3.1 Method evaluation
The separation of imatinib and IS was satisfactory and free of interfering peaks in the biological matrix. The retention time of imatinib and IS was 1.4 and 2.5 min, respectively (Figure 2). There was a good linear relationship between the concentration of imatinib (10-2000 ng/mL) in plasma and the ratio of peak area with R2 greater than 0.99. The intra-batch and inter-batch precision, expressed as RSD%, about 10, 30, 300 and 1800 ng/mL QC samples were less than 4.72%. Accuracy for intra-batch and inter-batch assays ranged from 105.07-108.60% (Table 1). The matrix effect for 30, 300 and 1800 ng/mL imatinib and IS were 106.36±3.98%、104.38±2.37%、102.22±3.39%、99.92±2.01%, along with RSD% less than 3.8%. The recovery were 100.76±3.08%, 104.21±3.30%, 102.68±3.65% and 99.01±4.39%, respectively. The RSD values were less than 3.6%. The stability of imatinib at different conditions conformed to the verification guidelines of biological sample quantitative analysis methods, and the RSD values were less than 5.83%. The accuracy ranged from 92.17 to 105.96%. In addition, work-solution stability of imatinib for 14 days was also assessed and the accuracy was within ±15% (Table 2). The average accuracy was 107.36% for 10-fold dilutions with RSD values less than 3.2%. The imatinib Cssmin would not be affected as plasma samples processed with dilution pre-treatment.