(307b) Modeling of Breakage and Dissolution of Crystals with Varying Particle Size Distribution within the Quality by Design Framework

An important aspect of the Quality by Design (QbD) framework for pharmaceutical product development and manufacture is to develop a full understanding of how material attributes and process parameters affect critical quality attributes, which in turn determine product performance. Theory and modeling based on first principles approach can be a very effective tool to accomplish this task. In this contribution, we attempt to demonstrate the use of mechanism-based mathematical modeling in developing a fundamental understanding of both the effect of the crystal size variability of an active pharmaceutical ingredient (API) and the impact of different encapsulation process conditions on dissolution rate of a direct encapsulated drug product. During development of a drug product capsules, dissolution rate was shown to correlate with the crystal size distribution (CSD) of the API. A dissolution model [1] was developed for predicting API dissolution profiles in agitated vessels from a known CSD of API. The model assumes that dissolution can be internal or external mass transfer limited, particles are spherical (equivalent spheres), and dissolution occurs under sink or non-sink conditions. The model was employed for a priori prediction of dissolution of API and validated against dissolution profiles of different batches. Since some particle size reduction (breakage) was observed during encapsulation, a population balance (PB) model [2] of the crystal breakage was developed to predict the extent of particle size reduction. The PB model assumes a size dependent breakage kernel and a uniform breakage distribution function. A procedure was developed for merging the breakage model with the dissolution model. A simple ?trial and error? procedure was subsequently used to determine the breakage parameters by fitting the combined model to experimental dissolution profiles of capsule plugs of different densities. A quantitative correlation between the plug density and the breakage rate constant was developed to describe the higher extent of crystal breakage at more aggressive (higher tamping) encapsulation conditions. The correlation, together with the combined model was used for prediction of dissolution of capsules prepared at various encapsulation tamping conditions from API of arbitrary CSDs. As both the dissolution model and the breakage model are based on first principles, it was possible to use the models for direct evaluation of the model parameters by fitting to existing experimental data without need of any additional statistically designed experiments (DOE). It saved a lot of additional experimental work. Another advantage of models based on first principles is that they can be used also for prediction of the breakage and dissolution for interpolated and (to some extent) also extrapolated CSD and encapsulating conditions.

References 1. Shan G. et al.: Chem. Eng. J. 88 (2002) 53-58. 2. Reid K.: Chem. Eng. Sci. 20 (1965) 953-958.