W.J. Polacheck, M. Rathod, K.P. Kim, C. Cao
University of North Carolina at Chapel Hill,
United States
Keywords: microphysiological systems, organs-on-chip, microfluidics, pharmacokinetics, NAMs
Summary:
As New Approach Methodologies (NAMs) gain regulatory momentum for Investigational New Drug (IND) applications, engineers are uniquely positioned to collaborate with clinical pharmacologists and preclinical teams in the development, quantification, and implementation of these methodologies. Microfluidic devices and techniques that are useful in these translational settings have unique design requirements that necessitate an alternative prioritization of parameters and fabrication techniques than is typical for academic device development. For example, to inform pharmacokinetic models and to replace animals in early toxicity screening, throughput and repeatability must be prioritized over complexity and physiological mimicry. Understanding this design space requires close collaboration with clinical pharmacologists to understand unmet needs and metrics of success. The potential upside to these collaborative efforts is genuinely transformational as regulatory agencies increasingly push for NAMs to replace animals in preclinical testing. Here, we present recent collaborative work among biomedical and mechanical engineers with clinical pharmacologists toward the development of integrated microfluidic and computational platforms to improve preclinical modeling of large molecules, including antibody therapeutics. Specifically, we present several advances in microfluidics for improved physiological mimicry, including blood vessel-on-chip devices fabricated from human donors. These devices, in which cells, extracellular matrix, and perfusate, have been derived from donors, have been employed to model rare vascular diseases including vascular malformations and vascular Ehlers-Danlos syndrome. We then discuss efforts to scale up the manufacture and implementation of these devices, including automation by liquid handling robot. Finally, we discuss challenges and opportunities of integrating data from these platforms with advanced, AI-informed physiologically based pharmacokinetic models (PBPK) for predicting native tissue exposure and toxicity in preclinical drug development.