Scientists at Imperial College London have developed a blood-based approach that could one day help doctors predict how an illness will progress and how a patient may respond to treatment. The method, called VeloCD, has already shown promise in a proof-of-concept study covering infectious and chronic disease, including acute fever, flu, COVID-19 and inflammatory bowel disease. The findings were published on May 6, 2026, in Nature Communications. Source
How the new test works
The technique measures key markers in blood that reveal gene-expression activity in response to illness. Instead of relying only on a snapshot of a patient’s current condition, the approach looks for patterns that may indicate whether a disease is likely to improve or worsen over the next few hours or days. Researchers adapted RNA velocity, a method originally developed for single cells, and applied it to whole blood samples to generate prognostic clues.
According to the Imperial team, the tool could eventually support faster triage in hospitals by helping clinicians identify who needs further care and who may be safe to send home with treatment. The researchers say the work lays the groundwork for a future prognostic test and that a working clinical version could be available within as little as five years if further development and validation are successful.
Evidence from patient and challenge-study data
To test the concept, the team analysed real-world data from several large studies, including blood samples from almost 400 children admitted to hospital with fever across nine European countries. In that dataset, whole-blood RNA sequencing identified more than 2300 markers linked to mild, moderate and severe illness, while a reduced set of 59 markers was enough to predict which children were most likely to deteriorate.
The approach was also tested using data from Imperial College London’s human challenge programme, where healthy adults were exposed to influenza or SARS-CoV-2 under controlled conditions. Early blood samples were able to help predict whether participants would go on to develop flu or COVID-19, even before infection could be confirmed by PCR.
Potential applications in routine care
Researchers also reported that VeloCD showed potential for highlighting complications of HIV and tuberculosis, as well as treatment response in inflammatory bowel disease. The team has filed a patent for the method and says it could ultimately be developed into a commercially available hospital test.
Professor Aubrey Cunnington said the aim is to give medics a tool that can predict the course of illness and help deliver the right treatment at the right time. The study authors describe the approach as a possible step toward more personalised and proactive medicine, though they note that additional work is still needed before any clinical use.
The project involved collaborators at UCL, the University of Cape Town, Queen Mary University of London, the PERFORM consortium and the human challenge programme at Imperial College London. Funding came from the European Union’s Horizon 2020 programme, the UK Research and Innovation Medical Research Council and Engineering and Physical Sciences Research Council, and the UK Department for International Development, among others.
If validated in larger studies, the method could offer clinicians a new way to anticipate disease trajectories rather than only reacting after symptoms worsen. For hospitals facing pressure to make quick decisions, that shift could prove especially significant.
