How a Genetic Testing Breakthrough Is Helping Doctors Catch Almost Any Infection
In late December 2019, a team of virologists across several Chinese cities began testing swabs from hospital patients suffering mysterious, severe pneumonia. All had visited the Huanan Seafood Market in Wuhan before falling ill — yet when standard tests were run, nothing matched any known pathogen.
The breakthrough came from metagenomic testing, a method that scans for every microorganism present in a sample rather than checking against a list of known microbes. It revealed the mystery pathogen closely resembled the coronavirus family, which had previously caused outbreaks such as Middle East Respiratory Syndrome. The scientists had identified Covid-19.
“It was a gamechanger compared with previous pandemics,” says Niranjan Nagarajan, an associate professor at the National University of Singapore. Pinpointing the virus allowed diagnostics, treatments and vaccines to be developed at unprecedented speed. That remains metagenomics’ most famous success, but the technique is now spreading quietly through hospitals worldwide, helping diagnose sepsis and other mystery illnesses that would otherwise take far longer to identify.
Consider a patient arriving confused and feverish on a neurological ward — symptoms that could signal meningitis or encephalitis, both potentially fatal. The right treatment depends on whether a virus, bacteria or an overactive immune response is to blame, yet doctors often must guess. “You need to make a decision immediately,” says Charles Chiu, an infectious diseases professor at the University of California San Francisco. “Currently, with patients who are critically ill, doctors treat them for infection and for autoimmune disease” simultaneously — a contradictory approach, since immune-suppressing drugs can worsen an actual infection. “But there’s often no other choice,” he adds.
Chiu’s team has built a metagenomic test specifically for cerebrospinal fluid, the liquid surrounding the brain and spine, which has already caught cases doctors would otherwise have missed. One patient diagnosed with meningitis and liver failure was due for a transplant when the test instead revealed hepatitis E — a treatable condition rarely screened for in wealthy countries. “We gave her an antiviral medication, she got better, and we spared her from undergoing the transplant,” Chiu says.
Until recently, speed held the technology back. “Doing the sequencing and the analysis can take several days to a week,” Nagarajan notes. “The patient might die before you’re able to get a result.” Now DNA sequencing can take as little as one to two hours, and AI tools can sift through the resulting data far faster than before, making the approach newly practical.
That speed matters most for people with weakened immune systems — transplant recipients, cancer patients and those with autoimmune disease — who face sharply elevated risk once infection takes hold; studies suggest immunosuppressed sepsis patients are 23% more likely to die in hospital. “Conditions such as pneumonia or sepsis can be caused by hundreds of potential pathogens which results in a very long diagnostic odyssey,” Chiu says. “It can be a race against time.” AI helps by filtering out harmless background viruses and lab contamination, and by analysing gene activity to determine whether the immune system is actively fighting an infection or coexisting with it. “[This] gives us the ability to do precision medicine,” Chiu says.
Nagarajan is now running a Singapore-based project testing whether AI-enhanced metagenomics can speed treatment for immunocompromised patients. A pilot study across two London hospitals found the technique identified pathogens that routine tests miss roughly 30% of the time — in one case sparing a patient invasive surgery by revealing tuberculosis rather than suspected lung cancer, and in another flagging two cases of parvovirus B19 months before health agencies issued public alerts. Singapore labs are also testing sewage and air samples through AI models trained to spot emerging viruses, the kind of early warning that, in Italy, retrospective sewage analysis later showed could have flagged Covid-19 months before its first confirmed cases.
Source: BBC News