Does cefiderocol heteroresistance explain the discrepancy between the APEKS-NP and CREDIBLE-CR clinical trial results?

Cefiderocol is a novel siderophore-cephalosporin conjugate antibiotic approved for treatment of Gram-negative bacterial infections. In-vitro laboratory testing found cefiderocol to be effective against many species, including carbapenem-resistant and multidrug resistant isolates¹. As part of the APEKS-NP clinical trial,² cefiderocol was evaluated in the treatment of health-care-associated pneumonia caused largely by carbapenem-susceptible, cephalosporin-resistant, extended-spectrum β-lactamase (ESBL) producers, or carbapenem-susceptible, non-ESBL strains (referred to from hereon as susceptible). Cefiderocol was non-inferior to meropenem in this trial, suggesting that cefiderocol is a potential option for the treatment of patients with nosocomial pneumonia, including those caused by multidrug-resistant Gram-negative bacteria.²

However, concern arose following publication of the CREDIBLE-CR trial,³ which involved infections caused by carbapenem-resistant Gram-negative bacteria. Cefiderocol had similar efficacy compared with the best available therapy for the treatment of pneumonia, urinary tract infections, and bloodstream infections. But despite 95% of isolates in the trial demonstrating a minimum inhibitory concentration of 4 μg/mL or less, cefiderocol was associated with a higher rate of all-cause mortality, particularly in infections with Acinetobacter.³

The incongruity between the performance of cefiderocol in APEKS-NP and CREDIBLE-CR is a major outstanding question in the field.^4–8 As stated by Heil and Tamma,⁴ “how do we reconcile the seemingly conflicting mortality data from these studies?”. Cefiderocol is now being used increasingly as a last-line agent against carbapenem-resistant strains (appendix), despite concerns in treating such strains raised by the CREDIBLE-CR study and uncertainty about when to rely on this new antibiotic. This uncertainty heightens the need to elucidate the basis of the discrepancy of the results between APEKS-NP and CREDIBLE-CR, such that the scope of cefiderocol utility can be defined and best guide patient care.

We previously revealed a correlation between cefiderocol heteroresistance among carbapenem-resistant isolates and the increased all-cause mortality observed in the CREDIBLE-CR trial.⁹ Heteroresistance is a phenomenon in which only a minor subpopulation of cells are resistant to a given antibiotic.¹⁰ In the presence of a given antibiotic, the resistant cells are selected, predominate, and can cause treatment failure during in vivo murine infection.¹⁰ On the basis of these previous results, we hypothesised that frequency of cefiderocol heteroresistance could explain the discordant findings of the APEKS-NP and CREDIBLE-CR trials.

Here, we investigated the frequency of cefiderocol heteroresistance among susceptible or cephalosporin-resistant, carbapenem-susceptible bacteria that were predominant in the APEKS-NP trial.

The resistant subpopulation of cells in heteroresistance can be detected using the population analysis profile (PAP) test (appendix).^9,10 Using PAP on isolates collected by the Emory Antibiotic Resistance Center’s Investigational Clinical Microbiology Core, we observed that susceptible isolates exhibited no or low rates of cefiderocol heteroresistance (appendix). Cephalosporin-resistant bacteria mostly exhibited increased rates of heteroresistance, but lower than those of carbapenem-resistant strains. These differences in rates of cefiderocol heteroresistance correlated with the mortality data from the APEKS-NP and CREDIBLE-CR trials, across the bacterial species tested (appendix). These data suggest that the lower rates of cefiderocol heteroresistance in susceptible and cephalosporin-resistant isolates that predominated the APEKS-NP trial might explain the enhanced efficacy of the drug in that study compared with the CREDIBLE-CR trial.

Cefiderocol is a welcome and crucial addition to the antibiotic armamentarium. However, our findings raise concern that cefiderocol could often fail in treating infections caused by some species of carbapenem-resistant bacteria, especially Acinetobacter baumannii, because of high rates of heteroresistance. Due to the very low numbers of resistant cells in cefiderocol heteroresistance (often <1 cell in 10 000), this pheno-type was largely undetected by recommended antibiotic susceptibility testing among carbapenem-resistant isolates,⁹ and a similar phenomenon was observed among the susceptible and cephalosporin-resistant isolates in this study. As the use of cefiderocol therapy increases, undetected heteroresistance should be carefully considered and monitored as the utility of this antibiotic is established.