In Vitro Activities of Aztreonam-Avibactam, Eravacycline, Cefoselis, and Other Comparators against Clinical Enterobacterales Isolates: a Multicenter Study in China, 2019

ABSTRACT Aztreonam-avibactam, eravacycline, and cefoselis are three novel antimicrobial agents for the treatment of serious infections caused by Gram-negative bacteria. We evaluated the in vitro activities of the above-mentioned three antimicrobial agents against clinical Enterobacterales isolates. A total of 1,202 Enterobacterales isolates, including 10 genera or species, were collected from 26 hospitals that cover seven regions of China. The susceptibilities of the 30 antimicrobial agents were interpreted based on the combination of U.S. Food and Drug Administration and Clinical and Laboratory Standards Institute guidelines. The results indicated that all Enterobacterales isolates showed high susceptibility to aztreonam-avibactam (98.25%), eravacycline (85.69%), and cefoselis (62.73%). The first two antimicrobial agents also demonstrated potent activities against multidrug-resistant and carbapenem-resistant Enterobacterales independent of antimicrobial resistance mechanisms. The rates of susceptibility to aztreonam-avibactam, eravacycline, and cefoselis were lowest in Morganella spp. (84.42%), Proteus spp. (33.65%), and Escherichia coli (40.14%), respectively. In general, the lower rates of susceptibility to eravacycline and cefoselis were in the older inpatient group. The strains isolated from urinary tract exhibited the lowest rate of susceptibility (78.97%) to eravacycline, and the lowest rate of susceptibility (45.83%) to cefoselis was observed in nervous system specimens. The strains isolated from intensive care unit (ICU) wards showed significantly reduced susceptibility to cefoselis compared with those isolated from non-ICU wards. The MIC values of aztreonam-avibactam and ceftazidime-avibactam have poor consistency (weighted kappa = 0.243), as did eravacycline and tigecycline (weighted kappa = 0.478). Cefoselis and cefepime showed highly similar activities against Enterobacterales (weighted kappa = 0.801). Our results support the clinical development of aztreonam-avibactam, eravacycline, and cefoselis to treat infections caused by Enterobacterales. IMPORTANCE Infections caused by multidrug-resistant (MDR) Enterobacterales, especially carbapenem-resistant Enterobacterales (CRE), have been a challenging clinical problem due to the limited therapeutic options. Therefore, the need to develop novel antimicrobial agents and evaluate their activities against Enterobacterales in vitro is urgent. Our results show that the novel antimicrobial agents aztreonam-avibactam and eravacycline retain activities against MDR and CRE isolates, including carbapenemase producers and non-carbapenemase producers. Further analysis combined with clinical information on the strains tested revealed that no significant differences were observed in susceptibility rates of strains with different demographic parameters to aztreonam-avibactam. Age, specimen source, and department were associated with the susceptibility of strains to eravacycline and cefoselis (P ≤ 0.01). Compared with ceftazidime-avibactam, aztreonam-avibactam has its advantages and limitations against Enterobacterales. The potent activity of eravacycline against Enterobacterales was higher than that of tigecycline. Cefoselis and cefepime showed a highly consistent activity against Enterobacterales.

rates of strains with different demographic parameters to aztreonam-avibactam. Age, specimen source, and department were associated with the susceptibility of strains to eravacycline and cefoselis (P # 0.01). Compared with ceftazidime-avibactam, aztreonamavibactam has its advantages and limitations against Enterobacterales. The potent activity of eravacycline against Enterobacterales was higher than that of tigecycline. Cefoselis and cefepime showed a highly consistent activity against Enterobacterales.
KEYWORDS aztreonam-avibactam, eravacycline, cefoselis, multidrug-resistant Enterobacterales, carbapenem-resistant Enterobacterales, antimicrobial susceptibility testing A ntimicrobial resistance is becoming an increasing threat endangering global public health. Multidrug-resistant (MDR) pathogens can increase the disease burden and cause severe infections associated with high morbidity and mortality (1). Enterobacterales are among the most prominent pathogens associated with antimicrobial resistance, as the pathogens are common causes of community-associated and health care-associated infections. Carbapenems possess the broadest antimicrobial spectrum of activity and are the favored last-resort drugs for treatment of MDR Enterobacterales infections. Worryingly, the emergence and worldwide dissemination of carbapenem-resistant Enterobacterales (CRE) pose one of the most challenging threats to human health (2). The main mechanisms of carbapenems resistance in CRE include (i) production of carbapenemases and (ii) production of extended-spectrum b-lactamases (ESBLs) or AmpC b-lactamases combined with the loss of either of the major outer membrane proteins (OMPs) porin and upregulated efflux pumps (3). In light of this, developing novel antimicrobial agents to combat the threat of these emerging resistant bacteria is urgent (4).
Aztreonam belongs to the member of the monobactam class and was approved by the U.S. Food and Drug Administration (FDA) in 1986 (5). Although aztreonam is stable to hydrolysis by metallo-b-lactamases (MBLs) compared with other b-lactams, it can also be hydrolyzed by most clinically relevant serine lactamases, such as ESBLs, AmpC b-lactamases, and Klebsiella pneumoniae carbapenemases (KPCs) (6). Its combination with avibactam, a novel non-b-lactam b-lactamase inhibitor which can effectively inhibit serine lactamases, has potential utility against MDR bacterial pathogens and CRE (7,8). Eravacycline is a novel broad-spectrum synthetic tetracycline antibiotic comprising a tetracycline core with two novel modifications, a fluorine atom at the C-7 position and apyrrolidinoacetamido group at the C-9 position (9). Eravacycline inhibits bacterial protein synthesis through binding to the 30S ribosomal subunit and retains antibacterial activities against isolates with tetracycline ribosomal protection proteins or most tetracycline efflux proteins, the same as tigecycline (10). Cefoselis is a fourth-generation cephalosporin with a 1-hydroxyethyl-5-aminopyrazole moiety at position 3 of the cephem ring (11). Due to its structural properties, cefoselis has a higher antibacterial activity than the third-generation cephalosporins against Enterobacterales (12).
Awareness of local antimicrobial susceptibility data plays a guiding role in selecting drugs to treat infections. Currently, only a few reports have been published in China on the in vitro antimicrobial activities of aztreonam-avibactam, eravacycline, and cefoselis against Enterobacterales isolates. This multicenter study in China was conducted to evaluate the in vitro activities of these three antimicrobial agents against a large collection of clinical Enterobacterales isolates recovered from hospitalized patients and to understand the demographic parameters which affect the susceptibilities of Enterobacterales isolates to these agents.

RESULTS
Among the 1,202 Enterobacterales isolates in this study, Escherichia (including only Escherichia coli) was the most abundant genus (n = 284), followed by Klebsiella (n = 243), Salmonella (n = 169), Enterobacter (n = 140), Proteus (n = 104), Serratia (n = 97), Citrobacter (n = 82), Morganella (n = 77), Providencia (n = 5), and Edwardsiella (n = 1). These isolates were collected from different specimens, including blood (n = 241) and respiratory tract (n = 191), abdominal cavity (n = 94), skin and soft tissue (n = 108), urinary tract (n = 214), gastrointestinal tract (n = 187), genital tract (n = 119), and nervous system (n = 48). The frequency of MDR isolates in various regions in China ranged from 43.24% to 55.56% (49.83% overall). The frequency of CRE isolates was highest in eastern China (25.56%), followed by southwestern China (13.51%), central China (10.36%), northwestern China (9.13%), northeastern China (6.80%), southern China (6.54%), and northern China (5.98%) (Fig. 1A). Among 119 CRE isolates, a total of 88 carbapenemases produced by 87 isolates were detected (1 K. pneumoniae isolate produced two carbapenemases, KPC-2 and NDM-1), including 43 KPCs (43 KPC-2), 40 NDMs (22 NDM-1, 16 NDM-5, 1 NDM-4, and 1 NDM-9), 3 IMPs (1 IMP-4, 1 IMP-8, and 1 IMP-26), and 2 OXAs (1 OXA-181 and 1 OXA-232). Through ESBL screening and confirmation tests, we detected 307 ESBL screen-positive isolates and 748 screen-negative isolates. We assessed the in vitro activities of 30 antimicrobial agents against all Enterobacterales isolates tested, including MDR and CRE isolates, and the results are listed in Table 1. Overall, all Enterobacterales isolates retained high rates of susceptibility to aztreonam-avibactam (98.25%), amikacin (95.76%), ceftazidime-avibactam (95.76%), and meropenem (92.26%). Only 13.14% and 24.46% of isolates in our study were susceptible to ampicillin and cefazolin, respectively. Other antimicrobial agents showed susceptibility rates ranging from 41.01% to 89.02%. The most active agents tested against MDR and CRE isolates were aztreonam-avibactam (susceptibility rates of 96.49% and 92.44%, respectively), amikacin (susceptibility rates of 91.49% and 67.23%, respectively), ceftazidime-avibactam (susceptibility rates of 91.49% and 59.66%, respectively), tigecycline (susceptibility rates of 83.81% and 82.35%, respectively), colistin (susceptibility rates of 74.12% and 80.67%, respectively), eravacycline (susceptibility rates of 78.96% and 78.99%, respectively), and fosfomycin (susceptibility rates of 75.79% and 77.00%, respectively) ( Fig. 1B and Table 1). To further investigate the in vitro activities of the three novel antimicrobial agents, aztreonam-avibactam, eravacycline, and cefoselis, against MDR and CRE isolates, we analyzed the MIC distributions and the cumulative percentage of these three novel antimicrobial agents against three organism groups, (i) all Enterobacterales, MDR, and CRE isolates, (ii) serine carbapenemase producers, MBL producers, and non-carbapenemase-producing CRE (the K. pneumoniae isolate coproducing KPC and NDM belongs to both the serine carbapenemase producer and MBL producer groups), and (iii) ESBL screen-positive isolates and screen-negative isolates (Fig. 2). Aztreonam-avibactam inhibited 97.67% of isolates at 4/4 mg/mL, including 95.33% of MDR isolates and 90.76% of CRE isolates. Moreover,   (Fig. 2). Among CRE isolates, both aztreonam-avibactam and eravacycline had high activities against carbapenemase (serine carbapenemases and MBLs) and non-carbapenemase producers, with all susceptibility rates exceeding 68%. Further observation revealed that eravacycline exhibited similar MIC distributions against both carbapenemase producers and non-carbapenemase producers, while the MIC distributions of aztreonam-avibactam and cefoselis against these two types of CRE differed. Non-carbapenemase producers showed a comparatively lower rate of susceptibility to aztreonam-avibactam (81.25%) and a relatively higher rate of susceptibility to cefoselis (12.50%) than did carbapenemase producers (Fig. 2). Similar to the case with ESBL screen-negative isolates, both aztreonam-avibactam and eravacycline were also highly active against ESBL screenpositive isolates, with MIC 90 values of #1/4 mg/mL and 1 mg/mL, respectively. Inevitably, ESBL screen-negative isolates showed very high rates of susceptibility to cefoselis (93.92%) (Fig. 2). Figure 3 shows the rates of susceptibility of different species and genera of Enterobacterales to the three novel antimicrobial agents. The number of Providencia spp. and Edwardsiella spp. was too small to be included in the analysis. Overall,  Table S1 in the supplemental material). For eravacycline, Morganella spp. and Proteus spp. exhibited significantly lower susceptibility rates than did other genera or species (P # 0.01  (14). Clinical information for the 1,202 hospitalized patients was included in this study to assess the effects of different demographic parameters (gender, region, age, specimen source, and department) associated with the resistance of Enterobacterales isolates to the three novel antimicrobial agents, aztreonam-avibactam, eravacycline, and cefoselis. No significant differences in rates of susceptibility to aztreonam-avibactam were observed among different demographic parameters (Table 2). For eravacycline and cefoselis, gender and region had no significant effect on the susceptibility rates of the corresponding isolates, but the isolates showed significantly different susceptibilities to eravacycline and cefoselis in the different age and specimen source groups. With the increasing ages of the inpatients, Enterobacterales isolates had reduced susceptibility to eravacycline and cefoselis. The one exception to this pattern was the isolates from patients age group #14 years, which showed the lowest rate of susceptibility to cefoselis. Among different specimen sources, the isolates from the urinary tract showed the lowest rate of susceptibility to eravacycline (78.97%), and the lowest rate of susceptibility to cefoselis (45.83%) was observed in the nervous system. Compared with non-ICU wards, the isolates collected from ICU wards had reduced susceptibility only to cefoselis among the three novel antimicrobial agents.
A scatterplot with the distribution of MIC values was used to assess the potency of the three novel antimicrobial agents (aztreonam-avibactam, eravacycline, and cefoselis) versus three comparators (ceftazidime-avibactam, tigecycline, and cefepime) against 1,202 Enterobacterales isolates (Fig. 4). Because of the high antibacterial activities of aztreonam-avibactam and ceftazidime-avibactam, most of the MIC values of the two antimicrobial agents were #1/4 mg/mL. It also should be noted that a few (n = 43) ceftazidimeavibactam MIC values were .16/4 mg/mL when the corresponding aztreonam-avibactam MIC values were #1/4 mg/mL. In contrast, there are 35 cases in which MIC values of aztreonam-avibactam were higher than those of ceftazidime-avibactam. The analysis of linear weighted kappa further indicated that the MIC values of aztreonam-avibactam and ceftazidime-avibactam had poor consistency against isolates tested (weighted kappa = 0.243). The comparison of antibacterial activities of eravacycline with tigecycline also exhibited diversity (weighted kappa = 0.478). Most of the tigecycline MIC values were higher than the corresponding MIC values of eravacycline (2-to 4-fold in many cases). For cefoselis and cefepime, although a few MIC values of cefoselis were higher than that of cefepime, 80.78% of strains shared identical cefoselis and cefepime MIC values. A linear weighted kappa value of 0.801 showed good agreement between cefoselis and cefepime MIC values.

DISCUSSION
In recent years, the prevalence of CRE has significantly increased in the world (15). In this study, CRE rates varied by geography and the highest CRE rate was 25.56% in eastern China, the most economically developed region of China. Yet therapeutic options for CRE infections have been limited. A few agents, such as colistin, tigecycline FIG 3 Susceptibility rates of different species, including Escherichia coli, Klebsiella spp., Salmonella spp., Enterobacter spp., Proteus spp., Serratia spp., Citrobacter spp. and Morganella spp., to aztreonam-avibactam, eravacycline, and cefoselis. The same letters for a given antimicrobial agent indicate that no significant differences were observed in susceptibility rate at a P value of #0.05.
In Vitro Activities of Novel Antimicrobial Agents Microbiology Spectrum and novel b-lactam-b-lactamase inhibitor combinations, ceftazidime-avibactam, meropenem-vaborbactam, and imipenem-relebactam, may remain active against these organisms. However, ceftazidime-avibactam, meropenem-vaborbactam, and imipenem-relebactam are effective against KPC-producing CRE isolates, while they have limited effectiveness against OXA-and MBL-producing CRE isolates (16). In this study, aztreonam-avibactam and eravacycline showed considerable effectiveness against MDR and CRE isolates. The aztreonam-avibactam MIC 90 values among CRE and MDR isolates were only 4/4 mg/mL and #1/4 mg/mL, respectively, and eravacycline MIC 90 values were  (19)(20)(21). In addition, an unexpected observation was that non-carbapenemase-producing CRE strains (81.25%) were slightly less susceptible to aztreonam-avibactam than carbapenemase producers (serine carbapenemase producers, 95.56%, and MBL producers, 97.67%) (Fig. 2). An evaluation study for the in vitro efficacy of a broad spectrum of antibiotics against CRE from the Arabian Peninsula also revealed a lower rate of susceptibility of non-carbapenemase producers to aztreonamavibactam than that of carbapenemase producers (85.4% versus 96.9%) (22). The relatively low susceptibility of non-carbapenemase-producing CRE isolates to aztreonam-avibactam may be due to the production of CMY-type AmpC b-lactamases, as Ma et al. found that the production of CMY-42 is correlated with aztreonam-avibactam resistance (23). Aztreonam-avibactam, as a novel b-lactamase-inhibitor combination, is currently under clinical development (ClinicalTrials registration no. NCT01689207). Since aztreonam-avibactam is not used in the clinic, our results showed that the corresponding strains in different demographic parameter groups showed no significant differences in rates of susceptibility to aztreonam-avibactam. Eravacycline was approved by FDA and the European Medicines Agency (EMA) in 2018, but the approval is still pending in China. The strains with only two demographic parameters, the age of patients and specimen source, have significant differences in susceptibilities to eravacycline. Our results showed that the rate of susceptibility of isolates from urinary specimens to eravacycline was the lowest, followed by those from genital tract and skin and soft tissue specimens. This was due to relatively higher isolation rates of Proteus spp. and Morganella spp., which had low rates of susceptibility to eravacycline in urinary (27.10%), genital tract (21.85%), and skin and soft tissue specimens (20.37%) compared to less than 11% in other specimens (data not shown). Eravacycline was approved for the treatment of complicated intra-abdominal infections (cIAI) (24,25). In this study, eravacycline exhibited a high activity against strains collected from the abdominal cavity, with a susceptibility rate of 86.17%. Cefoselis, as a novel fourth-generation cephalosporin, was widely applicable for respiratory, urogenital, and abdominal infections (11,26). The susceptibility rates of corresponding strains from the respiratory tract, urinary tract, genital tract, and abdominal cavity to cefoselis were 63.35%, 51.87%, 69.75%, and 52.13%, respectively. A significant difference was observed in the activity of cefoselis against Enterobacterales isolates between ICU (susceptibility rate of 45.86%) and non-ICU wards (susceptibility rate of 64.83%). This may be related to the fact that rates of isolation of MDR (63.91%) and CRE (19.55%) strains from ICU wards were both significantly higher than those from non-ICU wards, at 48.08% and 8.70%, respectively (Table S2). A metaanalysis revealed that length of stay in the ICU wards is one risk factor related to infection and/or colonization with ESBL-producing bacteria (27). A study by Hu et al. also showed that the rate of isolation of carbapenem-resistant K. pneumoniae from ICU wards was significantly higher than that from non-ICU wards (28).
Ceftazidime-avibactam, as the only avibactam-based combination currently available for clinical use, represented an important advance in the treatment for infections caused by CRE (29). Aztreonam-avibactam has a broader spectrum of activity than ceftazidime-avibactam, with additional coverage for MBLs. In this study, ceftazidimeavibactam MIC values for 43 isolates were .16/4 mg/mL, while corresponding aztreonam-avibactam MIC values were #1/4 mg/mL. As expected, 95.35% (41/43) of these isolates produced MBLs. However, aztreonam-avibactam may not replace ceftazidimeavibactam. The production of KPC-21 in combination with four extra amino acids in penicillin-binding protein 3 (PBP3) is able to confer on strains resistance to aztreonamavibactam, but they remain susceptible to ceftazidime-avibactam (30). Our results also showed that partial MIC values (n = 35) of aztreonam-avibactam were higher than the corresponding MIC values of ceftazidime-avibactam. The consistency test also showed possible different resistance mechanisms of the strains to the two antimicrobial agents. Therefore, the avibactam-based agents both have their limitations, and combination with ceftazidime-avibactam and aztreonam may be a good treatment strategy for CRE infection (31). Our results revealed that eravacycline has demonstrated 2-to 4-fold greater potency in vitro than tigecycline. The possible reason for this is that the few key distinctions between eravacycline and tigecycline, a fluorine atom at the C-7 position and a pyrrolidinoacetamido group at the C-9 position, take effect. Another reason may be that tigecycline has been used in the clinic for several years, while eravacycline is still under approval in China. Both cefoselis and cefepime belong to the fourth-generation cephalosporins. As expected, the antibacterial activity of cefoselis against Enterobacterales was similar to that of cefepime in this study. Our results align with those of the previous studies by other investigators (12,32).
In summary, this study evaluated the in vitro activities of aztreonam-avibactam, eravacycline, and cefoselis against 1,202 Enterobacterales, including 119 CRE and 599 MDR isolates, from multiple centers. The results of this large investigation support the clinical development of the three novel antimicrobial agents for the treatment of Enterobacterales infections.

MATERIALS AND METHODS
This study relied on the surveillance program that involved seven subcenters, covering northern China, eastern China, southern China, central China, northeastern China, southwestern China, and northwestern China. The isolates included in the study were randomized and were required to meet the following criteria: the host of the isolates was clinically diagnosed as having a possible infection and only one isolate per patient was included. With reference to the inclusion criterion, each participating center was required to collect nonduplicated Enterobacterales isolates from the following specimen sources of inpatients between January 2019 and December 2019: respiratory tract, abdominal cavity, blood, skin and soft tissue, urinary tract, gastrointestinal tract, genital tract, and nervous system. Isolates were sent to the central laboratory. The central laboratory confirmed the species identification of all isolates tested using matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry.
A total of 1,202 Enterobacterales isolates were collected from 26 hospitals in 19 provinces and municipalities that cover seven subcenters, including northern China (n = 234), eastern China (n = 90), southern China (n = 153), central China (n = 222), northeastern China (n = 147), southwestern China (n = 148), and northwestern China (n = 208). All antimicrobial susceptibility testing was conducted in a central monitoring laboratory. Aztreonam-avibactam, eravacycline, cefoselis, and 27 other antimicrobial agents were tested using the broth microdilution method (Gram-negative bacteria susceptibility test card; Dier, China) specified by Clinical and Laboratory Standards Institute (CLSI) standards (33). Aztreonam-avibactam and ceftazidime-avibactam were both tested with avibactam at a fixed concentration of 4 mg/mL. The susceptible breakpoints of aztreonam, cefepime, and cefoperazone defined by the CLSI were applied for aztreonam-avibactam, cefoselis, and cefoperazone-sulbactam. The tigecycline and eravacycline susceptibilities were interpreted based on FDA guidelines. The susceptibility rate of colistin was replaced by the nonresistance rate. The susceptibilities of the remaining 24 antimicrobial agents were interpreted according to current CLSI guidelines (34).
MDR Enterobacterales isolates were defined as showing acquired resistance to at least one agent in three or more antimicrobial classes according to the combination of CLSI and FDA criteria (tigecycline and eravacycline: FDA criteria, and other antimicrobial agents: CLSI criteria). Enterobacterales isolates categorized as CRE were defined as displaying resistance to any carbapenem (imipenem, meropenem, and ertapenem) according to CLSI interpretive criteria. Imipenem was not applied to Proteus spp., Providencia spp., or Morganella spp. due to their intrinsically elevated MIC values. Subsequently, CRE isolates were tested for the presence of carbapenemase genes (i.e., bla KPC , bla OXA , bla NDM , bla VIM , bla IMP , bla GIM , and bla SPM ) through PCR followed by DNA sequencing. In accordance with CLSI guidelines (34), all Enterobacterales isolates were phenotypically screened for possible ESBL production using ceftazidime, ceftriaxone, and aztreonam MICs (any one MIC value of $2). Confirmation tests were then conducted for possible ESBL-producing isolates by testing with clavulanate in combination with ceftazidime and cefotaxime. Quality control of ESBL confirmatory test was done using K. pneumoniae ATCC 700603 as a positive control and E. coli ATCC 25922 as a negative control (33). "ESBL screen negative" was defined as meeting at least one of the following criteria: (i) for the isolates, the ceftazidime, ceftriaxone, and In Vitro Activities of Novel Antimicrobial Agents Microbiology Spectrum aztreonam MIC values were all #1 mg/mL, and (ii) the isolates demonstrated a negative result by the ESBL confirmatory test and were nonresistant to ceftazidime, ceftriaxone, and aztreonam. "ESBL screenpositive isolates" were defined as Enterobacterales with a positive result for ESBL confirmation tests. All statistical analyses were performed using Statistical Package for the Social Sciences version 23.0 software (SPSS Inc., Chicago, IL, USA). Pearson's chi-square (x 2 ) test or correction for continuity was carried out to compare the categorical variables, and a P value of #0.05 was considered statistically significant. Linear weighted kappa analysis was used for the consistency test for two comparable agents. A weighted kappa coefficient greater than 0.60 indicates strong consistency, a coefficient less than 0.4 indicates poor consistency, and a coefficient of 0.4 to 0.6 indicates moderate consistency.
Ethics statement. No human subjects participated in this study. The study was approved by the human research ethics committee of Peking Union Medical College Hospital (PUMCH; no. JS-2581).

SUPPLEMENTAL MATERIAL
Supplemental material is available online only. SUPPLEMENTAL FILE 1, PDF file, 0.1 MB.