Coupling of Trastuzumab chromatographic profiling with machine learning tools: A complementary approach for biosimilarity and stability assessment

Streamlining the Characterization of Disulfide Bond Shuffling and Protein Degradation in IgG1 Biopharmaceuticals Under Native and Stressed Conditions

Post translational modifications (PTMs) have been shown to negatively impact protein efficacy and safety by altering its native conformation, stability, target binding and/or pharmacokinetics. One PTM in particular, shuffled disulfide bonds, has been linked to decreased potency and increased immunogenicity of protein therapeutics. In an effort to gain more insights into the effects of shuffled disulfide bonds on protein therapeutics' safety and efficacy, we designed and further optimized a semi-automated LC-MS/MS method for disulfide bond characterization on two IgG1 protein therapeutics-rituximab and bevacizumab. We also compared originator vs. biosimilar versions of the two therapeutics to determine if there were notable variations in the disulfide shuffling and overall degradation between originator and biosimilar drug products. From our resulting data, we noticed differences in how the two proteins degraded. Bevacizumab had a general upward trend in shuffled disulfide bond levels over the course of a 4-week incubation (0.58 ± 0.08% to 1.46 ± 1.10% for originator) whereas rituximab maintained similar levels throughout the incubation (0.24 ± 0.21% to 0.51 ± 0.11% for originator). When we measured degradation by SEC and SDS-PAGE, we observed trends that correlated with the LC-MS/MS data. Across all methods, we observed that the originator and biosimilar drugs performed similarly. The results from this study will help provide groundwork for comparative disulfide shuffling analysis by LC-MS/MS and standard analytical methodology implementation for the development and regulatory approval of biosimilars.