Sorbitol crystallization-induced aggregation in frozen mAb formulations.

Sorbitol crystallization-induced aggregation of mAbs in the frozen state was evaluated. The effect of protein aggregation resulting from sorbitol crystallization was measured as a function of formulation variables such as protein concentration and pH. Long-term studies were performed on both IgG1 and IgG2 mAbs over the protein concentration range of 0.1-120 mg/mL. Protein aggregation was measured by size-exclusion HPLC (SE-HPLC) and further characterized by capillary-electrophoresis SDS. Sorbitol crystallization was monitored and characterized by subambient differential scanning calorimetry and X-ray diffraction. Aggregation due to sorbitol crystallization is inversely proportional to both protein concentration and formulation pH. At high protein concentrations, sorbitol crystallization was suppressed, and minimal aggregation by SE-HPLC resulted, presumably because of self-stabilization of the mAbs. The glass transition temperature (Tg ') and fragility index measurements were made to assess the influence of molecular mobility on the crystallization of sorbitol. Tg ' increased with increasing protein concentration for both mAbs. The fragility index decreased with increasing protein concentration, suggesting that it is increasingly difficult for sorbitol to crystallize at high protein concentrations.

Beyond glass transitions: studying the highly viscous and elastic behavior of frozen protein formulations using low temperature rheology and its potential implications on protein stability.

PURPOSE: A novel application of oscillatory shear rheology was used to directly monitor global phase behavior of protein formulations in the frozen state and study its correlation with physical instability of frozen protein formulations. METHODS: Oscillatory rheology was used to measure changes in rheological parameters and to identify mechanical softening temperature (Ts*) and related properties of an IgG2 mAb formulation. Rheological measurements were compared to DSC/MDSC. Physical stability of IgG2 formulations was monitored by SE-HPLC. RESULTS: Rheological parameters and Ts* of an IgG2 formulation were sensitive to physical/morphological phase changes during freezing and thawing. Ts* of the frozen formulation was a function of concentration of protein and excipient. Complex modulus, G*, and phase angle, δ, for IgG2 at 70 mg/mL in a sucrose-containing formulation showed the system was not completely frozen at -10°C, which correlated to stability data consistent with ice-induced protein aggregation. CONCLUSIONS: We report the first application of oscillatory shear rheology to study phase behavior of IgG2 in a sucrose-containing formulation and its correspondence with physical stability not explained by glass transition (Tg'). We provide a mechanism and data suggesting that protein instability occurs at the ice/water interface.