Claudia Suenderhauf (1), Mats Karlsson (2), Stephan Krähenbühl (1), and Felix Hammann (1, 3)
(1) Division of Clinical Pharmacology & Toxicology, Department of Biomedicine and Clinical Research, University and University Hospital Basel, Switzerland, (2) Department of Pharmaceutical Biosciences, Faculty of Pharmacy, Uppsala University, (3) Division of Clinical Pharmacology & Toxicology, Department of Internal Medicine, University Hospital Bern, Switzerland
Introduction
The lipopeptide antibiotic daptomycin is primarily used in the treatment of systemic infections with gram-positive bacteria, usually administered as a 30 minute infusion. It exhibits dose-linear pharmacokinetics, and is primarily eliminated unchanged via the kidneys. Efficacy correlates with the ratio of area under the curve (AUC) over minimum inhibitory concentration (MIC), and thus varies with the targeted organism’s sensitivity to the drug. An AUC:MIC > 800 is considered bacteriocidal, and a ratio of > 400 and < 800 bacteriostatic [1]. Because of this relationship, daptomycin is often subject to therapeutic drug monitoring (TDM). At the University Hospital Basel (UHBS), this is currently done by sampling 2 and 24h post-dose to calculate AUC 0-24h, Cmax, and Cmin after steady-state has been reached, a modification of Begg’s method [2].
Objectives
- -create a pharmacokinetic model for a prioriand a posterioridose optimization
- -generate dosing nomograms from simulation to guide clinicians without prompt access to a pharmacometric model
Methods
Patients
Samples were collected retrospectively from measurements made during routine daptomycin TDM from January 2014 until December 2017 at the UHBS. Available covariates included demographic data, chemistry and hematology labs, infection specific data (type, site, organism, MIC where available), and clinical outcome at time of discharge from the hospital.
Data Analysis
Population pharmacokinetic analysis was carried out using NONMEM (Version 7.4.3; Icon Development Solutions, http://www.iconplc.com, Ellicott City, MD, USA). The first order conditional estimation with eta-epsilon interaction (FOCE-I)was used throughout all runs. We selected models based on goodness-of-fit statistics, graphical analysis with visual predictive checks, and model plausibility.
Results
A total of 32 patients were enrolled, totaling 111 samples from 1-7 different occasions. None were below the limit of quantification and all samples were used in modeling. The final model was a one-compartment model with linear elimination (volume of distribution (Vd) 13.9 L (intra-individual variability (IIV): 31%) and clearance (CL) 0.48 L/h (IIV: 36%)) and a proportional residual error (0.24). These results are in agreement with previously reported daptomycin PKs (e.g. [3]). Estimated glomerular filtration rate (eGFR, Cockroft-Gault) was positively and serum albumin negatively correlated with CL. Patients on ICU had an additional 0.31 L/h in CL. We found no covariate effects on Vd.
We generated dosing nomograms by simulating concentration profiles at steady state for a broad range of doses (2-14 mg/kg) and computing AUC 0-24h for typical patients (weight 70 kg, albumin 19.8 g/L) on the wards and on ICU. Although the dose-toxicity relationship for rhabdomyolysis, a dreaded though rare adverse effect of daptomycin, remains controversial, some practitioners prefer to keep the concentration at 24h below 24 mg/L [4]. In the nomograms we indicated with a dashed line where these concentrations arose in our simulations.
Conclusions
This retrospective analysis of routine TDM at a large university hospital gave similar results for what is already known about the PK of daptomycin. Given that daptomycin is administered mostly in severe or even life-threatening situations where there may not be enough time to wait for a well-founded dose recommendation, we believe that nomograms can help make more informed decisions. They are interpretable by non-pharmacometricians, familiar to clinicians, and cheaply available. Most importantly, they can convey the practically relevant aspects of the often multidimensional relationships captured by pharmacometric models more intuitively than a set of differential equations can.
References:
[1] European Committee on Antimicrobial Susceptibility Testing Steering, C., EUCAST Technical Note on daptomycin.Clin Microbiol Infect, 2006. 12(6): p. 599-601.
[2] Begg, E.J., M.L. Barclay, and S.B. Duffull, A suggested approach to once-daily aminoglycoside dosing.Br J Clin Pharmacol, 1995. 39(6): p. 605-9.
[3] Di Paolo, A., et al., Population pharmacokinetics of daptomycin in patients affected by severe Gram-positive infections.Int J Antimicrob Agents, 2013. 42(3): p. 250-5.
[4] Bhavnani, S.M., et al., Daptomycin exposure and the probability of elevations in the creatine phosphokinase level: data from a randomized trial of patients with bacteremia and endocarditis.Clin Infect Dis, 2010. 50(12): p. 1568-74.
Reference: PAGE 28 (2019) Abstr 8908 [www.page-meeting.org/?abstract=8908]
Poster: Drug/Disease Modelling - Infection