DISCUSSION
The PK/PD parameters of β-lactam/β-lactamase inhibitor combos are challenging to determine. This is because some β-lactamase inhibitors have antibacterial activity in addition to protecting β-lactams from β-lactamases, and the antimicrobial effects of various combinations are interdependent. There is a need to establish accurate analytical methods to determine the PK/PD parameters of β-lactam/β-lactamase inhibitor combinations because inappropriate analytical methods or models with low predictive accuracy risks creating a dissociation between susceptibility data in vitro and actual clinical efficacy. The current research shows how the MICi concept may be used in conventional PK/PD analysis to produce PK/PD parameters that are more practical, versatile, and accurate than those produced usingf T>CT.
The first generation β-lactamase inhibitor combination piperacillin/tazobactam has been tested for the MIC of piperacillin in a fixed concentration of tazobactam of 4 mg L-1, at a dose ratio of 8:1. On the other hand, amoxicillin/clavulanate is tested for susceptibility at a fixed 2:1 ratio, although several formulations with ratios of 2:1, 4:1 and 7:1 are used and may not correlate with efficacy. This in vitro and in vivo difference in combination ratios may be due to the fluctuating susceptibility of β-lactam drugs in response to changing inhibitor concentrations over time is ignored during in vivo analysis. Optimisation off T>CT is based on the assumption that the susceptibility of β-lactam drugs is constant in the presence of β-lactamase inhibitors. However, MICs for clinical strains are not incorporated into the dosing design and do not adequately reflect changes in β-lactam drug susceptibility caused by different β-lactamase inhibitor concentrations.
In the present study, the MIC curve for aztreonam varied significantly in the therapeutic concentration range of nacubactam (approximately 0.1-10 mg L-1) (Fig. 2). It confirmed the importance of the PK/PD parameters for β-lactam drugs/β-lactamase inhibitors should consider changes in susceptibility under different β-lactamase inhibitor concentration conditions. On the other hand,f T>MICi can reflect changes in β-lactam susceptibility and changes in β-lactamase inhibitor concentration. In order to compare the accuracy of these two PK/PD parameters, an attempt was made to predict clinical efficacy in humans using each analysis method. The results showed that for K. pneumoniae ATCC BAA-2473 and K. pneumoniae MSC 21444, the bactericidal effect simulated byf T>CT tended to be estimated lower than f T>MICi (orange area of 2 log10-kill in Tables 2 and 3). This may be due to underestimating the PK/PD predictions based onf T>CT. Higher aztreonam concentrations are clinically effective even when nacubactam concentrations are below the CT value. Therefore,f T>MICi is considered a more accurate PK/PD parameter than f T>CT. However, the f T>MICi approach requires time-consuming susceptibility testing andE max modelling analysis.
It is worth noting that analyses based onf T>MICi have practical versatility because they do not need to consider all factors. The MIC curve reflects enzyme type, inhibitory activity, antimicrobial activity, and the additive and synergistic effects of antimicrobial agents. Indeed, the analysis using f T>MICi showed a good correlation (R 2=0.868) for three β-lactamase enzyme-producing strains of different genotypes, confirming the in vivo validity of the f T>MICi-based analysis. On the other hand, a similar analysis usingf T>CT confirmed that the CT values of nacubactam depended on the strain or the dose of aztreonam used in combination (Fig. 7). These results suggest that PK/PD analysis using f T>CTlacks practical versatility in determining a single target value for various clinical isolates.
The novel DBO-based β-lactamase inhibitor, nacubactam, is a potent inhibitor of Ambler class A (e.g. KPC and ESBL) and C (AmpC) β-lactamases but weakly inhibits class D (OXA) β-lactamases (Morinaka et al., 2015),(Morinaka et al., 2016). In addition, nacubactam exhibits antibacterial activity through binding to the bacterial penicillin-binding protein (PBP)-2. Furthermore, unlike first-generation β-lactamase inhibitors (e.g. sulbactam, clavulanic acid, tazobactam), second-generation DBO-based β-lactamase inhibitors, including nacubactam, do not have a β-lactam structure and are not subject to β-lactamase degradation. Therefore, they are expected to be a new treatment option for carbapenemase (Metallo-β-lactamase)-producing bacteria, including Class B (NDM and IMP) (Livermore et al., 2015),(Mushtaq et al., 2018), because no other established treatment is available. In the current checkerboard MIC study, aztreonam/nacubactam showed good combination efficacy against Class B and Class D β-lactamase-producing bacteria such as NDM-1, IMP-6, and OXA-48 β-lactamase producing K. pneumoniae (Table 1). Furthermore, the MIC of nacubactam against all bacterial species in the presence of 4 mg L-1 aztreonam, which reflects the minimum nacubactam concentration to achieve the microbiological breakpoint (Clinical and Laboratory Standards Institute, 2021), can be easily achieved in clinical practice, indicating aztreonam/nacubactam in vivo potent antimicrobial effect is obtainable. Indeed, the following two lines of evidence in this study support the potent β-lactamase inhibition and antibacterial activity of nacubactam against β-lactamase-producingK. pneumoniae . Firstly, aztreonam/nacubactam combination therapy was effective in a mouse model against NDM-1 and IMP-6, OXA-48-producingK. pneumoniae in a nacubactam dose-dependent manner, despite aztreonam monotherapy being ineffective. Secondly, the dose fractionation study showed a concentration-dependent bactericidal effect of nacubactam irrespective of the number of doses administered. Thus, the present study provides the first evidence of a satisfactory therapeutic effect of nacubactam against refractory β-lactamase-producing K. pneumoniae , particularly the in vivo antibacterial effect of nacubactam against NDM-producing bacteria.
There are two drawbacks to this research. The first is that all PD trials only measured activity against fixed inoculum levels (approximately 106 CFU mL-1). The effect of greater inoculum levels on aztreonam/nacubactam drug efficacy is unknown. Due to the inoculum effect, several antimicrobials have lower antimicrobial effectiveness at higher initial inoculum levels (Brook, 1989). Some β-lactam/β-lactamase inhibitor combinations, on the other hand, have been less influenced by the inoculum effect. Ceftazidime/tazobactam, for example, showed decreased action at high bacterial levels, but ceftazidime/avibactam was said to be unaffected by inoculum levels (Tam et al., 2021). The causes for this are unknown. However, the inoculum effect could be related to avibactam’s reversible inhibitory action and resistance to enzymatic hydrolysis. Because nacubactam inhibits in the same way that avibactam does, it may be less vulnerable to the inoculum effect. On this point, more evidence is required. Finally, clinical target values were predicted using PK data from healthy adults. Although it would be ideal for simulating more clinically relevant settings using PK data from patients with infections, PK data for nacubactam in patients with infections is not yet publically available. Nacubactam PK data in patients with infections are awaited.