The Role of Population PK/PD Analysis During the Implementation of a Bridging Strategy for Linezolid

Authors: Antal E, Grasela TH, Bergstrom T, Bruss J, Wong E
Conference: DIA
Keywords: anti-infectives, antibiotic, Asia, AUC:MIC ratio, dosing recommendations, drug development process, ethnic, exposure-response, infectious disease, Linezolid, logistic regression, Michaelis-Menten, multi national, NONMEM, PD, pharmacodynamic, pharmacokinetic, PK, safety, sparse sampling
Division: Cognigen

Abstract

Objectives: To describe the population PK/PD analysis process to support a bridging strategy for regulatory approval of linezolid in Japan/Asia Pacific region.

Methods: Pharmacokinetic data from Phase I healthy volunteers was used to develop a population PK model using the computer program NONMEM. This model was then used to estimate the population pharmacokinetic parameters of linezolid, and their variabilities, using 3,238 linezolid concentrations from 655 patients enrolled in Phase II trials.

NONMEM was used to obtain estimates of exposure for each patient and logistic regression analysis was performed to identify relationships between exposure and efficacy and safety outcomes. Pharmacokinetic information from Japan/Asia Pacific volunteers was used to demonstrate similarity in pharmacokinetics between Japan/Asia Pacific and US/EU volunteers. The PK/PD relationships between exposure and safety and efficacy outcomes were used to develop dosing recommendations in Japan/Asia Pacific patients.

Results: The mean weight-corrected clearance estimates were 1.14 (0.27) and 1.38 (0.52) mL/min/kg for the Japan/Asia Pacific and US/EU subjects, a difference of 20%. This difference is well within the range of clearance values observed across the population PK database for US/EU. These findings suggest that the ethnic difference in linezolid PK was not substantial. Overall, there were no clinically significant differences in the hematologic or hepatic response to similar linezolid exposure between Japan/Asia Pacific and US/EU subjects. The subsequent population PK/PD analyses of the Phase II efficacy and safety data supported the use of the 600 mg twice daily regimen for linezolid in Japan/Asia Pacific region.

Conclusions: Population PK/PD analyses using the NONMEM computer program to analyze sparse sampling data from large populations represent a powerful tool for pooling data to estimate pharmacokinetic parameters and the magnitude of variability in patient populations. This is a particularly valuable tool in the implementation of bridging strategies because it permits the pooling of large numbers of patients, each of whom has had limited sampling performed (two or three samples) in order to assess similarity of exposures and outcomes across populations and to develop dosing recommendations.

Drug Information Association (DIA); Hong Kong, China; November 2000

By Edward Antal, Thaddeus H. Grasela, Tamie Bergstrom, Jon Bruss, and Ellick Wong