Data sharing and linking are the main hurdles in the quest for new diagnoses and therapies
One of our missions is to directly and indirectly help patients by aggregating and linking both private and public data. A direct help is to facilitate awareness about health and disease by providing proper information about prevention, diagnosis and treatment, amongst other things.
The creation of an interconnected network of information sources that benefits all stakeholders in healthcare is a broader and indirect way to help. It shapes a landscape where data flows are used efficiently for the greater good of the patient. A specific case is the domain of rare or orphan diseases where Linked Data and open data become indispensable to continue the advancements in patient care and research.
February 8-9 last, the International Rare Diseases Research Consortium (IRDiRC) organized their third conference in Paris. This organization has already achieved some great goals in their six-year existence. It all started with the joint idea of the European Commission and the US National Institute of Health to coordinate their research funding on rare diseases and to make major investments in rare diseases research. After six years of IRDiRC, we noticed in Paris that most of the attendees know one another and are dedicated to collaborating on different levels to find new diagnoses and treatments for rare diseases. Here’s our report.
There was a great vibe at the conference. The opening session was dedicated to the achievements of the first six years of IRDiRC with talks by Dr. Ruxandra Draghia-Akli (European Commission), Dr. Francis Collins (National Institutes of Health (NIH)), Dr. Paul Lasko, (McGill University, Canada) and the IRDiRC president Dr. Christopher Austin (National Center for Advancing Translational Sciences (NCATS)).
Makoto Suematsu, from the Japan Agency for Medical Research and Development (AMED), explained in an entertaining way that data sharing is one of the main hurdles to overcome in the quest for new diagnoses and treatments for rare diseases. A vision we wholeheartedly share, as making data sharing easier is our bread and butter.
The majority of the subsequent talks explained the current initiatives to collaborate and exchange data. Due to the sensitivity of the data, some collaborations are limited to clinical geneticists in order to pinpoint whether gene variants are a factor in patients with a rare disease. For example, Dr. David Valle (John Hopkins University, USA) explained the recent progress made in Mendelian Genomics research with a focus on public resources such as PhenoDB, GeneMatcher, Geno2MP, MyGene2, and OMIM. His colleague Dr. Ada Hamosh (Johns Hopkins University, USA) explained the Matchmaker Exchange initiative that combines a number of these resources with other public databases such as DECIPHER and the Monarch Initiative.
Day two of the conference started with parallel sessions and it was an honor to present our data linking and search application DISQOVER. The (free) public version of our platform can be used to search across 120+ public data sources containing disease, gene, gene variant and other related data types. During our talk, we gave a demonstration of a local installation of DISQOVER only containing a mockup dataset with Electronic Health Records of rare disease patients from a hypothetic university hospital combined with internal biobanking and genetic testing data. We clearly demonstrated that the internal data remains in the local setup while an automatic connection is made to the public DISQOVER system to enrich this data with links to literature, drugs, medicines, genes, clinical studies, etc.
Below you can see a hypothetical patient with limited clinical data and with links to other different data types, such as the diagnosed disease, the gene harboring the detected variant in a patient sample, the hospital (organization), the sample information and the genetic variant.
If you follow the ‘gene variant’ link, you dynamically go from the local and thus private data to the remote and public data, resulting in an overview of variant-specific data aggregated from different public sources.
If desired, you can highlight the origin of each data point.
We concluded our short demo with a call to participate in projects to bring more rare disease-related data online. We have an ongoing plan to include additional rare disease data sources in the public DISQOVER version.
The final session of this two-day conference was dedicated to setting the goals of IRDiRC for the next ten years. This panel discussion was started by Chairman Dr. Christopher Austin who listed the current challenges in the field. Eroom’s law states that the cost of developing a new drug roughly doubles every nine years. This is even more relevant for orphan drugs and really hampers the introduction of new therapies for rare diseases.
How can we solve that? What should we strive for in the coming years?
First, diagnostics need to gear up. The newest screening technologies combined with data linking and proper annotation should lead to cheaper, shorter and higher quality diagnoses. Finding the right cure for a patient with a rare disease helps the patients but also creates a market for the drug which could, hopefully, lead to less expensive drugs. Patients having a previously identified rare disease (genotype-phenotype) should be diagnosed within one year of coming to subspecialty medical attention. In addition, diagnostic pipelines for currently unsolvable cases should be improved via a globally coordinated approach.
Secondly, we need more affordable treatments. If we want this and we assume that the budgets for treating orphan indications won’t increase in the major markets (EU, US and Japan), this requires that the production cost of a new drug should decrease. This means that the development pipelines should be shorter, less expensive and with higher success rates. Lastly, the regulatory process should be more straightforward, less bureaucratic and with better data processing. There is a need for a definition of standards that will allow the testing, adoption and payment for non-regulated interventions. These could be equally or more effective than regulated interventions.
An overarching vision
Finally, we need an overarching vision and IRDiRC could be the right platform for this. The current ecosystem will not solve these issues and the barriers are located on the operational, scientific and regulatory planes. We can only achieve these goals if we collaborate and share data. We, at ONTOFORCE, are ready to play our part in the exciting times to come.