J. Yin1, L.B.S. Aulin1, C. Moser2, O. Ciofu2, P.H. van der Graaf1, N. Høiby2, W. Hengzhuang2, J. G. C. van Hasselt1
1. Leiden Academic Centre for Drug Research, Leiden University, Leiden, Netherlands 2. Department of Clinical Microbiology, University Hospital, Rigshospitalet, Copenhagen, Denmark
Objectives:
Chronic lung infections can be challenging to treat with conventional antibiotic dose regimens due to formation of microbial biofilms [1]. The reduced antibiotic susceptibility of biofilms is mediated by several mechanisms including decreased antibiotic exposure within the biofilm and altered drug susceptibility phenotypes. Pseudomonas aeruginosa is one important pathogen associated with chronic lung infections, which can readily transition into a biofilm phenotype. Currently, pharmacokinetic targets to derive antibiotic dose regimens are based on assays that use planktonic cultures. However, when studying optimization of biofilm-associated infections it is important to assay antibiotic activity specifically for biofilms. Several assays to quantify antibiotic activity in biofilms are available, including the alginate bead assay [2]. The alginate bead assays are based on the generation of small alginate beads, mimicking the biofilm aggregates observed in chronic biofilm-associated infections. Pharmacokinetic pharmacodynamic (PK-PD) models quantifying dynamic relationships between drug exposure and bacterial growth or kill kinetics are increasingly used to optimize treatment strategies with antibiotics. However, these modeling approaches have primarily focused on analysis of static or dynamic time kill assays of planktonic cultures exposed to antibiotics. In the current contribution we develop a PK-PD model to characterize kill kinetics in the alginate bead biofilm assay for P. aeruginosa PAO1 biofilms treated with colistin.
Methods:
Data: Planktonic bacteria (P. aeruginosa PAO1) were immobilized in spherical alginate beads (50-100 µm), as previously described [2]. The beads were exposed for 24 hours to different concentrations of colistin ranging from 1-265 mg/L, and at different time points viable bacteria were quantified in triplicate using plating.
Model development: We tested different model structures formulated as sets of ordinary differential equations (ODEs) to describe the observed kill kinetics. A nonlinear mixed effect modeling approach was used to analyze the data. Prior to analysis bacterial counts were log10-transformed. In a first step, a natural growth model was developed. For antibiotic concentration-effect relationships we consider both linear and Hill equations.
Results:
The natural growth kinetics could be described using capacity-limited growth function, with Bmax quantifying the maximum number of bacteria, estimated at 9.45 log10 CFU/mL (relative standard error, RSE 1%). The model for the full dataset including colistin exposures included two sub-populations of bacteria: drug sensitive (S) and resistant (R), where resistant indicated a reduce antibiotic susceptibility. A first-order transfer rate (Ksr) from the S to R was estimated at 0.0003 h-1 (RSE 38%) and included if drug is present. We estimated bacterial baseline concentrations (B0) at the start of the experiment at 5.86 log10 CFU/mL (RSE 2%). B0 was assumed to only consist of bacterial sub-population S. A single common growth rate (Kg) was identified for both S and R, estimated at 0.767 h-1 (RSE 4%). A common death rate was fixed to a previously reported value. We included two separate slope models to describe the drug dependent kill for S and R sub-populations, with slope terms for S and R estimated at 0.427 (RSE 16%) and 0.002 mg/L (RSE 7%), respectively.
Conclusions:
We successfully developed a full dynamical model to quantify the bacterial growth kinetics of P. aeruginosa PAO1 against colistin. The model will be used to investigate the efficacy of clinical colistin dose regimens [3].
References: [1] Ciofu O, Tolker-Nielsen T, Jensen PØ, Wang H, Høiby N. Antimicrobial resistance, respiratory tract infections and role of biofilms in lung infections in cystic fibrosis patients Adv Drug Deliv Rev. 2015 May;85:7-23. doi: 10.1016/j.addr.2014.11.017.
[2] Pedersen, S. S., Shand, G. H., Hansen, B. L. & Hansen, G. N. Induction of experimental chronic Pseudomonas aeruginosa lung infection with P . aeruginosa entrapped in alginate microspheres. 203–211 (1990).
[3] Boisson, M. et al. Comparison of Intrapulmonary and Systemic Pharmacokinetics of Colistin Methanesulfonate ( CMS ) and Colistin after Aerosol Delivery and Intravenous Administration of CMS in Critically Ill Patients. 58, 7331–7339 (2014).
Reference: PAGE 28 (2019) Abstr 8936 [www.page-meeting.org/?abstract=8936]
Poster: Drug/Disease Modelling - Infection