Unstructured Formulation Data Analysis for the Optimization of Lipid Nanoparticle Drug Delivery Vehicles

Designing nanoparticle formulations with features tailored to their therapeutic targets in demanding timelines assumes increased importance. In this context, nanostructured lipid carriers (NLCs) offer an excellent example of a drug delivery nanosystem that has been broadly explored in the treatment of glioblastoma multiforme (GBM). Distinct fundamental NLC quality attributes can be harnessed to fit this purpose, namely particle size, size distribution, and zeta potential. These critical aspects intrinsically depend on the formulation components, influencing drug loading capacity, drug release, and stability of the NLCs. Wide variations in their composition, including the type of lipids and other surface modifier excipients, lead to differences on these parameters. NLC target product profile involves small mean particle sizes, narrow size distributions, and absolute values of zeta potential higher than 30 mV. In this work, a wealth of data previously obtained in experiments on NLC preparation, encompassing, e.g., results of preliminary studies and those of intermediate formulations, is analyzed in order to extract information useful in further optimization studies. Principal component analysis (PCA) and partial least squares (PLS) are performed to evaluate the influence of NLC composition on the respective characteristics. These methods provide a rapid and discriminatory analysis for establishing a preformulation framework, by selecting the most suitable types of lipids, surfactants, surface modifiers, and drugs, within the set of investigated variables. The results have direct implications in the optimization of formulation and processes.