Upscaling of the hot-melt extrusion process: comparison between laboratory scale and pilot scale production of solid dispersions with miconazole and Kollicoat IR.

Since only limited amount of drug is available in early development stages, the extruder design has evolved towards smaller batch sizes, with a more simple design. An in dept study about the consequences of the differences in design is mandatory and little can be found in literature. Miconazole and Kollicoat IR were used as model drug and carrier for this study. Two series of solid dispersions were made with a laboratory scale (internal circulation-simple screw design) and a pilot scale extruder (continuous throughput-modular screw design). Efforts were made to match the operating parameters as close as possible (residence time, extrusion temperature and screw speed). The samples were analyzed with modulated DSC straight after production and after exact 24h and 15 days storage at -26 °C. The kinetic miscibility of the samples prepared with the laboratory scale extruder was slightly higher than the samples prepared with the pilot scale extruder. As the solid dispersions with high drug load were unstable over time, demixing occurred, slightly faster for the samples prepared with the laboratory scale extruder. After 15 days, the levels of molecular mixing were comparable, pointing to the predictive value of samples prepared on laboratory scale.