Solving Medical Mysteries at the Stanford Center for Undiagnosed Diseases

Posted By Donna Alvarado, Medical Editor

Presented by Jon Bernstein, MD PhD
Associate Professor of Pediatrics
Matthew Wheeler, MD PhD
Assistant Professor of Medicine

February 28, 2018

Every so often, you might wake up feeling sick one morning with a fever, aches, and pains. Is it a cold or flu? Often the symptoms fade and disappear before you see a doctor.

For some people, the symptoms persist—for weeks, even months—and don’t stop. If symptoms grow severe, they may eventually go to a doctor and get tests—only to get no answers and no relief. Some of them are able to consult specialists and get advanced tests and procedures: all to no avail.

It’s these undiagnosed patients who are the mission of the Stanford Center for Undiagnosed Diseases, said Matthew Wheeler, MD PhD, an assistant professor of medicine and medical director at the center. The center is trying to end what he called the “diagnostic odyssey” that these patients often undergo, Dr. Wheeler said in a public lecture at the Stanford Health Library.


Before coming to the center, patients have often undergone years of illness and failed medical attempts to identify their disease. By the time they reach the center, their frustration has accumulated to the point where they may feel like the figure in a famous painting by Edvard Munch titled “The Scream,” Dr. Wheeler said.

“Eventually, the patients are like this,” Dr. Wheeler said as he showed an image of the screaming figure on a screen to the audience. “Frankly, the physicians are like this, too.”

Research only

The center accepts only those undiagnosed patients who are eligible to participate in research studies that focus on rare diseases. There are about 6,000 different rare diseases known so far. The center provides state-of-the-art diagnostic tools, like genetic testing called whole genome sequencing.

Genome sequencing and other cutting-edge tests are done to uncover any molecular abnormalities that may underlie a disease, said Dr. Wheeler, who described the program in a public lecture at the Stanford Health Library.

Some participants at the Stanford Center for Undiagnosed Diseases may also get additional advanced diagnostic tests that profile their metabolic and immune system function to uncover more clues to their illness.

What is rare?

What constitutes a “rare” disease can vary from one place to another. In the United States, rare diseases are defined as 1 in 1,625, said Jon Bernstein, MD PhD, a geneticist and associate professor of pediatrics who is also a medical director at the Stanford center. In Europe, rare is usually defined as 1 in 2,000.

While that may be an individually rare condition, if we count all these cases overall they collectively add up to tens of millions of people. “Millions of children are born with them every year,” Dr. Bernstein said.

Stanford’s center is 1 of 12 sites in the nationwide Undiagnosed Disease Network (UDN), a federally funded project that also has links to similar centers around the world. The network has been sharing information and strategies for improving patient care for about a decade.

Not all diagnosed

So far, the UDN has received more than 3,200 applications from undiagnosed patients across the U.S. Of these, the network has accepted and evaluated more than 1,000 participants. At this time, 298 of 1,089 participants have obtained a diagnosis so far, Dr. Wheeler said.

To pursue a diagnosis, it may not be enough merely to obtain a patient’s test results showing a genetic variation or immune system pattern. Doctors have to dig deeper to analyze the relationship of that genetic variation or pattern with specific symptoms.

“We collect as much information as we can about their genome,” Dr. Bernstein said. “We collect all the information about their health.”

Advanced tests

Among the advanced tests the center doctors may use for patients include:

  • Genome sequencing
  • Tests of the immune system, including key biochemicals called cytokines
  • Tests of the body’s metabolic processes, especially metabolites, the biochemical products of metabolism that influence energy production in cells
  • Online worldwide searches for information on patients with similar symptoms and genetic variations. Are there other cases like this? Clues may be found in:
    • Case reports
    • Journal studies
    • National (or international) databases

A decade of fevers

One 70-year-old man came to the Stanford center after enduring a decade of recurrent fevers, rashes, pain, and weakness that other doctors couldn’t diagnose. He had seen other specialists at multiple medical centers and had many tests, procedures, and biopsies that didn’t find the cause.

After coming to the Stanford center, doctors there ran state-of-the-art diagnostic tests that profiled biochemicals called cytokines produced by the body’s immune system cells. Cytokines can influence the immune system in ways that lead to fevers and other symptoms.

Using the results of those tests and others, Stanford doctors combed worldwide reports of similar test results and symptoms to match the patient. That eventually led them to discover a diagnosis: Schnitzler’s syndrome. This syndrome is a rare disorder with similar but variable symptoms that come and go in patients for many years.

With a diagnosis came treatment that finally gave the man relief from his symptoms. That’s the kind of outcome that the Stanford center seeks for as many patients as possible.

Benefits ahead

Even those who don’t find a diagnosis soon after coming to the Stanford center may get benefits in years ahead. That’s because their test results may prove valuable when newer tests and discoveries shed more light on their condition in the future.

“We’d really like to take this to the next place. How can we do this better?” said Dr. Bernstein.

  • One way is to improve lab efficiency in analyzing all the genetic data obtained from a patient. Current methods are still “painstaking,” he said, and can take several days to look through all the genetic variants in one person that might be significant.
  • Another way is to develop better computerized systems for extracting medical information from patient medical records. A program called ClinPhen automates extraction of patient information from electronic medical records to speed up genetic disease diagnosis.
  • A third way is to improve literature searches that track all research studies and case reports published in medical journals. A new program called AMELIE is combing medical publications, using automated software, to find articles about genetic variations relevant to an individual patient.

Shedding light

Using all these advanced tools isn’t a guarantee that any one patient will get a diagnosis. But sometimes, patients who don’t get answers at first may get help years later as new information emerges.

“I’ve had patients I haven’t seen in 5 or 6 years,” Dr. Bernstein said. “Things didn’t make sense for a diagnosis before, but now there’s new information that sheds light,” he said.

For more information:
Stanford Center for Undiagnosed Diseases

Matthew Wheeler, MD PhD

Jon Bernstein, MD PhD

Stanford Health Library can do the searching for you. Send us your medical questions.