NEW YORK – A European initiative to make a million genomes accessible to researchers around the region by 2022 is grappling with legal questions about the best way to achieve that. As data is held at the national level — meaning domestic restrictions on cross-border data sharing apply — investigators involved in the effort are exploring mechanisms to make data available, while acknowledging that scaling up and sustaining the initiative will take place beyond 2022.
Regina Becker, a research scientist at the University of Luxembourg, is concerned with the ethical and legal challenges of the project, called the 1+ Million Genomes (1+MG) initiative. She said that its goal of making a million genomes available for cross-border research in Europe, based on technical, semantic, and legal interoperability, will require a "massive effort" of member states, especially given the legal hurdles that exist to secure the privacy of European citizens. The project is now working to obtain a clearer idea on the ultimate framework for 1+MG, as well as how they might implement that architecture.
"This will require interim solutions where we are able to start operations while we work on a sustainable approach with all countries involved," said Becker, who also discussed the topic at the virtual European Society of Human Genetics meeting earlier this month. "Ultimately we hope to find ways that, once implemented, will pave the way for the 1+MG and beyond."
The 1+MG initiative emerged in April 2018. Representatives of 22 EU countries as well as the UK and Norway have signed the declaration, which pledged to have a million genomes sequenced and accessible by 2022. A few countries, including France, Ireland, Poland, and Switzerland, are official observers to the effort but have not yet signed the declaration.
While the concept of making a million genomes shareable to embolden European genetics research seemed achievable three years ago, the legal framework for doing so remains a challenge. Different countries attach different legal qualifiers to sharing genomic data. For instance, countries in Northern Europe tend to link data sharing for research purposes to a public interest, whereas those in southern European countries often require the consent of those who provided samples to repositories before sharing that data. According to Becker, 1+MG is now exploring ways to make such diverse systems interoperable while minimally interfering in any national legislation.
One proposed solution is that where national law limits data use, those datasets could come with attached caveats on data use. This is called a flexible approach. However, researchers could still find such restrictions burdensome, and there may be cases where different potentially conflicting rules apply to the same datasets. As such, the initiative is also exploring the idea of harmonizing legislation around data sharing, a common approach, where all rules governing genomic data sharing would be the same across all signatory countries.
The main target is to work in alignment with the European General Data Protection Regulation, or GDPR, which allows the EU member states to introduce additional restrictions for sharing genetic and healthcare data as well as other rules and requirements, such as safeguards for scientific research processing, Becker said.
According to Becker, the 1+MG effort has turned to previous EU studies for guidance. Several studies, she noted, have been commissioned by the European Commission to explore these issues. The Department of Justice, for example, investigated all national provisions based on opening clauses of the GDPR, and the Department for Health and Food Safety focused on implementing the GDPR for health data. While helpful, these studies provided only a part of the jigsaw of the European data protection landscape of sharing health and genomic data for secondary use.
"All reports proved to be a valuable element in gaining an overview," said Becker, "but none was sufficient for the full picture." She added that under current circumstances there is no way for 1+MG to get around national legislation protecting genetic data, and that the initiative itself cannot remake the European legal framework.
Yet, she noted that 1+MG is not operating in isolation. The EU is currently undertaking the creation of a European Health Data Space, which should promote the exchange and access to different kinds of health data, including genomics data. The EHDS will be built in a way that protects individual data and is aligned with the GDPR. This effort will include the introduction of new European legislation around the cross-border use of health data. The European Commission is engaging member states to realize the aims of the EHDS, an effort that overlaps with 1+MG's ambitions.
"We have the vision of 1+MG and EHDS integrating for secondary use of health and genomic data," remarked Becker. "We very much hope that the design of the EHDS will allow such integration."
As such, what 1+MG is hoping to achieve by 2022 is not seamless interoperability but to have an ethical and legal framework adopted internally to support future interoperability. "The next step is the implementation," Becker noted. "However, already having the framework will be a great achievement in my opinion, considering the current challenges in data sharing."
To support 1+MG, the European Commission last year awarded €4 million ($4.7 million) to bankroll an associated coordination and support project, called the Beyond 1 Million Genomes. The goal at its time of formation was to connect research infrastructures, academics, and other stakeholders to set legal guidance, best practices, and recommendations to realize the 1+ Million Genomes initiative.
B1MG commenced in June 2020 and will run through May 2023. Part of its purview is to support the aims of 1+MG by defining requirements, putting guidelines in place, and developing models to support long-term operations. The ultimate goal is to sustain secure cross-border technical infrastructure to support data access across borders and future implementation of genomics into healthcare.
This will include recommendations for standardizing sequencing data as well as linked phenotyping and clinical information. In addition to the legal challenges that Becker has touched on, there are also language issues, for instance, such as the 27-plus languages contained in electronic health records. Establishing and sustaining technical infrastructure is also not without its challenges — but the project is moving forward by focusing on a proof of concept with a workflow of standards and services to access genomic data for rare disease, much of which will be applicable across the other use cases: cancer, common complex disease, and infectious disease.
Serena Scollen, head of human genomics and translational data at ELIXIR, the European life sciences infrastructure for biological information, is the lead investigator on B1MG. She said via email last week that the effort now has a live stakeholder portal, via which it can distribute and request feedback on initial data standards and quality guidelines as well as the technical specifications and implementation guidelines.
This year, the B1MG project also carried out virtual country visits to Estonia, Finland, and the UK to observe how different countries involved in 1+MG use and manage genomics data. Regarding the legal challenges the project faces, Scollen stressed that they will be overcome.
"Personalized medicine is the future, however complex and uncertain it may look now," said Scollen. "We have seen recently how many of the hurdles around infrastructure and data sharing have been overcome in an infectious disease public health crisis." She said that policymakers and stakeholders need to work together to realize the goals of 1+MG. "We need — as researchers, industry, funders, governments, and citizens — to come together to enable the collection of and access to large, rich datasets," Scollen said.