Category: Clinical Pharmacology
Purpose: Crystalline camptothecin-20-O-propionate hydrate (CZ48) is a potential anticancer agent and a prodrug of camptothecin (CPT). A new micronized formulation of CZ48 was developed and the bioavailability and pharmacokinetics (PK) were evaluated under the Food and Drug Administration guidelines for clinical trials by Cao Pharmaceutical Inc. In previous PK analysis, CZ48 profiles were typical and reached steady-state quickly by 24 h in all patients. However, those of the active metabolite CPT were atypical and CPT concentratons continued to accumulate over 4 to 5 days. CPT is responsible for the anticancer activity and toxicity, and thus it is crucial to monitor the CPT levels and project CPT profiles during the treatment course to ensure the safe administration of CZ48.
Methods: Fourteen patients were enrolled in the study with the new formulation of CZ48 in Phase I clinical trial. Patients were given 3 doses (50 mg of CZ48) for prescreening, and blood samples were withdrawn prior to and at 0.5, 1, 2, 4, 8, 10, and 24h post the first dose. After wash out period of more than 1 week, the qualified patients were administrated with CZ48 (50 mg) twice daily. The PK samples were withdrawn prior to and at 0.5, 1, 2, 4, 8, 24, 48, 72 and 96h post the first dose. Population PK modeling and simulation, using actual dosing and sampling times, were developed using Phoenix (version 8.0). Model selection was based on the likelihood ratio test on the difference of criteria (-2LL) and comparison of the quality of fit plots.
Results: Two PK models (models B and C) were established for the best prediction of CZ48 and CPT PK profiles (Figure 1). The models consisted of one compartment each for CZ48 and CPT, and an enterohepatic cycling (EHC) compartment was connected with CPT by the nonlinear PK process. The difference of model C from B was with another added elimination process for EHC compartment. The PK parameters from the population PK model analysis were derived, with accuracy of different diagnostic plots predicted versus observed concentrations and weighted residuals versus time plots. The observed CPT concentrations in eleven (11) out of the fourteen (14) patients were projected by the simulation, based on model B or model C. The representative simulation profiles of randomly selected patients (0084 and 0089) from models B and C were shown in Figure 2.
Conclusion: A population PK model that could predict the PK of CZ48 and CPT in patients after dosing was developed. The reliability of the analysis results was checked by careful examinations of diagnostic plots. Incorporating a non-linear transfer between CPT and EHC compartments resulted in a significantly improved model fitting. The model could be used to predict and simulate CZ48 and CPT concentration profiles with multiple doses, offering the capability of predicting the CPT levels of individual patients for safety screening and adjusting individual patient’s dose in the clinical trial.
Yang Wang– Postdoctoral Fellow, University of Houston, Houston, Texas
Diana Chow– Professor of Pharmaceutics and Director, Institute of Drug Education and Research, University of Houston, Houston, Texas
Zhisong Cao– Webster, Texas