Blood test approach shows potential to predict disease progression and treatment response

An international team of scientists led by Imperial College London has developed a new blood test method, called VeloCD, that can predict how a patient’s illness will progress and how well they will respond to treatment¹.

Tested using whole-blood RNA sequencing data from patients and healthy volunteers across Europe, Africa and the UK, the approach could help hospitals rapidly triage patients, identify those most likely to deteriorate, and guide personalized treatment decisions.

What is VeloCD and how does it work?

When people become ill, combinations of genes in their immune cells are switched on and off, producing RNA markers that can be detected in the blood. Previous research has shown that patterns of these RNA markers can help identify the cause of illness, such as distinguishing between bacterial and viral infections.

In the new study, researchers adapted a method called RNA velocity, originally developed for studying single cells, to analyze whole blood samples. The enhanced approach, named VeloCD, measures not only which genes are active, but also whether their activity is increasing or decreasing. This allows scientists to infer the likely future clinical state of a patient without needing repeated blood samples over time.

Using state-of-the-art bioinformatics and computational modeling, the team compared changes in gene expression patterns in an individual’s blood with those seen in other people whose clinical outcomes were already known, such as mild, self-resolving illness or severe disease requiring intensive care. This enabled them to predict whether a patient was on a trajectory towards recovery or deterioration, and how they might respond to specific treatments.

Predicting outcomes in children with fever

To validate the concept, the researchers analyzed real-world data from several large studies, including the EU-funded PERFORM study. This dataset included blood samples from almost 400 children admitted to hospitals in nine European countries with acute fever.

Whole-blood RNA sequencing revealed that more than 2,300 RNA markers formed reliable patterns associated with mild, moderate and severe illness. By focusing VeloCD on a subset of just 59 of these markers, the team could predict whether a child was likely to progress to mild, moderate or severe disease.

Crucially, the method identified children at highest risk of deterioration who were most likely to require intensive care. This demonstrates the potential of VeloCD as a prognostic tool to support early risk stratification and clinical decision-making in paediatric infectious disease.

Forecasting flu and COVID-19 infection after exposure

The team also tested VeloCD using data from the human challenge programme at Imperial College London, in which healthy adult volunteers were exposed to influenza or SARS-CoV-2 under controlled conditions.

Blood samples collected at an early time point (Day 2 after exposure) were analyzed using VeloCD. The method accurately predicted which individuals would go on to develop flu or COVID-19 infection and which would remain uninfected, even before infection could be confirmed by PCR testing.

These findings highlight the potential of blood RNA velocity analysis to detect early immune responses to viral exposure and to forecast infection risk before standard diagnostic tests become positive.

Assessing complications and treatment response in chronic disease

Beyond acute infections, the researchers demonstrated that VeloCD could be applied to other conditions. In further validations, the method showed promise for highlighting complications of HIV and tuberculosis based on RNA markers in the blood.

The team also used VeloCD to predict how patients with inflammatory bowel disease would respond to therapy. By analyzing blood samples taken after the first dose of treatment, the method could indicate how well a patient was likely to respond to a full course of therapy. This suggests that VeloCD could support more personalized treatment strategies in chronic inflammatory conditions.

Potential impact on hospital triage and personalized medicine

According to the Imperial College London team, the findings show that VeloCD can generate clinically meaningful predictions of future clinical states from a single blood sample. The researchers have filed a patent for the method and believe it could be developed into a commercially available prognostic test for use in hospitals.

In practice, such a test could be used to rapidly triage patients in emergency departments or hospital wards, helping clinical staff decide who needs closer monitoring, intensive care or early intervention, and who can be safely discharged with appropriate treatment.

Publication, availability and next steps

The researchers have made the VeloCD tool available on GitHub to support further research and validation. They emphasize that additional work is needed to develop and clinically validate diagnostic and prognostic tests based on this approach. With a focused development program and successful trials, a working clinical test could be available within as little as five years.

References

1. Dunican C, Wilson C, Habgood-Coote D. et al. Predicting trajectories of illness using RNA velocity of whole blood. Nat Commun 17, 3652 (2026).

Want the latest science news straight to your inbox? Become a SelectScience member for free today>>