PROJECT SUMMARY Up to 40% of ARDS patients lose their battle with acute respiratory distress syndrome (ARDS). Many of these patients succumb because ongoing ventilator-induced lung injury (VILI) overwhelms the innate capacity of the lung to repair itself. On the other hand, many ARDS patients go on to recover, which means that their reparative capacities eventually prevail. Since there are currently no medical therapies for ARDS, minimizing its mortality amounts to achieving an optimal balance between spontaneous tissue repair versus the generation of VILI. However, one-size-fits-all strategies for managing ARDS have had limited success in reducing mortality, and the heterogeneity of ARDS is such that empirical searches are very unlikely to lead to optimal ventilation strategies. Progress must thus be based on a fundamental understanding of VILI mechanisms, so understanding the balance between injury and repair in VILI remains a pressing unmet medical need. We have shown in vivo that the timing of the ventilatory cycle is critically important to the prevention of cyclic airway recruitment and derecruitment (RD) that leads to VILI. Furthermore, we have shown in vitro that, following onset of RD, the physical integrity of an epithelial monolayer remains normal for a period of time before decreasing quasi- exponentially, which is reminiscent of cancer survival curves governed by multi-hit mechanisms. Accordingly, our overarching hypothesis is that the progression of VILI involves a multi-hit mechanism leading to dysfunction of the blood-gas barrier, and that eventual recovery from VILI is determined by whether the mechanical ventilation regimen being applied allows barrier dysfunction to be counteracted by barrier healing. The overall goal of our research is to determine in detail how both the progression and resolution of VILI are influenced by the current status of lung injury and by the relative contributions of RD versus tissue over-distension and identify modes of ventilation that can adapt successfully to a patient's own pathophysiology. Analysis of experimental findings will be incorporated into a computational model to develop biopredictors that can forecast whether injury or recovery will prevail in a given situation based on physiologic measurements at the bedside. We will achieve this goal by measuring how barrier function in epithelial cell monolayers either degrades or improves following application of controlled levels of atelectrauma, and by determining how lung injury in pig models of ARDS either further develops or resolves with different mechanical ventilation strategies that vary in their degrees of injuriousness. Computational modeling informed by the experiment results will be used to explore energy dissipation within the respiratory cycle as a biopredictor of patient outcome. By taking this approach, we expect to advance the design of optimal ventilation strategies for ARDS that are adaptive and personalized to a given patient.