DEEPAK K. SHARMA,1 PETER OMA,1 CLARK MERCHANT,1 DAVE KING,1 DAVE THOMAS1
1ProteinSimple Technologies Inc., 115 Terence Matthews Cr., Ottawa, Ontario, Canada, K2M2B2
Purpose:
Characterization of sub-visible particle populations in protein formulations is a regulatory requirement during formulation development. Analytical method development and validation recommends the use of linearity studies to help determine the suitability of a particular analytical technique. Many protein formulations are highly sensitive to dilution and sample preparation techniques, and therefore may produce an inherently nonlinear subvisible particle population. The purpose of this research is to illustrate this effect using a particle counting system (Micro-Flow Imaging).
Methods:
Protein sample ‘A’ was produced, consisting of a subvisible protein particle population induced within a stable, high concentration mAb formulation through stress mechanisms. This stock solution was diluted with buffer to obtain dilutions ranging from 1/100 to 1/819,200 (~100,000 particles/ml and ~20 particles/ml respectively at the extremes) and measured using a DPA4100 MFI particle counting/imaging system. A relatively less stable high concentration protein formulation ‘B’ was also selected and treated in a similar fashion to sample ‘A’, and the effect of sample dilution was quantified following the linearity measurements using MFI. Further testing was conducted to explore how subtleties in sample handling prior to testing can affect protein aggregate populations and subsequent particle counts.
Results:
Sample linearity was assessed by analyzing protein samples at different dilutions. After serial dilution, the particle population in formulation ‘A’ was observed to be highly linear, yielding an R^2 value of 0.9969. The linearity measurements of formulation ‘B’ were observed to be highly sensitive to the sample dilution and handling, exhibiting considerable variation in protein particle measurement. Time based measurement studies also showed significant variations, with ~15% RSD at time zero, 4% after one hour, 3% after two hours, and ~1% thereafter, depending on the dilution factor.
Conclusions:
Diluting some protein formulations will introduce an instability that introduces an inherent nonlinearity to the subvisible particle population. Furthermore, with such a formulation, sample handling and introduction can dramatically affect the measured subvisible particle population. Caution should therefore be exercised when conducting and interpreting linearity studies, in order to avoid reaching an erroneous conclusion regarding linearity and reproducibility of a particular analytical technique.
PRESENTED AT:
2010 AAPS National Biotechnology Conference, San Francisco CA, USA
2010 BioProcess International Europe Conference & Exhibit, Vienna, Austria
2010 ProteinSimple_Sample Dilution and Protein Particle Linearity.pdf
2010 ProteinSimple_Sample Dilution and Protein