Comparisons of Two Different Analytical Methodologies for the Characterization of Sub-Visible Particles in Therapeutic Protein Formulations

Abstract

The US and EU pharmacopeias require sub-visible particle (SbVP) testing of parenteral drug products by either light obscuration (LO) or light microscopy. According to the USP/EP, the LO method requires the use of four measurements consuming a total volume of 25 ml per sample. This large sample volume makes the compendia test not only cost-intensive for biopharmaceutics, but in certain cases impractical, especially in early stage development when clinical supplies are often limited. The first part of this work therefore presents a set of experimental data for evaluation of a small scale, lower volume LO method (1.2 ml per sample). Limitations for detecting various translucent particles such as glass splinters, silicone oil droplets and non-soluble protein aggregates have been considered for both of these LO methods. In addition, a relatively new technology - Micro Flow Digital Imaging (MDI) - has been introduced, which may be a viable add-on to measure SbVP. The purpose of the second phase of this research was to evaluate the advantages and limitations of MDI by directly comparing LO and MDI, and also by comparing two different MDI instruments (i.e., Micro Flow Imaging (MFI) and FlowCam), with regards to subvisible particle counts, size distribution and morphology parameter analysis. As protein subvisible particles vary widely in shape and size, it is difficult to quantitatively describe the limitations of MDI and LO for actual samples of a protein therapeutic drug. Moreover, real-life protein formulations can differ in optical properties such as turbidity, color, refractive index which may potentially influence SbVP detection. During this work, MDI technologies have been improved for the use on a routine basis with protein samples (e.g., FlowCam counting ability was improved by adding a syringe pump for controlled flow, and MFI optical system was enhanced with the so called set-point 3). The effect of solution optical properties on SbVP particle sizing and counting was addressed subsequently using the latest MDI instrument and three different LO instruments. In addition, translucent particles such as glass and irregular shaped particles in development as "false protein standard" (to mimic properties of real life proteinaceous particles) were evaluated in this work.