Ceftolozane/tazobactam (C/T) was associated with a higher clinical cure rate for multidrug-resistant (MDR) Pseudomonas aeruginosa infections and lower all-cause mortality than aminoglycosides/polymyxins but was not distinguished from ceftazidime/avibactam (CZA) in a network meta-analysis1 (NMA) funded by C/T’s manufacturer, Merck. Lead author Hannah Collings, MS, director of Adelphi Values in Bollington, United Kingdom, and colleagues explained that they conducted the systematic literature review and NMA to assess real-world effectiveness of C/T in the absence of head-to-head trials and in view of the variability in comparative effectiveness study designs and methodology.
"Given these limitations, there is a clear need for a well-designed evidence synthesis to guide clinical decisions for patients with serious infections caused by highly resistant gram-negative pathogens like MDR P aeruginosa," the investigators indicated.
The NMA did not determine differences between the agents for either microbiological cure or duration of hospital stay. In addition, Collings and colleagues noted that there were insufficient comparator studies to statistically distinguish C/T from CZA on other clinical outcomes.
"The results of the NMA did not provide conclusive evidence for either C/T or CZA being more favorable compared with the other for any outcome explored," they acknowledged. "Larger prospective studies with standardized outcome measures are needed to further inform clinical decision-making."
From a review of 318 comparative effectiveness studies, 42 were included in a meta-analysis and 16 in the NMA. The meta-analysis of data from both single-arm and multiarm trials provided a single pooled estimate of efficacy across outcomes. The NMA was conducted with either common fixed effects (FE) or random effects (RE), depending on which model best fit each outcome data set, to rank the treatments by efficacy.
What You Need to Know
In the network meta-analysis, C/T was associated with higher clinical cure rates and lower all-cause mortality compared with aminoglycosides/polymyxins but showed no statistically significant advantage over CZA.
The analysis found no meaningful differences between C/T and CZA for clinical cure, microbiological cure, hospital length of stay, or mortality, largely due to heterogeneity across studies and insufficient comparator data.
Variability in study design, patient populations, and definitions of clinical cure, combined with the exclusion of highly resistant infections from many randomized trials, underscores the need for larger, well-designed prospective studies to guide treatment decisions for MDR P aeruginosa.
Collings and colleagues noted that the studies not only differed in the populations, treatment comparators, and end points but also that there were different definitions of clinical cure. These included survival at 30 days after therapy completion, resolution of infection signs and symptoms, success at day 30, clinical success at day 14, and 90-day survival. For the analysis, achievement of clinical cure was accepted by whichever measure was applied.
The investigators reported that the aminoglycosides/polymyxins (eg, colistin) demonstrated (on the RE model) lower odds of achieving clinical cure than C/T (OR, 0.308; 95% CI, 0.168-0.515) and a higher chance of contributing to all-cause mortality (OR, 1.651; 95%CI, 1.114-2.502). CZA did not evidence statistically significant difference from CT for clinical cure (OR, 0.920; 95%CI, 0.565-1.595).
Although C/T was determined to have lower odds of all-cause mortality than aminoglycosides/polymyxins, the mortality rate associated with MDR P aeruginosa infection treated with C/T remains substantial: estimated at 22% within 30 days with both the FE and RE models (CI, 20%-24% and 18%-26%, respectively).
Among the challenges the investigators confronted in conducting meta-analyses to rank treatments for MDR P aeruginosa infection is that most noninferiority randomized controlled trials (NI-RCTs) exclude patients with suspected or documented infections caused by pathogens resistant to the active comparator.
"This has significant implications for clinical practice, as data from NI-RCTs do not offer clinicians enough information to determine the best therapies for treating highly resistant gram-negative infections, such as MDR P aeruginosa," Collings and colleagues observed.
Reference
1. Collings H, Stone M, Chrysostomou M, et al. A systematic literature review and (network) meta-analysis of the effectiveness of ceftolozane/tazobactam versus aminoglycosides/polymyxins and ceftazidime/avibactam for treating adult patients with multidrug-resistant Pseudomonas aeruginosa infections. Antimicrob Agents Chemother. 2026;70(2):e0073525. doi:10.1128/aac.00735-25
