Trough plasma concentration of imatinib reflects BCR-ABL kinase inhibitory activity and clinical response in chronic-phase chronic myeloid leukemia: a report from the BINGO study.

Pharmacokinetic (PK) factors have been suggested to be involved in the unfavorable clinical responses of chronic myeloid leukemia (CML) patients treated with imatinib. The purpose of this study was to clarify prognostic implications of PK factors in CML patients treated with imatinib. The plasma trough (C(min)) level of imatinib and serum α(1)-acid glycoprotein (AGP) level were measured on two different days in 65 CML patients treated with imatinib for more than 12 months. We further examined whether the C(min) level of imatinib actually reflects inhibitory activity against BCR-ABL kinase using the plasma inhibitory activity (PIA) assay. Since the differences of five patients were statistically rejected by the Smirnov-Grubbs' test, we excluded them for further analysis. The C(min) level was strongly associated with the achievement of MMR at the 12th month, and ROC analysis demonstrated C(min) levels and their discrimination potential for major molecular response (MMR) with the best sensitivity (63.2%) and specificity (68.2%) at a C(min) threshold of 974 ng/mL. The α(1)-Acid glycoprotein (AGP) level was within the normal range in 57 of 60 patients, indicating little impact of AGP on our study. There was a weak correlation between PIA against phospho (P)-BCR-ABL and the C(min) level of imatinib (r(2) = 0.2501, P = 0.0007), and patient plasma containing >974 ng/mL imatinib sufficiently inhibited P-BCR-ABL. These results collectively indicated that maintaining ∼1000 ng/mL of C(min) was clinically and biologically important for the optimal response in CML patients treated with imatinib. A prospective intervention study is required to establish PK-based management in CML patients treated with imatinib.