Pharmacokinetics of Imatinib Mesylate and Development of Limited Sampling Strategies for Estimating the Area under the Concentration–Time Curve of Imatinib Mesylate in Palestinian Patients with Chronic Myeloid Leukemia

Background and Objective Imatinib is a tyrosine kinase inhibitor used in the treatment of chronic myeloid leukemia (CML). The area under the concentration–time curve (AUC) is a pharmacokinetic parameter that symbolizes overall exposure to a drug, which is correlated with complete cytogenetic and treatment responses to imatinib, as well as its side effects in patients with CML. The limited sampling strategy (LSS) is considered a sufficiently precise and practical method that can be used to estimate pharmacokinetic parameters such as AUC, without the need for frequent, costly, and inconvenient blood sampling. This study aims to investigate the pharmacokinetic parameters of imatinib, develop and validate a reliable and practical LSS for estimating imatinib AUC 0–24 , and determine the optimum sampling points for predicting the imatinib AUC after the administration of once-daily imatinib in Palestinian patients with CML. Method Pharmacokinetic profiles, involving six blood samples collected during a 24-h dosing interval, were obtained from 25 Palestinian patients diagnosed with CML who had been receiving imatinib for at least 7 days and had reached a steady-state level. Imatinib AUC 0–24 was calculated using the trapezoidal rule, and linear regression analysis was performed to assess the relationship between measured AUC 0–24 and concentrations at each sampling time. All developed models were analyzed to determine their effectiveness in predicting AUC 0–24 and to identify the optimal sampling time. To evaluate predictive performance, two error indices were employed: the percentage of root mean squared error (% RMSE) and the mean predictive error (% MPE). Bland and Altman plots, along with mountain plots, were utilized to assess the agreement between measured and predicted AUC. Results Among the one-timepoint estimations, predicted AUC 0–24 based on concentration of imatinib at the eighth hour after administration ( C 8 -predicted AUC 0–24 ) demonstrated the highest correlation with the measured AUC ( r 2 = 0.97, % RMSE = 6.3). In two-timepoint estimations, the model consisting of C 0 and C 8 yielded the highest correlation between predicted and measured imatinib AUC ( r 2 = 0.993 and % RMSE = 3.0). In three-timepoint estimations, the combination of C 0 , C 1 , and C 8 provided the most robust multilinear regression for predicting imatinib AUC 0–24 ( r 2 = 0.996, % RMSE = 2.2). This combination also outperformed all other models in predicting AUC. The use of a two-timepoint limited sampling strategy (LSS) for predicting AUC was found to be reliable and practical. While C 0 / C 8 exhibited the highest correlation, the use of C 0 / C 4 could be a more practical and equally accurate choice. Therapeutic drug monitoring of imatinib based on C 0 can also be employed in routine clinical practice owing to its reliability and practicality. Conclusion The LSS using one timepoint, especially C 0 , can effectively predict imatinib AUC. This approach offers practical benefits in optimizing dose regimens and improving adherence. However, for more precise estimation of imatinib AUC, utilizing two- or three-timepoint concentrations is recommended over relying on a single point.