Population cell life span models for effects of drugs following indirect mechanisms of action.

Pharmacokinetic/pharmacodynamic (PK/PD) models for hematological drug effects exist that assume that cells are produced by a zero- or first-order process, survive for a specific duration (cell lifespan), and then are lost. Due to the fact that delay differential equations (DDE) are needed for cell lifespan models, their software implementation is not straightforward. Our objective is to demonstrate methods to implement three different cell lifespan models for dealing with hematological drug effects and to evaluate the performance of NONMEM to estimate the model parameters. For the basic lifespan indirect response (LIDR) model, cells are produced by a zero-order process and removed due to senescence. The modified LIDR model adds a precursor pool. The LIDR model of cytotoxicity assumes a three-pool indirect model to account for the cell proliferation with capacity-limited cytotoxicity followed by maturation, and removal from the circulation. A numerical method (method of steps) implementing DDE in NONMEM was introduced. Simulation followed by estimation was used to evaluate NONMEM performance and the impact of the minimization algorithm (first-order method vs. first-order conditional estimation method) and the model for residual variability on the estimates of the population parameters. The FOCE method combined with log-transformation of data was found to be superior. This report provides methodology that will assist in application of population methods for assessing hematological responses to various types of drugs.

Single-dose pharmacokinetics of sodium ferric gluconate complex in iron-deficient subjects.

STUDY OBJECTIVES: To determine the single-dose pharmacokinetics of intravenous sodium ferric gluconate complex in sucrose injection (SFGC) in iron-deficient human volunteers, and to assess iron transport. DESIGN: Open-label, randomized study. SETTING: Clinical research facility. SUBJECTS: Fourteen iron-deficient men and women. INTERVENTIONS: Subjects were randomized to receive a single intravenous dose of either SFGC 62.5 mg administered over 30 minutes or SFGC 125 mg over 60 minutes. Five days later, the same subjects were rerandomized to receive a second intravenous dose of SFGC, either 62.5 mg administered over 4 minutes or 125 mg over 7 minutes. MEASUREMENTS AND MAIN RESULTS: Blood samples were collected at predefined times before, during, and up to 72 hours after the infusion to determine the single-dose pharmacokinetics of SFGC. Assays were performed for both total iron and transferrin-bound iron, from which drug-bound iron could be calculated. Urine was collected over 24 hours before dosing and for 24 hours after the start of infusion to determine the renal elimination of iron. Clearance of SFGC from serum was rapid and far exceeded rates reported for iron dextran. Pharmacokinetic parameters were unaffected by dose or infusion rate. Serum iron derived from SFGC did not exceed the binding capacity of transferrin. Serum iron from SFGC became rapidly available (< 24 hrs) as transferrin-bound iron, but only after passage through another compartment, presumably the reticuloendothelial system (RES). At least 80% of the administered iron was transported to bone marrow within 24 hours after infusion. CONCLUSIONS: Iron derived from SFGC appears to be rapidly transferred to a bioavailable iron compartment as transferrin-bound iron after digestion in the RES. At the doses administered in this study, liberation of potentially toxic, free iron was not detectable.

Population pharmacokinetic and pharmacodynamic modeling of etanercept using logistic regression analysis.

OBJECTIVE: Our objective was to develop a population pharmacokinetic and pharmacodynamic model of etanercept in patients with rheumatoid arthritis, with the American College of Rheumatology response criterion of 20% improvement (ACR20) used as a binary clinical outcome variable. METHODS: Concentration-time profiles from 25 subjects, administered 25 mg subcutaneous etanercept twice weekly for 24 weeks, were pooled with data from 77 subjects, enrolled in a 24-week, randomized, double-blind study comparing 25 mg and 50 mg subcutaneous etanercept twice weekly. The cumulative area under the concentration-time curve (AUC) was used as the exposure variable, and ACR20 was the binomial clinical outcome. ACR20 data from another 80 placebo-treated patients enrolled in a randomized, double-blind phase III study were used to describe the placebo time course of ACR20. A logistic regression analysis with NONMEM was applied to describe the exposure-response relationship, and the 95% confidence intervals (95% CIs) were constructed by bootstrapping 1000 times. RESULTS: The population mean apparent clearance was 0.117 L/h (95% CI, 0.108-0.130 L/h) for white female patients and 0.138 L/h (95% CI, 0.118-0.163 L/h) for white male patients. Interindividual variability and interoccasion variability were 41.1% and 27.6%, respectively. The mean absorption half-life was 20.9 hours, and the elimination half-life was 95.4 hours. An improved response profile in male patients was shown, but the multiplicative factor between slope on cumulative AUC between male and female patients was not statistically significant (1.69; 95% CI, 0.37-9.99). The model-predicted percentage of patients achieving ACR20 at 6 months after dosing of 25 mg subcutaneously twice weekly was 54.9%, comparable to the observed 52.9%. CONCLUSION: The population pharmacokinetic analysis confirmed that etanercept is slowly absorbed and eliminated after subcutaneous administration. The logistic model linking cumulative AUC with ACR20 adequately characterized the time course of clinical improvement in patients with rheumatoid arthritis receiving etanercept.