Purpose: The use of mouse serial sampling technique in PK assessment enables one pharmacokinetic profile per small animal, significantly contributes to the 3R principles (replacement, reduction, refinement) (1, 2). The mouse serial sampling technique involves the collection of a few microliters of whole blood from mouse, which is diluted in buffer, centrifuged and stored as the supernatant until further analysis. The sample matrix thus obtained is equivalent to diluted plasma (1). To support a GLP toxicity study in mouse for Drug-X (monoclonal antibody therapeutic) using serial sampling technique, we developed a quantitative ligand binding assay on the Gyrolab platform to quantify Drug-X. In addition to regular method validation stipulated in the bioanalytical guidance (3), current method validation was also designed to address: a) if the method could accurately quantify Drug-X that is spiked into mouse whole blood and is processed following mouse serial sampling procedure; and b) stability of Drug-X in processed mouse blood during storage.
Methods: In mouse serial sampling, 20 µL of mouse whole blood was collected from the tail vein, mixed with 390 µL of serial sampling buffer, and centrifuged for 10 minutes at 1500 g (1). The supernatant (processed mouse blood) was collected, stored at -70°C. Assuming a hematocrit factor of 50%, the processed mouse blood has a dilution factor of approximate 1:40 from pure plasma. In method validation, above procedures were used to process mouse whole blood (1). Calibration standards and QCs were prepared in 100% pooled mouse plasma and diluted at MRD 1:40 prior to assay. The method validation was designed to be a typical validation for ligand binding assay in mouse plasma (3), with additional matrix selectivity tested in mouse whole blood, and dilution linearity and sample stability evaluated in processed mouse blood.
Results: The method was validated successfully as a quantitative ligand binding assay on Gyros platform. The quantitative range of this method was 151 ng/mL to 4840 ng/mL in 100% mouse plasma. The inter- and intra- assay precision and accuracy met acceptance criteria (%CV < 9.6; %RE between -8.6% and 9.9%). The recovery of Drug-X spiked into 10 lots of mouse plasma and 10 lots of whole blood met targeted acceptance criteria (%RE between -4.6% and 6%), suggesting minimum interference from whole blood and from mouse serial sampling procedure. In addition, the dilution linearity in mouse plasma and in processed whole blood also met acceptance criteria (%RE-3.1% between and 8.2%) with dilution factor up to 1/650000, which demonstrated integrity of sample dilutions from processed mouse blood. Lastly, Drug-X spiked samples were shown to be stable in mouse plasma and in processed diluted whole blood when exposed to room temperature for 19 hours and when subjected to up to 6 freeze‑thaw cycles. Long term storage stability tests are ongoing and have established 6-month stability for Drug-X spiked in both mouse plasma and processed mouse blood.
Conclusion: We have successfully validated a ligand binding assay for the quantitation of a monoclonal antibody biotherapeutics in mouse blood samples collected by mouse serial sampling technique. The validated method was demonstrated to be accurate, precise, and has been successfully used in bioanalytical support of GLP mouse study. The experience shared in this study could serve as a model process for bioanalytical method validation when capillary microsampling technique is used.
Ihor Djura– Senior Associate Scientist, Pfizer Inc., Andover, Massachusetts
Soma Basak– Andover, Massachusetts
Brianna Donnelly– Andover, Massachusetts
Katrina Olson– Andover, Massachusetts
Rosemary Lawrence-Henderson– Andover, Massachusetts
Marcela Araya Roldan– Andover, Massachusetts
Alison Joyce– Pfizer Inc., Andover, Massachusetts
Ying Wang– Pfizer Inc., Andover, Massachusetts
Boris Gorovits– Senior Director PDM-NBE, Pfizer Inc., Andover, Massachusetts