Pathway-Dependent Inhibition of Paclitaxel Hydroxylation by Kinase Inhibitors and Assessment of Drug–Drug Interaction Potentials

Yedong Wang, Meiyu Wang, Huixin Qi, Peichen Pan, Tingjun Hou, Jiajun Li, Guangzhao He, Hongjian ZhangDrug 2014 · DOI

TL;DR

This study investigates how kinase inhibitors, commonly used in cancer therapy, can interfere with the metabolism of paclitaxel, a widely used chemotherapy drug. By examining the inhibition of key enzymes (CYP2C8 and CYP3A4) responsible for paclitaxel breakdown, the researchers identified specific inhibitors that can alter paclitaxel's metabolic profile, potentially leading to significant drug-drug interactions and impacting treatment efficacy and safety.

What Problem Does This Paper Address?

Paclitaxel is a cornerstone in treating various cancers, but its effectiveness can be hampered by significant inter-patient variability in its response and toxicity, particularly neurotoxicity. This variability is partly attributed to how paclitaxel is metabolized in the body, primarily by two cytochrome P450 enzymes: CYP2C8 and CYP3A4. As cancer treatment increasingly involves combination therapies, particularly with novel small molecule kinase inhibitors, understanding potential drug-drug interactions (DDIs) is crucial. Kinase inhibitors, by their nature, can inhibit these metabolic enzymes, potentially altering the levels of paclitaxel and its metabolites. This study aims to systematically evaluate the inhibitory potential of 12 different kinase inhibitors on paclitaxel's CYP2C8 and CYP3A4 metabolic pathways and to predict the clinical implications of these interactions.

Methods

The researchers employed a multi-faceted approach to assess the interactions:

  • In Vitro Enzyme Inhibition Assays: Human liver microsomes (HLMs) were used to study the metabolism of paclitaxel. The formation of two key metabolites, 6a-hydroxypaclitaxel (CYP2C8-mediated) and p-3'-hydroxypaclitaxel (CYP3A4-mediated), was simultaneously monitored using liquid chromatography-tandem mass spectrometry (LC-MS/MS).
  • Reversible Inhibition Studies: The inhibitory potency (Ki values) of 12 kinase inhibitors against both CYP2C8 and CYP3A4 were determined.
  • Time-Dependent Inhibition (TDI) Studies: For inhibitors showing strong reversible inhibition, their ability to cause time-dependent inactivation of CYP3A4 was investigated. This involved pre-incubation of the inhibitor with HLMs before adding paclitaxel. Axitinib's inactivation kinetics (kinact and K₁ values) were specifically characterized.
  • Molecular Docking Simulations: To understand the structural basis of inhibition, molecular docking was used to predict how kinase inhibitors bind to the active sites of CYP3A4 and CYP2C8.
  • UV-Visible Spectral Analysis: This technique was used to characterize the binding type (e.g., Type I or Type II) of inhibitors to the P450 enzymes, providing further insight into their interaction mechanisms.
  • Drug-Drug Interaction (DDI) Prediction: Mathematical models (static models) were used to predict the change in paclitaxel's area under the curve (AUC) in the presence of kinase inhibitors. These models incorporated reversible and time-dependent inhibition data, as well as the hepatic extraction ratio (EH) of paclitaxel.

Datasets

The study utilized:

  • Human Liver Microsomes (HLMs): Pooled HLMs from 20 different organ donors were used to ensure consistency.
  • Kinase Inhibitors: A panel of 12 commercially available small molecule kinase inhibitors were tested.
  • Paclitaxel and its Metabolites: Paclitaxel, 6a-hydroxypaclitaxel, and p-3'-hydroxypaclitaxel were used as substrates and standards.
  • NADPH: A necessary cofactor for CYP enzyme activity.

No new benchmark datasets were introduced; the study focused on evaluating known compounds and established in vitro systems.

Key Insights and Ideas

  • Pathway-Dependent Inhibition is Key: Not all kinase inhibitors affect paclitaxel metabolism equally. Some preferentially inhibit CYP2C8 (e.g., axitinib), while others target CYP3A4 (e.g., bosutinib), highlighting the importance of understanding which pathway is affected.
  • Nilotinib is a Potent Dual Inhibitor: Nilotinib emerged as a particularly strong inhibitor for both CYP2C8 and CYP3A4, suggesting a high potential for significant DDIs.
  • Time-Dependent Inactivation of CYP3A4 is Common: Many kinase inhibitors not only reversibly inhibit CYP3A4 but also irreversibly inactivate it, a mechanism that can lead to more profound and prolonged DDIs.
  • Structural Insights into Binding: Molecular docking and spectral analysis revealed that strong inhibition often involves specific hydrogen bonding and arene-cation interactions within the enzyme's active site, with Type II binding spectra being common for many inhibitors.
  • Hepatic Extraction Ratio Matters for Prediction: Accurately predicting DDIs requires considering the victim drug's (paclitaxel's) hepatic extraction ratio (EH). Ignoring EH can lead to overestimation of interaction severity.

Key Results

  • Potent Reversible Inhibition: Nilotinib showed the lowest Ki values for both CYP2C8 (0.10 µM) and CYP3A4 (0.28 µM).
  • Pathway Selectivity: Axitinib was more potent against CYP2C8, while bosutinib was more potent against CYP3A4.
  • Axitinib's Inactivation Kinetics: Axitinib exhibited a Ki of 0.93 µM and a kinact of 0.0137 min⁻¹ for CYP3A4 inactivation, suggesting it forms reactive intermediates.
  • DDI Prediction: Using a static model that incorporated EH, nilotinib was predicted to cause the largest increase in paclitaxel AUC (up to 3.8-fold), while other potent inhibitors also showed significant predicted interactions.
  • Correlation between Docking and Inhibition: Moderate correlations were observed between docking scores and experimental Ki values for CYP3A4, but weaker for CYP2C8, suggesting room for improvement in docking accuracy for this enzyme.

Limitations and Future Work

  • In Vitro to In Vivo Extrapolation: While predictive models were used, they are simplifications. Clinical DDI data for paclitaxel and kinase inhibitors are limited, making direct validation challenging.
  • Individual HLM Variability: Although pooled HLMs were used, individual donor variability in enzyme activity can influence results.
  • Metabolite Ratios and Neurotoxicity: The study highlights that altered metabolite ratios could be clinically relevant, especially concerning paclitaxel-induced neurotoxicity, but further investigation into these specific relationships is needed.
  • Further Characterization of CYP2C8 Inactivation: The paper notes that while time-dependent inhibition of CYP3A4 was common, further investigation into CYP2C8 inactivation mechanisms is ongoing.
  • Broader Range of Inhibitors and Combinations: The study focused on 12 inhibitors; future work could expand this panel and explore more complex drug combinations.

The authors emphasize that despite the lack of reported clinical DDIs, the strong in vitro inhibition warrants further evaluation, particularly concerning the impact on metabolite formation and its link to neurotoxicity.

Summary updated

BibTeX
@inproceedings{wang2014pathway,
  author = {Yedong Wang and Meiyu Wang and Huixin Qi and Peichen Pan and Tingjun Hou and Jiajun Li and Guangzhao He and Hongjian Zhang},
  title = {{Pathway-Dependent Inhibition of Paclitaxel Hydroxylation by Kinase Inhibitors and Assessment of Drug–Drug Interaction Potentials}},
  booktitle = {Drug},
  year = {2014},
  doi = {10.1124/dmd.113.053793},
}
Figure