A population pharmacokinetic model of intravenous telavancin in healthy individuals to assess tissue exposure

Abstract

Non-compartmental analysis of telavancin microdialysis data indicated a sustained exposure in soft tissues and that unbound plasma concentrations were underestimated in vitro. The objective of the present evaluation was to develop a population pharmacokinetic model of telavancin to describe its plasma protein binding, its distribution into muscle, and subcutaneous tissue and to predict pharmacokinetic/-dynamic target attainment (PTA). Total plasma concentrations and microdialysate concentrations (plasma, subcutaneous, and muscle tissue) were available up to 24 h (plasma microdialysate, up to 8 h) post-dose from eight healthy subjects after a single intravenous infusion of 10 mg/kg telavancin. Population pharmacokinetic modeling and simulations were performed using NONMEM. A two-compartment model with saturable protein binding best described plasma concentrations. Plasma unbound fractions at steady state were 23, 15, and 11% at 100, 50, and 10% of the maximum predicted concentrations respectively. Distribution into muscle and subcutaneous tissue was non-linear and described appropriately by one additional compartment each. Based on total plasma concentrations, predicted median (95% confidence interval) values of AUC/MIC (MIC 0.125 mg/L, clinical breakpoint for MRSA) at steady state were 4009 [3421–4619] with a PTA of 96 [78–100] %. The fAUC/MIC in muscle was 496 [227–1232] with a PTA of 100 [98–100] %. The %fT_>MIC was approximately 100% in plasma and interstitial space fluid of muscle and subcutaneous tissues up to an MIC of 0.25 mg/L. The model provided a new hypothesis on telavancin plasma protein binding in vivo. Proposed pharmacodynamic targets in plasma and muscle are achieved with currently approved doses of 10 mg/kg daily.