Comparative Modeling Approaches in the Development and Evaluation of Silica Nanoparticles as 5-Fluorouracil Carriers using PLS, DoE and GBR
摘要
The objective of the study was to investigate the impact of formulation variables on the physicochemical properties of silica-based nanoparticles as carriers for 5-Fluorouracil (5-FU).
MethodsNanoparticles were synthesized via the sol–gel process involving the hydrolysis of tetraethyl orthosilicate (TEOS) alone or in combination with 3-aminopropyltriethoxysilane (APTES). Twenty-two formulations were prepared to explore the effects of precursor type, water volume, 5-FU amount, and pH of the hydrolysis medium on particle size (PS), polydispersity index (PDI), drug content (DC), and zeta potential (ZP) using statistical modeling by Partial Least Squares (PLS) multivariate analysis, Design of Experiments (DOE) and Gradient Boost Regressor (GBR). Stability studies in buffers simulating gastrointestinal conditions (pH 1, 4.5, and 7.4) were also performed.
ResultsNanoparticle sizes ranged from 48 to 1089 nm, PDI from 0.017 to 1, DC from 1.71 to 18.97 µg 5-FU/mg nanoparticles, with ZP values from -7.22 mV to 37.3 mV. Using the PLS, GBR and DOE, statistically significant formulation factors influencing nanoparticles` properties were identified. Stability studies conducted in buffers simulating gastrointestinal conditions demonstrated variations in PS (47 nm to 3333 nm), PDI (0.022 to 1), and ZP (-12.23 mV to 53.1 mV) and significant formulation factors were also identified via PLS, DOE and GBR modeling.
ConclusionSilica nanoparticles with encapsulated 5-FU, were developed using three different modeling approaches through systematical examination of different formulation and process factors, and their impact on the nanoparticle physicochemical properties. Each model offers unique benefits and limitations, influencing its suitability for various stages in nanoparticle formulation.