Systematic development and optimization of a cIEF method for the charge variants analysis of basic monoclonal antibodies using the Analytical Quality by Design strategy

Monoclonal Antibodies (mAbs) are the most important biopharmaceuticals, chemically defined as glycoproteins and derived from a monoclonal cell line. As complex proteins, mAbs exhibit inherent structural heterogeneity as primarily due to post-translational modifications, such as deamidation, glycation, sialylation, methionine oxidation, etc. Therefore, different charged isoforms, often referred to as charge variants, are inextricably present in mAbs; their characterization and monitoring is a critical requirement to the quality control, and it has been considered as a Critical Quality Attribute (CQA). Capillary Isoelectric Focusing (cIEF) is a gold-standard technique for the high-resolution separation of mAb charge variants mainly addressed at the pI assessment. The separation by cIEF is influenced by several parameters and in the present work, Analytical Quality by Design (AQbD) approach was used to develop a Platform Analytical Procedure (PAP) able to analyze molecules that are sufficiently similar with respect to the attributes is intended to measure. Infliximab, a murine mAb mainly used in the treatment of inflammatory diseases, was selected as a model compounds. The Analytical Target Profile required a PAP able to ensure accurate measurement of mAbs isoforms pI with a Bias lower than 4%. The risk analysis step allowed the identification of several potential Critical Procedure Parameters (pCPPs) e.g., urea and methyl cellulose concentrations, cathodic and anodic stabilizer concentrations, and mainly the composition of the carrier ampholytes (CAs), the most important components establishing the pH gradient throughout the capillary. The ratio and type of CAs, namely ampholyte pH 3-10, pH 5-8 and pH 8-10.5 were recognized as Critical Procedure Parameters (CPP). The optimization of the composition was carried out by the mixture design, an approach in which the factors under study are considered as the components of a mixture, and the response depends on their relative proportions rather than on their absolute amounts. By means of Monte Carlo simulation, the Method Operable Design Region, accounting for model uncertainty and experimental variability, was defined with a probability level of 5%. Under the found AQbD conditions, the optimized cIEF method was validated for accuracy of the pI determination, linearity, accuracy and precision of the infliximab quantitation. Interestingly, selectivity was demonstrated by the ability of the cIEF system in differentiating between therapeutic proteins based on their specific charge heterogeneity. Despite being analyzed under identical experimental conditions, bevacizumab and daratumumab exhibited discrete and well-resolved isoform profiles confirming the ability of the optimized cIEF method to be a robust horizontal platform for quality control of mAbs.