Background
Carbapenem-resistant Enterobacterales, such as Klebsiella pneumoniae and Escherichia coli, are considered critical priority pathogens by the World Health Organization. These pathogens produce an extensive range of β-lactamases, including New Delhi metallo-β-lactamases (NDM) which render most β-lactams inactive. The newly approved β-lactam/β-lactamase inhibitor combination aztreonam/avibactam shows promise in treating these pathogens; however, penicillin-binding protein (PBP) modifications can impact the efficacy of this combination. Understanding this impact on bacterial killing and regrowth is important for the selection of effective antibiotic dosing regimens.
Aim
To develop a mechanism-based model (MBM) which incorporates PBP and other bacterial characteristics to describe the full time-course of bacterial killing and regrowth in response to aztreonam/avibactam.
Methods
Twelve NDM-producing clinical isolates, eight E. coli isolates, harbouring four amino acid insertions to PBP3, denoted either YRIN or YRIK, and four K. pneumoniae isolates with wild-type PBP3, were studied. Each isolate was exposed for 168h to a simulated regimen of aztreonam/avibactam (1.5/0.5g 3-h infusion, q6h) in a dynamic in vitro hollow fibre infection model. Growth controls were also studied. Total viable and resistant counts were determined. An MBM was developed using S-ADAPT (version 1.57) facilitated by SADAPT-TRAN with the importance sampling algorithm (pmethod=4) to estimate all pharmacodynamic model parameters simultaneously.
Results
All controls showed the same growth kinetics. The MBM included three subpopulations and inhibition of successful replication by aztreonam. Avibactam was described as decreasing the concentration of aztreonam required to achieve 50% of maximum inhibition. Total viable and resistant subpopulation counts were simultaneously modelled across isolates, with population fits well describing the total and resistant bacterial counts for all treatments. The same drug effect parameters were used to characterise all 12 isolates. The impact of YRIN or YRIK insertion at PBP3 was described by increased concentrations of aztreonam required to achieve half-maximal inhibition, compared to isolates with wild-type PBP3. A β-lactamase function in the model reduced the estimated concentration of aztreonam in the periplasmic space for all isolates.
Conclusions
The MBM well described the effects of different PBP modifications in Enterobacterales when challenged with aztreonam/avibactam; enabling a step towards personalised medicine targeted to characteristics of the infecting pathogen.