The Development of a Population Pharmacokinetic (PK) Model for Linezolid

Authors: Cirincione B, Phillips L, Grasela TH, Stalker D, Jungbluth GL
Conference: ICAAC
Keywords: anti-infectives, FO, infectious disease, nonlinear, nonlinearity, NONMEM V, Pharmacia, Phase 1, Phase 2, Upjohn
Division: Clinical Pharmacology and Pharmacometrics (CPP)

Abstract

Introduction: Linezoid (lzd) is an oxazolidinone antibiotic under development for infection due to gram-positive bacteria. It is rapidly and extensively absorbed following oral administration and initial PK studies have suggested non-linear elimination.

Purpose: To develop a PK model using Phase I data for application to Phase II data.

Methods: Full profile blood samples following single and multiple doses of ltz and interval trough samples were collected from healthy volunteers in an open-label, three-way crossover study evaluating oral doses; 125 mg, 375 mg, 625 mg of lzd. Samples were pooled with full profile samples from volunteers enrolled in an open-label study of 375 mb given via oral and intravenous administration. A total of 1937 lzd concentrations (cps) collected from 31 subjects were analyzed using a one compartment model with first-order absorption (Ka). Elimination was modeled as a first-order (Ke) plus a Michaelis Menton (MM) (Km, Vm) pathway with competitive inhibition from a hypothetical factor (Kf).

Results: The combined linear and MM model adequately fit the single and multiple-dose data across lzd doses. The parameter (% SEM) estimates were as follows: Mean Vd was 0.672 (2.2) L/kg, mean Ka was 4.52 (13.6) hr-1, mean Vm and Km were 38.0 (45.3) mg/hr and 466 (45.9) mg respectively, Ke was 0.0745 (9.0) hr-1, the equilibrium rate constant (Kf) for the hypothetical factor. Modest interindividual variability was noted for Ka, Km, K, V, and Kf. A mean prediction error of -4.8% indicating a trend to over predict the cps, was noted when these estimates were used to predict the lzd cps obtained in Phase II trials. Simulations of the predicted peak and trough cps over 60 days of lzd BID dosing suggest that steady-state is achieved after 3 days with no evidence of excessive accumulation.

Conclusion: A one compartment PL model with combined linear and non-linear elimination adequately describes the PLK of lzd. The non-linearity is not expected to result in excessive accumulation with dosing to SS.

Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC), San Francisco, September 1999

By B Cirincione , Luann Phillips , Thaddeus H. Grasel , D Stalker, and G Jungbluth