Purpose: Various amino acids are known as bulking agents and cryoprotectants for lyophilized protein formulations including arginine in combination with counter ions like chloride and phosphate. Typically the amino acids are utilised in mixtures with stabilising sugars like sucrose. In addition, multivalent acids (e.g. citric acid and phosphoric acid) are known to raise both Tg’ and Tg of amorphous arginine systems, which has a positive effect on the freeze drying process and the storage stability of the formulation.
Consequently, arginine salts with different mono- and multivalent counter ions were studied as sugar free systems with respect to performance during the freeze drying process as well as the stability of a monoclonal antibody during freeze drying and after 3 month storage at 50°C.
Methods: All formulations containing L-arginine and different acids as counter ions, succinic acid, lactobionic acid, HCl, citric acid and phosphoric acid, were prepared with a total arginine content of 4 wt.% at pH 6 and 2 mg/ml mAb. Sucrose (7 wt.%) in 15 mM histidine buffer pH 6 containing 2 mg/ml mAb was used as reference.
Formulations (1.5 mL) containing arginine citrate, lactobionate and phosphate as well as the reference were freeze dried (FTS LyoStar III, SP Scientific) in 2R vials with lyophilisation stoppers according to the following protocol: frozen to 50°C at 0.5 K/min; primary drying at 0.06 mbar and 20°C shelf temperature for 50 h; secondary drying at 0.06 mbar and 40°C shelf temperature for 8.3 h. Formulations containing arginine chloride and succinate were kept at 0.04 mbar for 55 h and 50°C shelf temperature for 5 h during secondary drying.
Tg’ was determined by DSC (DSC 821e, Mettler Toledo). Freeze dried samples were analysed for visual appearance, residual moisture (RM) by Karl Fischer analysis (AQUA 40.0, Analytik Jena AG), morphology by XRD analysis (3000TT, Seifert), and Tg and crystallisation events by DSC (Polyma 214, Netzsch).
After reconstitution of the freeze dried products size exclusion chromatography (BEH SEC 200Å, Waters), turbidity analysis (Nephla turbidimeter, Dr. Lange), and sub visible particles analysis (SVSS C35, PAMAS) were performed. Lyophilisates were stored at 50°C for 3 months and subsequently analysed.
Results: The Tg’ values of -29°C for Arg-phosphoric acid up to -26°C for Arg-lactobionic acid were higher compared to that of the sucrose reference of -33°C (Table 1). Samples containing Arg-HCl and succinic acid showed Tg’ values of -46°C and -38°C resp. which might be critical during the freeze drying process. All lyophilisates exhibited an elegant cake structure (Figure 1). The Tg values of 76°C for Arg-HCl up to 113°C for Arg-citric acid were markedly higher than that of the sucrose reference formulation of 64°C. Only samples containing HCl and sucrose showed a recrystallisation peak at higher temperatures. After 3 months storage at 50°C Arg-HCl samples crystallised. Overall, the residual moisture level of the formulations was below 1.0% and stayed below 1.6% over storage.
Reconstitution was completed for all samples within 30 seconds. After preparation Arg-citric acid, -HCl, and -succinic acid lyophilisates showed a lower sub-visible particle count compared to that of the sucrose based formulation (Figure 2). After 3 months storage, an increase of the sub visible particle count can be found in Arg-citric acid, -lactobionic acid, -phosphoric acid, and -succinic acid formulations. The turbidity increased from 3 – 7 FNU after freeze drying to 4 - 15 FNU after 3 months. The lowest turbidity was found for Arg-HCl and the highest for Arg-lactobionic acid. After freeze-drying, SEC results were similar for all formulations with 0.25% low molecular weight species (LMWS) and 1% high molecular weight species (HMWS). After 3 months storage, LMWS levels increased to 1.5% and HWMS to 9% for Arg-citric acid and Arg-phosphoric acid, all other formulations showed an increase by 1% HMWS.
Conclusion: Both arginine and sucrose were able to act as cryoprotectants and resulted in elegant cakes. The arginine systems showed higher Tg values compared to the sucrose formulation and both systems had comparable HMWS and LMWS levels. The sub visible particle count was lower for Arg-citric acid, -HCl, and -succinic acid.
After storage, arginine lyophilisates containing multivalent counter ions or lactobionic acid showed increased particle counts while particle formation was not observed in sucrose and Arg-HCl products. Only Arg-citric acid and -phosphoric acid resulted in an increase in HWMS upon storage.
In conclusion, freeze drying of antibodies in sugar-free arginine formulations is possible with low residual moisture levels and high Tg values. The low Tg’ value of Arg-HCl is critical for the lyophilisation process, but during stability testing at 50°C Arg-HCl showed better results compared to sucrose as well as all other arginine based products, which showed increased antibody aggregation after 3 months storage. Other proteins and lower storage temperatures will be evaluated.
Wolfgang Friess– Ludwig Maximilians University, Munich, Bayern