(164b) Practical Application of a Mechanistic Model to Scale up Fluid Bed Drying in Drug Product Development

Authors: 
Gavi, E., F. Hoffmann - La Roche Ltd
Ghosh, P. K., F. Hoffmann - La Roche
Page, S., F. Hoffmann - La Roche
Fluid bed drying is one of the core processes in pharmaceutical manufacturing of oral dosage forms. For several decades models based on mechanistic understanding of the fluid bed drying process have been used. In practice however, in the pharmaceutical industry extensive experimentation is still carried out during drug product development, to investigate fluid bed drying of wet granules at different operating conditions and equipment scales. Typically experimentation is guided by Design of Experiments and results are studied by multivariate analysis. This procedure does not take advantage of the existing fluidized bed drying theory that is already available, and that would allow to reduce the number of experiments performed, at the same time providing better process understanding and more reliable scaling to different operating conditions [1].

In this work a case study is presented, the goal of which was to show that it is possible to use a mechanistic model to scale up fluid bed drying of granules containing active pharmaceutical ingredients (API) from the laboratory to the pilot scale. To this end, two different formulations containing two model APIs were chosen and granulated via high shear wet granulation. The obtained wet granules were then dried in a fluid bed dryer at the laboratory scale. The fluid bed drying model available in the software gSolids v.4.1 (PSE, London, UK) [2, 3], was calibrated for the two formulations with the laboratory scale experimental results. The model was then used to predict the fluid bed drying process of the two formulations at pilot scale. Experiments were run at the pilot scale and model predictions could be confirmed without further adjustments.

It was so demonstrated that it is possible to scale up the fluid bed drying process of API containing granules from laboratory to pilot scale by using a mechanistic model. A workflow was established based on the experience gathered in the case study that details experimental data needed for model set up and calibration and possible solutions to overcome challenges in using the model to predict the process at pilot scale.

It can be concluded that using a mechanistic model constitutes a viable alternative to empirical modelling allowing the saving of API during drug product development by reducing experimentation at large scale.

References

[1] Kemp, I. (2011) Drying models, myths and misconceptions. Chem. Eng. Technol., 34 (7), 1057-1066.

[2] Burgschweiger, J., Groenwold, H., Hirschmann, C. And Tsotsas, E. (1999) From hygroscopic single particle to batch fluidized bed drying kinetics. The Canadian Journal of Chemical Engineering 77, 333-341.

[3] Burgschweiger J, Tsotsas E (2002) Experimental investigation and modelling of continuous fluidized bed drying under steady-state and dynamic conditions. Chem. Eng. Sci., 57, 5021-5038.