Demonstration of Advanced Data Mining Tools for Optimization of Pellets Employing Modified Extrusion-pelletization Technique

Abstract

Background: The multi particulate drug delivery system is preferred due to its numerous advantages but the batch to batch consistency and to achieve desired physical properties are the major challenges in the formulation of such dosage form. Objective: The objective of the present study was to explore the concept of quality by design for the development of galantamine HBr controlled release pellets using a modified palletization technique. Methods: Compritol 888 and Ethocel were chosen as hydrophobic release retardants, while Avicel was chosen as pelletization aid. A compatibility study was conducted between the drug and excipients. Drug loaded extrudes were prepared by using a mixture of isopropyl alcohol, and dichloromethane. Before converting the wet extrudes into pellets, pregelatinized starch was sprinkled on them to improve the physical properties of the pellets. The pellets were characterized for size, shape, and flow. The critical evaluation parameter was the drug dissolution pattern in distilled water. The dissolution data were treated with advanced data mining techniques. The in-vivo profile was predicted employing pharmacokinetic parameters of the drug and in-vitro drug release data of optimized batch pellets. Results: The failure mode and effect analysis revealed that the amount of Compritol 888 ATO and Ethocel were the most critical formulation parameters. The results of FTIR and DSC revealed compatibility between the drug and the excipients. The spherical pellets exhibited good flow. The drug dissolution studies of the batches, prepared according to the central composite design, revealed modified drug release. Multiple regression analysis and analysis of variance were performed to identify statistically significant factors. Contour plots demonstrated the impact of the amount of Compritol 888 and ethyl cellulose. The Design-Expert software was used to identify optimized formulation. The predicted in-vivo plasma concentration-time profile revealed the modified drug release up to 12h. Conclusions: Compritol and Ethocel were able to retard the drug release up to 12 hrs in distilled water. The innovative finding of this study is the use of a dry binder (pregelatinized starch) to improve the characteristics of pellets. Other dry binders are expected to show a similar effect. The newer processing technique can be of use in the industry.