Clinical Evaluation of a Point of Care (POC) Assay to Identify Phenotypes in the Acute Respiratory Distress Syndrome

Acute respiratory distress syndrome (ARDS) is an inflammatory condition that results in severe respiratory failure and the need for mechanical ventilation. It is a syndrome with significant global burden and accounts for approximately 24% of mechanically ventilated patients in intensive care units. It is estimated to account for approximately 75000 deaths annually in the USA alone.

Despite decades of research, mortality due to ARDS remains high at 35-46%, with increasing mortality in patients with more severe lung injury. ARDS survivors have significant long term comorbidity with reduced quality of life even 5 years after disease resolution. Various pharmacological agents such as β2 agonists, statins, keratinocyte growth factor and aspirin have been investigated as potential therapies to prevent or treat ARDS, however to date there is no effective pharmacological therapy for ARDS and current treatment strategy is largely supportive.

One reason for the lack of specific pharmacological therapy is likely due to the clinical and biological heterogeneity. It is essential to rapidly identify patients with specific therapy responsive traits to improve our chance of identifying a specific therapy.

Rationale for the Study

ARDS phenotypes have different outcomes and response to therapy.

The clinical and biological heterogeneity in ARDS makes it essential to identify homogenous phenotypes when investigating potential therapies.

A retrospective analysis of the clinical and biological data-set collected as part of two large multicentre studies (ARMA and ALVEOLI) using latent class analysis has identified at least two ARDS phenotypes. Furthermore these two phenotypes could be differentiated using a parsimonious data-set including the presence of shock, metabolic acidosis and a higher inflammatory status (IL-6 and sTNFr1). The hyper-inflammatory phenotype demonstrated significantly worse outcomes when compared to the hypo-inflammatory phenotype with higher mortality and less ventilator free and organ failure free days. In the ALVEOLI study, where low PEEP was compared to high PEEP strategy, the two phenotypes demonstrated a differential response to PEEP suggesting the potential for using this phenotypic classification in identifying a therapy responsive trait.

In addition, in a secondary analysis of the HARP-2 study, a multicentre study investigating the potential of simvastatin as an anti-inflammatory therapy for ARDS, the presence of a hyper- and hypo-inflammatory phenotype was confirmed. The hyper-inflammatory phenotype had a higher 28 day mortality, fewer ventilator free days and organ failure free days. Survival of patients classified as hyper-inflammatory and randomised to simvastatin was improved.

Implementation of a precision medicine approach to identify patients with a therapy response trait is crucial to identify specific therapies to prevent or treat ARDS. Development of a Point of Care (POC) assay for IL-6 and sTNFr1 for prospective confirmation of the inflammatory phenotypes using the parsimonious data-set in patients with ARDS will support a precision medicine approach for this condition.

A POC assay will support precision medicine for ARDS

Studies that show no benefit from an intervention could occur as a result of a variety of reasons including a) the intervention was ineffective, b) the study design was poor or c) patient heterogeneity. Reduction of patient heterogeneity to identify patients with common biological processes will enable the selection of patients with a higher likelihood of therapy response in clinical studies. The identification and institution of therapy for critically ill patients with ARDS needs to occur rapidly in view of the nature of the disease and development of an accurate POC assay is likely to be an essential component in the discovery of effective therapies.