Contributed Commentary by Robert Schlaberg
March 23, 2018 | In the research community, metagenomics has yielded new insights into microbial communities and emerging pathogens, giving scientists reams of novel information about everything from optimal soil for crop production to new means of tracking criminal activity. Now, clinical labs are beginning to adopt metagenomics, with consequences that may be just as game-changing for diagnosing and treating patients with infectious diseases.
In recent years, clinical labs have rapidly shifted from culture-based and single-organism tests to panel-based tests for many applications. This approach delivers medically relevant answers faster, compared to the conventional method of serial testing for possible culprits over days or even weeks. Still, though, even the broadest panel-based tests can only be used to find a preselected subset of possible pathogens.
Metagenomics turns the conventional approach on its head. Instead of a test to determine whether one of a handful of organisms is present in the patient’s sample, sequencing-based metagenomics begins with the broadest possible query: what’s in this sample? The list of every organism detected can be quickly generated and prioritized with state-of-the art bioinformatic tools to home in on the pathogen most likely to be causing the patient’s symptoms.
To be sure, metagenomics is not appropriate for all cases. Patients who are well-served with existing molecular diagnostics are not good candidates for this kind of test. But for complex cases where conventional tests are unlikely to deliver answers, metagenomics offers clinical labs a path forward to generate medically actionable results.
Hypothesis-Free Testing
Clinical metagenomics lays the groundwork for hypothesis-free testing in a healthcare setting. The shift from testing for preselected sets of pathogens to unbiased analysis of a sample makes it possible to identify rare causes, pathogens that are difficult to culture, new strains, complex co-infections, and other situations that can confound conventional diagnostic testing.
The concept is very similar to how metagenomics has been used for research. A patient’s sample is prepared and loaded onto a next-generation DNA sequencer for shotgun analysis of any non-human nucleic acids detected. Sequences are then compared to massive databases of quality-controlled microbial reference sequences using validated interpretive criteria, and matches are used to determine the individual members of the microbial population in the sample.
Metagenomics results offer a comprehensive view of DNA and RNA in a sample, so naturally these tests will detect any benign microbes likely to be found in many types of samples (the microbiome) in addition to the pathogen(s) responsible for the infection. A key component for any successful clinical metagenomics test is an accurate and reliable process to identify and prioritize potentially relevant microbes before results are reported back to a physician.
How It Works
To illustrate the concept, let’s take a look at pneumonia. One of the biggest challenges in treating patients with pneumonia is that this syndrome can be caused by many different pathogens. More than 1,000 bacterial, viral, and fungal pathogens can cause pneumonia in a susceptible patient. This poses challenges for diagnostic testing and is part of the reason why as many as 60% of patients tested for pneumonia never get an answer about what caused the infection.
That would be an academic problem if it weren’t so closely linked to the need for more effective treatment based on accurate diagnosis, especially in severely ill patients. Here in the U.S., more than 50,000 people (mostly adults) die from pneumonia each year. Around the world, the disease burden is particularly high for children; almost 1 million kids younger than age 5 die annually from pneumonia. Surely we can do better.
I believe that situations like this call for creative new approaches to diagnosis, and shifting to a hypothesis-free testing protocol for complex cases may offer a major advance in improving outcomes and reducing costs to the healthcare system.
Moving Forward
Whether it’s pneumonia or any other difficult-to-diagnose infectious syndrome, unclear diagnostic results are a significant contributor to the overuse of antibiotics — and, by extension, the rise of antibiotic resistance. Now that clinical metagenomics offers a more comprehensive means of tracking down the cause of an infection, it’s time to re-evaluate many medical situations where conventional testing leaves too many ambiguities. Recent developments in next-generation sequencing and DNA/RNA analysis techniques provide exciting opportunities to help deliver better treatment to patients in need.
The use of clinical metagenomic testing to complement molecular diagnostics and other conventional assays will help medical teams solve complex cases, particularly among high-risk patients such as the elderly and immunocompromised. This should make it possible to improve care metrics across the board, from shorter hospital stays to lower readmission rates, with the very feasible goal of saving lives.
Robert Schlaberg is an Assistant Professor of Pathology at the University of Utah School of Medicine and a Medical Director at ARUP Laboratories. He is also Co-Founder, Chief Medical Officer at IDbyDNA. Robert co-developed Taxonomer, the metagenomics data analysis platform, utilized by IDbyDNA as its central technology. Robert is a board certified Clinical Pathologist and Medical Microbiologist. He can be reached at rschlaberg@idbydna.com.