Time-dependent inactivation of cytochrome P450 3A by imatinib: implications for its clinical pharmacokinetics and drug interactions

It has been reported that oral clearance (CL/F) of imatinib decreased by ~26% at steady-state compared to that at day 1 in the patients with gastrointestinal stromal tumor or chronic myeloid leukemia.  The mechanism underlying this time-dependent change on oral clearance remains unclear.  In the present study, we report that imatinib inhibited CYP3A, the enzyme that is partially responsible for the drug’s own metabolism, in an NADPH, preincubation time, and concentration-dependent manner.  k_inact (maximal rate of enzyme inactivation) and apparent K_I (inhibitor concentration achieving half the maximal rate of inactivation) were determined using both human liver microsomes (HLM) and pooled, cryopreserved human hepatocytes.  Incubation of fresh human hepatocytes with imatinib (48 hours) resulted in a significant loss of CYP3A activity, but had no impact on CYP3A4 mRNA.  The apparent k_inact and K_I values derived from human hepatocytes were used to predict clinical DDIs associated with CYP3A inhibition by imatinib using a population-based simulator.  The prediction correlated well with the reported in vivo DDI data. Moreover, two cyanide conjugates of imatinib formed in HLM supplemented with NaCN were structurally characterized by LC-MS/MS to elucidate the mechanism of time-dependent CYP3A inhibition.  These findings provide a possible explanation for the time-dependent decrease in oral clearance of imatinib and suggest that the reported lack of effect on the steady-state pharmacokinetics of imatinib by acute ritonavir treatment is also due to inactivation of CYP3A by imatinib.  In addition, our predictions indicate that imatinib may contribute to the clinical DDIs with CYP3A substrates (e.g., simvastatin) mainly through the mechanism of time-dependent CYP3A inhibition rather than the reversible inhibition as previously reported.