How the Marathon of Hope Cancer Centres Network is supporting responsible data sharing to advance precision oncology

Adrian Thorogood, TFRI's Data Governance Manager, explains how the Network is addressing barriers to ensure data sharing is consistent with applicable ethical and legal norms and international best practices.

By Adrian Thorogood, Data Governance Manager, Terry Fox Research Institute

Sharing clinical and genomic data between research and healthcare institutions is vital to advance precision oncology in Canada. It is the only way to establish datasets that are large and representative enough to perform innovative research in this area, and it also helps to enable collaborations and improve scientific rigor and reproducibility.

Unfortunately, sharing health data across institutions and provinces can be difficult. Some reasons for this are obvious: institutions have obligations to protect patient data, and are rightfully concerned about the legal and reputational risks of potential data breaches. Genomic sequencing data introduce additional privacy concerns, as they are biologically unique to an individual, and can reveal sensitive health information about patients as well as their families.

In addition to legal and privacy issues, technical barriers also hamper data sharing. Different hospitals, for example, use different software, vocabulary or formats to manage their data, which complicates the integration and re-use of data between institutions.

Overcoming these challenges and facilitating data sharing to accelerate the implementation of precision oncology in Canada is one of the main goals of the Marathon of Hope Cancer Centres Network (MOHCCN). The Network, led by the Terry Fox Research Institute (TFRI) and the Terry Fox Foundation, aims to establish a resource of genomic and clinical data from 15,000 cancer patients treated across the country. This resource will be made available to Network researchers and to the broader research community in order to enable data-driven discoveries that ultimately improve the diagnosis and treatment of cancer patients.

During its first few years, Network experts focused on producing guidelines to ensure all data generated by the Network met the same high-quality standards. More recently, the Network has produced a series of principles, policies, and procedures to ensure that data are shared effectively and responsibly. To date, over 600 Network researchers and clinicians from 34 institutions across the country have agreed to share their data according to these rules. Recent achievements include the creation of:

  • A consent guideline that ensures that prospective patients are provided with transparent information about participating in precision cancer research and data sharing, supporting meaningful consent from patients to contribute their data to this research initiative.
  • A consent element checklist that helps researchers and Research Ethics Boards assess existing consents and determine if previously collected samples and data can be ethically contributed to the Network.
  • Data access procedures that ensure that data are only made available for ethics-approved cancer studies subject to robust security and confidentiality safeguards. At the same time, the procedures ensure researchers have streamlined and impartial access to data from across the country.

Another innovative way that the MOHCCN is facilitating data sharing is through its commitment to creating a federated data system. Data sharing traditionally involves moving data into a central repository, and then distributing copies to researchers. In a federated approach, data never leaves the site of origin. Through the magic of data standards and cloud technology, researchers will be able to securely access data or train artificial intelligence models across the entire Network. This approach can eliminate the risks and legal complexities associated with data transfers, reinforce local accountability towards patients, and support trust in broad data sharing.

The Network has already done work to establish a technical platform that will be installed at member sites and allows data to be securely ingested and queried. Efforts are now underway to pilot Network data access procedures and technical implementations to begin routinely sharing all the rich data that the Network has already generated. We are also starting to develop additional data governance layers needed for sharing data outside the Network, including for new types of uses like AI testing and training.

As TFRI’s Data Governance Manager, I’m passionate about promoting data sharing to accelerate scientific progress and improvements in patient care. By working with experts across the country, we can ensure that the Network’s principles, policies, and procedures for data sharing are consistent with applicable ethical and legal norms and international best practices. Tackling cancer is a common global challenge, so I am also passionate about promoting connections between the Canadian and international cancer research communities to advance research and patient care here and around the world.

One of the things I appreciate the most about working with MOHCCN is the effort the Network puts into inclusive and participatory governance, on bringing together diverse perspectives from clinicians, bioinformaticians, ethicists, lawyers and patient representatives across the country to determine the best ways to share and protect data as a community. Because, in the end, governance is not just about figuring out what to do (or not do) with data; it is also about figuring out who gets to do the governing.

About the author

Adrian Thorogood is TFRI’s Data Governance Manager. He is passionate about enabling the effective and responsible sharing of data across research networks in order to accelerate science and improve patient care. As a lawyer and legal researcher, Adrian has supported numerous international precision medicine initiatives, including the Global Alliance for Genomics and Health, and the EU 1+ Million Genomes Initiative. He has also contributed to innovation in the broader field of data governance through publications in leading scientific, legal, and bioethics journals. Adrian holds a joint common law / civil law degree from McGill University, and a Master of Laws from the University of Toronto.

"By working with experts across the country, we can ensure that the Network’s principles, policies, and procedures for data sharing are consistent with applicable ethical and legal norms and international best practices."