International Journal of Infectious Diseases 123 (2022) 157–165

Contents lists available at ScienceDirect

International Journal of Infectious Diseases

journal homepage: www.elsevier.com/locate/ijid

A phase III, multicenter, double-blind, randomized clinical trial to

evaluate the efficacy and safety of ceftolozane/tazobactam plus
metronidazole versus meropenem in Chinese participants with
complicated intra-abdominal infections
Yihong Sun 1, Jia Fan 2,∗, Gang Chen 3, Xiaofei Chen 4, Xiaoling Du 4, Ye Wang 4, Hui Wang 4,#,
Fang Sun 4, Matthew G. Johnson 5, Mekki Bensaci 5, Jennifer A. Huntington 5,
Christopher J. Bruno 5,∗∗
1
Department of General Surgery, Zhongshan Hospital, Fudan University, Shanghai, China
2
Department of Liver Surgery and Transplantation, Zhongshan Hospital, Fudan University, Shanghai, China
3
Department of Hepatobiliary Surgery, The First Hospital of Kunming, Kunming, China
4
MSD China, Building A, Headquarters Park Phase II, No. 1582 Gumei Rd, Xuhui District, Shanghai 200233 China
5
Merck & Co., Inc., Rahway, NJ, USA

a r t i c l e i n f o a b s t r a c t

Article history: Objectives: This study aimed to evaluate the efficacy and safety of ceftolozane/tazobactam (C/T) plus
Received 3 June 2022 metronidazole versus meropenem plus placebo for the treatment of complicated intra-abdominal infec-
Revised 29 July 2022
tion (cIAI) in Chinese adult participants.
Accepted 5 August 2022
Methods: In this phase III clinical trial (NCT03830333), Chinese adult participants with cIAI were ran-
domized 1:1 to receive C/T plus metronidazole or meropenem plus placebo. The primary objective was
Keywords: to assess C/T plus metronidazole for noninferiority versus meropenem for clinical response rate at the
Antibacterial test of cure (TOC; 28 ± 2 days after study start) visit in the clinically evaluable population. Secondary
China endpoints included clinical and microbiologic responses at the TOC and end-of-treatment (≤24 hours af-
Ceftolozane
ter last dose) visits and adverse event rates.
Tazobactam
Results: Clinical cure at the TOC visit in the clinically evaluable population was 95.2% and 93.1% for C/T
Meropenem
plus metronidazole and meropenem, respectively (between-treatment difference: 2.1% [95% confidence
interval: -4.7%, 8.8%]); thus, noninferiority was met. Clinical responses at the TOC and end-of-treatment
visits and microbiologic responses at the TOC visit were consistent with the primary efficacy endpoint.
Safety was comparable between study treatment groups.
Conclusion: In Chinese adult participants with cIAI, C/T plus metronidazole was noninferior to
meropenem, with comparable safety.
© 2022 Merck Sharp & Dohme LLC., a subsidiary Merck & Co., Inc., Rahway, NJ, USA and The Author(s).
Published by Elsevier Ltd on behalf of International Society for Infectious Diseases.
This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/)

Introduction

Antibacterial resistance is a global threat to public health

(Centers for Disease Control and Prevention, 2019; World Health
∗
Correspondence to: Jia Fan, Liver Cancer Institute, Zhongshan Hospital, Fudan Organization, 2018; Xiao and Li, 2016), representing a significant
University, 180 FengLin Road, Shanghai 20 0 032, China. clinical and economic burden (Cassini et al., 2019; Cosgrove, 2006;
∗∗
Christopher J. Bruno, Merck & Co., Inc., 126 E. Lincoln Ave., Rahway, NJ 07065 Falagas et al., 2014; Martin et al., 2018; McCann et al., 2020a,
USA.
2020b; Tabak et al., 2019). For patients with complicated intra-
E-mail addresses: fan.jia@zs-hospital.sh.cn (J. Fan), christopher.bruno@merck.com
(C.J. Bruno).
abdominal infection (cIAI), current treatment options are limited
#
Current affiliation: Shanghai Fosun Pharmaceutical (Group) Co. Ltd, Shanghai, for some bacterial pathogens owing to antibacterial resistance,
China. leading to elevated mortality, which highlights the need for

https://doi.org/10.1016/j.ijid.2022.08.003
1201-9712/© 2022 Merck Sharp & Dohme LLC., a subsidiary Merck & Co., Inc., Rahway, NJ, USA and The Author(s). Published by Elsevier Ltd on behalf of International
Society for Infectious Diseases. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Y. Sun, J. Fan, G. Chen et al. International Journal of Infectious Diseases 123 (2022) 157–165

additional treatment options (Zhang et al., 2018b). The incidence Randomization and blinding
of multidrug-resistant (MDR) gram-negative pathogens in com-
plicated urinary tract infections and cIAI in China is high, with An interactive voice response system/integrated web response
48% of Escherichia coli and 26% of Klebsiella pneumoniae isolates system (IVRS/IWRS) was used to allocate participants to C/T plus
expressing extended-spectrum β -lactamase (ESBL) enzymes metronidazole or meropenem plus placebo in a blinded man-
(Zhang et al., 2019). The prevalence of drug resistance among ner. Stratification used for treatment allocation/randomization
Pseudomonas aeruginosa isolates in China has increased over accounted for anatomic site of cIAI (bowel [small or large] vs
time, with nearly one-third of P. aeruginosa isolates from China other sites). The number of participants with localized compli-
expressing MDR phenotypes (Peng et al., 2015). cated appendicitis was limited to up to approximately 50% of the
Ceftolozane/tazobactam (C/T), a combination antibacterial agent randomized population. In addition, the IVRS/IWRS ensured that
comprising the antipseudomonal cephalosporin ceftolozane and enrollment of participants with CrCl of 30-≤50 ml/min was limited
the established β -lactamase inhibitor tazobactam, is approved to up to approximately 15% of the randomized population. Partic-
(in combination with metronidazole) in >70 countries, includ- ipants were randomized in a 1:1 ratio to C/T plus metronidazole
ing the United States, for the treatment of adults with cIAI, or meropenem plus placebo (Supplementary Appendix Figure A1).
complicated urinary tract infections, including pyelonephritis, and Double blinding was used. The participant, the investigator, and
hospital-acquired/ventilator-associated bacterial pneumonia ZER- sponsor personnel or delegate(s) in this study were all blinded.
BAXA [ceftolozane and tazobactam] (US Prescribing informa- The study drug was prepared and/or dispensed in a blinded
tion, 2022). In the previously reported global phase III ASPECT-cIAI fashion by an unblinded pharmacist/qualified site personnel and
clinical trial, C/T plus metronidazole was demonstrated to be non- infusion bags were covered in an opaque sleeve. Placebo for
inferior to meropenem, with a comparable safety profile, for the metronidazole infusions was given to participants assigned to the
treatment of adult participants with cIAI (Solomkin et al., 2015). meropenem arm to maintain blinding. To maintain study blinding,
Currently, there are limited data on the use of C/T plus metron- placebo for meropenem was administered to participants in the
idazole in Chinese patients. Here, we report the efficacy and safety C/T group with CrCl 30-≤50 ml/min due to a dose adjustment
of C/T plus metronidazole versus meropenem for the treatment of of meropenem. The frequency of placebo administrations was
cIAI in adult participants of Chinese descent. adjusted according to participants’ CrCl to ensure double blinding.

Study treatment

Methods Participants were administered C/T 1.5 g (ceftolozane

1 g/tazobactam 0.5 g) plus metronidazole 0.5 g or meropenem
Study design 1 g plus placebo as 60-minute intravenous (IV) infusions every
8 hours for 4-14 days (Supplementary Appendix Figure A1). Doses
This study was a randomized, multicenter, double-blind, phase in participants with CrCl of 30-≤50 ml/min (estimated using the
III efficacy and safety study of C/T plus metronidazole versus Cockroft-Gault formula) were reduced to C/T 0.75 g (ceftolozane
meropenem plus placebo in adult participants with cIAI (Clinical- 0.5 g/tazobactam 0.25 g) plus metronidazole 0.5 g IV every 8
trials.gov: NCT03830333), completed at 21 sites across 14 provinces hours and meropenem 1 g plus placebo IV every 12 hours.
throughout China.
Procedures

Baseline assessments were performed ≤24 hours before ran-

Participants domization. Intra-abdominal specimens were collected at baseline
during the time of the initial procedure and before debridement,
The study was designed to enroll 268 participants of Chinese removal, or disinfection of the primary site of infection. Specimens
descent (aged ≥18-≤75 years) with a diagnosis of cIAI, wherein for assessment of baseline isolates were collected from aspirates
there was evidence of intraperitoneal bacterial infection and (using a needle or syringe), tissue samples, or biopsy samples.
systemic infection requiring surgical intervention within 24 hours Blood cultures were collected from participants who failed pre-
(before or after) of the start of study drug administration. Key vious antibacterial therapy, had hospital-acquired infections, or
exclusion criteria included the following: cIAI managed by staged demonstrated signs of severe sepsis. When clinically indicated (e.g.,
abdominal repair, wherein the fascia was not closed or infec- cases requiring repeat intra-abdominal intervention), postbaseline
tion source control was unlikely to be achieved; serious underly- cultures for intra-abdominal and blood pathogens were collected
ing medical conditions; severe renal impairment (creatinine clear- at any study visit through the test of cure (TOC; day 28 ± 2 days)
ance [CrCl] <30 ml/min or requirement for peritoneal dialysis, visit. Adverse events (AEs) were monitored throughout the study.
hemodialysis or hemofiltration, or oliguria [<20 ml/h of urine out- Quantitative measurements of in vitro activity of C/T plus
put over 24 hours]); use of systemic antibacterial therapy for >24 metronidazole and meropenem plus placebo against bacterial
hours during the previous 72 hours before the first dose of study species isolated from specimens at baseline were performed using
drug, unless there was a documented treatment failure with such broth microdilution. Minimum inhibitory concentration (MIC) val-
therapy, defined as the presence of clinical plus operative findings ues and study drug susceptibility were reported based on the 2018
or clinical plus radiographic findings indicating ongoing or wors- Clinical Laboratory Standards Institute breakpoints (Clinical and
ening infection despite systemic treatment for ≥48 hours, and op- Laboratory Standards Institute, 2018). Screening for ESBL enzymes
erative intervention or reintervention (if previous surgical proce- was performed according to standard MIC panels; ESBL-positive
dure) intended within 24 hours of the first dose of study drug; isolates were defined as those showing a decrease by three or
or presence of a condition not manageable by surgical interven- more two-fold MIC dilutions for a drug when tested in combina-
tion or cured only with surgical intervention in the absence of sys- tion with clavulanic acid versus its MIC when tested alone.
temic antibacterial therapy. Additional details regarding inclusion Whole genome sequencing using next-generation sequencing
and exclusion criteria are provided in Supplementary Appendix was performed using the Illumina MiSeq System at Covance Cen-
Table A1. tral Laboratory Services (Shanghai, China) to identify the predomi-

158
Y. Sun, J. Fan, G. Chen et al. International Journal of Infectious Diseases 123 (2022) 157–165

nant resistance genes or alleles present within ESBL-positive Enter- not performed. Participants with missing evaluations for clinical
obacterales and all P. aeruginosa isolates. Sequencing analysis was or microbiologic endpoints were considered to be indeterminate
performed using Illumina Sequencing Analysis Viewer (SAV; ver- for the ITT population. Participants with an indeterminate clinical
sion 2.8.1) software. Sequencing quality control was evaluated us- outcome were excluded from the CE, microbiologically evaluable;
ing SAV Top 5, which was used as a starting point for analyzing the subset of CE participants who had ≥1 baseline intra-abdominal
the quality of an Illumina sequencing run. pathogen identified that was susceptible to study drug), and EME
populations; however, participants who discontinued because of
Endpoints lack of efficacy were considered to have a clinical response of fail-
ure and were presumed to have a microbiologic response catego-
The primary efficacy endpoint was clinical response rate at rized as persistence. Safety was assessed by clinical review of AEs,
the TOC visit in the clinically evaluable (CE) population (par- laboratory tests, and vital signs.
ticipants who met the protocol definition of cIAI, adhered to
study procedures, and had a clinical outcome at the TOC visit). Results
Secondary endpoints included clinical response at TOC in the
intention-to-treat (ITT) population (all randomized participants), Participants
clinical response at the end-of-treatment (EOT) visit in the ITT and
CE populations, per-participant and per-pathogen microbiologic Between March 20, 2019 and October 14, 2020, 268 participants
responses at the TOC visit in the expanded microbiologically were enrolled and randomized to the ITT population to receive C/T
evaluable (EME) population (all randomized participants who met plus metronidazole (n = 134) or meropenem (n = 134; Figure 1).
all CE population criteria and had ≥1 baseline intra-abdominal Baseline characteristics were generally balanced across treatment
pathogen identified, regardless of susceptibility to study drug), and groups (Table 1 [ITT population]; Supplementary Appendix Ta-
AE rates (all participants as treated [APaT] population). ble A3 [CE population]). The majority of participants in the C/T
Clinical response at the TOC visit was categorized as cure plus metronidazole and meropenem groups were male (59.0% and
(complete resolution or significant improvement in signs and 63.4%, respectively) and the median age was 48.0 years and 55.0
symptoms of the index infection without further requirement for years, respectively. Median Acute Physiology and Chronic Health
additional antibacterial therapy, surgical procedure, or drainage Evaluation II scores were 5.0 in both treatment groups; three
procedure), failure (evidence of persistent or recurrent infection participants in the meropenem group and none in the C/T plus
within the abdomen requiring additional intervention, need for metronidazole group had CrCl 30 - <50 ml/min. The most com-
additional antibacterial therapy, postsurgical wound infection, mon site of infection was the appendix in both treatment groups.
or death related to cIAI), or indeterminate response (study data
were not available for evaluation for any reason, including death Pathogens at baseline
unrelated to the index infection or extenuating circumstances that
precluded classification as either cure or treatment failure). Of the 221 participants in the CE population, 127 (57.5%) had ≥1
Microbiologic responses at the TOC visit were categorized as pathogen isolated from baseline intra-abdominal cultures and thus
eradication, presumed eradication, persistence, persistence acquir- made up the EME population. Among these 127 participants, most
ing resistance, presumed persistence, or indeterminate response had monomicrobial infections at baseline (C/T plus metronidazole:
and were defined as described in Supplementary Appendix Table 41/54 [75.9%]; meropenem: 63/73 [86.3%]). The most commonly
A2. A favorable per-participant microbiologic outcome required a isolated pathogens were gram-negative aerobes (E. coli, n = 78
favorable microbiologic response for each baseline pathogen; if the [61.4% of participants]; K. pneumoniae, n = 22 [17.3%]; P. aerugi-
microbiologic response for any baseline pathogen was unfavorable, nosa, n = 5 [3.9%]) and gram-positive aerobes (Streptococcus angi-
the participant was considered to have had a microbiologic failure. nosus group, n = 11 [8.7%]); all other pathogens were identified in
≤5 participants in both treatment groups.
Statistical analysis Of the 139 participants in the microbiologic ITT population, the
most commonly isolated baseline pathogens were Enterobacterales
The study was designed to demonstrate that the clinical re- (n = 124 isolates); E. coli and K. pneumoniae were the most fre-
sponse rate in the CE population at the TOC visit for C/T plus quently isolated gram-negative pathogens, with 82 and 28 isolates,
metronidazole was noninferior to meropenem using planned en- respectively (Table 2). Among all Enterobacterales isolates, 97.4%
rollment of 268 participants (134 per arm). Anticipated enrollment and 100% were susceptible to C/T and meropenem, respectively.
was 200 participants (100 per arm) in the CE population, giving The susceptibility rate of P. aeruginosa isolates was the same for
80% power to demonstrate noninferiority between treatments us- both study drugs, with 83.3% of isolates susceptible.
ing a one-tailed alpha of 2.5%. The power and sample size assumed A total of 32/176 (18.2%) isolates were ESBL-positive. Whole
a noninferiority margin of -12.5% and an underlying clinical re- genome sequencing was performed on 43 isolates (E. coli, n = 29;
sponse rate of 90% in the CE population for both treatments. The P. aeruginosa, n = 6; K. pneumoniae, n = 3; Citrobacter koseri, n = 1;
95% confidence intervals (CIs) for between-treatment differences Enterobacter asburiae, n = 1; Enterobacter cloacae, n = 1; Morganella
in the clinical response rate was calculated using the stratified morganii, n = 1; Proteus mirabilis, n = 1). AmpC only (n = 18 iso-
Miettinen and Nurminen method with Cochran-Mantel-Haenszel lates), β -lactam resistance gene blaACT7 (n = 14 isolates), combina-
weighting. C/T plus metronidazole was considered noninferior to tion of AmpC and AmpC1 (n = 9 isolates), and β -lactam resistance
meropenem if the lower bound of the two-sided 95% CI of the gene blaCTX-M14 (n = 5 isolates) were the most commonly identified
between-treatment difference in clinical response rate at the TOC antibacterial resistance genes.
visit in the CE population was larger than -12.5%. Consistency of
the treatment effect across various subgroups was determined us- Efficacy
ing the estimate (with a nominal 95% CI) of the between-group
treatment effects for the primary endpoint within select categories. Clinical cure at the TOC visit was achieved in 95.2% and 93.1%
The secondary efficacy analyses followed the same approach as of the CE population who received C/T plus metronidazole and
the primary endpoint. Secondary efficacy objectives were intended meropenem, respectively (between-treatment difference: 2.1% [95%
as estimation objectives only; therefore, noninferiority testing was CI: -4.7%, 8.8%]); thus, the study criterion for noninferiority was

159
Y. Sun, J. Fan, G. Chen et al. International Journal of Infectious Diseases 123 (2022) 157–165

Fig. 1. Participant disposition.

a
The intention-to-treat population was defined as the all participants as treated population.
CE, clinically evaluable; C/T, ceftolozane/tazobactam; TOC, test of cure.

Table 1
Participant baseline characteristics (ITT population).

C/T + metronidazole Meropenem + placebo

Characteristics (N=134) (N=134)

Male, n (%) 79 (59.0) 85 (63.4)

Age, n (%)
≤65 y 108 (80.6) 103 (76.9)
66-75 y 26 (19.4) 31 (23.1)
Median (range), y 48.0 (18-75) 55.0 (18-74)
APACHE II score
Mean (SD) 4.9 (3.1) 5.1 (2.8)
Category, n (%)
<10 119 (88.8) 126 (94.0)
≥10 15 (11.2) 8 (6.0)
Baseline creatinine clearance
Median (range), ml/min 105.1 (50.4-316.0) 93.0 (42.8-252.3)
Category, n (%)
30 to <50 ml/min 0 3 (2.2)
>50 ml/min 134 (100) 131 (97.8)
Anatomic site of current infectiona , n (%)
Appendix 67 (50.0) 66 (49.3)
Non-appendix
Gallbladder 37 (27.6) 39 (29.1)
Liver 9 (6.7) 10 (7.5)
Duodenum 9 (6.7) 5 (3.7)
Other 6 (4.5) 8 (6.0)
Stomach 4 (3.0) 5 (3.7)
Jejunum/ileum 3 (2.2) 3 (2.2)
Diverticular 2 (1.5) 4 (3.0)
Peritoneum 1 (0.7) 0 (0.0)
Colon 0 (0.0) 2 (1.5)
Procedure type
Percutaneous aspiration 15 (11.2) 20 (14.9)
Laparoscopy 102 (76.1) 98 (73.1)
Laparotomy 17 (12.7) 16 (11.9)
a
Investigator may have chosen more than one site.
APACHE, Acute Physiology and Chronic Health Evaluation; C/T, ceftolozane/tazobactam;
ITT, intention-to-treat.

160
Y. Sun, J. Fan, G. Chen et al. International Journal of Infectious Diseases 123 (2022) 157–165

Table 2
Baseline pathogens with susceptibility to study treatment (MITT population).

Baseline pathogens N1 C/T + metronidazole, n/N2a , b (%) Meropenem + placebo, n/N2a (%)

Gram-negative aerobes
All Enterobacterales 124 114/117 (97.4) 117/117 (100)
Citrobacter freundii complex 1 1/1 (100) 1/1 (100)
C. koseri 1 1/1 (100) 1/1 (100)
Enterobacter cloacae complex 2 1/2 (50.0) 2/2 (100)
Escherichia coli 82 80/80 (100) 80/80 (100)
Klebsiella oxytoca 3 3/3 (100) 3/3 (100)
K. pneumoniae 28 26/27 (96.3) 27/27 (100)
Morganella morganii 1 1/1 (100) 1/1 (100)
Proteus mirabilis 3 1/2 (50.0) 2/2 (100)
Aeromonas hydrophila 1 0 1/1 (100)
Pseudomonas aeruginosa 6 5/6 (83.3) 5/6 (83.3)
Gram-positive aerobes
Aerococcus viridans 1 0 1/1 (100)
Bacillus 1 0 1/1 (100)
Streptococcus anginosus group 13 12/12 (100) 12/12 (100)
S. bovis group 1 1/1 (100) 1/1 (100)
S. mitis group 2 2/2 (100) 2/2 (100)
S. salivarius group 1 1/1 (100) 1/1 (100)
Gram-negative anaerobes
Bacteroides fragilis group 1 0 1/1 (100)
Gram-positive anaerobes
Actinomyces naeslundii 1 0 1/1 (100)
Propionibacterium acnes 1 0 1/1 (100)
a
Susceptibility criteria for each pathogen were defined per the Clinical Laboratory Standards Institute Performance Standards for Antimicrobial Susceptibility
Testing, 28th edition (Clinical and Laboratory Standards Institute, 2018).
b
If the pathogens for the C/T susceptibility breakpoint were not defined, baseline pathogens were considered susceptible to C/T if the MIC was ≤8 μg/ml.
C/T, ceftolozane/tazobactam; MIC, minimum inhibitory concentration; MITT, microbiologic intention-to-treat; n, number of pathogens within a specific category;
N1, number of baseline pathogens; N2, number of pathogens with a baseline MIC available.

met (Figure 2A). Between the two treatment groups, point es- meropenem, respectively. Overall, 67 (50.0%) participants in the
timates of clinical responses at the TOC visit favored C/T plus C/T plus metronidazole group and 68 (50.7%) in the meropenem
metronidazole across most of the predefined subgroups in the CE group experienced ≥1 AE (Table 4). The most commonly reported
population (Figure 2B). (occurring in ≥5% of participants in either treatment group) AEs
Clinical response rates at the TOC visit in the ITT population in the C/T plus metronidazole and meropenem groups were cough
were lower than in the CE population for both treatment groups (5.2% vs 6.7%), diarrhea (6.0% vs 8.2%), and pyrexia (6.7% vs 10.4%).
(C/T plus metronidazole: 85.1% vs meropenem: 89.6%; between- Similar incidences of serious AEs and discontinuations due to AEs
treatment difference: -4.4% [95% CI: -12.6%, 3.7%]; Figure 3A). Out- were reported across treatment groups.
comes for the EOT visit in the ITT and CE populations (between- Incidences of drug-related AEs were similar between-treatment
treatment difference: -1.5% [95% CI: -8.0%, 4.8%] and 1.4% [95% CI: groups, with abdominal distension (1.5% vs 0%), abnormal hep-
-3.8%, 6.7%], respectively; Figures 3B and 3C) were consistent with atic function (0% vs 2.2%), diarrhea (1.5% vs 2.2%), nausea (1.5% vs
the primary efficacy endpoint. 0.7%), peripheral swelling (2.2% vs 1.5%), and rash (0.7% vs 1.5%)
Clinical responses by baseline pathogen at the TOC visit were reported in ≥1% of participants in the C/T plus metronidazole or
comparable across treatment groups (Figure 4), with most par- meropenem groups (Table 4). No drug-related serious AEs were re-
ticipants achieving clinical cure against isolates identified in the ported in the C/T plus metronidazole group, and three (2.2%) were
CE population (between-treatment difference [95% CI]: E. coli, - reported in the meropenem group. No deaths were reported in the
1.9% [-16.5%, 9.4%]; K. pneumoniae, 9.1% [-19.0%, 38.5%]); among C/T plus metronidazole group, and one death was reported in the
the few participants with P. aeruginosa isolates, clinical cure was meropenem group. One participant from the meropenem group
achieved by 75.0% of participants who received C/T plus metron- discontinued because of a drug-related serious AE (atrial fibrilla-
idazole (n = 4) and 100% of participants who received meropenem tion).
(n = 1). Among the subgroup of participants in the ITT population
who were positive for ESBL-producing Enterobacterales pathogens
at baseline, rates of clinical cure were similar in those who re-
Discussion
ceived C/T plus metronidazole versus meropenem (84.6% vs 89.5%;
between-treatment difference: -4.9% [95% CI: -34.2%, 20.0%]).
The results of this phase III clinical trial indicated C/T plus
Rates of favorable per-participant microbiologic response at the
metronidazole was noninferior to meropenem for clinical response
TOC visit in the EME population were 94.4% and 93.2% in the C/T
at TOC in Chinese participants with cIAI, consistent with previ-
plus metronidazole and meropenem groups, respectively (between-
ously reported findings from the global ASPECT-cIAI trial (Solomkin
treatment difference: 1.2% [95% CI: -9.2%, 10.4%]; Table 3). Rates of
et al., 2015). High rates of clinical cure were observed in the CE
favorable per-pathogen microbiologic responses were also compa-
and ITT populations, with >85% of participants who received C/T
rable across the treatment groups (Table 3).
plus metronidazole achieving clinical cure at the EOT and TOC vis-
its. In addition, most participants who received C/T plus metron-
Safety idazole had favorable microbiologic responses at TOC for both per-
participant and per-pathogen assessments. Overall, these efficacy
In the APaT population, mean (SD) treatment duration was findings support using C/T plus metronidazole in Chinese patients
5.7 (2.6) days and 5.9 (2.7) days for C/T plus metronidazole and with cIAI.

161
Y. Sun, J. Fan, G. Chen et al. International Journal of Infectious Diseases 123 (2022) 157–165

Fig. 2. Clinical response at the TOC visit in the (A) CE population and (B) predefined subgroups of the CE population.
a
Percentage difference estimates ([C/T plus MTZ] - [MEM plus pbo]) and 95% CIs were based on the Miettinen and Nurminen method with Cochran-Mantel-Haenszel weight-
ing stratified by anatomic site of infection (bowel [small or large] vs other sites of cIAI).
b
Point estimate and corresponding CI were based on the unstratified Miettinen and Nurminen method.
c
CIs were included for parameters with four or more participants in one or more treatment groups.
d
ESBL status: positive was defined as participants who had baseline infecting ESBL-producing Enterobacterales; negative was defined as participants who had baseline
infecting non-ESBL-producing Enterobacterales; not tested was defined as participants who had baseline Enterobacterales but were not tested.
APACHE, Acute Physiology and Chronic Health Evaluation; CE, clinically evaluable; cIAI, complicated intra-abdominal infection; C/T, ceftolozane/tazobactam; ESBL, extended-
spectrum β -lactamase; IA, intra-abdominal; MEM, meropenem; MTZ, metronidazole; N1, number of participants within a specific subgroup; pbo, placebo; TOC, test of cure.

Table 3
Favorable microbiologic response at the TOC visits in the EME population.

C/T + metronidazole (N=54) Meropenem + placebo (N=73) % Difference estimate (95% CI)

Per-participant, n (%) 51 (94.4) 68 (93.2) 1.2 (-9.2 to 10.4)a

Per-pathogen,
Gram-negative aerobes, n/N1 (%) 47/50 (94.0) 57/60 (95.0) -1.0 (-11.9 to 8.7)b
All Enterobacterales, n/N1 (%) 45/48 (93.8) 57/60 (95.0) -1.2 (-12.5 to 8.5)b
Escherichia coli 30/32 (93.8) 44/46 (95.7) -1.9 (-16.5 to 9.4)b
Klebsiella pneumoniae 11/11 (100) 10/11 (90.9) 9.1 (-19.0 to 38.5)b
Pseudomonas aeruginosa 3/4 (75.0) 1/1 (100) -
Gram-positive aerobes, n/N1 (%) 8/10 (80.0) 15/17 (88.2) -8.2 (-42.2 to 20.0)b

Favorable microbiologic response included eradication and presumed eradication (few participants had follow-up cultures; presumed eradi-
cation/persistence was based on clinical response). Indeterminate responses (not shown) were considered unfavorable.
a
Percentage difference estimates ([C/T plus metronidazole] - [meropenem plus placebo]) and 95% CIs were based on the Miettinen and
Nurminen method with the Cochran-Mantel-Haenszel weighting stratified by anatomic site of infection (bowel [small or large] vs other sites
of cIAI).
b
Percentage difference estimates ([C/T plus metronidazole] - [meropenem plus placebo]) and 95% CIs were based on unstratified Miettinen
and Nurminen method.
cIAI, complicated intra-abdominal infection; C/T, ceftolozane/tazobactam; EME, expanded microbiologically evaluable; n, number of partici-
pants with the specified pathogen; N1, number of baseline pathogens; TOC, test of cure.

Consistent with other clinical trials assessing cIAI in China, the E. coli, and K. pneumoniae. Given the prevalence of E. coli and K.
most commonly isolated pathogens at baseline were E. coli, K. pneumoniae among patients with cIAI across China (Yang et al.,
pneumoniae, and P. aeruginosa. (Chen et al., 2010; Qin et al., 2017; 2020; Zhang et al., 2018b), favorable clinical and microbiologic re-
Wang et al., 2021). In the current trial, nearly 20% of baseline sponses observed herein support the use of C/T plus metronidazole
pathogens were ESBL-positive, compared with 7.2% in the global within the region.
pivotal trial (Solomkin et al., 2015). These findings were consistent ESBL-positive pathogens pose a global threat to the continued
with recent trends noting a regional uptick in drug-resistant efficacy of antibacterial agents (Hu et al., 2010; Zhang et al.,
pathogens in China and underscore the regional and global need 2016). As the incidence of ESBL-positive E. coli and K. pneumoniae
for novel antibacterial agents targeting MDR gram-negative bac- isolated from patients with complicated urinary tract infections
teria (Jean et al., 2022; Zhang et al., 2021, 2018a, 2018b). Notably, and cIAI increases, rates of these pathogenic variants in China
treatment with C/T plus metronidazole was associated with a have been reported as high as 48% and 26%, respectively (Yang
favorable clinical response against all gram-negative aerobes, et al., 2013; Zhang et al., 2019). Although the susceptibility of
including commonly isolated pathogens, such as Enterobacterales, ESBL-positive E. coli to common antibacterial agents remained

162
Y. Sun, J. Fan, G. Chen et al. International Journal of Infectious Diseases 123 (2022) 157–165

Fig. 3. Clinical response at (A) the TOC visit in the ITT population, (B) the EOT visit in the ITT population, and (C) the EOT visit in the CE population.
a
Indeterminate responses (not shown) were considered unfavorable.
b
There were a total of 11 participants with missing or indeterminant clinical response outcomes at TOC (C/T plus MTZ, n=7; MEM plus pbo, n=4).
c
Percentage difference estimates ([C/T plus MTZ] - [MEM plus pbo]) and 95% CIs were based on the Miettinen and Nurminen method with Cochran-Mantel-Haenszel weight-
ing stratified by anatomic site of infection (bowel [small or large] vs other sites of cIAI).
CE, clinically evaluable; cIAI, complicated intra-abdominal infection; C/T, ceftolozane/tazobactam; EOT, end of treatment; ITT, intention-to-treat; MEM, meropenem;
MTZ, metronidazole; pbo, placebo; TOC, test of cure.

Fig. 4. Clinical response by pathogen at the TOC visit in the CE population.

a
Percentage difference estimates ([C/T plus MTZ] - [MEM plus pbo]) and 95% CIs were based on the Miettinen and Nurminen method with Cochran-Mantel-Haenszel weight-
ing stratified by anatomic site of infection (bowel [small or large] vs other sites of cIAI).
b
Includes Citrobacter freundii complex, C. koseri, Enterobacter aerogenes, E. cloacae complex, other Enterobacter spp., K. oxytoca, Morganella morganii, Proteus mirabilis, Aeromonas
hydrophila, Aerococcus viridans, Bacillus, Enterococcus avium, E. faecalis, E. faecium, E. raffinosus, Pseudomonas aeruginosa, Rothia mucilaginosa, Staphylococcus aureus, S. caprae,
S. epidermidis, S. haemolyticus, S. hominis, S. saprophyticus, S. sciuri, S. warneri, Streptococcus anginosus group, S. bovis group, S. intermedius, S. mitis group, S. salivarius group,
Bacteroides fragilis group, Actinomyces naeslundii, and Propionibacterium acnes.
CE, clinically evaluable; cIAI, complicated intra-abdominal infection; C/T, ceftolozane/tazobactam; MEM, meropenem; MTZ, metronidazole; pbo, placebo; TOC, test of cure.

consistent between 2012 and 2014 (Zhang et al., 2017), reports port using C/T plus metronidazole as a potential therapeutic option
of ESBL-positive K. pneumoniae isolates from cIAIs between for Chinese patients with cIAI with ESBL-positive Enterobacterales
2015 and 2017 in China indicated increasing evidence of resis- pathogens.
tance to common antibacterial agents (i.e., amikacin, imipenem, C/T plus metronidazole demonstrated a favorable safety profile
piperacillin-tazobactam, ertapenem, and cefepime) (Zhang et al., and was generally well tolerated in Chinese participants, consistent
2021). Among participants in the current trial, favorable clinical with findings from the global pivotal trial (Solomkin et al., 2015).
response against ESBL-positive Enterobacterales was achieved with Drug-related AEs leading to study discontinuation were low in
C/T plus metronidazole, indicating C/T plus metronidazole was both groups, and no deaths were reported in the C/T plus metron-
comparable to meropenem in this population. These results sup- idazole group.

163
Y. Sun, J. Fan, G. Chen et al. International Journal of Infectious Diseases 123 (2022) 157–165

Table 4
Summary of AEs.

AE category, n (%) C/T + metronidazole (n=134) Meropenem + placebo (n=134) % Difference, estimate (95% CI)a

≥1 AE 67 (50.0) 68 (50.7) -0.7 (-12.6 to 11.2)

Drug-related AEb 15 (11.2) 14 (10.4) 0.7 (-7.0 to 8.5)
Serious AE 7 (5.2) 7 (5.2) 0.0 (-5.9 to 5.9)
Serious drug-related AEb 0 3 (2.2) -2.2 (-6.4 to 0.6)
Death 0 1 (0.7)c -0.7 (-4.1 to 2.1)
Discontinued drug owing to an AE 3 (2.2) 3 (2.2) 0.0 (-4.4 to 4.4)
Discontinued drug owing to a drug-related AEb 3 (2.2) 3 (2.2) -
Discontinued drug owing to a serious AE 0 1 (0.7) -
Discontinued drug owing to a serious drug-related AEb 0 1 (0.7) -
Most common AEs (≥5% in either treatment group)
Cough 7 (5.2) 9 (6.7) -
Diarrhea 8 (6.0) 11 (8.2) -
Pyrexia 9 (6.7) 14 (10.4) -
Most common drug-related AEs (≥1% in either treatment group)
Abdominal distension 2 (1.5) 0 (0.0) -
Abnormal hepatic function 0 (0.0) 3 (2.2) -
Diarrhea 2 (1.5) 3 (2.2) -
Nausea 2 (1.5) 1 (0.7) -
Peripheral swelling 3 (2.2) 2 (1.5) -
Rash 1 (0.7) 2 (1.5) -
a
Percentage difference estimates ([C/T plus metronidazole] vs [meropenem plus placebo]) and 95% CIs were based on the Miettinen and Nurminen method for protocol-
defined events, which were defined as having ≥4 participants within ≥1 treatment group.
b
Determined by the investigator to be related to the drug.
c
Death was categorized as related to the study treatment in all safety analyses; however, the relationship to the study treatment was determined to be unknown by the
investigator.
AE, adverse event; C/T, ceftolozane/tazobactam.

This study had limitations that should be considered along- Funding

side the findings. Microbiologic response was presumed based
on clinical response because postbaseline intra-abdominal cul- Funding for this research was provided by Merck Sharp &
tures were unavailable due to the lack of repeat intra-abdominal Dohme LLC., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA
interventions among study participants. Because of clinical mi- (MSD). Medical writing and/or editorial assistance was provided by
crobiologic practices at participating sites, anaerobes were not Jessica Deckman, Ph.D., CMPP, of The Lockwood Group, Stamford,
routinely isolated from cIAI specimens; however, the most com- CT, USA. This assistance was funded by MSD.
monly isolated pathogens, E. coli, K. pneumoniae, and P. aeruginosa,
were consistent with the previous reports (Chen et al., 2010; Qin Ethics
et al., 2017; Solomkin et al., 2015; Wang et al., 2021). The study
was designed to test for noninferiority in the CE population only. The study was conducted in accordance with the Declaration of
Although the clinical response rate at TOC was lower in the ITT Helsinki for experiments involving humans and was approved by
population in the C/T plus metronidazole group, the study was not the appropriate institutional review boards and regulatory agen-
designed to test for noninferiority between treatment groups in cies. Written informed consent was obtained from all participants
this population. The difference in clinical response rates between and documented accordingly.
the treatment groups in the ITT population was driven by the
higher number of participants with missing or indeterminant Acknowledgments
responses (classified as clinical failures) in the C/T plus metronida-
zole group compared with the meropenem group (seven vs four, The authors thank the participants for taking part in this
respectively). Participants with missing or indeterminant responses study. In addition to the site personnel, the authors also thank
were excluded from the CE population. all study investigators: Yihong Sun (Zhongshan Hospital of Fudan
Overall, these results demonstrated that C/T plus metronida- University, Shanghai, China), Dianrong Xiu (Peking University
zole was noninferior to meropenem for clinical response in Chi- Third Hospital, Beijing, China), Baoan Qiu (Navy General Hospital,
nese participants with cIAI. The safety profile of C/T plus metron- Beijing, China), Mingzhang Li (Baotou Central Hospital, Baotou,
idazole was favorable within this population and was consis- China), Xuehua Li and Guijie Liu (Liaocheng People’s Hospital,
tent with meropenem and with the results from the global piv- Liaocheng, China), Mengtao Zhou and Xiaolei Chen (The First
otal trial. Taken together, C/T plus metronidazole may provide Affiliated Hospital of Wenzhou Medical University, Wenzhou,
another option for antibacterial treatment of cIAI in China, po- China), Zhengjun Qiu (Shanghai General Hospital, Shanghai,
tentially addressing the need for treatment options for MDR China), Guoqing Tao (Wuxi People’s Hospital, Wuxi, China), Jian
pathogens. Lei (The First Affiliated Hospital of Guangzhou Medical University,
Guangzhou, China), Haiming Chen (The First Affiliated Hospital
Declarations of competing interest of Nanchang University, Nanchang, China), Weidong Jia (Anhui
Provincial Hospital, Hefei, China), Tuerganaili Aji (The First Af-
XD, YW, and FS are employees of MSD, China. XC and HW are filiated Hospital of Xinjiang Medical University, Urumqi, China),
former employees of MSD, China. MGJ, MB, JAH, and CJB are em- Tienan Bi (Taizhou Hospital of Zhejiang Province, Taizhou, China),
ployees of MSD. MSD, China and MSD employees may own stock Yongjie Zhao (Tianjin People’s Hospital, Tianjin, China), Gang Chen
and/or hold stock options in Merck & Co., Inc., Rahway, NJ, USA. (The First Hospital of Kunming, Kunming, China), Kailun Zhou and
YS, JF, and GC have no competing interests to declare. Jincai Wu (Hainan General Hospital, Haikou, China), Ping Chen

164
Y. Sun, J. Fan, G. Chen et al. International Journal of Infectious Diseases 123 (2022) 157–165

(Subei People’s Hospital, Yangzhou, China), Xudong Wang (The Martin A, Fahrbach K, Zhao Q, Lodise T. Association between carbapenem resistance
Second Hospital of Jilin University, Changchun, China), Sanrong Xu and mortality among adult, hospitalized patients with serious infections due to
Enterobacteriaceae: results of a systematic literature review and meta-analysis.
(Affiliated Hospital of Jiangsu University, Zhenjiang, China), Jiazeng Open Forum Infect Dis 2018;5:ofy150.
Xia (Wuxi No. 2 People’s Hospital, Wuxi, China), Zhonglin Ni (The McCann E, Sung AH, Ye G, Vankeepuram L, Tabak YP. Contributing factors to
2nd Affiliated Hospital of Wenzhou Medical University, Wenzhou, the clinical and economic burden of patients with laboratory-confirmed car-
bapenem-nonsusceptible Gram-negative respiratory infections. Infect Drug Re-
China), Weiyan Liu (Central Hospital of Minhang District, Shanghai, sist 2020a;13:761–71.
China), Jiaqing Cao (The Second Affiliated Hospital of Nanchang McCann E, Sung AH, Ye G, Vankeepuram L, Tabak YP. Contributing factors to the
University, Nanchang, China), Qicheng Lu (The First People’s Hos- clinical and economic burden of patients with laboratory-confirmed carbapen-
em-nonsusceptible Gram-negative urinary tract infections. Clinicoecon Out-
pital of Changzhou, Changzhou, China), and Guoxin Li (Southern
comes Res 2020b;12:191–200.
Medical University Nanfang Hospital, Guangzhou, China). Peng Y, Shi J, Bu T, Li Y, Ye X, Chen X, et al. Alarming and increasing prevalence
of multidrug-resistant Pseudomonas aeruginosa among healthcare-associated in-
fections in China: A meta-analysis of cross-sectional studies. J Glob Antimicrob
Author contributions Resist 2015;3:155–60.
Qin X, Tran BG, Kim MJ, Wang L, Nguyen DA, Chen Q, et al. A randomised,
double-blind, phase 3 study comparing the efficacy and safety of cef-
All authors are responsible for the work described in this tazidime/avibactam plus metronidazole versus meropenem for complicated in-
paper. All authors were involved in at least one of the following: tra-abdominal infections in hospitalised adults in Asia. Int J Antimicrob Agents
conception, design of work or acquisition, analysis, interpretation 2017;49:579–88.
Solomkin J, Hershberger E, Miller B, Popejoy M, Friedland I, Steenbergen J, et al.
of data, and drafting the manuscript and/or revising/reviewing the
Ceftolozane/tazobactam plus metronidazole for complicated intra-abdominal in-
manuscript for important intellectual content. All authors provided fections in an era of multidrug resistance: results from a randomized, dou-
final approval of the version to be published. All authors agree ble-blind, phase 3 trial (ASPECT-cIAI). Clin Infect Dis 2015;60:1462–71.
Tabak YP, Merchant S, Ye G, Vankeepuram L, Gupta V, Kurtz SG, et al. Incremen-
to be accountable for all aspects of the work in ensuring that
tal clinical and economic burden of suspected respiratory infections due to
questions related to the accuracy or integrity of any part of the multi-drug-resistant Pseudomonas aeruginosa in the United States. J Hosp Infect
work are appropriately investigated and resolved. 2019;103:134–41.
Wang HJ, Xing XZ, Qu SN, Huang CL, Zhang H, Wang H, et al. A randomized con-
trolled trial comparing the efficacy of tigecycline versus meropenem in the
Supplementary materials treatment of postoperative complicated intra-abdominal infections. Ann Palliat
Med 2021;10:1262–75.
World Health Organization. Global action plan on antimicrobial resistance. Geneva:
Supplementary material associated with this article can be World Health Organization; 2018.
found, in the online version, at doi:10.1016/j.ijid.2022.08.003. Xiao Y, Li L. China’s national plan to combat antimicrobial resistance. Lancet Infect
Dis 2016;16:1216–18.
Yang Q, Zhang H, Wang Y, Xu Y, Chen M, Badal RE, et al. A 10 year surveillance
References for antimicrobial susceptibility of Escherichia coli and Klebsiella pneumoniae in
community- and hospital-associated intra-abdominal infections in China. J Med
Cassini A, Högberg LD, Plachouras D, Quattrocchi A, Hoxha A, Simonsen GS, et al. Microbiol 2013;62:1343–9.
Attributable deaths and disability-adjusted life-years caused by infections with Yang Q, Zhang H, Yu Y, Kong H, Duan Q, Wang Y, et al. In vitro activity of
antibiotic-resistant bacteria in the EU and the European Economic Area in 2015: imipenem/relebactam against Enterobacteriaceae isolates obtained from in-
a population-level modelling analysis. Lancet Infect Dis 2019;19:56–66. tra-abdominal, respiratory tract, and urinary tract infections in China: Study for
Centers for Disease Control and Prevention. Antibiotic resistance threats in the Monitoring Antimicrobial Resistance Trends (SMART), 2015–2018. Clin Infect Dis
United States. Atlanta: United States Department of Health and Human Services. 2020;71:S427–35.
Centers for Disease Control and Prevention 2019. ZERBAXA® (ceftolozane and tazobactam). US Prescribing information. Rahway, NJ,
Chen Z, Wu J, Zhang Y, Wei J, Leng X, Bi J, et al. Efficacy and safety of tige- USA: Merck Sharp & Dohme Corp.; 2022.
cycline monotherapy vs. imipenem/cilastatin in Chinese patients with compli- Zhang H, Johnson A, Zhang G, Yang Y, Zhang J, Li D, et al. Susceptibilities of
cated intra-abdominal infections: a randomized controlled trial. BMC Infect Dis Gram-negative bacilli from hospital- and community-acquired intra-abdominal
2010;10:217. and urinary tract infections: a 2016–2017 update of the Chinese SMART study.
Clinical and Laboratory Standards Institute (CLSI). Performance standards for antimi- Infect Drug Resist 2019;12:905–14.
crobial susceptibility testing. 28th ed. Wayne: CLSI; 2018 M100Suppl. Zhang H, Kong H, Yu Y, Wu A, Duan Q, Jiang X, et al. Carbapenem susceptibilities of
Cosgrove SE. The relationship between antimicrobial resistance and patient out- Gram-negative pathogens in intra-abdominal and urinary tract infections: up-
comes: mortality, length of hospital stay, and health care costs. Clin Infect Dis dated report of SMART 2015 in China. BMC Infect Dis 2018a;18:493.
2006;42:S82–9. Zhang H, Yang Q, Liao K, Ni Y, Yu Y, Hu B, et al. Update of incidence and antimi-
Falagas ME, Tansarli GS, Karageorgopoulos DE, Vardakas KZ. Deaths attributable crobial susceptibility trends of Escherichia coli and Klebsiella pneumoniae isolates
to carbapenem-resistant Enterobacteriaceae infections. Emerg Infect Dis from Chinese intra-abdominal infection patients. BMC Infect Dis 2017;17:776.
2014;20:1170–5. Zhang H, Zhang G, Yang Y, Zhang J, Li D, Duan S, et al. Antimicrobial resistance com-
Hu B, Ye H, Xu Y, Ni Y, Hu Y, Yu Y, et al. Clinical and economic outcomes associ- parison of Klebsiella pneumoniae pathogens isolated from intra-abdominal and
ated with community-acquired intra-abdominal infections caused by extended urinary tract infections in different organs, hospital departments and regions of
spectrum beta-lactamase (ESBL) producing bacteria in China. Curr Med Res Opin China between 2014 and 2017. J Microbiol Immunol Infect 2021;54:639–48.
2010;26:1443–9. Zhang J, Zhou K, Zheng B, Zhao L, Shen P, Ji J, et al. High prevalence of ESBL-produc-
Jean SS, Lee YL, Liu PY, Lu MC, Ko WC, Hsueh PR. Multicenter surveillance of ing Klebsiella pneumoniae causing community-onset infections in China. Front
antimicrobial susceptibilities and resistance mechanisms among Enterobac- Microbiol 2016;7:1830.
terales species and non-fermenting Gram-negative bacteria from different in- Zhang Y, Wang Q, Yin Y, Chen H, Jin L, Gu B, et al. Epidemiology of carbapenem-re-
fection sources in Taiwan from 2016 to 2018. J Microbiol Immunol Infect sistant Enterobacteriaceae infections: report from the China CRE network. An-
2022;55:463–73. timicrob Agents Chemother 2018b;62 e01882–17.

165