Effect of sorbitol and residual moisture on the stability of lyophilized antibodies: Implications for the mechanism of protein stabilization in the solid state

Abstract Purpose To investigate the effect of plasticizers on the stability of protein formulations in the solid state and to apply these results to a study of mechanisms of protein stabilization by sugars in the solid sate. Methods The IgG1 antibody was formulated with either sucrose or trehalose alone or a mixture of sorbitol with sucrose or trehalose. After lyophilization, the pure protein and sucrose formulations were equilibrated at different relative humidities giving residual moistures from less than 1% to 5% for sucrose systems and up to 17% for pure protein systems. All the samples were stored at 50°C for up to1 month and at 40 and 25°C for up to 6 months. Aggregation and chemical degradation were monitored by size exclusion chromatography (SEC) and ion exchange chromatography (IEX), respectively. The secondary structure was characterized by FTIR using second derivative analysis of Amide I region. Structural relaxation times, τ, an indication of molecular mobility in the glassy matrix, were characterized using the thermal activity monitor (TAM). The τ values of the recombinant human monoclonal antibody (rhuMab) formulation with various water contents were also measured in this study and compared with stability data taken from the literature (Breen ED, Curley JG, Overcashier DE, Hsu CC, Shire SJ, 2001, Pharm Res 18:1345–1353). Results The structural relaxation time, τ, decreased sharply with increasing water content. However, the stability data suggest a minimum in degradation rate at 2%–3% water content. Addition of a small amount of sorbitol to a sucrose-based formulation resulted in greater retention of native structure, smaller relaxation time, but improved stability. However, with the trehalose-based formulations, addition of sorbitol had no effect on protein structure (FTIR), but the decrease in relaxation time and the improvement in stability were qualitatively similar to the corresponding data obtained with the sucrose-based formulation. Conclusion Glass dynamics as measured by τ could not explain the stability results. Stability correlated best with the preservation of native structure for sucrose-based formulations, but with the trehalose-based formulation, neither structural relaxation time nor extent of native structure was predictive of stability. However, it is possible that the β-relaxations rather than the α-relaxation (i.e., the τ we measured) is critical to the stability. Plasticizers like glycerol may decrease τ for “α-motion” but increase τ for “β-motion” and stabilize proteins (Cicerone MT, Tellington A, Trost L, Sokolov A, 2003, BioProcess Inter 1:1–9). © 2005 Wiley-Liss, Inc. and the American Pharmacists Association