PROJECT ABSTRACT Nutrition is an essential component in the care of critically ill patients with acute respiratory distress syndrome (ARDS) who are often limited in volitional intake and who incur new deficits in physical function secondary to their illness. Despite well-characterized associations between inadequate nutrition and adverse outcomes, large multicenter trials of specific nutrition strategies have failed to consistently demonstrate clinical benefit with any particular approach. The landmark NHLBI-funded Early versus Delayed Enteral Nutrition [EDEN] trial investigated low-level (trophic) versus full enteral feeds in 1000 patients and demonstrated increased gastrointestinal complications with full enteral feeds without any clinical benefit. Subsequently, current guidelines recommend low-level feeds for all ARDS patients, however, ARDS is increasingly recognized as a heterogeneous disease and a one-size-fits-all approach may not be appropriate. An essential first step in individualizing nutrition in ARDS is identifying the clinical and biologic variables that contribute to heterogeneity of treatment effect whereby individuals or groups vary in response to treatment. Thus, the overall goal of this proposal is to characterize differential responses to nutrition by leveraging data and biospecimens from the EDEN trial. In Aim 1, this proposal will investigate whether ARDS subphenotypes differed in response to low- level versus full enteral nutrition in EDEN. Recent studies have identified two distinct ARDS subphenotypes that differ in prognosis, host response, and potentially in response to treatments. ARDS subphenotypes have not been investigated with regards to nutrition. Thus, Aim 1 will perform biologic assays to facilitate ARDS subphenotype identification and will test for differences by subphenotype in clinical outcomes, in intestinal permeability, and in incretin hormones. When released at physiologic levels in response to enteral nutrients, incretins have beneficial effects on insulin release and glucose metabolism, but when released at supraphysiologic levels in response to intestinal inflammation, incretins may worsen tolerance to nutrition by reducing gastric motility. Incretins are not well characterized in ARDS. In Aim 2, this proposal will use computational approaches that directly model individual treatment effects based on patient covariates allowing for investigations of treatment response that result from a complex interaction between baseline demographics, severity of illness, endocrine hormones, and inflammation at an individual patient level. Successful completion of the proposed Aims will provide new knowledge on biologic responses to nutrition strategies, identify mechanisms that contribute to heterogeneity of treatment effect at an individual level, and provide direction for the next generation of precision nutrition studies in ARDS.