17 May 2017
In a perspective paper released today in the New England Journal of Medicine (NEJM), members of the Global Alliance for Genomics and Health (GA4GH) Clinical Cancer Genome Task Team call on the international community to put more resources – people, data, infrastructure and policy change – toward cancer genomic data sharing.
“Cancer does not respect national borders, so the international community needs to work together to make headway on this disease that kills over 8.5 million people worldwide each year,” said Professor Mark Lawler of Queen’s University Belfast, who is senior author of the paper and co-chairs the Clinical Cancer Genome Task Team with Charles Sawyers of the Howard Hughes Medical Institute and Memorial Sloan-Kettering Cancer Center, New York. “Creating a functional data ecosystem is a critical component of the Precision Medicine Initiative and our ability to share data across institutions will greatly empower this initiative," said Sawyers.
The article is the fourth in a series of academic papers published by the task team, including a blueprint for data sharing in cancer published in Nature Medicine in May 2016. This paper brought forward a series of recommendations for effective technical, clinical, ethical and legal approaches to international data sharing.
Results from the first comprehensive global survey of clinical cancer sequencing initiatives were published by the task team this month in the Annals of Oncology. “The survey demonstrates that there is great interest in data sharing and that people recognize it’s value,” said Lawler. “But they also see certain barriers and there is still some resistance.”
Such barriers include non-standardized methodologies and approaches to data collection and analysis, a lack of “thoughtful integration” of data across centres and nations, absence of an effective mechanism for academic recognition, and limited dedicated funding.
“We cannot underestimate the value of collecting data longitudinally. In cancer, it is critical not only to link genomic and clinical data at diagnosis, but also to follow the patient over time, uncovering clues that help us to judge the success of therapies but also combat treatment resistance,” said Lawler. In the NEJM perspective paper, he and his colleagues highlight the need for a dedicated funding mechanism that allows national governance structures to work together to support data sharing.
For instance, the U.S. Cancer Moonshot has set forth the need to bring together genomic, epidemiologic, and clinical data to improve patient outcomes.
"Effective responsible data sharing is critical if we’re to really use the research data to help us in our aspiration to improve outcomes in this deadly disease,” said Sawyers, who is a member of the Cancer Moonshot Blue Ribbon Panel.
“While cancer as a whole is common, it is actually more of a rare disease if we think about each person’s unique genotypic and phenotypic characteristics,” said Melissa Haendel of Oregon Health & Science University, who is also a co-author of the paper. “This is fundamentally the reason that we need global patient data sharing.”
The Cancer Moonshot panel on data sharing, which included Haendel and other members of the GA4GH community, aimed in part to promote the idea that patient generated data, clinical data, and research data all need to be utilized together and developed with interoperability in mind from the start. Such an aspiration depends on both effective collaboration as well as dedicated resources. “This requires technical, legal, social, regulatory, and financial solutions to sharing. What we are calling for is ecosystem-level change and commitment,” said Haendel.
“We hope that our ‘Call to Action’ for a 'Global Cancer Knowledge Network' energizes the community to act decisively and encourages our national institutions to dedicate resources to embed data sharing in all our cancer activities,” Lawler said.