Supercooled biopreservation protocol design using probabilistic safety analysis

Aqueous supercooling has shown tremendous promise as a method for low temperature biopreservation, but the omnipresent risk of ice formation has largely hindered its translation outside the laboratory. In this work, we present a framework for the safety-informed design of preservation protocols that utilizes nucleation rate data to predict the probability for nucleation as a function of temperature and time. The analysis leverages the fact that, though random, nucleation kinetics can be described by Poisson statistics. Furthermore, acknowledging that system-to-system variability poses a significant challenge for the clinical translatability of supercooling technology, the analysis framework rigorously incorporates experimental uncertainty and variability into the safety predictions. Finally, we demonstrate how this temperature-dependent probabilistic safety analysis can then be married temperature-dependent biophysical relations to quantitatively incorporate biological aspects of biopreservation into the protocol design and optimization pipeline. In total, this work provides a method by which to engineer supercooled preservation protocols for safety and efficacy in the face of both kinetic and conditional uncertainty.