Category: Formulation and Quality
Purpose: The development of peptide therapeutics has been growing in the pharmaceutical industry. With the development of these new complex biopharmaceuticals, the need of utilizing orthogonal or complementary analytical tools to characterize formulations and their stability are in demand. Peptide instability, like aggregation can result in loss of potency and immunogenicity. Detection of small peptide aggregates can be challenging depending on the formation mechanisms of aggregates. Size exclusion chromatography (SEC) is often used for detecting particles smaller than 100 nm, but the potential column adsorption was reported to disrupt the sample aggregates. In contrast, the use of FFF/MALS has the capacity to evaluate such aggregates in their native “states” due to the decreased sample shear and lack of column interactions. The goal of this study is to demonstrate the need for orthogonal and complementary tools and to show the abovementioned analytical bias with SEC and FFF/MALS, and other complementary methods like DLS and fluorescence microscopy.
Methods: Experiments were performed using a Wyatt field flow fractionation system coupled with a Dawn Helos-ll multiangle light scattering system (FFF/MALS). An Optilab T-rex RI detector, an Agilent UV detector and an Eclipse Dualtec controller were added on to the system. DLS measurements were performed on a Malvern, Zetasizer, nano series. FFF/MALS methods were developed and samples were initially characterized in 3 different mobile phases: PBS, pH 3.0 and pH 4.0 water. Formulations containing “Peptide X” were prepared in pH 3 and pH 4 water. After mobile phase and diluent optimization, peptide aggregates were separated by FFF/MALS and SEC confirming size and characters. DLS studies followed to confirm aggregates by measuring particle size and distribution in the sample formulations. DLS studies utilizing 1.0% and 2.5% HPBCD into the formulation media helped to understand mechanistic reasons for aggregation. Final evaluation of the formulations utilizing fluorescence microscopy will be carried out to determine the mechanism of Peptide X aggregation.
Results: Peptide X samples were prepared and stored @ 4°C for 14 days, then set at RT for 3 days to monitor aggregates. The FFF/MALS analysis showed that Peptide X, which has a molecular weight of ~ 4 kDa, formed tetramers and/or pentamers in the pH 3 media and higher-mers in the pH 4 media. Utilizing FFF/MALS, the molecular weight calculated was ~20 kDa in pH 3 and pH 4 across the peak at 9.2 and 9.6 minutes, respectively. The pH 4 sample had an additional peak at 23 minutes with a calculated MW of ~ 1 x 107 g/mol, indicating the presence of larger aggregates. These aggregates were not detected when the PBS was used as the mobile phase. Moreover, SEC showed that the majority of Peptide X in these formulations is monomer with a low percentage of dimers and trimers. DLS measurements were performed at two concentrations to evaluate concentration effect with respect to aggregation. Both 1 and 10 mg/ml samples in pH 3 & pH 4 showed that the majority of particles have an average diameter ranging from 4.3 to 5.6 nm and additionally there was a population of larger aggregates (100 nm) in both pH samples at 10 mg/ml. HPBCD was added to the 10 mg/ml Peptide X formulation in pH 3 media at 1.0% and 2.5% and the measured diameter of the majority ( >87%) of the particles by DLS were reduced to 2.75 nm and 3.05 nm in the presence of 1% and 2.5% HPBCD, respectively. Considering the periphery diameter of HPBCD (1.5 nm) reported in USP and the measured hydrodynamic diameter of 2 nm of neat HPBCD solutions, the data suggest that the aggregations formed in these formulations are reversible and the entity incorporated into HPBCD is of the size about 1 – 1.5 nm in diameter. These sizes also suggest that the incorporated entity was probably monomeric based on Jensen who reported that insulin has a hydrodynamic radius (rh ) of 1.5 nm, given the fact that Peptide X also has a lower MW.  Jensen, et al., Insulin diffusion and self-association characterized by real-time UV imaging and Taylor dispersion analysis, J. Pharm & Biomedical Analysis (2014)
Conclusion: The data from the FFF/MALS and DLS studies show that there are reversible aggregates in the samples. These aggregates were either not detected with SEC or could be de-structured when using a degrading mobile phase in FFF. These results show the importance of utilizing orthogonal and complementary methods in characterizing and monitoring the stability of peptide therapeutics.
 Jensen, et al., Insulin diffusion and self-association characterized by real-time UV imaging and Taylor dispersion analysis, J. Pharm & Biomedical Analysis (2014)
 Demeule, et al., New methods allowing the detection of protein aggregates, mAbs (2009)
Chen-Ming Lee– Ridgefield, Connecticut