Rapid genomic test developed to diagnose any human infection

A single genomic test effectively detects a broad range of infection-causing pathogens in the central nervous system and lungs, two studies in human patients have shown.

Infections such as meningitis, encephalitis or pneumonia can become life threatening if not treated quickly. To diagnose these infections, several dedicated tests are often required, taking days to weeks. In the new studies, authors from University of California, San Francisco (UCSF) report the capacity of a technique called metagenomics next-generation sequencing (mNGS) to single-handedly detect the pathogens causing these infections, in patients who are suspected of having an infection but are difficult to diagnose

'By replacing multiple tests with a single test, we can take the lengthy guesswork out of diagnosing and treating infections,' said senior author Professor Charles Chiu, professor of laboratory medicine and infectious diseases at UCSF.

mNGS analyses the nucleic acids present in a sample, both DNA and RNA. After removing sequences from human origin, it matches the results to over 68,000 genomic sequences of known pathogens. The approach was first developed as a clinical test in 2016 to aid the diagnosis of unexplained neurological conditions, and was outlined in the journal Genome Research.

The first study, published in Nature Medicine, reports the test can be used to diagnose infections in the central nervous system. Researchers analysed 4828 samples of cerebrospinal fluid, the liquid surrounding the brain and spinal cord, tested between 2016 and 2023 at the UCSF clinical microbiology laboratory. Nearly one in seven of the samples tested positive for one or more pathogens, including bacteria, DNA and RNA viruses, fungi and parasites.

To compare the performance of the assay with other diagnostic tests, the authors extracted clinical information from 1164 samples from UCSF cases. On average, the patients had 20.2 microbiological tests performed before diagnosis. mNGS alone was found to detect the pathogen responsible for the infection in one in five of the samples tested in a subset.

The test's sensitivity, a metric reporting the ability to identify a pathogen correctly, was higher than standard microbiology testing of cerebrospinal fluid or standard blood tests involving the detection of antibodies against a pathogen in the body.

The second study, published in Nature Communications, used mNGS to identify respiratory infections. The authors adapted the test to respiratory fluid (swab) samples and introduced automated machinery to make it faster. These modifications reduced turnaround time from just under four days in the spinal fluid assay to 14-24 hours for the swab assay, with less than two hours of 'hands-on' time.

'Our goal was to have the entire process completed within 12 to 24 hours, giving a same-day or next-day result,' said Professor Chiu.

Rapid detection of respiratory viral infections could be beneficial in cases of pandemic threat. To test the ability of mNGS to detect new viruses or strains affecting humans, the authors generated swabs containing viral sequences of viruses such as SARS-CoV-2 or influenza. Then, they tested if mNGS could be used to recognise them by matching the results to the genomes of other viruses infecting plants or animals. They found that the test could hypothetically identify all of them.

Both the cerebrospinal fluid and respiratory versions of the mNGS test have received breakthrough device designation from the US Food and Drug Administration.