OmicsXchange episode 3: How Data Sharing Advances Rare Disease Research: An Interview with Heidi Rehm

28 Feb 2020

29 February 2020 marks the 13th International Rare Disease Day. Established by the European Organisation for Rare Disease, this day of observance aims to raise awareness and improve access to treatment. We’re here with human geneticist Heidi Rehm to talk about the opportunity of genomic and health-related data sharing to advance rare disease research.

Angela Page: Welcome to the OmicsXchange — I’m Angela Page. Tomorrow marks the 13th International Rare Disease Day. Held each year on the last day of February since 2008, this day of observance aims to raise awareness and improve access to treatment. We’re here with human geneticist Heidi Rehm to talk about the opportunity of genomic and health-related data sharing to advance rare disease research. Heidi is incredibly dedicated to responsible genomic data sharing and she has been with GA4GH since the beginning. In 2008, she joined our executive team as a Vice-Chair. For her day job, Heidi serves as the Chief Genomics Officer at Massachusetts General Hospital, and as the Medical Director of the Clinical Research Sequencing Platform at the Broad Institute, where she also Co-leads the Center for Mendelian Genomics. This may sound like a lot, but it only scratches the surface of Heidi’s contributions to this community. You can read her full bio on our website, www.ga4gh.org. Welcome, Heidi.

Heidi Rehm: Thank you, it’s a pleasure to be here.

Angela Page: To get us started, why don’t you provide us with a little background on the scope of rare disease around the globe today?

Heidi Rehm: So rare disease—the term rare disease—is a bit of a misnomer because while each disease is rare in itself, the collection of rare diseases together are actually quite common. In fact, there are over 300 million people worldwide who are living with a rare disease. Some of these diseases individuals are clearly born with. Other diseases come later on in life and may only be adult onset. They’re still different types of rare diseases, which are afflicting millions of people worldwide.

Angela Page: And what is the current landscape of rare disease research?

Heidi Rehm: There’s a lot of people doing research on rare diseases–some of those researchers may target specific disease areas like cardiovascular disorders, or neurodevelopmental disorders. Other groups are studying rare diseases in aggregate, finding new methods to solve many different types of disorders. So there’s a lot going on in terms of rare disease research, but there’s still a lot to be done. Today when we test a patient that we suspect has a genetic rare disease, only about quarter of the time do we actually find the cause of disease and those individuals. And so we know we are missing many of the causes of rare disease and still have a lot of research to do.

Angela Page: You’re involved with a few different rare disease projects like the Broad Center for Mendelian Genomics that I mentioned at the beginning. Can you describe the aim of this project?

Heidi Rehm: So our Broad Center for Mendelian Genomics is one of four different centers that are funded in the U.S. to do broad research into rare diseases and discover new genetic causes of rare disease. So all four of these centers, including ours at the Broad, bring in samples from collaborators all around the world from patients with rare disease, and then we apply techniques like exome sequencing; genome sequencing; RNA analysis; we also use new methods for trying to find different types of variants that are either in genes or outside of genes and these non coding regions that are harder to look at, and we apply lots of different strategies to try to figure out what variant in what gene might be causing disease in a given patient. So sometimes we find that the patient has a variant in a known gene, and we return that information. Other times we find variants in new genes and have to build up evidence in order to implicate those genes in disease.

Angela Page: You also helped launch the Matchmaker Exchange to facilitate this kind of research. Can you explain what this is?

Heidi Rehm:This is a project that started a number of years ago where we realized that there were lots of groups around the world doing research in rare diseases. But if you only have one or a handful of patients with a particular rare disease, it’s difficult to build enough evidence to implicate a new gene in disease. And therefore, we found that if we could bring different groups together, we would be able to build more evidence for these individual new genes we were finding. So we gathered as a group in one of the major genetics meetings a number of years ago, and agreed that we would build a platform that would allow us to match different patients with different candidate genes with each other across this platform.

The Matchmaker Exchange is actually one of the first platforms that works as a federated network. And what that means is that each group can store their own data, in their own database, in their own country. And some countries have laws that prevent that data from exiting the country. But they can store their data in their own database, and then through APIs—or application programming interfaces—they can send a query to all of the other databases and say, “I have a patient with a mutation in this new gene, does anyone have a match?” It’s kind of like Go Fish. And then the other databases can automatically respond and say, “Oh yes, I have a patient with a mutation in that same gene.” And the system sends emails to each collaborator to signify the match, and then the two collaborators share more details of their cases.

Angela Page: And what sorts of data are being shared?

Heidi Rehm: You know, what is the disease that patient had? What are specific phenotypes and clinical features of that disease? And they figure out if the phenotypes match, and if they do, that helps build evidence that that particular gene they matched on may actually be the real cause of disease. Sometimes, for very rare phenotypes with just specific features, iIt may only require one match with one other patient to build enough evidence. Other matches, other diseases that are more common like autism, other neurodevelopmental disorders, are more nonspecific and may require more matches. Sometimes through the Matchmaker Exchange, we’ve matched 20 different patients, all with mutations in the same gene and similar phenotypes, that allows us to then implicate that gene in disease.

Angela Page: So how do patients benefit from projects like the Matchmaker Exchange and the Center for Mendelian Genomics?

Heidi Rehm: So the patients often have a disorder that they’ve been seeking an answer to: “Why does my child or why do I have this disease?” And they constantly are going back to physicians trying to get answers. And sometimes they want to know whether they are at risk of having a second child with the same disorder. Sometimes they just want to know what’s in store for their child who has the disease. Is it going to get worse? Is it going to stay the same? Are new things going to develop? These are incredibly anxiety-provoking questions that these families want to know and want to have answers to. And so there’s so much anxiety over whether it was their fault, what’s to come, whether another child they have would have this disease also. And we call that a diagnostic odyssey that these families are on. So when we can find the cause of their disease, go back to them and say, “This is the cause—it wasn’t your fault.You know, it’s because of a de novo mutation that arose, and you didn’t cause that. Or it’s a recessive disorder, you and your husband are both carriers, and you each contributed a mutation, it’s not your fault. This is what happens when rare disease happens.” This can help these families take that blame away from themselves, which they shouldn’t have, but we know they do often. So that’s one thing that can help, is figuring out why they had it.

The other questions they have around the disease itself, and is it going to get worse. And by giving them an answer, they can then find other families who also have the same cause of disease, and they might have older children or be adults at that stage and be able to go to them and say what’s happened over your lifetime? What’s known about this disorder? And therefore they can anticipate what the future holds for them. And so that’s important to understand the disease better.

And then of course, most importantly, what we want for all of these patients is that their diseases can be treated, managed better, sometimes even cured. And although it is rare that we fully cure a rare disease for most these patients, in a lot of cases, we have ways of better managing or treating the symptoms if we understand the cause of them. And by identifying these causes, pharmaceutical companies can work to come up with treatments and management strategies that are more targeted to their underlying disease, and that can have beneficial outcomes for these patients.

Angela Page: So just having the diagnosis is helpful and empowering for both the patient and their family. Do you have an estimate of how many new genes have been identified through the Matchmaker Exchange?

Heidi Rehm: So we do maintain a website called matchmakerexchange.org. And we have statistics there on the number of cases from which candidates have been pulled, how many gene candidates are in Matchmaker Exchange, and publications that we are aware of that resulted from matchmaking across the federated network. My guess is we have well over 100 genes that have been identified through matchmaking on this federated network. And many of those are in the process of being published and aren’t yet up on the website, because sometimes it can take a while to get more functional data to provide sufficient evidence to get a publication out even if we’re pretty sure we’ve got the right gene.

Angela Page: Over 100 genes, that’s a lot. Does that mean that when two researchers match with each other, it usually leads to the discovery of a new gene?

Heidi Rehm: I will say most matches are not productive in that most times when you match, the patient’s phenotypes aren’t a match. And it was a variant that looked suspicious, but in the end wasn’t the real cause of disease. But sometimes, the lack of matches actually rule out the gene and allow the researcher to move on to other things. And so even when a match isn’t productive in implicating a new gene, it can be productive in getting rid of a candidate that’s not a good candidate and being able to move on to new candidates that might be, using other methods.

Angela Page: So you have a unique perspective because you’re a genomics researcher, you run major global data initiatives, and you run a lab that does genetic testing for patients. How have these experiences shaped your belief in the power of international data sharing?

Heidi Rehm: So I’ve seen many examples of patients who would have gotten a result that is either wrong or simply uncertain significance of a variant that is really not helpful for patients simply because the evidence to help classify that variant and give a diagnosis was in someone else’s lab that wasn’t shared. So you know, as I ran clinical laboratories and have been doing genetic testing for patients over the last twenty years, I have watched many patients not receive an answer or even a right answer. And so that has led me to really focus on global data sharing, where we can bring information together. Often that information is not published and not accessible to anyone. But by sharing it in public databases, we can bring that data to each other and use it to help accurately diagnose patients with different genetic disorders that then help those patients get the answers that they need.

Angela Page: So how can GA4GH have a role in shaping that effort?

Heidi Rehm: So the Global Alliance plays a critical role in helping us share data because if we’re each collecting our data in different formats, with different standards, different terminologies, then when we go to share our data, it’s ineffective. We can’t put it in the same categories. We can’t understand what each other’s data means if it’s all been formatted and structured and defined differently. So through the Global Alliance, we develop the standards that define the different terms and fields and data structures that we need so that when we combine our data, it’s a productive combining. And we can build the evidence across many different groups that we need to actually understand genetic variants and the genes that are associated with disease and the many types of information needed to help patients improve their lives.

Angela Page: Fantastic. Thank you so much, Heidi.

Heidi Rehm: Your welcome. It was a pleasure.

Thank you for listening to the OmicsXchange—a podcast of the Global Alliance for Genomics and Health. The OmicsXchange podcast is produced by Stephanie Li and Caity Forgey, with music created by Rishi Nag. GA4GH is the international standards org for genomics, aimed at accelerating human health through data sharing. I’m Angela Page and this is the OmicsXchange.