A Modeling Approach to Inform In Vitro Hypoxia Platforms for Patient-Specific Analyses

Hypoxia is a consequence of diseases such as obstructive sleep apnea (OSA) and can be detrimental to tissue function, with in vitro platforms commonly used to study the associated impacts. However, to create a realistic experimental simulation of hypoxia, the dissolved oxygen concentration at the in vitro cell layer must mimic in vivo values. Considering the limitations associated with mass transfer and continuous oxygen monitoring in standard hypoxia chambers, mathematical modeling can be leveraged to inform experimental design for more accurate and clinically relevant in vitro representations.

A model of oxygen transfer in a standard hypoxia chamber with a cell culture set-up was calibrated and validated using experimental data. To demonstrate a direct application with clinical data, the validated model was coupled with a novel optimization approach to create a "patient-specific" gas controller protocol capable of reproducing a model-derived OSA patient profile at the in vitro cell layer.

The validated model was successful in independently capturing the effect of endothelial cell oxygen consumption in the media. Furthermore, the predicted "patient-specific" gas controller protocol resulted in an in vitro oxygen profile which mimicked the oxygen values and fluctuation timings seen in a model-predicted arterial oxygen pattern for an OSA patient. Combining the analyses of this study, media height, evaporation, and spatial position were determined to be important factors in experimental design.

This work demonstrates the utility of applying mathematical modeling for experimental design and presents a means to bridge the gap between the clinical and experimental fields with a "bedside to bench" approach.