Evaluation of an Automated Immunoassay for Quantification of Imatinib in Plasma

INTRODUCTION: Imatinib mesylate is a first-line tyrosine kinase inhibitor used for the treatment of chronic myeloid leukemia (CML). Therapeutic drug monitoring (TDM) of imatinib is an important tool for evaluating patient adherence to therapy, particularly in the setting of therapeutic failure, and for dose optimization. It has been proposed that the therapeutic range for pre-dose (trough) plasma imatinib concentration (C₀) is 1000 - 3000 ng/mL (Biol Pharm Bull 38:645-54, 2015). TDM for imatinib is not widely available in the United States. Automated analysis of imatinib concentrations in plasma may improve accessibility of testing to support routine TDM.

OBJECTIVE: The objective of this study was to evaluate an automated immunoassay for quantification of imatinib in plasma.

METHODS: Imatinib concentrations were determined using a laboratory developed test based on reagents provided by Saladax Biomedical, Inc. (Bethlehem, NJ), and a Beckman Coulter AU5800 automated chemistry analyzer. Accuracy at 4 concentrations (350, 1000, 1600 and 2700 ng/mL), linearity over 11 concentrations (251 - 3315 ng/mL) and precision at 3 concentrations (604, 1120, 1932 ng/mL; 4 replicates per concentration, 2 runs per day over 4 days) were evaluated using fortified plasma samples. Residual authentic plasma samples from 21 CML patients treated with imatinib were also obtained based on protocols approved by the University of Utah Institutional Review Board. The samples were split for a method comparison with the Oregon Health & Science University in Portland, OR, where imatinib quantification is performed by liquid chromatography tandem mass spectrometry (LC-MS/MS). Statistical analysis was performed by EP Evaluator.

RESULTS: The immunoassay was found to be linear with a p=0.143 and a regression equation of y = 1.016x - 29.5. Mean recovery (n=20 results) ranged from 85.1% at 350 ng/mL to 90.3% at 2700 ng/mL. Within run, between run and total imprecisions, as measured by percent coefficient of variation (%CV) were <3.5% (n=32 per concentration). Qualitative agreement of patient sample results was 100%; concentrations of the patient specimens using the immunoassay ranged from 453 - 2543 ng/mL for positive samples (n=18), and 534 - 2929 ng/mL by LC-MS/MS. There were 3 samples that were negative by both methods. The correlation coefficient (R) was 0.9911 and regression equation was y = 0.901x - 35.3. When compared to the LC-MS/MS method, the immunoassay was associated with a 12.9% negative bias.

CONCLUSION: The automated immunoassay described here for determination of plasma imatinib concentrations performed with excellent linearity, accuracy and precision, and with an approximate 13% negative bias as compared to LC-MS/MS.