A model-based optimization strategy to achieve fast and robust freeze-drying cycles

Abstract

Freeze-drying is a time and cost-intensive process. The primary drying phase is the main target in a process optimization exercise. Biopharmaceuticals require an amorphous matrix for stabilization, which may collapse during primary drying if the critical temperature of the formulation is exceeded. The risk of product collapse should be minimized during a process optimization to accomplish a robust process, while achieving an economical process time. Mechanistic models facilitate the search for an optimal primary drying protocol. We propose a novel two-stage shelf temperature optimization approach to maximize sublimation during the primary drying phase, without risking product collapse. The approach includes experiments to obtain high-resolution variability data of process parameters such as the heat transfer coefficient, vial dimensions and dried layer resistance. These process parameters variability data are incorporated into an uncertainty analysis to estimate the risk of failure of the protocol. This optimization approach enables to identify primary drying protocols that are faster and more robust than a classical approach. The methodology was experimentally verified using two formulations which allow for either aggressive or conservative freeze-drying of biopharmaceuticals.