Biomedical research is rapidly progressing and often produces large amounts of data and resources that should be widely accessible to researchers and clinicians to maximize output. Sharing of data can be particularly challenging as laws and regulations from across jurisdictions must be followed in addition to scientific best practices.  Consequently, there is an urgent need for a systematic, structured, computer-readable way to capture, communicate, and manage biomedical data and resource sharing.

The Automatable Discovery and Access (ADA) Task Force, co-organized by the Global Alliance for Genomics and Health (GA4GH) and IRDiRC, was established to explore machine readability of consent and its impact on data use and accessibility. The Task Force, co-chaired by Anthony Brookes (University of Leicester) and John Wilbanks (Sage Bionetworks), consisted of over 50 members from academia, industry, and non-profit sectors.  The resulting Automatable Discovery and Access Matrix (ADA-M) was released in late 2016. A key feature of ADA-M is the possibility to create structured metadata “Profiles” for any regulatory condition. Therefore, the system is flexible in terms of the information it can convey, while at the same time structured to provide the required conditions and restrictions to allow for the sharing of data across different regulatory jurisdictions. These issues, the work done by the Task Force, and potential applications were described in the Genomic Medicine publication.

 

Since its release, ADA-M has been integrated in whole or in part by many organizations, including: Australian Genomics Health Alliance, the Solve-RD project, members of the Health Data Research UK Institute and the US Broad Institute’s DUOS project. It has also been actively evaluated by the EU Biobanking and BioMolecular resources Research Infrastructure (BBMRI), the Canadian Care4Rare-SOLVE and the UK Tissue Directory and Coordination Center. The authors emphasized that ADA-M “is designed to expand, evolve and adapt with use”, making its global use by researchers crucial in promoting international collaboration. Many more challenges remain to optimize biomedical data sharing but initiatives like ADA-M, that provides standardized and interoperable system, are key to accelerate the process.

For more details, read the full article.

A recent article published in the European Journal of Human Genetics summarized the achievements of RD-Connect, NeurOmics and EURenOmics, the three flagship projects financed by the European Commission between 2012-2017 as a response to the unmet needs of the rare disease community. Aimed to help advance –omics research and data sharing, aligned with the goals of IRDiRC, the cooperation between the three projects illustrates a solid basis on how to build and reinforce global collaboration to improve both diagnosis and therapy development.

Among the key achievements of this initiative:

• Identification of over 120 new disease genes by both NeurOmics and EURenOmics, most of which have been published. Researchers from both projects also developed targeted Next Generation Sequencing (NGS) panels to accelerate the analysis of known genes associated with the disease of interest. These NGS panels have been used for diagnosis of over 700 patients for NeurOmics and 4,000 cases for EURenOmics. Both projects also identified novel biomarkers for diagnostic purposes and initiated some therapeutic trials.
• The RD-Connect Genome-Phenome Analysis Platform which by December 2017, contained exome and genome data of over 3,000 patients, mainly with neuromuscular or neurodegenerative rare disease, and includes data generated by both NeurOmics and EURenOmics. The amount of data is rapidly increasing and the platform now also contains patients with other types of rare diseases. Additionally, it integrates numerous tools and online data analysis services  (Exomiser, GA4GH “beacon” application programming interface, runs of homozygosity, etc.) that are easy to use, even for researchers without bioinformatics experience.
• Implementation of Findable, Accessible, Interoperable and Reusable (FAIR) Guiding Principles to ensure data harmonization between registries and enable computer-assisted data analyses.
• Development of new bioinformatics tools to assist researchers in omics analysis and therapeutic target identification: the UMD-Predictor, the Human splicing finder and the VarAFT. To date, these tools have contributed to more than 1,000 research studies.
• Connection to other collaborative European projects such as ELIXIR and the Biobanking and Biomolecular Research Infrastructure Large Prospective Cohorts (BBMRI-LPC).

RD-Connect, NeurOmics and EURenOmics have greatly contributed to the progress in rare diseases research. Many more challenges remain, but they can be tackled only by supporting, reinforcing and expanding similar types of collaborative initiatives.

For details, please read the full article

On the occasion of Rare Disease Day 2018, the European Reference Networks (ERNs) for rare diseases celebrated their first anniversary. Within the past year, the impact of the ERNs on the quality of life of rare disease patients in Europe was highlighted in an articleby the European Commissioner for Health and Food Safety, Dr. Vytenis Andriukaitis.

The ERNs are unique and innovative cross-border cooperation platforms linking specialists for the diagnosis and treatment of rare or low prevalence complex diseases. Since their launch, the ERNs have treated more than 50 patients. Since collaboration is key to expanding knowledge and expertise, Dr. Andriukaitis expects the ERNs “to help hundreds if not thousands of patients” in the coming years. The ERNs are now reaching a critical mass of patients analyzed, diagnosed, and treated, thus enabling the acceleration of research, patient participation in large clinical trials, and increased sustainability of the healthcare system. In the future, if the ERN-model continues to be proven successful, Dr. Vytenis Andriukaitis suggests that expansion into other diseases might be attractive.

In a commentary published last month in Nature Reviews Drug Discovery, Kym Boycott and Diego Ardigo, Chairs of the IRDiRC Diagnostics and Therapies Scientific Committees, respectively, discussed diagnostic and therapeutic challenges for rare diseases and options on how to tackle them.

The introduction of next generation sequencing (NGS) in 2010 as a diagnosis method for patients was a great accelerator in rare disease gene discovery. However, a substantial proportion of patients defined as “complex” are often left without a molecular diagnosis, either because they are suffering from more than one rare disease or because the molecular mechanism underlying their disease cannot be detected by NGS. The pace of straightforward gene discovery by NGS has recently slowed down, mainly due to the “n-of-1” rule (i.e. the need to document pathogenic variations in the same gene in at least two unrelated individuals in order to confirm a novel gene discovery). A example of international collaboration is the Matchmaker Exchange, which connects federated databases to facilitate such discovery. However, for the “complex” patients, new methods or ‘omics’ approaches will be required to identify all pathogenic genetic lesions to inform therapy development.

The pace of orphan drug designations significantly increased after 2010. If this pace is maintained for the next ten years, it will still leave 90% of rare diseases without a treatment. Rethinking the traditional drug development process for rare diseases will be critical, including trial design for small populations, real-world evidence collection, stronger patient engagement for data collection and definition of clinical end-points. The cost of orphan drug development remains an important challenge and the authors express the need for alternative approaches to “achieve better care for patients, value for money payers and a return on investment for companies”. Opportunities lie ahead in the creation of strong partnerships with patients to collect post-approval, real-world data on treatment outcomes, and in the use of patient-led registries to both cut drug development costs and generate data on disease history/treatment outcomes.

The authors concluded on the observation that “the grand challenges for rare diseases will require international cooperation and stakeholder engagement at an unprecedented level”. It is with these challenges in mind that IRDiRC presented its new and bold vision for the next decade: “Enable all people living with a rare disease to receive an accurate diagnosis, care, and available therapy within one year of coming to medical attention”.

For details, please read the full article.

An article by Kiley et al. in the New England Journal of Medicine presents the specific vision of research funders (Wellcome Trust, Medical Research Council, Cancer Research UK, and Bill and Melinda Gates Foundation) for optimizing the value of data produced in studies, especially clinical trials, that they funded. As the need in medical research is always greater than the resources, the authors emphasize on the importance of “collaboration and cooperation among members of the global research community to maximize the effect of funded research”. However, to support this vision, funders of research have to actively incentivize and support data sharing.

There is currently a consensus that clinical trials results should be shared in a timely manner, with consideration of participants’ privacy and confidentiality; this is supported by a joint statement of the World Health Organization. The authors, as funders, identified three main challenges to data sharing: resources, equity and incentives.

As data sharing is not a cost-free process, funders need to support technical solutions to help researchers to access and (re)use data. Additionally, funders are increasingly asking researchers to include data management plans in their research proposals and the costs of these plans in the funding budget. To tackle concerns on inequity of data sharing for low- or middle-income countries, the authors suggest inclusion in data-access governance specific terms that “require users to contribute to training and capacity development or to share the resulting outcomes”.

The main challenge to data sharing remains that many researchers believe sharing can be disadvantageous for them. As such, funders must demonstrate the value of data sharing as well as other outputs (e.g. material or software) by taking them into account in the grant review process. Additionally, to understand the significance of a study in its field, the scientific content of an article should be re-valued compared to its publication metrics. The idea is somehow to redefine criteria used for the grant application process but also maybe to ask applicants to demonstrate how they support the values of open research. Finally, the authors support the idea of “data authorship” where researchers would be cited when their data are (re)used.

These lines of thought are essential to ensure the use of research data to their fullest potential and even more so in the case of rare disease where data are more scattered.

For details, please read the full article.