Formation of mannitol hemihydrate in freeze-dried protein formulations--a design of experiment approach.

Since the discovery of mannitol hemihydrate, this form of mannitol has been seen as potentially negative with regard to the stability of pharmaceutical formulations. The formation of mannitol hemihydrate is reported in several case studies; however, no systematic investigation has been performed so far. In this study, design of experiments was applied for response surface modelling of mannitol hemihydrate formation. The formulation parameters investigated in a composite face-centred design were the overall solid content, protein concentration, protein type and the ratio between mannitol and sucrose. Additionally, annealing as process parameter was included in a full factorial mixed design. For two proteins, models with a high goodness of fit (R(2): 0.82 and 0.93) and goodness of prediction (Q(2): 0.78 and 0.89) were achieved. Inclusion of the process parameter annealing resulted in models of similar quality. The successful application of design of experiments showed that the most prominent factors enhancing the formation of hemihydrate were a high protein concentration, low relative mannitol content and annealing at -20°C.