Wisse van Os, Markus Zeitlinger
Department of Clinical Pharmacology, Medical University of Vienna, Austria
Objectives: Lefamulin is the first pleuromutilin antibiotic for systemic application in humans, and currently only approved for bacterial community-acquired pneumonia (CAP). However, clinical development for acute bacterial skin and skin structure infections (ABSSSIs) is considered[1,2]. We aimed to evaluate whether the currently approved dosing regimen may be appropriate for treatment of soft tissue infections, using population pharmacokinetic (PK) modelling, probability of target attainment (PTA) analyses and in vitro surveillance data.
Methods: The PK data was obtained in a previously published Phase 1 trial in which 12 healthy males received a single dose of 150 mg lefamulin as intravenous infusion over 1 h[3]. PK samples were obtained over 24 h. Total plasma lefamulin concentrations were quantified, and microdialysis was used to measure unbound concentrations in the interstitial space fluid (ISF) of subcutaneous adipose and skeletal muscle tissue.
A total of 120 microdialysis samples from subcutaneous adipose tissue and 130 from skeletal muscle tissue were included, and 144 plasma PK observations. Observations below the quantification limit, comprising 2.2% of the original dataset, were excluded. PK data was analysed using nonlinear mixed-effects modelling. A previously established saturable protein binding model for lefamulin was used to convert total to unbound drug plasma concentrations[4]. Microdialysis data was modelled using an integrated dialysate-based approach, which considers the collection interval of microdialysate and does not require correcting for relative recovery of the microdialysis probes upfront[5]. To describe tissue distribution we evaluated models in which ISF concentrations are assigned and scaled to compartments of the plasma base model and models in which separate tissue compartments are estimated. Inter-individual variability (IIV) was evaluated on all structural model parameters.
Monte Carlo simulations were performed to calculate PTA across a range of minimum inhibitory concentrations (MIC). An unbound drug area under the curve (fAUC) to MIC ratio of 14, required for bacteriostasis in a neutropenic murine thigh infection model, was used as exposure target[6]. By weighting the PTA over the EUCAST MIC distributions[7], the cumulative fraction of response (CFR) was calculated for Staphylococcus aureus and Streptococcus pneumoniae.
Results: In accordance with other population PK models for lefamulin, plasma PK data was best described by a three-compartment model[4,8]. Tissue concentrations were best described with one extra compartment each. To stabilise model predictions, the volumes of distribution of these compartments were fixed to their physiological values[9]. A power function on drug concentrations in the central compartment was needed to capture the nonlinear distribution of drug into the tissues. Inclusion of IIV was supported on clearance, volume of distribution of the first peripheral compartment, all intercompartmental clearances, and the relative recovery parameters for microdialysate observations. Residual variability of the plasma, microdialysis and retrodialysis data was described separately using proportional error models.
For the currently approved dosing regimen of 150 mg twice daily as intravenous infusion over 1 h, and using the conventional cut-off point of 90% PTA, our simulations suggest a susceptibility breakpoint for soft tissue infections of 0.06 mg/L, regardless of whether unbound concentrations in plasma or ISF are used as exposure measure. This breakpoint splits the MIC distribution of both S. aureus and S. pneumonia. Correspondingly, CFR values for S. aureus were 86%, 61%, and 62% when using unbound drug in plasma, subcutaneous adipose tissue and skeletal muscle tissue as exposure measure, respectively. For S. pneumoniae the CFR was lower still: 71%, 45% and 46% for plasma, subcutaneous adipose tissue and skeletal muscle tissue, respectively. Simulating a doubling of the dose increased the CFR to ≥94% and ≥80% in all matrices for S. aureus and S. pneumoniae, but this dosing strategy would have to be supported by safety data.
Conclusions: We developed a model describing tissue exposure of lefamulin. Our PTA analyses suggest that drug exposure following the currently approved dosing regimen for CAP may not be optimal against pathogens commonly causing ABSSSIs.
References:
[1] Prince WT et al. (2013). Antimicrob. Agents Chemother. 57(5);2087-94
[2] Nabriva Therapeutics; https://www.nabriva.com/pipeline-research
[3] Zeitlinger M et al. (2016). J. Antimicrob. Chemother. 71(4);1022-26
[4] Rubino CM et al. (2015). Antimicrob. Agents. Chemother. 59(1);282-8
[5] Tunblad K et al. (2004). Pharm. Res. 21(9);1698-707
[6] Wicha WW et al. (2019). J. Antimicrob. Chemother. 74(S3);5-10
[7] EUCAST; https://www.eucast.org/mic_distributions_and_ecoffs/
[8] Zhang L et al. (2019). J. Antimicrob. Chemother. 74(S3);27-34
[9] Shah DK et al. (2012). J. Pharmacokinet. Pharmacodynam. 39(1);67-86
Reference: PAGE 30 (2022) Abstr 10149 [www.page-meeting.org/?abstract=10149]
Poster: Drug/Disease Modelling - Infection