The acute respiratory distress syndrome (ARDS) is a life-threatening condition of airspace inflammation and edema causing severely low blood oxygen levels that may complicate up to 24% of ICU admissions for mechanical ventilation. Mortality remains above 35% and no prevention or pharmacotherapy options exist. We previously identified and replicated a functional genetic variant in the interleukin-1 receptor antagonist (IL1RA) gene, encoding the protein IL1RA, that was strongly associated with lower ARDS risk and higher evoked plasma IL1RA levels. We also demonstrated a causal role for plasma interleukin-1 beta (IL1), the molecule inhibited by IL1RA, in mortality from septic shock. Finally, in a subgroup of subjects randomized to receive recombinant human interleukin-1 receptor antagonist (rhIL1RA) or placebo for sepsis, we demonstrated significant heterogeneity in the mortality treatment effect of rhIL1RA that was predictable by plasma IL1RA concentration. Taken together, our data indicated that some patients may be genetically protected from ARDS via more efficient IL1RA production, suggesting that the drug rhIL1RA may be an effective prevention or treatment strategy. However, because we proved that the effect of rhIL1RA during sepsis is not uniform, the current application will obtain critical information about the causal contribution of plasma IL1RA and IL1to ARDS risk and death from ARDS, will develop predictive models for ARDS risk in an effort to identify a target population for rhIL1RA intervention, and will determine the direct effects of rhIL1RA treatment on ex vivo perfused human lungs. Our long term objective is to evaluate rhIL1RA as a precision ARDS prevention and/or treatment strategy. Aim 1 will focus on ARDS risk, using a genetic instrumental variable technique to infer the causal effect of plasma IL1RA and IL1for ARDS risk, and developing and validating an ARDS predictive model that considers clinical factors and plasma IL-1 markers in 2 large sepsis cohorts. Aim 2 will focus on ARDS and sepsis mortality. We will use multiple genotypes as an instrumental variable to test the contribution of early or delayed plasma IL-1 markers for death from ARDS, and will develop and validate predictive models for ARDS mortality, using clinical data and plasma IL-1 markers measured in 2 large sepsis cohorts. In Aim 3, we will use a novel platform known as ex vivo lung perfusion (EVLP), whereby lungs declined for transplantation are ventilated and perfused in a highly controlled environment, to test whether rhIL1RA given as treatment or prevention to injured human lungs will improve oxygenation, lung compliance, and inflammation. The multidisciplinary team of investigators and consultants to enact these aims include an EVLP expert, 2 epidemiologists with expertise in lung injury prediction and molecular subphenotyping, a statistical geneticist who is expert in instrumental variable analysis, and the PI, a translational scientist who first ide...