Limited Sampling Strategies to Estimate the Exposition of Different Lipid Nanocaspules

The use of nanoparticles in the field of drug delivery and the optimization of pharmacokinetic profiles in vivo is a growing area of research. The aim of this study is to determine the key sampling times and the factors that exert a significant influence on the development of a pharmacokinetic model that can reliably predict the area under the curve (AUC) of lipid nanocapsules in rats using a limited sampling strategy (LSS).

This study was conducted in rats following intravenous injection of lipid nanocapsules (LNCs). Förster resonance energy transfer (FRET)-based quantification was used to monitor pharmacokinetics across ten initial time points. A limited sampling strategy (LSS) model was developed using principal component multiple linear regression, combined with recursive feature elimination and leave-one-out crossvalidation (RFECV-LOOCV).

The final model is based on three sampling time points (180, 360, and 600 min) and demonstrates strong predictive performance (R² = 0.952; root mean square error [RMSE] = 9.81%; low bias). It incorporates plasma concentrations, physicochemical properties of the nanoparticles, and individual animal characteristics.

This approach reduces the number of blood samples by 70% while maintaining high accuracy. The main limitations concern its generalizability to other formulations or species, which would require additional validation.