A global roundtable on Trusted Research Environments highlighted the importance of interoperability, dynamic consent, and governance in enabling secure genomic research. Participants emphasised that success should be measured by research and patient outcomes, not just security and compliance.

Earlier this year, CSIRO’s Dr Natalie Twine was invited to contribute to an international roundtable on Trusted Research Environments (TREs), organised by Public Policy Projects and Syntropy. The discussion brought together leaders from global genomics initiatives, including Genomics England, the Broad Institute, Genome Canada, GA4GH, Illumina and the 1+ Million Genomes initiative to explore the future of secure genomic data access and analysis.

The resulting report, Fulfilling the Promise of Trusted Research Environments, has now been published. Many of the themes discussed resonated strongly with CSIRO’s Digital OneHealth group through our work including development of TRECA, sBeacon and ConsentGuardian, which facilitate secure analysis of genomic and health data and support research collaboration.

Interoperability is more than connecting systems

Interoperability emerged as one of the most frequently discussed challenges. While it is often framed as a technical problem, the reality is more complex. Different metadata standards, phenotype definitions, consent models and governance processes can all create barriers, even when technical systems are capable of exchanging data.

As genomic datasets become increasingly distributed across institutions and jurisdictions, interoperability will require alignment not only of technology, but also of the policies and standards that sit around it.

Another recurring theme was the role of consent in enabling responsible data use.

In many organisations, consent remains a static document, often captured as a signed PDF, while genomic data may be reused across multiple research studies, clinical contexts and timeframes. This creates challenges for governance, secondary use and participant engagement. Approaches such as dynamic consent, used in our ConsentGuardian platform, offer a more flexible model, enabling consent preferences to be managed as structured, machine-readable information rather than static documentation.

Connecting governance is harder than connecting data

The roundtable also highlighted that many of the barriers to collaboration are not caused by technology. Differences in legal interpretation, ethics processes, risk tolerance and governance frameworks often create more friction than the underlying infrastructure.

An important insight from the discussion and our experience building TRECA for Indonesia’s BGSi is that progress may depend less on harmonising regulations themselves and more on developing infrastructure and standards that can operate across different governance models while respecting local requirements and data sovereignty.

Measuring impact beyond security and compliance

TREs are increasingly recognised as essential infrastructure for genomics research, but the discussion raised an important question: how should success be measured?

Security, compliance and user numbers are relatively easy to quantify. Much harder to measure are the outcomes that ultimately matter most: improved diagnosis, clinical translation, patient benefit and scientific discovery. As investment in TREs continues globally, developing meaningful measures of impact will become increasingly important for demonstrating their value.

Looking ahead

The discussion reinforced that the future of genomic research depends not only on generating data, but also on developing the infrastructure, standards and governance approaches needed to use it effectively.