Optimization of paclitaxel-loaded poly (d,l-lactide-co-glycolide-N-p-maleimido benzoic hydrazide) nanoparticles size using artificial neural networks

Abstract

The aim of this study was to find a model using artificial neural networks (ANNs) to predict PLGA-PMBH nanoparticles (NPs) size in preparation by modified nanoprecipitation. The input variables were polymer content, drug content, power of sonication and ratio of organic/aqueous phase (i.e. acetone/water), while the NPs size of PLGA-PMBH was assumed as the output variable. Forty samples of PLGA-PMBH NPs containing anticancer drug (i.e. paclitaxel) were synthesized by changing the variable factors in the experiments. The data modeling were performed using ANNs. The effects of input variables (namely, polymer content, drug content, power of sonication and ratio of acetone/water) on the output variables were evaluated using the 3D graphs obtained after modeling. Contrasting the 3D graphs from the generated model revealed that the amount of polymer (PLGA-PMBH) and drug content (PTX) have direct relation with the size of polymeric NPs in the process. In addition, it was illustrated that the ratio of acetone/water was the most important factor affecting the particle size of PLGA-PMBH NPs provided by solvent evaporation technique. Also, it was found that increasing the sonication power (up to a certain amount) indirectly affects the polymeric NPs size however it was directly affected in higher values.

• Artificial neural networks
• hydrophobic drug
• optimization
• paclitaxel
• particle size
• passive targeting
• PLGA-PMBH nanoparticles

Declaration of interest

The authors declare that they have no conflicts of interest to disclose.