Defining ARDS and sepsis sub-phenotypes
Research type
Research Study
Full title
Defining ARDS and sepsis sub-phenotypes: a re-analysis of two trials to inform stratified medicine approach with future trials
IRAS ID
239286
Contact name
Manu Shankar-Hari
Contact email
Sponsor organisation
Guy's and St. Thomas' NHS Foundation Trust
Duration of Study in the UK
3 years, 5 months, 30 days
Research summary
We study two common conditions affecting critically ill patients admitted to intensive care units, namely sepsis and ARDS. Both these conditions are triggered most often by the body’s response to
infection and/or injury. This causes multiple organs of the body to fail. About a third of patients admitted to hospital with these two illnesses do not survive, despite intensive care management.
Over the last two decades, doctors and scientists have tested numerous medical treatments for these two conditions. However, to date we do not have any specific treatment, especially medications that work in these conditions identified in clinical trials. One major reason for this lack of effective medical treatment is that patients with these conditions are very different to each other.
Therefore, we hypothesised that we need to identify sub-groups of patients with these two illnesses, who are similar and likely to respond to treatments tested. This project tests the hypothesis that such patient sub-groups exist using information from 2 trials that we already completed – HARP-2 trial in ARDS and VANISH trial in sepsis. This project also generates additional information using stored samples already collected from patients in these trials. This means that we do not inconvenience patients and keep the costs to a minimum.
To test this hypothesis, we first assess how the severity of illness varies in the patients included within these two trials. Then we assess how the response to the two treatments tested in the two trials (statin in HARP-2 trial and steroids in VANISH trial) varies with differences in severity of illness.
To understand why and how patients respond to treatments tested in the trials, we will measure blood levels of illness indicators in ARDS and sepsis. We then use well-established statistical analysis methods to identify patient sub-groups in ARDS and sepsis using the principle of clustering of patients with similar features. Amongst these sub-groups of patients with similar features, we plan to identify those sub-groups of patients with ARDS who respond to statin treatment and patients with sepsis who respond to steroids. Finally, we investigate why this difference in response happens and how it could be predicted on admission day to target treatment to those who are likely to benefit the most.
We have put together an expert group of clinicians, scientists and bio-statisticians to ensure our results are robust and inform future care of patients with these two serious illnesses.REC name
London - Westminster Research Ethics Committee
REC reference
18/LO/1079
Date of REC Opinion
29 Jun 2018
REC opinion
Favourable Opinion