In vitro pharmacokinetic/pharmacodynamic interaction model assessing the synergy between meropenem and fosfomycin in a clinical multi-resistant K. pneumoniae isolate.
A. Farooq (1), N. Kroemer (1), J.W. Decousser (2), S.G. Wicha (1)
(1) Dept. of Clinical Pharmacy, Institute of Pharmacy, University of Hamburg, Hamburg, Germany, (2) DYNAMYC Team – EA 7380, FACULTE DE SANTE, Université Paris-Est-Créteil Val-De-Marne, France
Objectives: Antimicrobial combination therapy offers an option for increasing efficacy and prevention against the emergence of resistance. Synergy between meropenem (MER) and fosfomycin (FOF) has been described, but results are often based on checkerboards with the insensitive turbidity criterion without precisely informing about the pharmacodynamic (PD) interactions. Comprehensive understanding of PD drug interactions can be achieved by applying semi-mechanistic PKPD models. Here we describe in vitro kill data from static Time-Kill-Experiments (TKE) and quantification of the synergistic PD interaction between MER and FOF using the general pharmacodynamic interaction (GPDI) model [1].
Methods: A clinical multi-resistant K. pneumoniae strain was evaluated against MER, FOF and MER-FOF. The in vitro evaluation included MIC determinations and static TKE over 30 h. Concentrations tested ranged from 0.5 - 512 mg/L and 0.5 – 1024 mg/L for MER and FOF, respectively. Nonlinear-mixed-effects modeling was conducted using NONMEM® 7.5. The PK of MER was corrected by implementing a degradation rate [2]. Antimicrobial effects of MER and FOF were described using slope or maximum effect (Emax) models. Combined effects were quantified using the GPDI model implemented in Bliss Independence. The GPDI model quantified PD interactions between MER and FOF as multidirectional shifts of the potency (EC50) or maximal effect (Emax) assigning the role of perpetrator or victim to each drug. The GPDI term is a function of the perpetrator drug concentration, which affects a PD parameter (either EC50 or Emax) of the victim drug. Accordingly, the PD interactions were described as monodirectional or bidirectional shifts of either EC50 or Emax of MER and/or FOF. The best model was chosen by the Akaike information criterion (AIC) [3] and inspection of the graphical model fit.
Results: In TKE the tested K. pneumoniae strain showed double resistance to both MER alone and FOF alone, and regrowth after initial killing was observed after 24 hours at 128 mg/L for MER and 1024 mg/L for FOF, respectively. Nevertheless, when combining MER-FOF at different concentrations (MER-FOF: 32 mg/L – 64 mg/L, 64 mg/L – 32 mg/L) substantial killing without regrowth was observed. The final PD model consisted of two bacterial compartments representing antibiotic-susceptible bacteria (S) and double-resistant bacteria (R), respectively. The effect of MER on S was described using a slope model while the effect of FOF on S and MER and FOF on R was best described using the Emax model. A monodirectional PD interaction model with MER as perpetrator drug significantly lowering the EC50 of FOF on R from 410 mg/L to 7.38 mg/L outperformed the opposite directionality model with FOF being the perpetrator and MER the victim drug (AIC: 1697.0 vs. 2157.3). Bidirectional PD interactions with both drugs being perpetrator at the same time were not supported by the data. When trying to quantify the combined effect as additivity using BI without interaction, combined effects were substantially underpredicted. The developed semi-mechanistic PKPD model described both the mono and combination data from low to high concentrations well, while also identifying the interaction-type (EC50 or Emax) and its’ directionality.
Conclusions: MER-FOF showed highly synergistic interactions against a double resistant strain thereby potentially re-sensitizing the strain to this combination of antibiotics. Our developed PKPD model successfully assessed and quantified the synergy between MER and FOF and its directionality using the GPDI model implemented in a longitudinal PKPD model. Simulation studies with different dosing regimen resulting in dynamic concentrations of both drugs will be performed to guide the translation of static TKE into dynamic hollow-fiber-infection-model experiments and eventually their in vivo evaluation.
References:
[1] Wicha, S. G., Chen, C., Clewe, O. & Simonsson, U. S. H. A general pharmacodynamic interaction model identifies perpetrators and victims in drug interactions. Nat. Commun. 8, 1–11 (2017).
[2] Wicha, S. G.,C. Kloft, Simultaneous determination and stability studies of linezolid, meropenem and vancomycin in bacterial growth medium by high-performance liquid chromatography. Journal of Chromatography B. 1028, 242-248 (2016).
[3] H. Akaike, “A New Look at the Statistical Model Identification,” IEEE Trans. Automat. Contr., vol. 19, no. 6, pp. 716–723 (1974).