Romain Aubry (1), Vidmantas Petraitis (2,3) , Ruta Petraitiene (2), Thomas J. Walsh (2,4), Lena E. Friberg (1)
(1) Department of Pharmacy, Uppsala University, Uppsala, Sweden. (2) Division of Infectious Diseases, Department of Medicine, Weill Cornell Medicine of Cornell University, New York, NY, USA. (3) Center for Discovery and Innovation, Hackensack Meridian Health, Nutley, New Jersey, USA. (4) Center for Innovative Therapeutics and Diagnostics, Richmond, VA, USA (citdx.org).
Introduction
Antimicrobial resistance is a rapidly spreading threat to global public health. Efficient drug development is required to increase the efficacy of antimicrobial agents and prevent emergence of resistance. However, translation of antimicrobial efficacy from preclinical to clinical settings is complex and challenging for drug development, given resource-demanding clinical trials and limited possibilities to study clinical symptoms in a preclinical setting (1).
Traditionally, 24h murine infection models are used to define pharmacokinetic-pharmacodynamic (PKPD) targets for antibiotics. A recently developed 14-day persistently neutropenic rabbit pneumonia model has shown promise to better replicate clinical pneumonia and characterize disease progression (2). Through PKPD-modelling, this rabbit model may improve the bridging of in vitro and in vivo studies, to the clinic, for optimization of dosing regimens, strengthening antimicrobial stewardship and improving patient outcome.
Objectives
The aim of this study was to translate the PK and PD of ceftazidime/avibactam (CAZ/AVI) for the treatment of Klebsiella pneumoniae carbapenemase (KPC) isolates, from in vitro time-kill curves to the persistently neutropenic rabbit pneumonia model using a PKPD-modelling approach.
Methods
Data from the rabbit pneumonia model, generated by Petraitiene et al. (2), were used for PK and PKPD-modelling. Three CAZ/AVI doses (60/15, 90/22.5 and 120/30 mg/kg) were infused intravenously to healthy New Zealand White rabbits as a single dose or in repeated administration (q8h). Based on the plasma samples, a PK model which included allometric scaling on parameters was built using NONMEM 7.4. The free fractions of CAZ and AVI in plasma were fixed to 85% and 92%, respectively.
The PD effect of CAZ/AVI was characterized using the previous model from Kristoffersson et al. based on in vitro time-kill curves (3). The model included a main population of bacteria, susceptible to both drugs, and a subpopulation with decreased susceptibility. Both populations consisted of two states each, a growing drug-susceptible state, and a non-growing, drug non-susceptible state. The killing effects of CAZ and AVI were described with linear (slope) models. In addition, AVI inhibited β-lactamase activity and enhanced the effect of CAZ.
The developed PK model was linked with the PD model described above to predict the lung CFU in neutropenic rabbits which were infected with two KPC clinical isolates (MIC of 1 and 2 mg/L in the presence of 4 mg/L of avibactam) and treated with a 120/30 mg/kg CAZ/AVI dose q6h for 7 or 14 days. The effect of CAZ and AVI was driven by free lung concentrations assuming penetration ratios of 24% and 22% of total plasma concentrations, respectively (4). Model predictions were compared to the lung CFU observed at 7- and 14-days. Further adjustments in bacterial growth and in antimicrobial efficacy were made to achieve in vitro/in vivo translation.
Results
The plasma PK profiles in healthy rabbits were adequately described by a two compartment PK model with interindividual variability in clearance and peripheral volume for CAZ, and in central and peripheral volumes for AVI. A rapid elimination of CAZ and AVI was observed with corresponding terminal half-lives of 0.62 h and 0.29 h, respectively. For a CAZ/AVI dosing regimen of 120/30 mg/kg q6h, unbound CAZ concentration was predicted to exceed the MIC (2 mg/L) during 77% (plasma) and 63% (lung) of the treatment interval. The percentage of time that unbound AVI concentration exceeded the concentration threshold (1 mg/L) was 34% in plasma and 17% in lung. Without adaptations of parameters, the efficacy of CAZ/AVI in rabbits was overpredicted using the PD model developed from in vitro data. The predictions improved when the killing rate constant for CAZ was re-estimated (46% reduction in the rabbit model) at 7 days, while for the 14-day data the model overpredicted the bacterial load in the lungs.
Conclusion
The time courses of CAZ and AVI plasma concentrations in rabbits were well described by two-compartment PK models. Bacterial killing over time was overpredicted by the PD model developed from in vitro data and a reduction of the CAZ effect was required to improve PD translation to rabbit model. Through further adaptations, the model will facilitate the characterization of in vitro/in vivo PKPD differences and strengthen reliable preclinical to clinical translation of antimicrobial drug effects.
References
- Friberg LE. 2021. Pivotal Role of Translation in Anti-Infective Development. Clinical Pharmacology & Therapeutics 109:856–866.
- Petraitiene R, Petraitis V, Kavaliauskas P, Maung BBW, Khan F, Naing E, Aung T, Zigmantaite V, Grigaleviciute R, Kucinskas A, Stakauskas R, Georgiades BN, Mazur CA, Hayden JA, Satlin MJ, Walsh TJ. 2020. Pharmacokinetics and Efficacy of Ceftazidime-Avibactam in the Treatment of Experimental Pneumonia Caused by Klebsiella pneumoniae Carbapenemase-Producing K. pneumoniae in Persistently Neutropenic Rabbits. Antimicrob Agents Chemother 64:e02157-19.
- Kristoffersson AN, Bissantz C, Okujava R, Haldimann A, Walter I, Shi T, Zampaloni C, Nielsen EI. 2020. A novel mechanism-based pharmacokinetic-pharmacodynamic (PKPD) model describing ceftazidime/avibactam efficacy against β-lactamase-producing Gram-negative bacteria. J Antimicrob Chemother 75:400–408.
- Berkhout J, Melchers MJ, van Mil AC, Seyedmousavi S, Lagarde CM, Nichols WW, Mouton JW. 2015. Pharmacokinetics and Penetration of Ceftazidime and Avibactam into Epithelial Lining Fluid in Thigh- and Lung-Infected Mice. Antimicrobial Agents and Chemotherapy 59:2299–2304.
Reference: PAGE 32 (2024) Abstr 10951 [www.page-meeting.org/?abstract=10951]
Poster: Drug/Disease Modelling - Infection