A powerful new way to diagnose mystery illnesses
When you visit a doctor, the hope is that they'll run a few tests and you'll leave with a diagnosis. For some patients, though, it doesn't work out that way. This could be for any number of reasons, but a big one is this: They have diseases that are so uncommon they might never have been seen before.
If you have a rare, undiagnosed disease, there hasn't historically been much medicine could do for you. Now an initiative called the "UDN Gateway" based out of Harvard Medical School is beginning to change that. It opened on Sept. 16, is part of a larger initiative called the Undiagnosed Diseases Network, and promises to use big data and information sharing to identify diseases that have previously been understood only as collections of mysterious symptoms.
"The whole idea here is to enable to anyone with an undiagnosed disease to apply, be evaluated, and get referred with all relevant medical data," says Isaac Kohane, one of three Harvard researchers leading the UDN Coordinating Center based at Harvard Medical School.
Rare diseases affect an estimated 2-4 percent of the population. Each disease itself afflicts few people, but there are a great many of them, what Kohane calls "a long tail of rare diseases." These diseases often have a genetic basis, either as uncommon recessive conditions or as the result of "de novo" genetic mutations not present in either parent. They can take varied and often debilitating form.
They also yield a common, frustrating clinical experience for sufferers, who bounce from one medical specialist to another without getting any answers. "Our fragmented health care system is traditionally very poor at data-sharing," explains Euan Ashley, professor at the Stanford University School of Medicine and co-chair of the Undiagnosed Diseases Steering Committee, in an email. "We are incredibly excited to have a custom-made system tailored to the needs of these most complicated of patients."
The UDN Gateway, which is funded by the National Institutes of Health, is a way to open up the process. It's an online portal where people with rare diseases enter their medical histories, get connected with specialists around the country, and potentially find other patients suffering from the same unexplained condition.
"The more people you show the case to, the more likely you are to find the diagnosis," says Rachel Ramoni of Harvard Medical School who runs the Gateway along with Kohane and Alexa McCray.
To apply through the Gateway, patients submit a letter from a doctor along with a basic medical history. If their application is accepted, they're referred to one of seven clinical sites around the country (one of which is Harvard Medical School and its three teaching hospitals, Brigham and Women's Hospital, Boston Children's Hospital, and Massachusetts General Hospital).
At a clinical site, a likely next step is whole genome sequencing, a diagnostic technique that over the last decade has significantly bolstered the effort to identify undiagnosed diseases. Whole genome sequencing extracts an individual's complete genetic code and allows researchers to scan for anomalies that might explain the patient's symptoms. Fifteen years ago whole genome sequencing cost around one billion dollars; today the price-tag is in the low-thousands and dropping, making the technique practical on a large-scale clinical basis.
From a whole genome sequence, researchers might identify a handful of candidate genes that could explain the condition. One next step is to experiment with those genes in fruit flies or zebra fish, in order to tease out which malfunctioning genes are capable of causing which symptoms. A different approach is to search for patients with similar genetic abnormalities and similar symptoms. This requires a tremendous database of genetic information and it's where the information-sharing power of the UDN Gateway really comes in.
"You find the first case of disease and think you've found the genetic cause of it, but you can't be certain you've found the cause until you find a second and third case," says Ramoni.
Determining the cause of a disease is often a victory in its own right for people who've spent years trying to figure out what's wrong with their bodies.
"It's hard for people to understand who have not had many years in the wilderness with an undiagnosed disease," says Kohane. "Just knowing the cause is a great source of comfort for the families."
Diagnosis is also a necessary first step to searching for treatments, though it's not enough by itself. Doctors can't run clinical trials on a single patient; once they find others suffering from the same condition, the possibilities multiply.
"It's about forming the community you need to advance research on the condition," says Ramoni. "Once you have a pool, then you can start doing interesting research."
In its first two weeks the UDN Gateway received 40 applications and 63 phone inquires. Over the next six months, researchers at the Harvard Medical School Coordinating Center plan to really begin publicizing the resource, so that doctors around the country—and eventually the world—know to refer patients when confronted with a mystery disease. "
"It is our hope that the Gateway will serve as a model for rare disease research, minimizing the practical distance between families and subject experts who may be anywhere in the world," writes David Adams, a pediatrician with the National Institutes of Health Undiagnosed Diseases Program.
The step of simply generating and sharing information sounds modest compared to the development of cutting-edge therapies. But, it's hard to underestimate the effect that putting patient information in front of more doctors and cross-referencing patient genetic data can have. The potential has already been demonstrated by the Undiagnosed Diseases Program, which NIH began as a pilot in 2008. That program grew to evaluate 100-150 undiagnosed patients a year. It achieved over a 30 percent diagnosis rate on some of the hardest unknown cases out there- all using the same approach that the UDN Gateway is now extending to the entire population.