Articles
Acute respiratory distress syndrome subphenotypes and differential response to simvastatin: secondary analysis of a randomised controlled trial

https://doi.org/10.1016/S2213-2600(18)30177-2Get rights and content

Summary

Background

Precision medicine approaches that target patients on the basis of disease subtype have transformed treatment approaches to cancer, asthma, and other heterogeneous syndromes. Two distinct subphenotypes of acute respiratory distress syndrome (ARDS) have been identified in three US-based clinical trials, and these subphenotypes respond differently to positive end-expiratory pressure and fluid management. We aimed to investigate whether these subphenotypes exist in non-US patient populations and respond differently to pharmacotherapies.

Methods

HARP-2 was a multicentre, randomised controlled trial of simvastatin (80 mg) versus placebo done in general intensive care units (ICUs) at 40 hospitals in the UK and Ireland within 48 h of onset of ARDS. The primary outcome was ventilator-free days, and secondary outcomes included non-pulmonary organ failure-free days and mortality. In a secondary analysis of HARP-2, we applied latent class analysis to baseline data without consideration of outcomes to identify subphenotypes, and we compared clinical outcomes across subphenotypes and treatment groups.

Findings

540 patients were recruited to HARP-2. One patient withdrew consent for the use of their data, so data from 539 patients were analysed. In our secondary analysis, a two-class (two subphenotype) model was an improvement over a one-class model (p<0·0001), with 353 (65%) patients in the hypoinflammatory subphenotype group and 186 (35%) in the hyperinflammatory subphenotype group. Additional classes did not improve model fit. Clinical and biological characteristics of the two subphenotypes were similar to previous studies. Patients with the hyperinflammatory subphenotype had fewer ventilator-free days (median 2 days [IQR 0–17] vs 18 [IQR 0–23]; p<0·0001), fewer non-pulmonary organ failure-free days (15 [0–25] vs 27 [21–28]; p<0·0001), and higher 28-day mortality (73 [39%] vs 59 [17%]; p<0·0001) than did those with the hypoinflammatory subphenotype. Although HARP-2 found no difference in 28-day survival between placebo and simvastatin, significantly different survival was identified across patients stratified by treatment and subphenotype (p<0·0001). Specifically, within the hyperinflammatory subphenotype, patients treated with simvastatin had significantly higher 28-day survival than did those given placebo (p=0·008). A similar pattern was observed for 90-day survival.

Interpretation

Two subphenotypes of ARDS were identified in the HARP-2 cohort, with distinct clinical and biological features and disparate clinical outcomes. The hyperinflammatory subphenotype had improved survival with simvastatin compared with placebo. These findings support further pursuit of predictive enrichment strategies in critical care clinical trials.

Funding

UK Efficacy and Mechanism Evaluation Programme and National Institutes of Health.

Introduction

Acute respiratory distress syndrome (ARDS) is a common and frequently fatal cause of respiratory failure among patients who are critically ill, with an incidence of nearly 200 000 cases per year in the USA alone, an estimated prevalence of 10% among critically ill patients worldwide, and a mortality of 30–40%.1, 2 ARDS is defined by clinical criteria including acute onset of hypoxaemia (ratio of the partial pressure of arterial oxygen [PaO2] to the fraction of inspired oxygen [FiO2] <300 mm Hg), bilateral chest radiographic opacities, and exclusion of cardiac failure as the sole cause of the syndrome.3 Since the first consensus definition of ARDS in 1988, experts have debated whether patients should be subdivided by natural history, clinical features, biology, or a combination of all three.4 During the ensuing three decades, positive trials of several supportive care interventions, including most notably lung protective ventilation, have led to decreases in ARDS mortality.5 However, over the same period, dozens of pharmacotherapies that seemed to show great promise in preclinical studies have failed in clinical trials.6, 7 An often-cited reason for this discouraging failure rate has been the considerable clinical and biological heterogeneity within ARDS, but objective data to guide a more precise approach to clinical trials have been lacking.

Research in context

Evidence before this study

Previous studies of patients with acute respiratory distress syndrome (ARDS), using data from US-based randomised controlled trials, have identified two distinct subphenotypes with differential responses to mechanical ventilation and fluid therapy. Whether these subphenotypes can be identified in non-US patient populations by use of different datasets and, more importantly, whether these subphenotypes respond differently to pharmacotherapies is unknown. We searched PubMed for any study reporting differential responses to pharmacotherapy by ARDS subphenotype published from database inception to March 7, 2018, using the terms (“ARDS” or “acute lung injury”) AND (subtype OR subphenotype OR endotype) and no language restrictions. We identified no previous studies.

Added value of this study

Our study reports the identification of two distinct ARDS subphenotypes in a secondary analysis of a UK-based randomised controlled trial of simvastatin for ARDS treatment. The hyperinflammatory subphenotype of ARDS had a survival benefit from simvastatin. To our knowledge, our study is the first to report a differential response to pharmacotherapy by molecular subphenotype in ARDS.

Implications of all the available evidence

Although other areas of medicine (eg, cancer and asthma) have made substantial progress through identification of biologically distinct subtypes of disease with differential treatment responses, critical care medicine has lagged behind. Our findings suggest that targeting specific biological subtypes of critical illness syndromes in clinical trials might yield progress after decades of negative pharmacotherapy trials in the intensive care unit.

Latent class analysis is a well validated statistical method that uses objective criteria to identify subgroups within a broader population. We previously applied latent class analysis in independent analyses of three cohorts of patients derived from National Heart, Lung, and Blood Institute ARDS Network randomised controlled trials. We observed strong evidence for two distinct and consistent subphenotypes of ARDS in all three cohorts, which included more than 2000 patients in total.8, 9 In all three cohorts, one subphenotype (hyperinflammatory) representing roughly 30% of patients with ARDS was consistently characterised by increased inflammatory biomarkers, more profound shock and acidosis, and significantly worse clinical outcomes. This subphenotype had a significantly different response to randomly assigned positive end-expiratory pressure and fluid management strategy, compared with the other subphenotype (hypoinflammatory).8, 9 These findings suggest that improved understanding of these subphenotypes could be crucial to future success in ARDS clinical trials.7 However, it is not known whether these ARDS subphenotypes are generalisable to non-US populations, whether they can be identified with less extensive datasets, and most importantly, whether they might respond differently to pharmacotherapies.

To test these questions, we designed a secondary analysis of a phase 2b randomised trial of simvastatin for treatment of ARDS (the HARP-2 study).10 On the basis of our previous research, we hypothesised a priori that latent class analysis of the HARP-2 cohort would identify two distinct subphenotypes of ARDS, with similar clinical and biological characteristics to those we had previously identified. We also hypothesised, on the basis of the anti-inflammatory effects of statins in laboratory and preclinical models of ARDS,11 that patients with the hyperinflammatory subphenotype would preferentially respond to simvastatin.

Section snippets

Study design and participants of HARP-2

We present secondary analysis of data from the previously reported HARP-2 trial. HARP-2 was a multicentre, randomised controlled trial done in general intensive care units (ICUs) at 40 hospitals in the UK and Ireland.10 Patients were eligible if they were intubated and mechanically ventilated and were within 48 h of the onset of ARDS, as defined by a ratio of PaO2 to FiO2 of 300 mm Hg or less, if bilateral pulmonary infiltrates consistent with pulmonary oedema were present on chest radiography

Results

Between Dec 21, 2010, and March 13, 2014, 540 patients were recruited to HARP-2 and randomly assigned to receive simvastatin or placebo. One patient withdrew consent for the use of their data, so data from 539 patients were analysed. Baseline characteristics of patients enrolled in the HARP-2 trial have been fully described elsewhere (appendix).10 Pneumonia was the most common contributing risk factor for ARDS (appendix). Overall, median number of ventilator-free days was 13 (IQR 0–22), and

Discussion

Our analyses have two novel findings with important implications for future clinical trials in ARDS. First, two distinct ARDS subphenotypes with features similar to those we have previously reported were identified for the first time in a non-US patient population, using a different and much smaller set of clinical and biomarker data than in previous studies. These findings indicate that these subphenotypes are consistent across geographical sites and are robust to variations in specific data

References (25)

  • BT Thompson et al.

    Acute respiratory distress syndrome

    N Engl J Med

    (2017)
  • KR Famous et al.

    Acute respiratory distress syndrome subphenotypes respond differently to randomized fluid management strategy

    Am J Respir Crit Care Med

    (2017)
  • View full text