Data-driven optimization of nanoparticle size using the prediction reliability enhancing parameter (PREP)

Abstract

The particle size of a nanoparticle plays a crucial role in regulating its biodistribution, cellular uptake, and transport mechanisms and thus its therapeutic efficacy. However, experimental methods for achieving a desired nanoparticle size and size distribution often require numerous iterations that are both time-consuming and costly. In this study, we address the critical challenge of achieving nanoparticle size control by implementing the Prediction Reliability Enhancing Parameter (PREP), a recently developed data-driven modeling-based product design approach that significantly reduces the number of experimental iterations needed to meet specific design goals. We applied PREP to effectively predict and control particle sizes of two distinct nanoparticle types with different target particle size properties: (1) thermoresponsive covalently-crosslinked microgels fabricated via precipitation polymerization with targeted temperature-dependent size properties and (2) physical polyelectrolyte complexes fabricated via charge-driven self-assembly with particle sizes and colloidal stabilities suitable for effective circulation. In both cases, PREP enabled efficient and precise size control, achieving target outcomes in only two iterations in each case. These results provide motivation to further utilize PREP in streamlining experimental workflows in various biomaterials optimization challenges.