Artificial neural networks applied to biodegradable nanoparticles for Noscapine delivery

Biopolymer based nanocarriers have been extensively studied as delivery systems for different therapeutics, due to their multiple favorable properties - e.g. small size, long circulation time, reduced opsonization, drug stability and controlled release [1-2]. In this study a series of biodegradable block copolymers based on polyethylene glycol (PEG) and poly lactide (PLA) were synthesized by ring opening polymerization (ROP). Successively, biocompatible PLA nanoparticles loading Noscapine were prepared using the nanoprecipitation method and characterized for their size, drug entrapment efficiency and morphology. Noscapine, an alkaloid derived from opium, has been widely used as antitussive agent for many years. However, more recently, it was discovered that it owns a tubulin binding activity. It affects the dynamics of microtubules resulting in arresting the metaphase of cell cycle which eventually leads to apoptosis of dividing cancer cells [3]. Artificial neural networks (ANNs) [4] were applied to predict particle size and Noscapine entrapment efficiency within the formed nanoparticles using different factors - i.e. copolymer molecular weight, ratio of polymer to drug and the number of blocks present in the copolymer backbone. Using these networks it was found that the copolymer molecular weight has the greatest effect on the nanoparticle size distribution. On the other hand, polymer to drug ratio was found to be the most influential factor on drug entrapment efficiency. ANNs may have a great impact on the design of PEG/PLA based copolymers, and they can be used to customize the formulations of nanoparticles that can fit required targets.