First 24-Hour-Long Intensive Care Unit Testing of a Clinical-Scale Microfluidic Oxygenator in Swine: A Safety and Feasibility Study.

Microfluidic membrane oxygenators are designed to mimic branching vasculature of the native lung during extracorporeal lung support. To date, scaling of such devices to achieve clinically relevant blood flow and lung support has been a limitation. We evaluated a novel multilayer microfluidic blood oxygenator (BLOx) capable of supporting 750-800 ml/min blood flow versus a standard hollow fiber membrane oxygenator (HFMO) in vivo during veno-venous extracorporeal life support for 24 hours in anesthetized, mechanically ventilated uninjured swine (n = 3/group). The objective was to assess feasibility, safety, and biocompatibility. Circuits remained patent and operated with stable pressures throughout 24 hours. No group differences in vital signs or evidence of end-organ damage occurred. No change in plasma free hemoglobin and von Willebrand factor multimer size distribution were observed. Platelet count decreased in BLOx at 6 hours (37% dec, P = 0.03), but not in HFMO; however, thrombin generation potential was elevated in HFMO (596 ± 81 nM·min) versus BLOx (323 ± 39 nM·min) at 24 hours ( P = 0.04). Other coagulation and inflammatory mediator results were unremarkable. BLOx required higher mechanical ventilator settings and showed lower gas transfer efficiency versus HFMO, but the stable device performance indicates that this technology is ready for further performance scaling and testing in lung injury models and during longer use conditions.

Predicting Clinical Outcomes of SARS-CoV-2 Drug Efficacy with a High-Throughput Human Airway Microphysiological System.

The average cost to bring a new drug from its initial discovery to a patient's bedside is estimated to surpass $2 billion and requires over a decade of research and development. There is a need for new drug screening technologies that can parse drug candidates with increased likelihood of clinical utility early in development in order to increase the cost-effectiveness of this pipeline. For example, during the COVID-19 pandemic, resources were rapidly mobilized to identify effective therapeutic treatments but many lead antiviral compounds failed to demonstrate efficacy when progressed to human trials. To address the lack of predictive preclinical drug screening tools, PREDICT96-ALI, a high-throughput (n = 96) microphysiological system (MPS)  that recapitulates primary human tracheobronchial tissue,is adapted for the evaluation of differential antiviral efficacy of native SARS-CoV-2 variants of concern. Here, PREDICT96-ALI resolves both the differential viral kinetics between variants and the efficacy of antiviral compounds over a range of drug doses. PREDICT96-ALI is able to distinguish clinically efficacious antiviral therapies like remdesivir and nirmatrelvir from promising lead compounds that do not show clinical efficacy. Importantly, results from this proof-of-concept study track with known clinical outcomes, demonstrate the feasibility of this technology as a prognostic drug discovery tool.