Feasibility of pharmacometabolomics to identify potential predictors of paclitaxel pharmacokinetic variability.

PURPOSE

Paclitaxel is a commonly used chemotherapy drug with substantial variability in pharmacokinetics (PK) that affects treatment efficacy and toxicity. Pharmacometabolomic signatures that explain PK variability could be used to individualize dosing to improve therapeutic outcomes. The objective of this study was to identify pretreatment metabolites or metabolomic signatures that explain variability in paclitaxel PK.

METHODS

This analysis was conducted using data previously collected on a prospective observational study of 48 patients with breast cancer receiving weekly 80 mg/m paclitaxel infusions. Paclitaxel plasma concentrations were measured during the first infusion to estimate paclitaxel time above threshold (T) and maximum concentration (C). Metabolites measured in pretreatment whole blood by nuclear magnetic resonance spectrometry were analyzed for an association with T and C using Pearson correlation followed by stepwise linear regression.

RESULTS

Pretreatment creatinine, glucose, and lysine concentrations were positively correlated with T, while pretreatment betaine was negatively correlated and lactate was positively correlated with C (all uncorrected p < 0.05). After stepwise elimination, creatinine was associated with T, while betaine and lactate were associated with C (all p < 0.05).

CONCLUSION

This study identified pretreatment metabolites that may be associated with paclitaxel PK variability demonstrating feasibility of a pharmacometabolomics approach for understanding paclitaxel PK. However, identification of more robust pharmacometabolomic predictors will be required for broad and routine application for the clinical dosing of paclitaxel.