Introduction

Piperacillin/tazobactam, a β-lactam/β-lactamase inhibitor combination, is commonly used for complicated urinary tract infections (UTIs). However, extended-spectrum β-lactamases limit their effectiveness. New therapeutic options are needed to combat β-lactam-resistant pathogens.

Cefepime, a fourth-generation cephalosporin, combined with enmetazobactam (formerly AAI101), a novel β-lactamase inhibitor, shows promise against β-lactamase-producing gram-negative pathogens. In vitro studies suggest this combination is more potent than piperacillin/tazobactam against extended-spectrum β-lactamase producers. Previous studies indicated that cefepime and enmetazobactam share similar pharmacokinetic profiles, including urinary excretion and half-lives.

Aim

• A randomized, phase 3 clinical trial was conducted to assess whether cefepime/enmetazobactam was non-inferior to piperacillin/tazobactam in efficacy and safety for treating complicated UTIs or acute pyelonephritis in adults. This trial was necessary for the potential registration of cefepime/enmetazobactam as a new treatment option.

Patient Profile

• Adults aged 18 or older with a clinical diagnosis of complicated UTI or acute pyelonephritis caused by gram-negative pathogens were included.

  Methods

  Study Design

• ALLIUM (A Phase 3, Randomized, Double-Blind, Multi-Center Study to Evaluate the Efficacy, Safety, and Tolerability of Cefepime-Enmetazobactam Compared to Piperacillin/Tazobactam in adults for the Treatment of Complicated Urinary Tract Infections, Including Acute Pyelonephritis). This study was designed to evaluate whether cefepime/enmetazobactam was non-inferior to piperacillin/tazobactam for treatment efficacy over 7 to 14 days.
• Randomization, Stratification, and Masking: Participants were randomly allocated to two treatment groups in a 1:1 ratio using a centralized computer-generated system. Stratification ensured balanced groups based on infection type, prior antibiotic use, and geographic region. Study drugs were prepared by an unblinded pharmacist and administered by research staff unaware of treatment assignments.
• Infection type included acute pyelonephritis, complicated urinary tract infection (with or without removable source), or other risk factors (e.g., anatomical abnormalities, neurogenic bladder, azotemia). Prior antibiotic use was categorized as short-acting antibiotics within 24 hours of baseline sampling or no prior antibiotics. Geographic regions included the Baltics, Eastern Europe, Latin America, South Africa, the United States, and Western Europe.

Interventions 

• Cefepime 2g/enmetazobactam 0.5g (n=520)
• Piperacillin 4g/tazobactam 0.5g (n=521)
• Both groups are treated via 2-hour infusion every 8 hours

Endpoints

Primary Outcome: The proportion of patients achieving clinical cure—defined as the complete resolution of baseline signs and symptoms present at screening—and microbiological eradication, indicated by a reduction of the qualifying baseline pathogen to less than 10³ CFU/mL in urine, was assessed on day 14.

Secondary Outcomes: Secondary outcomes included clinical cure and microbiological eradication at various time points (day 3, end of treatment, day 14, and day 21) and in specific subgroups based on age, sex, infection type, baseline creatinine clearance, prior antibiotic therapy, baseline diabetes, Charlson Comorbidity Index score, bacteremia, race, region, country category, and presence of extended-spectrum β-lactamase–producing pathogens.

Statistical Analyses: No interim analyses were performed. The stratified Newcombe method was used to calculate 95% confidence intervals (CIs) for the proportional difference in overall treatment success between groups, accounting for stratification factors (infection type, prior antibiotic therapy, and region). Non-inferiority was established if the lower CI limit exceeded -10%. Superiority was determined if the treatment difference was positive and the lower CI limit was greater than zero, without adjusting for multiple comparisons. Secondary endpoints were assessed for superiority using a two-sided 95% confidence interval (CI). A lower CI bound greater than zero indicated statistical significance. Due to multiple comparisons, secondary endpoint findings should be interpreted cautiously. Missing data and patient dropouts were considered treatment failures and included in overall success rate calculations. Data analysis was performed using SAS version 9.3. Post-hoc analyses employed the chi-square test with a significance level of p ≤ 0.05.

Result

• Patient Disposition: Between September 24, 2018, and November 2, 2019, a total of 1,041 patients were randomized, with 1,034 receiving at least one dose of the study drug. The median treatment duration was 8.0 days. 5.2% (27/516) and 4.1% (21/518) in the Cefepime/Enmetazobactam group and Piperacillin/Tazobactam didn't complete the therapy.
• A total of 678 patients were included in the primary set analysis with gram-negative bacteria susceptible to both study drugs in urine or blood. Most, 51.5% had acute pyelonephritis, while 21.8% had complicated UTIs with a removable source and 26.7% without. Bacteremia occurred in 11% of cefepime/enmetazobactam patients and 8.4% of piperacillin/tazobactam patients. Enterobacterales infections were common (95.6%), with E. coli being the primary pathogen (76.4%), followed by Klebsiella pneumoniae (9.7%), Proteus mirabilis (5.6%), Pseudomonas aeruginosa (3.5%), and Enterobacter cloacae (1.5%).
• Primary Outcome: At day 14, 79.1% of patients in the cefepime/enmetazobactam group achieved the primary outcome of clinical cure and microbiological eradication, compared to 58.9% in the piperacillin/tazobactam group—a difference of 21.2% (95% CI, 14.3%-27.9%) (Figure 1). This result demonstrates that cefepime/enmetazobactam was not only non-inferior but also superior to piperacillin/tazobactam.

Figure 1: Primary Outcomes: Clinical Cure and Microbiological Eradication Rates at Day 14

• Secondary Outcomes: At day 14, clinical cure rates were similar between groups: 92.5% in the cefepime/enmetazobactam group and 88.9% in the Piperacillin/Tazobactam group, with a treatment difference of 3.5% (95% CI, -1.0% to 8.0%). However, the cefepime/enmetazobactam group had a significantly higher microbiological eradication rate (82.9% vs. 64.9%; treatment difference, 19.0% [95% CI, 12.3%-25.4%]). In patients with extended-spectrum β-lactamase–producing pathogens, 73.7% of those in the cefepime/enmetazobactam group and 51.5% in the piperacillin/tazobactam group achieved both clinical cure and microbiological eradication by day 14, with a treatment difference of 30.2% (95% CI, 13.4%-45.1%). Improvements in the composite outcome were greater with cefepime/enmetazobactam at day 21 (10.7% [95% CI, 3.4% to 17.8%]), but not at day 3 (4.1% [95% CI, -0.6% to 8.9%]) or at the end of treatment (-1.3% [95% CI, -5.3% to 2.9%]).
• Among patients with ESBL-producing pathogens (20.9%), cefepime/enmetazobactam demonstrated superior efficacy compared to piperacillin/tazobactam for the composite outcome (73.7% vs. 51.5%) at day 14, resulting in a treatment difference of 30.2% (95% CI, 13.4%-45.1%).
• Patients Without Exclusion Based on Susceptibility: In analyses including all patients (regardless of resistance to both study drugs or missing minimum inhibitory concentration [MIC] determinations), the cefepime/enmetazobactam group showed a significantly higher composite outcome of clinical cure and microbiological eradication at day 14 (78.6%) compared to the piperacillin/tazobactam group (58.7%), with a treatment difference of 20.7% (95% CI, 14.1%-27.0%). The microbiological eradication rate was also higher for cefepime/enmetazobactam (82.7%) versus piperacillin/tazobactam (64.5%), with a treatment difference of 19.0% (95% CI, 12.8%-25.0%). Among all treated patients, the composite outcome of clinical cure and microbiological eradication at day 14 was significantly higher in the cefepime/enmetazobactam group (59.1%) compared to the piperacillin/tazobactam group (43.4%), with a treatment difference of 15.6% (95% CI, 9.5%-21.5%) (Figure 2). Notably, approximately 25% of patients in both groups lacked a qualifying baseline pathogen.

Figure 2: Results of all patients regardless of drug resistance or minimum inhibitory concentration

• Exploratory Outcomes and Post Hoc Analyses data: In the primary analysis set, microbiological recurrence was lower with cefepime/enmetazobactam compared to piperacillin/tazobactam. Clinical cure rates at day 14 were similar between the two groups in patients who received at least one dose. However, in the primary analysis set excluding patients with piperacillin/tazobactam, MIC >16 μg/mL and/or ESBL-producing Enterobacterales, cefepime/enmetazobactam showed significantly better outcomes compared to piperacillin/tazobactam.

Table 1: Data on exploratory outcomes and post hoc analysis on day 14

Outcome	Cefepime/ Enmetazobactam	Piperacillin/Tazobactam	Difference	95% CI
Exploratory Outcomes
Microbiological Recurrence (Day 14)	11.3%	29.4%	-	-
Post Hoc Analyses
Clinical Cure (Day 14)	91.3%	87.8%	3.70%	-0.06% to 7.5%
Clinical Cure (Excluding Resistant & Specific Pathogens	80.9%	60.7%	20.20%	12.4% to 27.7

Conclusion

Cefepime/enmetazobactam demonstrated superior efficacy over piperacillin/tazobactam in treating gram-negative complicated UTIs and acute pyelonephritis, as evidenced by the primary outcome of clinical cure and microbiological eradication. Further research is warranted to establish cefepime/enmetazobactam's optimal role in managing these infections.

Key Points/Discussion

• Resistance to traditional UTI drugs (aminoglycosides, fluoroquinolones, β-lactams) is increasing.
• Increasing resistance, especially due to extended-spectrum β-lactamases, has reduced the effectiveness of common UTI drugs like aminoglycosides, fluoroquinolones, and β-lactams. Carbapenem resistance has also risen. New therapies that spare carbapenems are needed.
• Cefepime/enmetazobactam was non-inferior and superior to piperacillin/tazobactam for clinical and microbiological cure in complicated UTI and acute pyelonephritis. It may be a suitable empirical therapy for suspected gram-negative complicated UTIs.
• Cefepime/enmetazobactam is a potential option for infections at risk for ESBLs, whereas other β-lactam/β-lactamase inhibitors (ceftazidime/avibactam, meropenem/ vaborbactam, imipenem/cilastatin/relebactam) are reserved for carbapenem-resistant infections.
• The recurrence rate was lower in cefepime/enmetazobactam compared to piperacillin/tazobactam.

JAMA 2022; 328(13): 1304-14