Differential inhibition of human CYP2C8 and molecular docking interactions elicited by sorafenib and its major N-oxide metabolite.

The tyrosine kinase inhibitor sorafenib (SOR) is being used increasingly in combination with other anticancer agents like paclitaxel, but this increases the potential for drug toxicity. SOR inhibits several human CYPs, including CYP2C8, which is a major enzyme in the elimination of oncology drugs like paclitaxel and imatinib. It has been reported that CYP2C8 inhibition by SOR in human liver microsomes is potentiated by NADPH-dependent biotransformation. This implicates a SOR metabolite in enhanced inhibition, although the identity of that metabolite is presently unclear. The present study evaluated the capacity of the major N-oxide metabolite of SOR (SNO) to inhibit CYP2C8-dependent paclitaxel 6α-hydroxylation. The IC50 of SNO against CYP2C8 activity was found to be 3.7-fold lower than that for the parent drug (14 µM versus 51 µM). In molecular docking studies, both SOR and SNO interacted with active site residues in CYP2C8, but four additional major hydrogen and halogen bonding interactions were identified between SNO and amino acids in the B-B' loop region and helixes F' and I that comprise the catalytic region of the enzyme. In contrast, the binding of both SOR and SNO to active site residues in the closely related human CYP2C9 enzyme was similar, as were the IC50s determined against CYP2C9-mediated losartan oxidation. These findings suggest that the active metabolite SNO could impair the elimination of coadministered drugs that are substrates for CYP2C8, and mediate toxic adverse events, perhaps in those individuals in whom SNO is formed extensively.

Inhibition of Hepatic CYP2D6 by the Active N-Oxide Metabolite of Sorafenib.

The multikinase inhibitor sorafenib (SOR) is used to treat patients with hepatocellular and renal carcinomas. SOR undergoes CYP-mediated biotransformation to a pharmacologically active N-oxide metabolite (SNO) that has been shown to accumulate to varying extents in individuals. Kinase inhibitors like SOR are frequently coadministered with a range of other drugs to improve the efficacy of anticancer drug therapy and to treat comorbidities. Recent evidence has suggested that SNO is more effective than SOR as an inhibitor of CYP3A4-mediated midazolam 1'-hydroxylation. CYP2D6 is also reportedly inhibited by SOR. The present study assessed the possibility that SNO might contribute to CYP2D6 inhibition. The inhibition kinetics of CYP2D6-mediated dextromethorphan O-demethylation were analyzed in human hepatic microsomes, with SNO found to be ~ 19-fold more active than SOR (Kis 1.8 ± 0.3 µM and 34 ± 11 µM, respectively). Molecular docking studies of SOR and SNO were undertaken using multiple crystal structures of CYP2D6. Both molecules mediated interactions with key amino acid residues in putative substrate recognition sites of CYP2D6. However, a larger number of H-bonding interactions was noted between the N-oxide moiety of SNO and active site residues that account for its greater inhibition potency. These findings suggest that SNO has the potential to contribute to pharmacokinetic interactions involving SOR, perhaps in those individuals in whom SNO accumulates.

Sorafenib N-Oxide Is an Inhibitor of Human Hepatic CYP3A4.

The multi-kinase inhibitor sorafenib (SOR) is clinically important in the treatment of hepatocellular and renal cancers and undergoes CYP3A4-dependent oxidation in liver to the pharmacologically active N-oxide metabolite (SNO). There have been reports that kinase inhibitors such as SOR may precipitate pharmacokinetic interactions with coadministered drugs that compete for CYP3A4-mediated biotransformation, but these occur non-uniformly in patients. Clinical evidence also indicates that SNO accumulates in serum of some patients during prolonged SOR therapy. In this study undertaken in hepatic microsomes from individual donors, we assessed the possibility that SNO might contribute to pharmacokinetic interactions mediated by SOR. Enzyme kinetics of CYP3A4-mediated midazolam 1'-hydroxylation in individual human hepatic microsomes were analyzed by non-linear regression and appropriate replots. Thus, SNO and SOR were linear-mixed inhibitors of microsomal CYP3A4 activity (Kis 15 ± 4 and 33 ± 14 µM, respectively). To assess these findings, further molecular docking studies of SOR and SNO with the 1TQN crystal structure of CYP3A4 were undertaken. SNO elicited a larger number of interactions with key amino acid residues located in substrate recognition sequences of the enzyme. In the optimal docking pose, the N-oxide moiety of SNO was also found to interact directly with the heme moiety of CYP3A4. These findings suggest that SNO could contribute to pharmacokinetic interactions involving SOR, perhaps in individuals who produce high circulating concentrations of the metabolite.

Differential effects of hepatic cirrhosis on the intrinsic clearances of sorafenib and imatinib by CYPs in human liver.

The tyrosine kinase inhibitors sorafenib and imatinib are important in the treatment of a range of cancers but adverse effects in some patients necessitate dosage modifications. CYP3A4 has a major role in the oxidation of sorafenib to its N-oxide and N-hydroxymethyl metabolites and also acts in concert with CYP2C8 to mediate imatinib N-demethylation. CYP3A4 expression and function are impaired in patients with advanced liver disease, whereas the functions of CYP2C enzymes are relatively preserved. We evaluated the biotransformation of sorafenib and imatinib in well-characterized microsomal fractions from 17 control subjects and 19 individuals with hepatic cirrhosis of varying severity. The principal findings were that liver disease impaired the microsomal oxidation of sorafenib to its major metabolites to 40-44% of control (P<0.01), whereas the N-demethylation of imatinib was relatively unimpaired. The impairments in sorafenib biotransformation were correlated with decreased serum albumin concentrations and increased serum bilirubin concentrations in patients with liver disease, but not with the overall grade of liver disease according to the Child-Pugh system. In contrast, there was no relationship between imatinib N-demethylation and clinicopathologic factors in liver disease patients. These findings were accounted for in terms of the differential roles of CYPs 3A4 and 2C8 in the intrinsic clearance of the drugs. CYP3A4 has the major role in the intrinsic clearance of sorafenib but plays a secondary role to CYP2C8 in the intrinsic clearance of imatinib. In agreement with these findings CYP2C protein expression and CYP2C8-mediated paclitaxel 6α-hydroxylation were unimpaired in cirrhotic livers. This information could be adapted in individualized approaches such as in vivo CYP3A4 phenotyping to optimize sorafenib safety and efficacy in cancer patients with liver dysfunction.

Cytochrome P450-Mediated Biotransformation of Sorafenib and Its N-Oxide Metabolite: Implications for Cell Viability and Human Toxicity.

The multikinase inhibitor sorafenib (SRF) is approved for the treatment of renal and hepatic carcinomas and is also undergoing evaluation in therapeutic combinations with other anticancer agents. SRF is generally well tolerated but produces severe toxicities in a significant proportion of patients by mechanisms that are largely unknown. It has been shown that cytochrome P450 (CYP) 3A4 has a major role in SRF biotransformation to the pharmacologically active N-oxide (SRF-Nox) and two other metabolites. In this study, we prepared the major metabolites of SRF and evaluated their further biotransformation by CYPs in relation to their capacity to produce cellular toxicity. CYP3A4 was also found to be the principal enzyme that mediated the secondary oxidation of SRF metabolites. However, the reduction of SRF-Nox to SRF was also found to be a significant reaction mediated by several CYPs, especially CYPs 2B6 and 1A1. In human liver-derived HepG2 cells, SRF effectively decreased ATP production to an extent greater than that of its metabolites. SRF also markedly altered the cell cycle distribution in HepG2 cells by decreasing the proportion in G0/G1 phase and increasing that in S and G2/M phases. In comparison, SRF metabolites minimally affected HepG2 cell cycle progression. These findings suggest that SRF, but not its metabolites, prevents cells from entering the cell cycle and also inhibits cycling cells from completing mitosis. Reduction of the major metabolite SRF-Nox back to SRF may mediate decreased cellular viability and contribute to adverse reactions in some individuals.

Anti-proliferative actions of N'-desmethylsorafenib in human breast cancer cells.

The multi-kinase inhibitor sorafenib is used for the treatment of renal and hepatic carcinomas and is undergoing evaluation for treatment of breast cancer in combination with other agents. Cytochrome P450 (CYP) 3A4 converts sorafenib to multiple metabolites that have been detected in patient plasma. However, recent clinical findings suggest that combination therapy may elicit inhibitory pharmacokinetic interactions involving sorafenib that increase toxicity. While sorafenib N-oxide is an active metabolite, information on the anti-tumor actions of other metabolites is unavailable. The present study evaluated the actions of sorafenib and its five major metabolites in human breast cancer cell lines. All agents, with the exception of N'-hydroxymethylsorafenib N-oxide, decreased ATP formation in four breast cancer cell lines (MDA-MB-231, MDA-MB-468, MCF-7 and T-47D). Prolonged treatment of MDA-MB-231 cells with N'-desmethylsorafenib, N'-desmethylsorafenib N-oxide and sorafenib (10 µM, 72 h) produced small increases in caspase-3 activity to 128-139% of control. Sorafenib and its metabolites, again with the exception of N'-hydroxymethylsorafenib N-oxide, impaired MEK/ERK signaling in MDA-MB-231 cells and modulated the expression of cyclin D1 and myeloid cell leukemia sequence-1, which regulate cell viability. When coadministered with doxorubicin (0.5 or 1 µM), sorafenib and N'-desmethylsorafenib (25 µM) produced greater effects on ATP production than either treatment alone. Thus, it emerges that, by targeting the MEK/ERK pathway, multiple sorafenib metabolites may contribute to the actions of sorafenib in breast cancer. Because N'-desmethylsorafenib is not extensively metabolized and does not inhibit major hepatic CYPs, this metabolite may have a lower propensity to precipitate pharmacokinetic drug interactions than sorafenib.