In an effort being compared to the international collaboration that created the standards for the World Wide Web, 70 major research and health care organizations from dozens of countries announced Wednesday that they will work together to make medicine more effective by building a framework to share genomic data.
At this point, the project is in its earliest stages—a signed letter of intent to join a global alliance and a white paper describing the efforts. But the work of the global alliance could profoundly alter medicine and research, by taking information that today is siloed away in individual hospital or research databases, and making it available to clinicians and researchers across the world. Seven hospitals and research institutions from the Boston area have signed the letter.
Dr. David Altshuler, deputy director of the Broad Institute, a genomics research center in Cambridge and one of the organizers of a January meeting where the idea began to be hashed out, explained that the lesson of the past decade of genome research has been that DNA is most informative when it is aggregated into very large databases, from which the underpinnings of risk for complicated diseases can be sifted out.
Imagine, Altshuler said, that a child is born at a local hospital with a severe, apparently genetic condition. For some fraction of children, careful analysis of their whole genome sequence may pinpoint a mutation in their DNA that is causing the disease. But then what? Many of those mutations are not yet known. Many are extremely uncommon.
“Even in a big city like Boston, you might see zero. Or occasionally see one. You can’t discover anything from one observation,” Altshuler said. A system of standards and technology designed by the alliance could mean that when such a child is born and their DNA is sequenced, the information could be compared against a national or international database that protects patients’ identity but makes that telling clue far more powerful.
For example, the child’s doctor could find out if other hospitals have seen the same condition and have experience on what therapies to try or avoid. A team of doctors facing a medical mystery might be able to connect with others who have already solved some of the problems.
The ability to compare the genomes of millions of people could alter medicine profoundly, in a change that Brigham and Women’s Hospital president Dr. Betsy Nabel compared to the introduction of antibiotics, vaccines, or organ transplantation. It could allow doctors to get a better grip on what any given genetic risk factor for a condition such as diabetes or cancer really means for any individual patient.
“What we are realizing is that it’s insufficient for each hospital to do it’s own sequencing and have it’s own data set, of maybe thousands of patients,” said Nabel, who heads one of the half dozen Boston hospitals that has signed the letter. “What we really need to do is pool our data across hospitals, so that we have hundreds of thousands and maybe even millions of patients.”
Most of the detailed specifics of how the global alliance’s genome and clinical data-sharing project will surmount the many technical, logistical, and ethical issues remain to be worked out over the next one to two years, but Nabel said it was the right time to start.
An initial January meeting in New York City included a wide variety of stakeholders with expertise that ranges from patient care to funding research to technology, with representatives from the National Institutes of Health and the chief technology officer of Disney and Pixar.
Nabel said that a primary concern will be giving patients control and a choice over whether they want to share any information at all, and to protect the privacy of those who did.
One concern may be that genomic information could be used to identify someone, especially if it is coupled with clinical information. Yaniv Erlich, a fellow at the Whitehead Institute for Biomedical Research in Cambridge, earlier this year reported in the journal Science that it was possible to identify supposedly anonymous participants who had their DNA sequenced for research using simple Internet searches and publicly available databases.
Erlich said he was enthusiastic about the new effort, which he thinks will create uniform standards and a system that will clarify what information should be shared, and how. One of the problems thus far, Erlich said, is that every institution has had its own ethical rules, policies, and technological systems.
On the flipside, creating a giant database of genomic information also increases the risk. With so much data held under a uniform standard, any flaw could affect many people’s most personal information.
But Erlich said that the scrutiny that the consortium will bring to the topic will hedge against problems, and would improve on the security gap created by the sometimes hodgepodge way that genome data is currently compiled and shared.
“If you do it right, and you create this large collaboration with many people and integrate their input, you’re reducing the probability of something going wrong,” Erlich said. “The way we’re doing it right now, it’s like we’re developing the process as we go. Here is a group of really accomplished scientists ... willing to sit together and to think about how to put these standards together.”