An improved size exclusion-HPLC method for molecular size distribution analysis of immunoglobulin G using sodium perchlorate in the eluent
Size exclusion (SE) high performance liquid chromatography (HPLC) is widely used for the molecular size distribution (MSD) analyses of various therapeutic proteins. We report development and validation of a SE-HPLC method for MSD analyses of immunoglobulin G (IgG) in products using a TSKgel SuperSW3000 column and eluting it with 0.4 M NaClO4, a chaotropic salt, in 40 mM phosphate buffer, pH 6.8. The chromatograms show distinct peaks of aggregates, tetramer, and two dimers, as well as the monomer and fragment peaks. In addition, the method offers about half the run time (12 min), better peak resolution, improved peak shape and more stable base-line compared to HPLC methods reported in the literature, including that in the European Pharmacopeia (EP). A comparison of MSD analysis results between our method and the EP method shows interactions between the protein and the stationary phase and partial adsorption of aggregates and tetramer on the stationary phase, when the latter method is used. Thus, the EP method shows lower percent of aggregates and tetramer than are actually present in the products. In view of the fact that aggregates have been attributed to playing a critical role in adverse reactions due to IgG products, our observation raises a major concern regarding the actual aggregate content in these products since the EP method is widely used for MSD analyses of IgG products. Our method eliminates (or substantially reduces) the interactions between the proteins and stationary phase as well as the adsorption of proteins onto the column. Our results also show that NaClO4 in the eluent is more effective in overcoming the protein/column interactions compared to Arg-HCl, another chaotropic salt. NaClO4 is shown not to affect the molecular size and relative distribution of different molecular forms of IgG. The method validated as per ICH Q2(R1) guideline using IgG products, shows good specificity, accuracy, precision and a linear concentration dependence of peak areas for different molecular forms. In summary, our method gives more reliable results than the SE-HPLC methods for MSD analyses of IgG reported in the literature, including the EP, particularly for aggregates and tetramer. The results are interpreted in terms of ionic (polar) and hydrophobic interactions between the stationary phase and the IgG protein.