Mathematical modelling may hold the key to saving patients from sepsis, a deadly complication that causes about 20 per cent of all deaths worldwide.
Sepsis occurs when the immune system goes awry, turning on itself during an infection and damaging the body’s organs. About a quarter of those who develop the condition die.
Viral sepsis was a major cause of deaths from severe Covid-19, while many deaths in historical pandemics like the 1919 influenza pandemic and the bubonic plague are thought to have resulted from sepsis.
Experts say understanding how sepsis develops and therefore how to prevent and treat it would help protect against the worst consequences and highest death tolls in future pandemics.
And since immune dysregulation from sepsis can linger, causing symptoms similar to post-viral syndromes like long Covid, learning to treat it could also benefit some chronic illness patients.
But to make it happen, more funding and larger studies will be needed, say researchers.
There are currently no treatments that tackle sepsis directly, but experts say a field called systems immunology could help predict and treat it. It uses mathematical and computational modelling to study the immune system in the context of the body’s other systems, identifying patterns in data that can tell us about the body’s reaction to sepsis in detail.
“We need to adopt a concerted approach to tackle sepsis,” said Prof Robert Hancock of the University of British Columbia, lead author of the article. “Only a very small amount of funding is currently invested in sepsis research and product development – and yet sepsis is as prominent a cause of death as heart disease and cancer, and the major cause of death in pandemics.”
The development of sepsis is complicated and hard to predict. Many different infections can cause sepsis, and its symptoms and progression vary between patients and over time in the same patient. Its early symptoms mimic other illnesses, which makes it difficult to diagnose and treat quickly, contributing to its high death toll.
Symptoms of sepsis – in pictures
Experts seek patterns in systems immunology to help determine the basis for the immune dysregulation that drives sepsis to devise new ideas to test and develop treatments and identify markers to catch sepsis early.
By analysing the data, scientists have identified changes to gene expression that act as early warnings for sepsis. They have also been able to identify five different subtypes of sepsis, which are caused by different kinds of immune dysregulation and have different prognoses.
But systems immunology analysis is not yet in widespread use, because it is expensive and demands significant volumes of data – so it is not yet known how diagnostics could translate into clinical results.
“In sepsis we lack the depth of information required to enable more effective systems immunology and machine learning approaches,” said Prof Hancock. “We hope to encourage the development of large, in-depth patient studies that will trigger a new generation of insights.”
