Nonstationary Pharmacokinetics of Caspofungin in ICU Patients

Abstract

Standard dosing of caspofungin in critically ill patients has been reported to result in lower drug exposure, which can lead to subtherapeutic 24-h area under the curve to MIC (AUC0-24/MIC) ratios. The aim of the study was to investigate the population pharmacokinetics of caspofungin in a cohort of 30 intensive care unit patients with a suspected invasive fungal infection, with a large proportion of patients requiring extracorporeal therapies, including extracorporeal membrane oxygenation (ECMO) and continuous renal replacement therapy (CRRT). Caspofungin was administered as empirical 70 mg antifungal therapy administered intravenously (i.v.) on the first day and at 50 mg i.v. on the consecutive days once daily, and the concentrations were measured after three subsequent doses. Population pharmacokinetic data were analyzed by nonlinear mixed-effects modeling. The pharmacokinetics of caspofungin was described by two-compartment model. A particular drift of the individual clearance (CL) and the volume of distribution of the central compartment (V1) with time was discovered and described by including three separate typical values of CL and V1 in the final model. The typical CL values at days 1, 2, and 3 were 0.563 liters/h (6.7% relative standard error [6.7%RSE]), 0.737 liters/h (6.1%RSE), and 1.01 liters/h (9.1%RSE), respectively. The change in parameters with time was not explained by any of the recorded covariates. Increasing clearance with subsequent doses was associated with a clinically relevant decrease in caspofungin exposure (>20%). The use of ECMO, CRRT, albumin concentration, and other covariates did not significantly affect caspofungin pharmacokinetics. Additional pharmacokinetic studies are urgently required to assess the possible lack of acquiring steady-state and suboptimal concentrations of the drug in critically ill patients. (This study has been registered at ClinicalTrials.gov under identifier NCT03399032.).

Keywords: caspofungin; population pharmacokinetics.

Copyright © 2020 Borsuk-De Moor et al.

Figures

FIG 1

Caspofungin concentration-time profiles for all patients in the study.

FIG 2

Distribution of height, weight, and age of patients in the analyzed population.

FIG 3

Summary of time-dependent covariates changing over days for the study population. The boxes cover the 5th to 95th percentile range, with a horizontal line denoting the median value.

FIG 4

Relationship between individual estimates of kappa (deviation of the individual estimate from the population mean on each occasion) for clearance and central compartment volume of distribution and time. The blue line represents the linear trend in the data.

FIG 5

(A) Individual estimates for ETA (deviation of the individual estimate from the population mean) in relation to time-independent continuous covariates. The lines indicate trends in the data (red, linear function; blue, loess smooth). (B) Individual estimates for ETA in relation to median values of time-dependent continuous covariates: SOFA score, PCT concentration, dialysis dose, and ultrafiltration rate. The lines indicate trends in the data (red, linear function; blue, loess smooth). (C) Individual estimates for ETA in relation to median values of time-dependent continuous covariates: elevated lung water index, cardiac output albumin concentration, and total serum protein concentration. The lines indicate trends in the data (red, linear function; blue, loess smooth). (D) Individual estimates for ETA in relation to categorical covariates.

FIG 5

(A) Individual estimates for ETA (deviation of the individual estimate from the population mean) in relation to time-independent continuous covariates. The lines indicate trends in the data (red, linear function; blue, loess smooth). (B) Individual estimates for ETA in relation to median values of time-dependent continuous covariates: SOFA score, PCT concentration, dialysis dose, and ultrafiltration rate. The lines indicate trends in the data (red, linear function; blue, loess smooth). (C) Individual estimates for ETA in relation to median values of time-dependent continuous covariates: elevated lung water index, cardiac output albumin concentration, and total serum protein concentration. The lines indicate trends in the data (red, linear function; blue, loess smooth). (D) Individual estimates for ETA in relation to categorical covariates.

FIG 6

Prediction-corrected visual predictive check (VPC) showing the simulation-based 90% confidence intervals around the 10th, 50th, and 90th percentiles of the pharmacokinetic data in the form of turquoise (50th) and violet (10th and 90th) areas. The corresponding percentiles from the observed data are plotted in black.

FIG 7

Probabilities of target attainment (PTA) of caspofungin based on preclinical targets for C. parapsilosis, C. glabrata, and C. albicans for the three clearance values estimated for three subsequent dosing occasions (first occasion in red, second occasion in green, and third occasion in blue) with colored area around the lines representing 90% confidence intervals based on bootstrap samples.