Meta-analysis of nanoparticulate paclitaxel delivery system pharmacokinetics and model prediction of associated neutropenia.

PURPOSE: Nanoparticulate paclitaxel carriers have entered clinical evaluation as alternatives to the Cremophor-based standard Taxol(®) (Cre-pac). Their pharmacokinetics (PK) is complex, and factors influencing their pharmacodynamics (PD) are poorly understood. We aimed to develop a unified quantitative model for 4 paclitaxel carriers that captures systems-level PK, predicts micro-scale PK processes, and permits correlations between carrier properties and observed PD.
METHODS: Data consisting of 54 PK profiles and 574 observations were extracted from 20 clinical studies investigating Cre-pac, albumin-(A-pac), liposome-(L-pac), and tocopherol-(T-pac) nanocarriers. A population-PK approach was used for data analysis. All datasets were simultaneously fitted to produce a unified model. Model-based simulations explored relationships between predicted PK and myelosuppression for each formulation.
RESULTS: The final model employed nonlinear drug-binding mechanisms to describe Cre-pac and a delayed-release model for A-pac, L-pac, and T-pac. Estimated drug-release rate constants (h(-1)): Cre-pac (5.19), L-pac (1.26), A-pac (0.72), T-pac (0.74). Simulations of equivalent dosing schemes ranked neutropenia severity (highest to lowest): T-pac~Cre-pac>L-pac~A-pac and predicted remarkably well the clinically-observed relationships between neutropenia and free drug exposure relative to a threshold concentration.
CONCLUSIONS: Paclitaxel disposition was well-described for all formulations. The derived model predicts toxicodynamics of diverse paclitaxel carriers.