In a Pinch: Cefiderocol for CRAB Infections

ABSTRACT

Carbapenem-resistant Acinetobacter baumannii (CRAB) can cause significant infections with limited treatment options available. Falcone et al. (https://doi.org/10.1128/aac.02142-21) describe a single-center retrospective study comparing clinical outcomes among patients with CRAB infections treated with cefiderocol-containing versus colistin-containing regimens. Patients who received cefiderocol-containing regimens had lower 30-day mortality, though there are several limitations raised here, which make interpretation and applicability difficult.

TEXT

In this issue of AAC, Falcone et al. (1) present a single-center retrospective analysis describing clinical outcomes among patients with carbapenem-resistant Acinetobacter baumannii (CRAB) infections when treated with cefiderocol (FDC)-containing regimens versus colistin (COL)-containing regimens. CRAB infections represent a significant public health concern, with limited available treatment options (2, 3). One difficulty in both treating and interpreting treatment studies is that A. baumannii is often recovered from the respiratory tract and wounds; thus, it is often unclear if it is a true pathogen or a colonizing organism. In addition, poor outcomes may not be due to infection from the organism, but rather host factors as it is a predominately nosocomial pathogen (4). Assuming a clear CRAB infection, there are also limited studies evaluating therapy directed against this organism, with no definitive “best therapy” (3). Recent guidance recommends high-dose ampicillin-sulbactam in combination with a second active agent for the treatment of moderate to severe CRAB infections and suggest that FDC may be used in combination for CRAB infections refractory to other antibiotics (5).

Two randomized controlled trials (RCT) evaluating FDC included patients treated for CRAB infections. In the phase 3 RCT, CREDIBLE-CR, the subgroup of patients with infections caused by CRAB who were treated with FDC had increased mortality compared to those treated with best available therapy (often polymyxin-based regimens) (49% versus 18%) (6). In the phase 3 RCT evaluating patients with nosocomial pneumonia (APEKS-NP), clinical outcomes were similar among patients with CRAB whether treated with FDC or with extended-infusion meropenem (7). However, both trials included small numbers of patients with CRAB and heterogenous comparator treatment regimens, making interpretation difficult.

In the present study by Falcone et al., the population was ill, with the vast majority being in the ICU (89% among both arms), slightly more than half on mechanical ventilation (53.2% in FDC-containing group versus 55.8% in COL-containing group), and a large proportion (40.4% in FDC-containing group, 37.7% in COL-containing group) having concurrent COVID-19. Unlike the prior RCTs, this study’s target population was patients with CRAB infections. The authors found a lower risk of 30-day all-cause mortality among those treated with FDC-containing regimens compared to COL-containing therapy, with the benefit driven by patients with bloodstream infection (BSI), as this was not observed in the subset of patients with ventilator-associated pneumonia (VAP). Strengths of this study include a critically ill population, the attempt to account for the non-randomized treatment assignment using inverse probability of treatment weighting (IPTW), adjusting the analysis by severity of illness and type of infection (BSI, VAP, other), and that susceptibility testing was assessed using multiple breakpoints (CLSI, EUCAST, FDA).

The primary outcome observed by Falcone and colleagues, two of whom disclose conflicts of interest with Shionogi—the manufacturer of FDC, was lower mortality among patients treated with FDC-containing regimens compared to COL-containing regimens. The majority of patients among the COL-containing arm received COL plus tigecycline (40/77, 52%), followed by COL monotherapy (13/77, 17%). The pharmacokinetics of tigecycline are unfavorable for BSI therapy, and tigecycline has been found to have increased mortality versus alternative agents when used to treat VAP (8, 9). In addition, very few patients received COL in combination with ampicillin-sulbactam (7/77, 11% all of whom had VAP), and when this combination was used the dose was lower than often recommended for CRAB infections—3 g every 6 h (5, 10–12). Other studies have demonstrated improved outcomes and mortality when ampicillin-sulbactam is included in treatment of CRAB and at relatively high doses (13, 14). Therefore, it is difficult to interpret whether worse outcomes were observed among those treated with COL-containing regimens due to the antibiotics received, or due to superior treatment when given FDC-containing regimens.

A concern raised from prior studies of FDC has been microbiologic failure and the development of resistance (both in CRAB and in other organisms). In CREDIBLE-CR, among the CRAB isolates, 13% (5/39) of isolates exposed to FDC showed at least a 4-fold increase in MIC (6). Indeed, prior siderophore-conjugated antimicrobials have not been successful in part due to development of resistance (15–17). Falcone et al. examined a secondary outcome of microbiologic failure and development of resistance to FDC from CRAB isolates. In this study, most patients were in the ICU and thus had weekly blood and respiratory surveillance cultures which was standard of practice in the study hospital, allowing for microbiologic failure rates to be followed in 120/124 patients. Although not statistically significant, microbiologic failure was seen more frequently among patients who received an FDC-containing regimen compared to COL-containing regimen (8/46, 17.4% versus 5/74, 6.8%; P = 0.079). Among the eight patients who received FDC and experienced microbiologic failure, all had BSI (8/27, 30% of BSI patients), and six of them received FDC monotherapy. Among the microbiologic failures, 50% developed resistance to FDC on repeat in vitro susceptibility testing. The frequency of observed microbiologic failure among FDC-containing treatment regimens, in conjunction with most patients receiving FDC monotherapy, raises concerns for the role of this agent as a monotherapeutic agent for these already highly resistant infections.

The encouraging signal of decreased mortality among patients with CRAB infections who received FDC-containing therapies juxtaposed with the study limitations raised here makes it challenging to determine how this study should impact clinical practice. Both this single-center retrospective study and the previous RCTs are limited by overall small numbers of patients and the inherent challenge of assessing clinical success with Acinetobacter infections, particularly with the high rate of concurrent SARS-CoV-2 infections in the present study. In addition, this study adds to concerns about microbiologic failure and development of resistance in siderophore-conjugated antimicrobials. An additional consideration regarding the use of FDC surrounds antimicrobial susceptibility testing. Accurate MIC testing requires specialized iron-depleted media and local validation for testing as it is not yet available on commercial automated platforms, which may require isolates to be sent out for testing, delaying results available to clinicians (18). Specifically for A. baumannii, testing is further complicated by errors in disk diffusion and difficult broth microdilution interpretation, and current CLSI breakpoints remain investigational (18, 19). While this study does provide further evidence for the potential role of FDC for treatment of CRAB infections, more data are still needed regarding long-term risk of resistance development and clinical outcomes (20). In the interim, FDC represents a salvage option that we believe should be used in combination with other active antimicrobials (e.g., polymyxin B, minocycline) for the treatment of CRAB infections.

The views expressed in this article do not necessarily reflect the views of the journal or of ASM.