Research

DNA Tests for Infectious Diseases

Using sequencing to precisely diagnose diseases allows doctors to precisely treat them.

By Aimee Swartz

Images courtesy: NIAID

In April 2014, 53-year-old Lee Benson of Phoenix began experiencing pain he could only describe as “strange.” After walking just a few blocks, he developed extreme soreness in his legs.

“It felt like I had hiked the Grand Canyon,” says the CEO of two aerospace maintenance companies and a business management software company. The pain eventually progressed to his shoulders, around his neck, and then migrated to the upper part of his rib cage. It got so bad that at certain times, he could not even raise his arm to shake hands. In search of answers, he went to the Mayo Clinic’s Arizona campus.

“They ran all kinds of tests,” he says. After working his way through the neurology and rheumatology departments, nothing was found except a positive result for ANA antibody and scleroderma 70. These can indicate scleroderma, a chronic autoimmune disease in which connective tissue in the skin, blood vessels, internal organs, and digestive tract thickens and tightens.

“The results were positive but not wildly positive,” says Benson, adding, “I have none of those symptoms whatsoever.”

A friend, who is a geneticist, suggested he get his blood sequenced to figure out the source of his mysterious pain. Some infectious diseases have symptoms that mimic those of autoimmune diseases — fever, headache, and muscle aches — although prescribing an immune-suppressing drug to a patient with a bacterial infection can be disastrous.


Paving the Way

Before next-generation sequencing, the only hope for identifying a possible infectious disease has been by culturing the bug. That method is both time-consuming and sometimes ineffective. For Benson, his case might have come too early. Next-generation sequencing is just now being used to diagnose pathogens, but it’s still in the research phase. Although his friend had ties with the University of California, San Francisco (UCSF), which has a lab that is using this technology to study infectious diseases, he was unable to get his genome sequenced.

“There is a limited number of folks who are accepted, and there is no way to do that other than to be put in sort of a lottery,” says Benson.

Joe DeRisi, a biochemist at UCSF, explains just how low the probability of someone in Benson’s situation is of getting his genome sequenced.

“We’re a small research lab, and the only reason at the moment that patients come to us is that one of those docs knows Michael or knows me,” he says. “We don’t have a systematic procedure in place where all patients will come to us and get sequenced. We would love that, but we would also be bankrupt, because we’re doing this all for free right now.”

Although patients have been contacting DeRisi’s lab for years in hopes it could help identify their illnesses, one patient’s story created an even bigger flurry of requests after its positive conclusion was published in the June 2014 issue of The New England Journal of Medicine. In 2013, for the first time, doctors were able to use the powerful new technology to diagnose a puzzling case of encephalitis that nearly killed 14–year–old Joshua Osborn.

Encephalitis is a rare condition in which the brain swells with fluid. It is most often caused by a viral infection, particularly in people with compromised immune systems, but it can also be caused by bacterial infections, noninfectious autoimmune diseases, and other illnesses (see sidebar, “Virus vs. Bacteria,”). The culprit behind Osborn’s deadly case was anyone’s guess.

It was not the first time Osborn’s life had been on the line. Raised in Cottage Grove, Wisconsin, the teen was born with severe combined immunodeficiency disease (SCID), the extremely rare genetic disorder also known as “bubble boy syndrome,” in reference to a Texas boy who lived his entire life inside a sterile plastic bubble. Babies born with SCID have such a weak immune system that they are vulnerable to any and all bacteria and viruses; even a common cold can be deadly. But thanks to a partial bone-marrow transplant at the age of five months that restored his immune system, Osborn had a relatively healthy childhood.

Then, in September 2012, he developed pink eye in both eyes, along with six days of fever and debilitating headaches. The strange symptoms resolved without treatment, but by April 2013, they returned, along with severe nausea and vomiting.

As his symptoms worsened, he was admitted to the hospital three times over three months where he was tested for an extensive number of infectious diseases. All the tests came back negative.


A Guessing Game

Osborn’s case illustrates how diagnosing serious infectious diseases is a bit like a guessing game. Pinpointing the cause of encephalitis is particularly challenging because it is so broad in scope. In fact, researchers at the California Encephalitis Project, a surveillance program that investigated 1,570 cases from 1998 to 2005, found that the cause of encephalitis was unknown in 63 percent of patients.

This is a problem because a diagnosis informs treatment choices. For example, viral encephalitis cannot be treated with antibiotics, just as bacterial encephalitis cannot be treated with antiviral medications. Delays in diagnosis can cause treatment delays.

The diagnostic process usually begins by guessing the most likely cause of the infection and then ordering a series of laboratory tests to confirm or rule out particular causes. Blood tests can detect antibodies or antigens to help diagnose some infectious diseases, as can cultures of samples isolated from the infection site. The problem with both is they take time that many patients do not have: Results can take days and, in the case of cultures, sometimes weeks.

Polymerase chain reaction (PCR) techniques have significantly improved doctors’ ability to rapidly detect pathogens that were previously difficult or impossible to identify by traditional methods. PCR allows doctors to look for the DNA of a virus or bacterium within a small sample of body fluids or tissue. However, PCR has many pitfalls, including less-than-ideal rates of both false-positives and false-negatives.

“Because most tests are based on direct detection on the organism, it can be challenging to diagnose pathogens that are present at low levels or those pathogens that are unculturable or grow poorly in culture,” says Charles Chiu, a pathologist and the associate director of the Clinical Microbiology Laboratory at UCSF.

Making matters worse, Chiu says, is that “it may be hard to obtain invasive material directly from the site of infection, such as bone biopsy in the setting of bone infection or brain biopsy in the setting of encephalitis.”

Without a diagnosis, selecting the correct treatment for Osborn was next to impossible.


A Race Against the Clock

In July 2013, Osborn’s condition became more severe. An MRI showed that his brain was dangerously inflamed, but neither a spinal tap nor a brain biopsy revealed the cause of his encephalitis.

Suspecting Osborn’s condition might not be due to an infection, but could be autoimmune in nature (where the body’s immune system turns inward to attack itself), doctors tried treating him with corticosteroids, normally an effective treatment for autoimmune encephalitis. This made things worse; his condition deteriorated into uncontrolled seizures. Puzzled, doctors put the boy in a medically induced coma in an attempt to buy time.

Having hit a dead end, doctors overseeing Osborn’s care consulted with infectious disease researchers at UCSF, who had been working on identifying pathogens based on their DNA for more than 10 years through genomic sequencing technologies. Could the same technology be used to help Osborn? His parents agreed to give it a try.

Genome sequencing has been increasingly used to detect, analyze, and stem the spread of mysterious disease outbreaks around the globe, from the peculiar outbreak of enterovirus D68 in children and teenagers across the U.S. (see sidebar “Making Sense of Mystery Illnesses”) to the Ebola virus outbreak that ravaged several West African countries.

Genome sequencing technology, says Chris Woods, a professor of medicine, pathology, and global health at the Duke Global Health Institute, “allows for more granular data generation and the potential to distinguish between closely related organisms for an improved understanding of pathogen transmission.”

The technology, Woods adds, is “extremely important” in assessing “high-consequence” pathogens, such as Ebola, dengue, and influenza. It can also provide detailed information about a specific strain — influenza H1N1 versus H3N2, for example — and suggest which kinds of drugs may be most effective and which ones infections are resistant to.

After receiving the sample of Osborn’s blood and spinal fluids, UCSF researchers, led by Chiu and DeRisi, performed an experimental sequencing technique that analyzes DNA sequences at rapid speed. Chiu then used special software he developed, named SURPI (Sequence-Based Ultra-Rapid Pathogen Identification), to rapidly match sample DNA with the DNA of organisms contained within ever-expanding national genome data-bases. This analysis, which required days or even weeks a few years ago, now can be completed in less than two hours.

Within 96 minutes of performing the new genomic sequencing technique, UCSF researchers had pinpointed the cause of Osborn’s illness as leptospirosis, an infection caused by a bacteria native to the Caribbean and warmer climes. Doctors suspect Osborn came in contact with it while on a fresh water swim in Puerto Rico the year before.

“Leptospira is an example of a bacterium that does not grow well in culture, so cultures of cerebrospinal fluid and serum were routinely negative,” explains Chiu. As dangerous as leptospira can be, it is easily treated with penicillin. Within days of being treated with high doses of the drug, Osborn emerged from his coma and improved in days.

Osborn’s remarkable recovery points not only to a healthy future but to the promise of using genomic sequencing for diagnosing even the most vexing infectious diseases. Chiu, DeRisi, and other UCSF researchers are now heading up a project to develop a test to diagnose the causes of encephalitis and, eventually, any infection.

Once validated, Chiu says, it will serve as a single test to detect the full spectrum of pathogens — virus, bacterium, fungus, or parasite — rather than testing for one suspected pathogen after another. Eventually, he believes it will become “part of routine clinical testing” for infectious disease diagnosis.

Until then, Lee Benson will have to wait.

“The doctors at the Mayo have told me that the good news is, it’s probably not life-threatening,” he says of his condition. An anti-inflammatory diet has helped mitigate his discomfort; where he once had pain levels reaching an 8 or 9 are now at a 4 or 5. His doctors did advise him that if the pain ever gets bad enough, the cause of it will likely surface.

“This type of diagnosis using genetic sequencing is the future,” says Benson. “It feels like everything before that is back in the caveman days.”