Stability of monoclonal antibodies after simulated subcutaneous administration

Abstract

Changes in the environment from the drug product to the human physiology might lead to physical and/or chemical modifications of the protein drug, such as in vivo aggregation and fragmentation. Although subcutaneous (SC) injection is a common route of administration for therapeutic proteins, knowledge on in vivo stability in the SC tissue is limited. In this study, we developed a physiologic in vitro model simulating the SC environment in patients. We assessed the stability of two monoclonal antibodies (mAbs) in four different protein-free fluids under physiologic conditions. We monitored protein stability over two weeks using a range of analytical methods, in analogy to testing purposes of a drug product. Both mAbs showed an increase of protein aggregates, fragments, and acidic species. mAb1 was consistently more stable in this in vitro model than mAb2, highlighting the importance of comparing the stability of different mAbs under physiologic conditions. Throughout the study, both mAbs were substantially less stable in bicarbonate buffers as compared to phosphate-buffered saline. In summary, our developed model was able to differentiate stability between molecules. Bicarbonate buffers were more suitable compared to phosphate-buffered saline in regards to simulating the in vivo conditions and evaluating protein liabilities.

Assessment of Antibody Stability in a Novel Protein-Free Serum Model

“…Analytical approaches to overcome this challenge include the enrichment of the protein of interest (e.g., immunoaffinity purification) [ 20 ] or labeling (e.g., fluorescence dye) [ 5 , 11 ]. Alternatively, the biological fluid can be substituted with an appropriate surrogate buffer [ 6 , 12 ].…”

“…HMWS decreased consistently over time with the exception of mAb2 which remained within the same range in AS over 14 days. Previous studies reported an increase in HMWS under simulated physiological conditions [ 5 , 12 ]. Comparison to other studies is challenging due to differences in the analytical approach, protein concentration, fluids, sampling pull points, and, molecules.…”

“…Although all fluids remained stable, the cloudy appearance of BSF could hamper certain analyses (e.g., visible particle inspection) and appeared less suitable as a physiological in vitro model compared to AS and HSF. Artificial fluids such as AS may form SbVPs due to salt precipitation and therefore would be less applicable for particle analysis [ 12 , 15 ]. Fluids such as AS and the prepared filtrates, which are devoid of macromolecules, enable direct analysis in fluids; however, the matrix composition and osmolality of the prepared filtrates are more reflective of the physiological environment compared to AS.…”

“…However, both methods can be laborious and potentially impact the intrinsic protein properties and stability. Instead, biological fluids can be substituted with buffer systems that enable analysis in fluids directly yet deviate markedly from the conditions found in vivo (e.g., phosphate-buffered saline) [ 6 , 12 , 13 ]. Alternatively, stripping the neat biological fluid of macromolecules can lead to higher resemblance of the physiological conditions encountered.…”

Protein Aggregates in Inhaled Biologics: Challenges and Considerations

“…It is important to note that the guidance for sub-visible particles is not designed to de-risk protein aggregation in vivo or immunogenicity concerns, but rather to avoid the risk of capillary occlusion 27 . Pre-clinical in vitro models are increasingly employed to evaluate the protein stability post injection 28 , 29 , 30 . For instance, a novel protein-free serum in vitro setup revealed faster degradation profiles for two monoclonal antibodies (mabs) compared to accelerated stability studies 30 .…”