Precision Medicine Adaptive Network Platform Trial in Hypoxemic Acute Respiratory Failure

Overview

The goal of this trial is to accelerate the development of pharmacological therapies for critical illness by identifying biological sub phenotypes in patients with acute respiratory distress syndrome (ARDS). The trial will stratify participants by biological markers into different sub phenotypes, then randomized 1:1:1 to active treatment 1, active treatment 2, or usual care. Initial stratification will be into hyperinflammatory and hypoinflammatory sub phenotypes in ARDS based on plasma biomarker profiles. Regular adaptive analyses will enable efficient identification of treatment effects within each sub phenotype, stopping interventions where there is evidence of efficacy or futility, and bringing in new interventions and potentially new sub phenotypes.

The primary objective of this trial is to accelerate the development of pharmacological therapies for critical illness by establishing an adaptive platform trial to test the efficacy of prioritized pharmacological interventions in patients with acute respiratory distress syndrome (ARDS). Participants will be recruited from multiple sites within the United States. ARDS is defined by: A known acute clinical insult or new or worsening respiratory dysfunction, and Receipt of respiratory support via invasive mechanical ventilation or non-invasive ventilation including continuous positive airway pressure, or high-flow nasal oxygen ≥30L/min, and Within the same 24-hour time period: Bilateral opacities on chest imaging not fully explained by effusions, lobar/lung collapse/atelectasis, or nodules, and Respiratory failure not fully explained by cardiac failure, fluid overload, pulmonary embolism, acute airways disease, or interstitial lung disease and Pa02/Fi02 ration \<40 kPa from arterial blood gases, or Sp02/Fi02 \<315 from pulse oximetry where Sp02 \<97. The time of onset of ARDS is when the last criterion in 3 is met. Developing effective pharmacological therapies for ARDS will improve patient outcomes in an area of significant unmet need as well as reduce health care costs. Survivors of ARDS experience reduced health-related quality of life, with substantial health care and societal costs. Secondary analyses of prior randomized clinical trials have suggested different treatment responses between biological sub phenotypes in patients with ARDS. A new paradigm in critical care suggests that de-emphasizing clinical syndromic definitions and focusing instead on sub phenotypes more closely linked to the host biological response is the key to identifying effective therapeutics. Using syndromic definitions does not provide information on which subgroups of patients are likely to respond effectively and safely to a given pharmacological treatment. Initially, the platform will use the hyper and hypoinflammatory sub phenotypes in ARDS that are characterized by differing levels of systemic inflammation, clinical features, and treatment responses. Hyper and hypoinflammatory phenotyping will be determined using validated assays for IL-6 and sTNFR1. These measurements will be combined with measurement of the lowest bicarbonate level within the proceeding 24 hours from a plasma or an arterial blood gas and the data used to determine phenotype in real time at the bedside using a validated algorithm. Thus, sub phenotyping will occur via collection and real-time processing of plasma biomarkers after inclusion criteria are met and prior to randomization. The ELLA Device (Bio-Techne) is a fully automated immune-analyzer that enables on-site simultaneous detection of analytes from a single sample of blood. As a minimal risk procedure for the participant and user, a sample of less than 1 mL of fresh blood from the consented participant is required. The results are generated typically within 60-80 minutes, and biomarker values for IL-6 and sTNFR1, along with the lowest blood bicarbonate level in the previous 24 hours, are entered into a logistic regression model to identify the probability of belonging to the hyperinflammatory sub phenotype. Participants and clinical teams will not be aware of the participant's real-time sub phenotype. This trial is open-label whereby participants, the clinical team and study team will not be masked to the intervention. Randomization will be balanced with equal ratios between usual care and each intervention participants are eligible for, i.e. a 1:1:1 ratio (usual care versus intervention (where eligible) by sub phenotype). There is no fixed sample size, but we have capped the sample size for the initial treatments and sub phenotype, and will stop enrollment when results indicate efficacy or futility. Every participant will have additional blood collected at trial enrollment (day 0), day 2, and day 6. Participants will be followed by the clinical research team daily whilst in the ICU. Once the participants have left ICU and been discharged to acute care, they will be followed prior to hospital discharge. They will also be followed up by telephone or electronically 90 days and 180 days after randomization (+14 days). Survival status will be entered at 365 days.