Antimicrobial Activities of Aztreonam-Avibactam and Comparator Agents against Enterobacterales Analyzed by ICU and Non-ICU Wards, Infection Sources, and Geographic Regions: ATLAS Program 2016–2020

Increasing antimicrobial resistance among multidrug-resistant (MDR), extended-spectrum β-lactamase (ESBL)- and carbapenemase-producing Enterobacterales (CPE), in particular metallo-β-lactamase (MBL)-positive strains, has led to limited treatment options in these isolates. This study evaluated the activity of aztreonam-avibactam (ATM-AVI) and comparator antimicrobials against Enterobacterales isolates and key resistance phenotypes stratified by wards, infection sources and geographic regions as part of the ATLAS program between 2016 and 2020. Minimum inhibitory concentrations (MICs) were determined per Clinical and Laboratory Standards Institute (CLSI) guidelines. The susceptibility of antimicrobials were interpreted using CLSI and European Committee on Antimicrobial Susceptibility Testing (EUCAST) breakpoints. A tentative pharmacokinetic/pharmacodynamic breakpoint of 8 µg/mL was considered for ATM-AVI activity. ATM-AVI inhibited ≥99.2% of Enterobacterales isolates across wards and ≥99.7% isolates across infection sources globally and in all regions at ≤8 µg/mL. For resistance phenotypes, ATM-AVI demonstrated sustained activity across wards and infection sources by inhibiting ≥98.5% and ≥99.1% of multidrug-resistant (MDR) isolates, ≥98.6% and ≥99.1% of ESBL-positive isolates, ≥96.8% and ≥90.9% of carbapenem-resistant (CR) isolates, and ≥96.8% and ≥97.4% of MBL-positive isolates, respectively, at ≤8 µg/mL globally and across regions. Overall, our study demonstrated that ATM-AVI represents an important therapeutic option for infections caused by Enterobacterales, including key resistance phenotypes across different wards and infection sources.


Introduction
Over the past few decades, the emergence of antimicrobial resistance in Enterobacterales (e.g., Escherichia coli, Klebsiella spp., Enterobacter spp., Proteus spp., Serratia marcescens, and Citrobacter spp.) has been a major global threat [1].Antimicrobial resistance in Enterobacterales occurs through a wide range of mechanisms, and the treatment of infections caused by multidrug-resistant (MDR) Enterobacterales, including extended-spectrum βlactamases (ESBL)-positive and carbapenem-resistant Enterobacterales (CRE), is highly challenging [2][3][4][5].ESBL production in Enterobacterales is associated with resistance to third-generation cephalosporins, leading to increased mortality, length of stay, and costs [6].CRE infections are associated with substantial healthcare burdens across both nosocomial and community settings [7,8].World Health Organization (WHO) has categorized third-generation cephalosporin-resistant Enterobacterales and CRE as critical pathogens ('priority 1') with urgent need for the development of new and effective drugs [9].
Avibactam, a non-β-lactam β-lactamase inhibitor, is capable of inhibiting Ambler class A and class C β-lactamases, and some class D carbapenemases [35].Phase 3 trials of avibactam, in combination with aztreonam, to treat infections caused by MDR Gramnegative isolates including those expressing MBLs along with one or more additional β-lactamases have been recently completed (NCT03580044 and NCT03329092, results not published yet) [36,37].The majority of previous surveillance studies on in vitro activity of aztreonam-avibactam (ATM-AVI) and comparator antimicrobial agents across regions in isolates from patients with respiratory tract infections (RTI), urinary tract infections (UTI), skin and soft tissue infections (SSTI), bloodstream infections (BSI), and intra-abdominal infections (IAI) have monitored their activity without the stratification of the infection sources [38][39][40][41].Only two studies have assessed the resistance profile of these antimicrobial agents stratified by infection sources [3,42].Importantly, these studies were either focused on specific geographic regions or pathogens or conducted over shorter study periods [3,42].In the current study, we performed a detailed analysis of a large contemporary collection of clinical Enterobacterales isolates collected between 2016 and 2020 from the Antimicrobial Testing Leadership and Surveillance (ATLAS) program, a global surveillance database [43].These isolates were collected from patients with RTI, UTI, SSTI, BSI, and IAI across Africa-Middle East (AfME), Asia-Pacific (APAC), Europe, Latin America (LATAM), and North America.This study characterized the activity of ATM-AVI and comparator antimicrobials against these isolates, including key resistant phenotypes prevalent among Enterobacterales stratified by wards (ICU and non-ICU), infection sources, and geographic regions.
Among MBL-positive isolates, ≥96.8% of isolates collected from wards (Tables 4 and S10) and ≥97.4% of isolates collected from infection sources (Tables 5 and S11) were inhibited by ATM-AVI at ≤8 µg/mL globally and across regions (data limited to a small number of isolates in AfME from IAI sources (n = 15) and North America from both wards and infection sources (n ≤ 140).Among comparator agents, applying the CLSI and EUCAST breakpoints, only colistin (available only per EUCAST) and tigecycline (except from RTI from Europe (per EUCAST)) demonstrated high susceptibility across wards (80.0-100%,Tables 4 and S10) and infection sources (80.0-100%,Tables 5 and S11) globally and in all regions (data limited to small number of isolates in AfME from IAI sources (n = 12) and North America from both wards and infection sources (n ≤ 7)).

Discussion
This study evaluated the in vitro antimicrobial susceptibilities of ATM-AVI and a panel of comparator agents against Enterobacterales isolates collected globally, across AfME, APAC, Europe, LATAM, and North America, from ICU and non-ICU wards and RTI, UTI, SSTI, BSI, and IAI infection sources between 2016 and 2020.The highest number of Enterobacterales (58.5-69.3%)and resistant isolates, including MDR, ESBL-positive, CRE, and MBL-positive isolates (47.5-70.0%),were collected from non-ICU wards globally and across regions.Among infection sources, the highest number of Enterobacterales (23.4%) and resistant phenotypes, including MDR, ESBL-positive, and MBL-positive isolates (25.1-26.0%),were collected from UTI sources globally, except CRE, for which the majority of isolates were from RTI sources (27.2%).Across regions, variability was observed for the highest number of isolates collected from different infection sources.Overall, ATM-AVI exhibited potent activity against Enterobacterales (MIC 90 0.12-0.5 µg/mL, ≥99.2 inhibited at ≤8 µg/mL) and all resistant phenotypes (MIC 90 0.25-4 mg/L, ≥98.5 inhibited at ≤8 µg/mL) from all wards and infection sources globally and across regions.Among comparator agents, amikacin, colistin, ceftazidime-avibactam, meropenem, and tigecycline were mostly active against all Enterobacterales (83.4-99.9%)and resistant phenotypes (79.5-99.9%),except for CRE and MBL-positive isolates, for which only colistin (except for MBL-positive isolates from Europe and LATAM) and tigecycline (80.0-100.0%)were notably active across wards and infection sources globally and in all regions.Altogether, these findings highlight the role of avibactam in potentiating the activity of aztreonam against Enterobacterales overall, including resistant phenotypes such as MDR, ESBL-positive, CRE, and MBL-positive isolates.
ATM-AVI inhibited ≥99.9% Enterobacterales isolates at ≤8 µg/mL across all infection sources globally, with MIC 90 value of 0.12 µg/mL observed for UTI, SSTI, BSI, and IAI sources and 0.25 µg/mL for RTI sources.Similar in vitro ATM-AVI activity was also observed irrespective of region, in which ≥99.7% Enterobacterales isolates from these infection sources were inhibited by ATM-AVI at ≤8 µg/mL (MIC 90 of 0.12-0.25 µg/mL) in AfME, APAC, Europe, LATAM, and North America.This is in agreement with a study from the SENTRY database between 2019 and 2020 on Enterobacterales isolates collected from Europe, which showed that ≥99.6% isolates were inhibited by ATM-AVI irrespective of infection sources [3].Overall, these findings suggest that ATM-AVI is potent against Enterobacterales isolates from various infection sources and its activity has been maintained over the years.Among the panel of comparator agents, amikacin, ceftazidime-avibactam, colistin, imipenem, meropenem, and tigecycline demonstrated high rates of susceptibility against Enterobacterales isolates irrespective of infection (79.6-99.9%)sources globally and across most of the regions, a pattern similar to that of isolates collected from wards.In contrast, the SENTRY study from Europe reported lower susceptibility rates for colistin (73.4-88.3%)and tigecycline (53.2-72.7%)compared to our study across infection sources [3].
Of note, our study identified 0.15% isolates from ICU wards and 0.07% from non-ICU wards with MIC > 8 µg/mL for ATM-AVI.Similarly, 0.11% isolates from RTI sources, 0.09% from UTI sources, 0.09% from SSTI sources, 0.08% from BSI sources, and 0.05% from IAI sources were also observed to have ATM-AVI MIC > 8 µg/mL.Potential resistance mechanisms responsible for such elevated MICs for ATM-AVI have been evaluated previously; a study from the INFORM database (2012-2017) in clinical isolates of Enterobacterales and a recent study assessing the reduced activity in clinical E. coli isolates suggested specific amino acid insertions in the penicillin-binding protein 3 (PBP3) sequence and an elevated expression of PER-type, VEB-type, and CMY-42 β-lactamases as potential resistance mechanisms to ATM-AVI [45,46], while other resistance mechanisms contributing to the reduction in ATM-AVI activity remain undefined and warrant further investigation [46].These emerging resistance mechanisms could affect the therapeutic potential of ATM-AVI and thus require continuous surveillance efforts.
Our study has a few limitations.A predefined number of isolates are collected for each species as part of the ATLAS program, hence the results of this study cannot be interpreted as prevalence or used for epidemiological data.The low number of samples for some resistant phenotypes and regions in this study should be taken into consideration while interpreting the findings.Agents such as meropenem-vaborbactam (approved in 2017) and ceftolozane-tazobactam (approved in 2014) were included in the antimicrobial panel of ATLAS in 2020 and hence have not been included in this study.Additionally, cefiderocol (approved in 2019) has not been added to the antimicrobial panel of ATLAS and could not be included in this study.Moreover, the current study does not ascertain the potential mechanisms of resistance in isolates with MIC > 8 µg/mL for ATM-AVI due to a lack of whole genome sequencing for isolates in the ATLAS program.Due to the unavailability of relevant data and the lack of granular information for ATM-AVI post 2020 in the ATLAS platform, we are unable to assess the potential impact of COVID-19 on the isolate distribution and susceptibility data included in this study.
In conclusion, the results from this study demonstrated potent antimicrobial activity of ATM-AVI against Enterobacterales from all wards and infection sources globally and across regions.Furthermore, sustained activity was observed against resistant phenotypes, including MDR, ESBL-positive, CRE, and MBL-positive isolates, from all wards and infection sources globally and in all regions.The results of this large and comprehensive surveillance analysis further support the clinical development of ATM-AVI for treatment of Enterobacterales infections, including those caused by resistant phenotypes, such as CRE and MBL-positive strains.Additionally, the results from this study emphasize that ATM-AVI may be an important addition to the limited therapeutic options available against such isolates.

Antimicrobial Susceptibility Testing
Minimum inhibitory concentrations (MICs) were determined by IHMA using the reference broth microdilution methodology per the Clinical and Laboratory Standards Institute (CLSI) guidelines [52].ATM-AVI and a panel of comparator antimicrobial agents including amikacin, aztreonam, cefepime, ceftazidime, ceftazidime-avibactam, ceftriaxone, ciprofloxacin, colistin, gentamicin, imipenem, levofloxacin, meropenem, piperacillin-tazobactam, and tigecycline were used for antimicrobial susceptibility testing.The susceptibility of antimicrobial agents were interpreted using CLSI M100 (33rd ed.) and European Committee on Antimicrobial Susceptibility Testing (EUCAST, version 13.0) breakpoints, wherever applicable [52,53].ATM-AVI was tested with avibactam at a fixed concentration of 4 mg/L.A provisional pharmacokinetic/pharmacodynamic susceptible breakpoint of ≤8 mg/L was used for ATM-AVI for comparison owing to lack of approved clinical breakpoints [35,41,44].For colistin, since the susceptible breakpoints per CLSI are not available, only EUCAST data were reported.Isolates of Morganella morganii, Proteus spp., Providencia spp., and Serratia marcescens were excluded from the analysis of colistin data because of their intrinsic resistance [54].For tigecycline, MICs were interpreted using FDA-approved breakpoints due to lack of MIC interpretative criteria per CLSI.For the analysis of tigecycline data, isolates of Morganella morganii, Proteus spp., and Providencia spp.were excluded due to intrinsic resistance [55].All antimicrobials were not tested in each year of the surveillance, hence varying numbers of isolates were recorded against the different antimicrobials.
All the data were collected and presented as percentage of susceptible (%S) isolates and MIC 90 based on CLSI and EUCAST guidelines for all identified organisms.No statistical analysis was performed as part of this study.In this study, susceptibility ≥80.0% was categorized as high.

Resistance Phenotypes Definitions
In this study, the CRE (CLSI/EUCAST) phenotype was defined as isolates of any Enterobacterales species resistant to meropenem per the ATLAS program.
Isolates of Enterobacterales with MICs to meropenem of ≥2 mg/L and/or a ceftazidimeavibactam MIC ≥ 16 mg/L and/or ATM-AVI MIC ≥ 16 mg/L were screened for the presence of extended-spectrum β-lactamases (ESBL) genes-bla SHV , bla TEM , bla CTX-M , bla VEB , bla PER , and bla GES -using multiplex PCR assays followed by full-gene DNA sequencing as previously described.Additionally, isolates of E. coli, K. pneumoniae, K. oxytoca, and P. mirabilis with a ceftazidime or aztreonam MIC ≥ 2 mg/L qualified for the above screening.
In this study, isolates carrying class B MBL genes-bla NDM , bla IMP , and bla VIM -(as data were available only for these three subsets on the ATLAS database) were identified by PCR and sequencing.

Table 1 .
Distribution of Enterobacterales isolates collected globally and across different regions stratified by wards and infection sources, 2016-2020.

Table 2 .
In vitro activity of ATM-AVI and comparator agents tested against Enterobacterales isolates across regions stratified by wards from 2016 to 2020.

Table 3 .
In vitro activity of ATM-AVI and comparator agents tested against Enterobacterales isolates across regions stratified by infection sources from 2016 to 2020.

Table 4 .
In vitro activity of ATM-AVI and comparator agents tested against resistance phenotypes of Enterobacterales isolates collected globally stratified by wards from 2016 to 2020.

Table 4 .
Cont.No breakpoints available from CLSI and EUCAST.Values expressed are indicative of the cumulative percentage of isolates inhibited at ≤8 mg/L for comparison purposes.e Susceptible category for colistin not available for CLSI breakpoints (only intermediate and resistant isolates are available).f Data for colistin do not include isolates of Morganella morganii, Proteus hauseri, Proteus mirabilis, Proteus vulgaris, Providencia alcalifaciens, Providencia rettgeri, Providencia spp., Providencia stuartii, and Serratia marcescens because of their intrinsic resistance.g Data for imipenem not available per EUCAST.h Data for tigecycline do not include isolates of Morganella morganii, Proteus hauseri, Proteus mirabilis, Proteus vulgaris, Providencia alcalifaciens, Providencia rettgeri, Providencia spp., and Providencia stuartii due to their intrinsic resistance.i Data for tigecycline were calculated based on FDA-approved breakpoints for CLSI.j EUCAST data for susceptibility to tigecycline are limited to E. coli and C.

Table 5 .
In vitro activity of ATM-AVI and comparator agents tested against resistance phenotypes of Enterobacterales isolates collected globally stratified by infection sources from to 2020.