Cefiderocol versus high-dose, extended-infusion meropenem for the treatment of Gram-negative nosocomial pneumonia (APEKS-NP): a randomised, double-blind, phase 3, non-inferiority trial

Background: Nosocomial pneumonia due to multidrug-resistant Gram-negative pathogens poses an increasing challenge. We compared the efficacy and safety of cefiderocol versus high-dose, extended-infusion meropenem for adults with nosocomial pneumonia. Methods: We did a randomised, double-blind, parallel-group, phase 3, non-inferiority trial in 76 centres in 17 countries in Asia, Europe, and the USA (APEKS-NP). We enrolled adults aged 18 years and older with hospital-acquired, ventilator-associated, or health-care-associated Gram-negative pneumonia, and randomly assigned them (1:1 by interactive response technology) to 3-h intravenous infusions of either cefiderocol 2 g or meropenem 2 g every 8 h for 7–14 days. All patients also received open-label intravenous linezolid (600 mg every 12 h) for at least 5 days. An unmasked pharmacist prepared the assigned treatments; investigators and patients were masked to treatment assignment. Only the unmasked pharmacist was aware of the study drug assignment for the infusion bags, which were administered in generic infusion bags labelled with patient and study site identification numbers. Participants were stratified at randomisation by infection type and Acute Physiology and Chronic Health Evaluation II (APACHE II) score (≤15 and ≥16). The primary endpoint was all-cause mortality at day 14 in the modified intention-to-treat (ITT) population (ie, all patients receiving at least one dose of study drug, excluding patients with Gram-positive monomicrobial infections). The analysis was done for all patients with known vital status. Non-inferiority was concluded if the upper bound of the 95% CI for the treatment difference between cefiderocol and meropenem groups was less than 12·5%. Safety was investigated to the end of the study in the safety population, which included all patients who received at least one dose of study drug. This trial is registered with ClinicalTrials.gov, NCT03032380, and EudraCT, 2016-003020-23. Findings: Between Oct 23, 2017, and April 14, 2019, we randomly assigned 148 participants to cefiderocol and 152 to meropenem. Of 292 patients in the modified ITT population, 251 (86%) had a qualifying baseline Gram-negative pathogen, including Klebsiella pneumoniae (92 [32%]), Pseudomonas aeruginosa (48 [16%]), Acinetobacter baumannii (47 [16%]), and Escherichia coli (41 [14%]). 142 (49%) patients had an APACHE II score of 16 or more, 175 (60%) were mechanically ventilated, and 199 (68%) were in intensive care units at the time of randomisation. All-cause mortality at day 14 was 12·4% with cefiderocol (18 patients of 145) and 11·6% with meropenem (17 patients of 146; adjusted treatment difference 0·8%, 95% CI −6·6 to 8·2; p=0·002 for non-inferiority hypothesis). Treatment-emergent adverse events were reported in 130 (88%) of 148 participants in the cefiderocol group and 129 (86%) of 150 in the meropenem group. The most common treatment-emergent adverse event was urinary tract infection in the cefiderocol group (23 patients [16%] of 148) and hypokalaemia in the meropenem group (23 patients [15%] of 150). Two participants (1%) of 148 in the cefiderocol group and two (1%) of 150 in the meropenem group discontinued the study because of drug-related adverse events. Interpretation: Cefiderocol was non-inferior to high-dose, extended-infusion meropenem in terms of all-cause mortality on day 14 in patients with Gram-negative nosocomial pneumonia, with similar tolerability. The results suggest that cefiderocol is a potential option for the treatment of patients with nosocomial pneumonia, including those caused by multidrug-resistant Gram-negative bacteria. Funding: Shionogi.