One test could identify the infection making seriously ill patients sick

A diagnostic technique capable of searching for a wide range of viruses, bacteria and fungi using a single patient sample could transform the way serious infections are diagnosed in NHS hospitals.

The metagenomics test, developed by researchers working at King’s College London and Guy’s and St Thomas’ NHS Foundation Trust, uses genetic sequencing to search broadly for microorganisms rather than requiring doctors to request separate tests for individual suspected infections. The approach has already been tested in intensive care patients with respiratory infections and is being piloted across multiple NHS hospitals.

For patients who are seriously ill, identifying the organism responsible for an infection can sometimes involve a series of different investigations. Traditional microbiology may use cultures, PCR tests and other techniques to look for particular bacteria, viruses or fungi, with clinicians often having to make decisions about which pathogens are most likely to be responsible.

Metagenomics approaches the problem differently. Rather than beginning with a suspected pathogen and designing a test to find it, scientists analyse genetic material contained within a clinical sample and compare the sequences they find against databases containing the genetic signatures of known microorganisms.

In simple terms, conventional testing can sometimes resemble asking a series of questions – is it influenza, is it Covid, is it a particular bacterium? Metagenomics attempts to ask a much broader question: what microorganisms are present?

Developing a test capable of answering that question quickly enough to influence treatment has been challenging. One major obstacle is that a clinical sample contains a huge amount of the patient’s own genetic material, which can overwhelm the much smaller amount belonging to the pathogen doctors are trying to find.

Researchers developed a technique to remove much of that human DNA while preserving genetic material belonging to bacteria, viruses and fungi. Human cells in the sample are physically broken apart using tiny beads before an enzyme breaks down the released human DNA. The remaining genetic material can then be sequenced and compared with databases of known pathogens.

The resulting process is considerably faster than many earlier approaches to metagenomic testing. Researchers reported that results could be produced within around seven hours of the laboratory receiving a sample, potentially making the technology useful for real clinical decisions rather than providing an answer days after treatment has already begun.

That speed can be particularly important in intensive care, where doctors may need to begin treatment before they know precisely what is causing an infection. Broad-spectrum antibiotics are sometimes used while clinicians wait for laboratory results, but unnecessary or inappropriate antibiotic use can contribute to side effects and antimicrobial resistance.

The technology was put into clinical practice during a pilot at Guy’s and St Thomas’, where researchers used metagenomic testing in real time for patients with respiratory infections in intensive care.

The results demonstrated why the approach has attracted interest. Metagenomic testing identified the pathogen responsible for infection in 30% of patients in whom standard testing had failed to find the cause. Results from the test also led clinicians to change antibiotic treatment in 30% of patients, while 15% were prescribed additional medication intended to modify their immune response and support recovery.

The potential significance goes beyond simply replacing several laboratory tests with one. Because metagenomic sequencing does not have to be targeted towards a particular suspected microorganism, it has the potential to identify unusual pathogens that clinicians may not initially have considered.

That could make the approach particularly useful when a patient is critically ill but routine investigations repeatedly fail to provide an explanation. Metagenomic sequencing is also being investigated internationally for difficult-to-diagnose infections affecting areas including the brain and central nervous system, as well as sepsis and unexplained fever.

It could potentially help identify emerging pathogens too. A conventional targeted test generally needs clinicians to know what they are looking for, whereas metagenomic sequencing has the capacity to detect genetic material from both known and previously unidentified organisms. That ability has made the technology of particular interest for infectious disease surveillance and investigating unusual outbreaks.

However, describing metagenomics as a single test capable of identifying any illness would go too far. The technology is primarily being developed as a tool for diagnosing infectious diseases, and finding genetic material from a microorganism does not automatically prove that it is responsible for a patient’s symptoms.

Samples can also become contaminated, very small amounts of a pathogen may be difficult to detect and results depend heavily on how samples are prepared, sequenced and analysed. Researchers have warned that different metagenomic laboratory processes can produce different results, making validation and standardisation particularly important before the technology can be used much more widely.

There is also not yet enough evidence to suggest metagenomic sequencing should simply replace the established diagnostic tests currently used throughout hospitals. A review published in The Lancet Infectious Diseases this year concluded that clinical metagenomic testing has potentially transformative capabilities, but researchers are still establishing where it produces meaningful improvements in patient treatment and outcomes.

Its immediate future may therefore be as an additional diagnostic tool, particularly for seriously ill patients whose infections remain unexplained or where rapidly identifying a pathogen could significantly change treatment.

What makes the technology potentially important is not that one laboratory test can diagnose every disease. It is that doctors may increasingly be able to search for a vast range of possible infectious causes at once, rather than having to choose which individual pathogens to look for first.

For a critically ill patient whose doctors know they are fighting an infection but do not yet know what is causing it, that difference could mean reaching the right treatment considerably sooner.

Stay informed with the stories shaping health and social care.