Dr. Thomas Wilckens, CEO of InnVentis, “One of the biggest challenges is how to avoid garbage in–garbage out and generate reliable, valid and comparable reference data…[and establish] standard operating procedures…tailored to meet the quality requirements of extremely sensitive technologies like metabolomics.”

The Collaborative Cancer Cloud makes it easier, faster and more affordable for developers, researchers and clinicians to determine how hundreds, even thousands, of genes interact to drive disease in individual patients. Working with advanced hardware from Intel, the three research institutions will initially focus on developing genomic pilot projects based on leading industry standard tools. They will identify novel analytic approaches using machine learning techniques against a collective set of molecular and imaging data in order to support big data analytics in a federated, aligned environment. The cloud is designed to scale to include unprecedented volumes of data and allow for secure, aggregated computation across distributed sites without losing local control of the data. Such cyber security allows an institution to maintain proper custody of its own datasets and protect both patient privacy and any institutional intellectual property that may result.

Eric Dishman, vice president and general manager, Health and Life Sciences Group at Intel, looks forward to integrating Intel technology to the Collaborative Cancer Cloud to “advance the processing, storage, networking, data security and advanced analytics necessary to push the boundaries of precision medicine and bring us closer than ever to truly making care personal.”

The expansion of this shareable knowledge base promises to help advance cancer research and contribute to improved outcomes for patients around the world. Long term, OHSU and Intel aim to open this federated, secure Collaborative Cancer Cloud platform to dozens of other institutions, which would accelerate the ability of clinicians and researchers around the world to understand the root causes of cancer and develop targeted, molecular treatments. In the future, these underlying technologies may be applied to cardiovascular diseases and neurological disorders, among others, to accelerate scientific discovery, yield new insights and inform treatment plans. A long-term goal of the collaboration is to reduce “big data” barriers and help make precision medicine widely available to patients.

David Housman, Ph.D., professor of biology and member of the Koch Institute at MIT, applauds efforts to obtain more precise information about the genetics of cancer patients and the genetics of their tumors in order to predict outcomes more accurately. “Unfortunately, as you gather masses of data, you also gather more irrelevant data,” he warned. “The problems are complex, and require focused people, who have great creativity and insight. For example, if you have a great race car and want to win a Grand Prix, you want to make sure you have a great driver who knows how to handle each challenging, tedious mile of the course.”

Brian Druker, M.D., director of the Knight Cancer Institute at OHSU, summarized the goals of the initiative: “The entire research community can benefit from insights revealed in large data cohorts. By securely sharing clinical and research data among institutions, while maintaining patient privacy, our goal is to turn a process that’s agonizing and uncertain for countless patients into a highly tailored, one-day diagnosis and treatment recommendation.“

Dishman reiterated the need for rapid turnaround. “The end goal is to empower researchers and doctors to help patients receive a diagnosis based on their genome and to arm clinicians with the data needed for a targeted treatment plan. By 2020, we envision this happening in 24 hours,” he said. 

This article was reprinted from Volume 20, Issue 14, of CWWeekly, a leading clinical research industry newsletter providing expanded analysis on breaking news, study leads, trial results and more. Subscribe »