Innovative Antibiotics: The Experience of Cefiderocol in Carbapenem-Resistant Gram-Negative Infections

Introduction

Cefiderocol, a parenteral siderophore cephalosporin, has been approved by both the U.S. Food and Drug Administration (2019) and the European Medicines Agency (2020) for the treatment of severe infections caused by multidrug-resistant (MDR) Gram-negative pathogens for which there are limited therapeutic options.¹

Cefiderocol is one of the most recent additions to the beta-lactam class. It is a key component in the therapeutic arsenal against MDR Gram-negative bacteria, which pose a critical and escalating challenge in infectious diseases and clinical microbiology.²

From a molecular standpoint, cefiderocol exhibits a unique chemical structure that confers stability against a broad spectrum of beta-lactamases, including extended-spectrum beta-lactamases (ESBLs) and carbapenemases, including metallo-beta-lactamases (MBLs), for which limited effective agents are currently available. The molecule’s innovation lies in its siderophore moiety, which binds to ferric iron ions and exploits bacterial iron transport systems to cross the outer membrane into the periplasmic space. This mechanism bypasses the limitation of passive diffusion characteristic of conventional beta-lactams that rely on porin channels, thereby enhancing intracellular delivery.³

Upon entering the periplasm, cefiderocol binds to penicillin-binding proteins (PBPs), thereby disrupting peptidoglycan synthesis and cell wall assembly, exerting bactericidal activity. It demonstrates a high affinity for PBP3 across a range of fermenting and non-fermenting Gram-negative species, as well as notable activity against PBP2 in Klebsiella pneumoniae and PBP1 in Pseudomonas aeruginosa. However, cefiderocol exhibits no in vitro activity against Gram-positive cocci or obligate anaerobes.^4,5

The clinical efficacy and safety of cefiderocol have been evaluated in four randomized clinical trials (RCTs). The multicenter, non-inferiority Phase II trial, APEKS-cUTI, compared cefiderocol with imipenem/cilastatin in the treatment of complicated urinary tract infections in hospitalized patients at risk of MDR Gram-negative infections. At the test of cure (7 days after treatment), cefiderocol showed a higher composite clinical and microbiological response rate of 73%, versus 55% in the imipenem/cilastatin group. This difference in response rates was statistically significant (18.6%). However, patients with carbapenem-resistant infections were excluded, which limited applicability to this population.⁶

The Phase III APEKS-NP trial demonstrated non-inferiority of cefiderocol compared with high-dose meropenem in patients with nosocomial pneumonia due to Gram-negative pathogens, reporting a 28-day all-cause mortality of 21% versus 20.5% (absolute risk difference 0.5%, 95% CI −8.7 to 9.8).⁷

The open-label, randomized, international phase III CREDIBLE-CR trial compared the efficacy and safety of cefiderocol with the best available therapy (BAT) for treating severe infections caused by carbapenem-resistant Gram-negative bacteria, including healthcare-associated and hospital-acquired pneumonia (HAP), ventilator-associated pneumonia (VAP), complicated urinary tract infections, and bloodstream infections (BSI). While overall clinical and microbiological outcomes were comparable between cefiderocol and BAT, a subgroup analysis revealed an increased mortality rate at the end of study in patients treated with cefiderocol and infected with carbapenem-resistant Acinetobacter spp. (49% vs. 18%). This was likely due to the cefiderocol group having a higher baseline risk profile, characterized by intensive care unit (ICU) admission at randomization, severe renal impairment, ongoing shock and shock episodes within 31 days prior to enrolment.⁸

Finally, the results of the most recent randomized trial, the GAME CHANGER trial, have been published. This study compared cefiderocol versus standard of care for the treatment of healthcare-associated Gram-negative BSI. With the exception of the APEKS-cUTI trial, this is the largest trial conducted to date, enrolling 504 patients. Cefiderocol was shown to be non-inferior in terms of 14-day mortality (using a 10% margin), including in a subgroup analysis of infections caused by carbapenem-resistant bacteria, in which 30-day all-cause mortality was 33% (21/64) in the cefiderocol group and 25% (16/63) in the standard-of-care group (absolute risk difference 7%, 95% CI –8 to 23). However, superiority was not demonstrated in either analysis. The authors conclude that further studies are needed, with particular focus on infections caused by MBL–producing pathogens, which may exhibit elevated cefiderocol MICs even without prior exposure to the drug.⁹

Infections caused by carbapenem-resistant Gram-negative bacteria are increasing worldwide and are associated with high mortality rates and severe clinical outcomes, likely due to limited availability of effective treatment options. Real-world data on the use of cefiderocol remain paramount in refining its clinical utility and optimal positioning within treatment algorithms for MDR Gram-negative infections, in particular carbapenem resistant Enterobacterales (CRE), carbapenem resistant Acinetobacter baumannii (CRAB), difficult to treat (DTR) P. aeruginosa, Stenotrophomonas malthophilia and other non-fermenting Gram-negative bacteria. This narrative review aims to summarize the existing real-life evidence on the use of cefiderocol in treating severe Gram-negative, carbapenem-resistant infections.

Susceptibility Testing

The reference method for cefiderocol susceptibility testing is broth microdilution. However, accurate assessment of cefiderocol activity remains challenging due to its unique “Trojan horse” mechanism of action. During infection, the host immune response limits free iron availability, leading to the upregulation of bacterial iron transport systems, which cefiderocol exploits to actively penetrate the bacterial cell. To better replicate in vivo conditions, susceptibility testing requires the use of iron-depleted, cation-adjusted Mueller–Hinton broth.¹⁰ This requirement poses practical challenges for many clinical microbiology laboratories, as the procedure is time-consuming and difficult to implement routinely.¹¹ Conversely, disk diffusion is more rapid and widely used; however, it has shown limited reliability and reproducibility, particularly for the evaluation of A. baumannii susceptibility.^12,13

The Clinical and Laboratory Standards Institute (CLSI) defines cefiderocol susceptibility breakpoints as MIC ≤ 4 µg/mL for Enterobacterales, P. aeruginosa, and A. baumannii, and MIC ≤ 1 µg/mL for Stenotrophomonas maltophilia, as shown in Table 1.¹⁴ In contrast, the European Committee on Antimicrobial Susceptibility Testing (EUCAST) provides breakpoints only for Enterobacterales and P. aeruginosa, which are considered susceptible with MIC ≤ 2 µg/mL.¹⁵ These differences between CLSI and EUCAST breakpoints further complicate the interpretation and comparison of results across studies.

Table 1 Cefiderocol Susceptibility Interpretative Criteria Proposed by CLSI and EUCAST

Methods

A comprehensive literature search was conducted in PubMed and ClinicalTrials.gov to identify studies reporting real-world use of cefiderocol. The search was performed using the keyword “cefiderocol” and included studies published up to October 2025. For the purpose of this review, real-life studies were defined as observational studies conducted in routine clinical practice outside the setting of randomized controlled trials.

Studies were included in the final analysis if they reported original data on the real-world use of cefiderocol for the treatment of infections caused by carbapenem-resistant bacteria. Articles were excluded if they were not written in English or if they were review articles. Study selection was performed independently by two reviewers.

The initial search retrieved 1154 records. After screening titles and abstracts, 160 articles were considered eligible for full-text assessment. Of these, 20 met the predefined inclusion criteria and were ultimately included in the analysis.

Acinetobacter baumannii

CRAB is the most alarming pathogen on the WHO’s priority list.¹⁶ It causes severe infections, including VAP and BSI, with mortality rates ranging from 40% to 70%.^17,18 CRAB-related infections are of particular concern as they most often affect critically ill patients, such as those admitted to ICU or burn units. Until a few years ago, the therapeutic approach to CRAB infections relied on the use of polymyxins in combination with other agents. However, this approach was associated with significant limitations, including serious side effects, particularly nephrotoxicity, and difficulties in achieving satisfactory PK/PD targets for microbiological eradication.¹⁹

Cefiderocol therefore represents a promising alternative to these treatment regimens. Although current international guidelines recommend a cautious approach to the use of cefiderocol for CRAB infections,^20,21 largely based on the results of the CREDIBLE-CR trial,⁸ real-world studies have provided additional data suggesting a potential role for this agent in selected clinical scenarios.

The first real-life data on the use of cefiderocol in CRAB infections come from studies conducted mainly in Italian centers during the SARS-CoV-2 pandemic. During this period, infections caused by MDR bacteria increased, likely due to reduced adherence to infection control measures and the inappropriate use of antibiotics.²²

Results from real-life studies focusing on CRAB are summarized in Table 2.

Table 2 Real-Life Studies Evaluating Cefiderocol Activity Against Carbapenem Resistant Acinetobacter baumannii (CRAB)

Pascale et al presented a case series of 107 patients admitted to the ICU with respiratory failure due to SARS-CoV-2, complicated by BSI or lower respiratory tract infections (LRTI) caused by CRAB. Of these patients, 39% were treated with cefiderocol. No statistically significant differences in 28-day mortality were observed between patients treated with cefiderocol and those receiving alternative regimens, including colistin-based therapies.²³

In a separate retrospective cohort study of 104 ICU patients with BSI or LRTI caused by CRAB during the pandemic, Oliva et al observed a lower 30-day mortality among patients treated with cefiderocol compared to those treated with colistin (22.2% vs 68.4%, p = 0.008). Furthermore, adverse events were much more frequent in the colistin-treated group (38.8% vs 10%, p < 0.0001).²⁴

Russo et al presented a case series involving 73 ICU patients with SARS-CoV-2 and VAP caused by CRAB, 19 of whom (26%) were treated with cefiderocol-containing regimens. They reported that cefiderocol therapy, particularly when combined with fosfomycin, was independently associated with improved 30-day survival.²⁵

The study published by Falcone et al found less favorable results regarding the use of cefiderocol in the treatment of VAP caused by CRAB. The study included 124 patients with BSI, VAP or other invasive CRAB infections in Italy during SARS-CoV-2 pandemic. Of these patients, 47 (37.9%) were treated with cefiderocol and 77 (62.1%) with colistin-containing regimens. Thirty-day all-cause mortality was higher in the colistin-treated group (55.8% vs. 34%, p = 0.018). Among patients with BSI, 30-day mortality was lower in the cefiderocol group (25.9% vs. 57.7%, p = 0.007, respectively). However, no significant differences were observed in patients with VAP. A higher rate of microbiological failure was observed in the cefiderocol group (17.4% vs. 6.8%), and the emergence of cefiderocol-resistant CRAB strains was reported. This study also confirmed that cefiderocol-containing regimens are less nephrotoxic than colistin-based regimens.²⁶

Another study by Bavaro et al demonstrated the efficacy of cefiderocol in treating CRAB-related BSI. This study included 118 patients: of these, 75 (63%) received colistin-containing regimens and 43 (37%) received cefiderocol-containing regimens. The 30-day all-cause mortality rate was significantly lower in the cefiderocol group (40% vs. 59%; p = 0.045).²⁷

Regarding VAP caused by CRAB, a single-center prospective observational study of 90 patients compared first-line treatment with colistin- and cefiderocol-based regimens. Overall, 50 patients received colistin-based intravenous regimens and 40 received cefiderocol-based regimens, both in combination with inhaled colistin. Clinical failure was lower in the cefiderocol group (25% vs. 48%, p = 0.02). Although 30-day mortality was also lower in the cefiderocol group, the difference did not reach statistical significance (35% vs. 52%). In multivariable Cox regression analysis, comorbidity burden independently predicted clinical failure (Charlson index aHR = 1.21, 95% CI = 1.04–1.42, p = 0.01), whereas timely targeted antibiotic therapy (aHR = 0.40, 95% CI = 0.19–0.84, p = 0.01) and cefiderocol-based first-line regimens (aHR = 0.38, 95% CI = 0.17–0.85, p = 0.02) were associated with a reduced risk of clinical failure.²⁸

In another study of 111 patients with severe CRAB infections (pneumonia and BSI), Mazzitelli et al found no significant differences in clinical or microbiological cure between patients treated with colistin-based regimens (51 patients, 45.9%) and those treated with cefiderocol in combination therapy. In the BSI subgroup, a significant reduction in 30-day mortality was observed in patients treated with cefiderocol (OR 0.121, p = 0.0109).²⁹

More recently, a single-center study of 45 patients with severe CRAB infections, mainly BSI and VAP, showed that early clinical success was more frequently achieved in patients treated with cefiderocol-based therapy compared with colistin-containing regimens. Furthermore, cefiderocol, even when used as monotherapy, was associated with favorable outcomes.³⁰

Two meta-analyses have been conducted based on these observational studies to better interpret the available data.

Gatti et al conducted a systematic review and meta-analysis including the CREDIBLE-CR trial and five observational studies (561 patients). Of these, 247 received cefiderocol-based regimens and 314 received non-cefiderocol-based regimens, predominantly colistin-based. The analysis showed a trend toward lower mortality among patients treated with cefiderocol. When the CREDIBLE-CR trial was excluded and only adjusted observational studies from Italy were considered, a significantly lower mortality rate was observed (N = 4; OR 0.53; 95% CI 0.39–0.71; I² = 0.0%). These findings were consistent across both BSI and pneumonia subgroups.³¹

Onorato et al conducted a meta-analysis including 18 studies (two RCTs, 13 cohort studies, and three case series) involving 733 patients treated with cefiderocol and 473 receiving BAT. The results suggested improved clinical outcomes in patients treated with cefiderocol compared with BAT, including lower 30-day mortality (particularly in BSI) and fewer adverse drug reactions. No significant differences were observed in clinical or microbiological failure rates.³²

Given the limited therapeutic options currently available for CRAB infections, real-world data provide supportive evidence for a potential role of cefiderocol. Reported 30-day mortality rates are generally lower, clinical success rates are encouraging, particularly in challenging scenarios such as BSI, and the drug is usually well tolerated. Nevertheless, most evidence comes from retrospective observational studies, which have inherent limitations, including potential confounding, especially indication bias; moreover, many of them were performed during the SARS-CoV-2 pandemic. Therefore, observed differences in mortality should be interpreted with caution. In addition, the previously mentioned RCTs have shown heterogeneous and, at times, conflicting results.^7,8 The recent development of sulbactam/durlobactam, specifically designed for CRAB, may offer an alternative first-line option; however, no head-to-head comparative data are currently available.³³ Maintaining multiple therapeutic options remains important, and prospective, randomized studies are needed to clarify the optimal use of cefiderocol, including monotherapy versus combination therapy and its impact on clinical and microbiological outcomes.

Pseudomonas aeruginosa

Infections caused by carbapenem-resistant and DTR P. aeruginosa (defined as resistance to all of the following: piperacillin–tazobactam, ceftazidime, cefepime, aztreonam, meropenem, imipenem–cilastatin, ciprofloxacin, and levofloxacin)³⁴ are associated with very high mortality, reaching rates above 40% in cases of VAP or BSI.³⁵ A major driver of these poor outcomes is the limited availability of effective therapeutic options capable of overcoming resistance.³⁶

In this context, cefiderocol represents an attractive option. In vitro studies have demonstrated potent activity against carbapenem-resistant P. aeruginosa, including isolates resistant to novel beta-lactam/beta-lactamase inhibitor (BL/BLI) combinations (such as ceftolozane–tazobactam), which are increasingly reported worldwide.^37–39

Since its approval, several real-life observational studies have described the use of cefiderocol for DTR P. aeruginosa infections, reporting encouraging outcomes. These studies, largely conducted in critically ill ICU patients across Europe and the United States, suggest favorable clinical responses, as shown in Table 3.^40–44

Table 3 Real-Life Studies Evaluating Cefiderocol Activity Against Pseudomonas aeruginosa

An early Italian case series (2021) reported a 30-day mortality of 35.3% and a clinical cure rate exceeding 70% in 17 patients with MDR P. aeruginosa infections treated with cefiderocol after failing prior antibiotic regimens.⁴¹ A French multicenter study evaluating new agents as last-line options for resistant pathogens reported a 30-day mortality of 21% in 19 patients with DTR P. aeruginosa treated with cefiderocol; favorable outcomes were also observed in infections caused by isolates resistant to ceftazidime–avibactam or ceftolozane–tazobactam.⁴⁰

Similarly, Satlin et al analyzed patients who received cefiderocol through the compassionate use program in Europe and the US.⁴³ Of 251 patients, 46 had a P. aeruginosa infections: the 28-day all-cause mortality was 24%, while the clinical response rate was 72%, comparable to that observed in pivotal trials, and independent of MIC values. Of note, most isolates resistant to ceftolozane–tazobactam and ceftazidime–avibactam, including MBL-producing strains, retained susceptibility to cefiderocol. Despite the complex clinical background of these patients, the drug was generally well tolerated.

The recently published results of the PERSEUS study further confirmed the role of cefiderocol against P. aeruginosa.⁴⁴ This multicenter Spanish cohort included patients treated with cefiderocol within the early access program, excluding those with CRAB infections. Among 261 patients, 174 had P. aeruginosa infection (nearly half of which were MBL-producing strains), and the 30-day mortality rate was 17.2%. Moreover, in multivariate analysis, P. aeruginosa infection itself was the only variable independently associated with a higher clinical cure rate.⁴⁴

Three additional recent large real-life studies have evaluated the activity of cefiderocol in patients with P. aeruginosa infections, most of which were carbapenem-resistant or DTR.^45–47 The first, a multicenter Spanish study, reported a favorable clinical response in 88.8% of patients, with a 30-day mortality rate of 28.6%, including 36 with XDR/DTR P. aeruginosa infection.⁴⁵ The second, a multicenter study conducted in the United States, observed a 30-day mortality rate of 13.1% among 82 patients with MDR P. aeruginosa.⁴⁶ The third, the interim analysis of the PROVE study, documented a clinical cure rate of 64.6%, with a 30-day mortality of 25.6% among 117 (48%) patients infected with P. aeruginosa.⁴⁷ However, the marked heterogeneity in patient populations and outcome definitions makes these findings challenging to interpret.

According to the most recent IDSA guidelines,²⁰ first-line therapy for DTR P. aeruginosa infections includes ceftolozane–tazobactam, ceftazidime–avibactam, and imipenem–cilastatin–relebactam, while cefiderocol is recommended as an alternative option. However, a recent study evaluating the in vitro activity of cefiderocol and novel BL/BLI combinations (ceftazidime–avibactam, imipenem–relebactam, and ceftolozane–tazobactam) against P. aeruginosa isolates collected in the United States and Europe between 2020 and 2023 (4% DTR) demonstrated that cefiderocol was active against 98.1% of DTR isolates (CLSI breakpoints), outperforming BL/BLI combinations.⁴⁸ These findings may support the use of cefiderocol as a potential first-line option in selected cases of DTR P. aeruginosa infection.

Overall, both in vitro and real-world evidence supports the activity of cefiderocol against this pathogen, and its role appears particularly relevant for the treatment of DTR P. aeruginosa and strains resistant to novel BL/BLI combinations.

Stenotrophomonas Maltophilia and Other Non-Fermenting Gram-Negative Bacteria

Stenotrophomonas maltophilia, Burkholderia cepacia complex (Bcc), Achromobacter spp and other non-fermenting Gram-negative bacteria are rare opportunistic pathogens that primarily affect severely immunocompromised patients. These patients often require invasive medical devices, such as central venous catheters, mechanical ventilation, or parenteral nutrition, or have underlying pulmonary conditions such as cystic fibrosis or bronchiectasis. These infections represent a clinical challenge, as it is often difficult to distinguish colonization from true infection requiring antibiotic treatment.⁴⁹ Furthermore, establishing an effective therapeutic regimen is often complicated by limited treatment options due to intrinsic and acquired resistance to multiple antibiotic classes.

Cefiderocol has demonstrated excellent in vitro activity against S. maltophilia, Bcc, and Achromobacter spp. isolates, with low MIC values in multinational surveillance studies. Takemura et al, compared the in vitro and in vivo antibacterial activity of cefiderocol with that of other approved antibacterial drugs against Achromobacter spp (N = 334) and Bcc (N = 425) isolates, including strains from the 5-year multinational SIDERO surveillance studies. Cefiderocol showed potent in vitro activity against Achromobacter spp. (MIC90 0.5 µg/mL) and Bcc (MIC90 0.5 µg/mL).⁵⁰

For S. maltophilia, in vitro data from large surveillance programs, such as the SENTRY Antimicrobial Surveillance Program, reported susceptibility rates of 100.0% (CLSI, 2021) and 97.9% (CLSI, 2022) to cefiderocol.³⁸

Real-world data on cefiderocol for these pathogens are very limited and mainly consist of case reports or small case series, as shown in Table 4. Vena et al described eight cases of S. maltophilia BSI treated with cefiderocol, reporting a clinical success rate of 62.5% (5/8). Among the remaining cases, clinical failure was due to death, although only one death was directly attributable to S. maltophilia infection.⁵¹

Table 4 Real-Life Studies Evaluating Cefiderocol Activity Against Stenotrophomonas maltophilia and Other Non-Fermenting Gram-Negative Bacteria

Data from a subgroup analysis of the observational PERSEUS study are also of interest. This study evaluated cefiderocol use within an early access program in Spain, including patients with Gram-negative infections other than Acinetobacter spp.⁵² Among 261 patients, 34 were included in this subgroup analysis: 20 had S. maltophilia infections and 14 had infections caused by other non-fermenting Gram-negative bacteria. Cefiderocol was rarely used as first-line therapy and was mainly administered after treatment failure or documented resistance to other agents. Combination therapy was used in 40.0% to 87.5% of patients. Clinical cure rates were 70.0% for S. maltophilia and 71.4% for other pathogens.⁵²

The 2024 IDSA guidelines recommend cefiderocol, as part of combination therapy (at least until clinical improvement is achieved), as a preferred option for the treatment of S. maltophilia infections²⁰ and the available real-world data, although limited, are consistent with this recommendation.

Enterobacterales

Cefiderocol represents a promising therapeutic option for CRE. This group of pathogens is highly heterogeneous, encompassing bacteria with diverse resistance mechanisms. In particular, carbapenem resistance may result from the production of various types of carbapenemases, or from other mechanisms such as the production of ESBLs or AmpC beta-lactamases combined with reduced porin expression or overexpression of efflux pumps (non–carbapenemase-producing CRE).⁵³ Regardless of the underlying resistance mechanisms, CRE infections are increasing globally and have been included by the WHO among the critical priority pathogens for the development of new antimicrobials.^2,54

Cefiderocol has demonstrated potent activity in this setting, retaining efficacy against all major types of carbapenemases, including serine carbapenemases (such as Ambler class A KPC and class D OXA-48) as well as MBL (class B, including NDM, VIM, and IMP).^11,55–57 This activity has also been confirmed by recent results of the SENTRY Antimicrobial Surveillance Program: among 252 MBL-producing Enterobacterales isolates collected in Europe and the USA between 2020 and 2023, in vitro susceptibility to cefiderocol was 87.7% and 63.9% according to CLSI and EUCAST breakpoints, respectively. Although MIC values were higher, susceptibility rates remained markedly greater than those observed with other comparator agents.⁵⁸ Similar findings were reported for 82 Enterobacterales isolates carrying multiple carbapenemase genes.⁵⁸

Currently available real-world in vivo data are often derived from studies including infections caused by various Gram-negative pathogens, with only a minority of patients having CRE infections; moreover, pathogen-specific outcomes are not always reported.^45,59,60

Nevertheless, multiple real-life studies have demonstrated the activity of cefiderocol against CRE infections and the main results are summarized in Table 5. An American study including 112 patients, of whom 28 (25%) had CRE isolates, reported a clinical success rate of 70% in both the overall population and the CRE subgroup, increasing to 80% in cases involving MBL-producing isolates, consistent with findings from pivotal trials.⁴⁶ In a multicenter Italian cohort including 200 patients treated with cefiderocol, 26 (13%) had Enterobacterales infections; the 28-day clinical cure rate was 77% and remained similar in patients with MBL-producing isolates.⁶¹

Table 5 Real-Life Studies Evaluating Cefiderocol Activity Against Carbapenem Resistant Enterobacterales (CRE)

Less favorable outcomes were reported in a previous Italian cohort of 142 patients, of whom 22 had infections caused by MBL-producing K. pneumoniae. In this study, the overall 30-day mortality rate was 37%, reaching 44% among patients with K. pneumoniae infections, with a microbiological cure rate of 47%. The authors suggested that resistance to cefiderocol may have contributed to the high mortality, particularly since susceptibility testing was available for only a minority of patients, among whom 7 of 22 isolates were resistant.⁶²

Furthermore, two recently published large studies also included patients with Enterobacterales infections: the previously mentioned PERSEUS study and the PROVE study (interim analysis).^44,47 In the PERSEUS study, cefiderocol was used in 26 patients with K. pneumoniae infections and in 12 with other Enterobacterales, with clinical cure rates of 69% and 75%, respectively.⁴⁴ The PROVE study, which included 244 patients, reported a clinical cure rate of 73% among the 26 patients with Enterobacterales infections.⁴⁷

Unlike other resistant pathogens, for CRE there are multiple treatment options available that have demonstrated good activity, including ceftazidime/avibactam, meropenem/vaborbactam, and imipenem/cilastatin/relebactam.²⁰ However, real-world data support cefiderocol as a valuable option, particularly for infections caused by pathogens with limited therapeutic alternatives due to resistance, given its activity against all classes of carbapenemases. Its role appears particularly relevant for MBL-producing isolates, for which the main available alternatives include aztreonam/avibactam or the combination of ceftazidime/avibactam plus aztreonam.

Figure 1 summarizes the role of cefiderocol in therapy across different pathogens.

Figure 1 Proposed role of cefiderocol for carbapenem-resistant Gram-negative pathogens based on evidence from observational real-world studies. Abbreviations: DTR, difficult-to-treat; MBL, metallo-beta-lactamase.

Cefiderocol PK/PD

Pharmacokinetic studies conducted in humans have demonstrated that cefiderocol exhibits linear pharmacokinetics across the investigated dose range.⁶³ The drug is predominantly eliminated via the renal route, with approximately 90% excreted unchanged in the urine. Metabolic pathways account for less than 10% of total drug elimination, while fecal excretion is negligible. The elimination half-life is approximately 2–3 hours. Cefiderocol demonstrates moderate plasma protein binding, with approximately 58% bound to circulating proteins, and is readily removed by certain renal replacement therapies.^63,64 Given its predominant renal clearance, renal function has a substantial impact on cefiderocol clearance and systemic exposure. Dose adjustments are required in patients with impaired glomerular filtration rate, those undergoing renal replacement therapy, and individuals with augmented renal clearance (eGFR >120 mL/min), in order to maintain adequate drug exposure.

Cefiderocol demonstrates time-dependent bactericidal activity, and its efficacy is best correlated with the percentage of the dosing interval during which free drug concentrations exceed the MIC (%fT>MIC).^11,63 In a recent PK/PD simulation study, the recommended dosing regimen reported in the product label (2 g every 8 hours administered as a prolonged 3-hour infusion) achieved the plasma target of 100% fT>MIC with a probability of target attainment (PTA) >90% for isolates with MIC values up to 2 mg/L.⁶⁵ However, lower target attainment was observed in epithelial lining fluid, suggesting that higher dosing strategies may be required in cases of pneumonia to ensure adequate drug exposure and reduce the risk of microbiological failure. The study also evaluated alternative dosing regimens, including short, prolonged, and continuous infusions. The highest PTA values were achieved with continuous infusion, which may therefore be preferable in severe infections such as VAP, particularly in patients with normal or augmented renal function, a clinical scenario frequently encountered in the ICU.^65,66

Moreover, in critically ill patients, more aggressive PK/PD targets have been proposed, such as 100% fT >4×MIC, as these have been associated with improved clinical outcomes and a reduced risk of resistance emergence. Standard PK/PD targets may not ensure sufficient pulmonary drug exposure, particularly in infections caused by CRAB.⁶⁷ In this context, and considering the favourable safety profile of cefiderocol, higher dosing strategies may be considered in selected cases, although supporting clinical evidence remains limited.

Another strategy to optimize PK/PD target attainment is the use of therapeutic drug monitoring (TDM), which has also been described for cefiderocol. TDM may be particularly valuable in patients receiving renal replacement therapy or extracorporeal support, as well as in individuals with obesity, conditions that may significantly increase the volume of distribution and alter drug exposure.⁶⁸

Finally, some experts advocate the administration of a loading dose, in line with recommendations from the Surviving Sepsis Campaign for beta-lactam antibiotics, although this approach is not currently included in the product information.⁶⁹

Emergence of Cefiderocol Resistance

As highlighted in this review, cefiderocol currently represents, in many clinical contexts involving infections caused by MDR Gram-negative bacteria, a “last-resort” agent. It is capable of overcoming several of the most common resistance mechanisms, including beta-lactamase production (notably MBLs), efflux pump activity, and porin mutations. However, the emergence of cefiderocol resistance, documented both in vitro and in vivo, is an increasing concern that cannot be overlooked.

Resistance rates are evolving over time and show substantial variability depending on geographical region and local epidemiology. In a study published in 2024, Kimbrough et al analyzed resistance profiles in more than 20,000 isolates collected between 2020 and 2021 across Europe and the US. The bacterial species included Enterobacterales, Pseudomonas spp., and Acinetobacter spp., and only 69 isolates (0.3%) were categorized as cefiderocol non-susceptible according to CLSI criteria (MIC > 4 mg/L). However, isolates were included regardless of resistance profile; for example, among Enterobacterales, carbapenem resistance prevalence was relatively low (2.8%).⁷⁰

Subsequent studies have, however, depicted a markedly different scenario. A meta-analysis published in 2023 by Karakonstantis et al, which evaluated data from 78 studies, reported overall low resistance rates according to EUCAST breakpoints (3% among Enterobacterales and 1.4% in P. aeruginosa). Nonetheless, resistance was substantially higher in specific subgroups, reaching 8.8% in A. baumannii, 38.8% in NDM-producing Enterobacterales, 44.7% in NDM-producing A. baumannii, and 36.6% in ceftazidime/avibactam-resistant Enterobacterales.⁷¹ Furthermore, a recent study from the United Kingdom evaluated 150 CRE isolates obtained from 136 patients, the majority of whom were colonized, while 16.9% had active infections. None of the patients had prior exposure to cefiderocol. In this cohort, the overall cefiderocol resistance rate (EUCAST breakpoints) was 48.7%, increasing to 69.7% among NDM-producing Enterobacterales, which accounted for 38% of all isolates.⁷² Although these findings are difficult to generalize, they highlight that, despite overall high susceptibility to cefiderocol, there are specific settings where markedly high rates of non-susceptibility have been reported.

The main resistance mechanisms described to date include: (1) enzymatic mechanisms associated with the production of MBLs or KPC variants (eg, KPC-31); (2) mutations in siderophore receptor pathways, initially described in P. aeruginosa; (3) mutations in penicillin-binding proteins; (4) alterations in porin channels; and (5) overexpression of efflux pumps.^73,74 Figure 2 illustrates the main resistance mechanisms.

Figure 2 Schematic overview of the principal resistance mechanisms associated with cefiderocol.

As recently described by Warecki et al,⁷⁵ the association between NDM production and cefiderocol resistance is being increasingly reported, despite cefiderocol being considered a preferred treatment option for infections caused by MBL-producing strains.²⁰ The underlying mechanism appears to be the ability of certain NDM variants, particularly NDM-1 and NDM-5, to hydrolyze cefiderocol at rates sufficient to compromise its activity, in contrast to what is observed with IMP- or VIM-producing strains.⁷⁶

In vivo emergence of resistance has been described in observational case series and case reports.^{26,41,77–81} Heteroresistance likely plays a key role in this process, whereby a small bacterial subpopulation acquires resistance under antibiotic pressure and is subsequently selected during therapy, particularly during prolonged exposure.⁷³

In these studies, cases of microbiological failure during treatment have been reported, although repeat MIC testing to confirm resistance development was not always performed. For example, in a study including 47 patients with CRAB infections, microbiological failure occurred in 8 patients, and in half of these cases an increase in MIC values consistent with resistance was documented.²⁶

Another important aspect emerging from in vivo studies is cross-resistance between cefiderocol and ceftazidime–avibactam. Cases have been reported in which patients treated with ceftazidime–avibactam developed resistance to both agents, even in the absence of prior cefiderocol exposure. This phenomenon was described by Bianco et al in 16 patients with infections due to KPC-producing K. pneumoniae,⁷⁸ by Tiseo et al in a patient with K. pneumoniae bacteremia carrying a KPC-3 variant (KPC-31),⁷⁹ and subsequently by Amadesi et al in six additional patients, all harboring KPC variants with mutations in the omega-loop region.⁸⁰

Finally, cefiderocol resistance has also been described in A. baumannii mediated by downregulation of TonB-dependent siderophore receptors such as PiuA.⁸¹

Although cefiderocol retains a high likelihood of therapeutic coverage in many settings, as demonstrated in a multicenter study conducted across 11 Italian ICUs,⁸² the emergence of resistance is an increasing concern. Continuous monitoring of local epidemiology, along with the implementation of optimized PK/PD strategies, is therefore essential to ensure the appropriate use of this antibiotic.

The combination of cefiderocol with beta-lactamase inhibitors may restore its activity against resistant bacteria, as suggested by promising results with cefiderocol–xeruborbactam (reducing MICs in Enterobacterales and A. baumannii) and cefiderocol–taniborbactam, which restored susceptibility in most cefiderocol-resistant, NDM-positive P. aeruginosa isolates; however, data on these combinations are currently extremely limited.^83,84

Use of Cefiderocol Among Immunocompromised Patients

Immunocompromised patients are among those most affected by MDR Gram-negative infections; however, in the registrational trials of cefiderocol, this population was often underrepresented, and subgroup-specific outcomes were usually not reported.^7,8 As a result, it is difficult to generalize the findings of major trials to immunocompromised patients, despite the fact that they are among those most likely to require treatment with cefiderocol.

Some real-life studies, including the Spanish PERSEUS study, reported outcomes for the immunocompromised subgroups included in their cohorts and suggested that immunosuppression does not appear to worsen clinical outcomes.^44,85 Among these, a cohort of 13 lung transplant recipients showed a 30-day mortality rate of 26%.⁸⁶ More recently, two real-world studies have specifically evaluated cefiderocol use in immunocompromised patients, reporting favorable outcomes in this subgroup.^87,88 The first, a French study including 114 immunocompromised patients treated with cefiderocol, reported a 28-day overall mortality rate of 37.7% and a clinical cure rate of 53.3%.⁸⁷ These results are consistent with those reported in randomized trials such as CREDIBLE-CR and in earlier real-world cohorts, suggesting that immunosuppression may not significantly impact clinical success or mortality.^8,59,62 Similar findings were reported in a post hoc analysis of the Italian multicenter CEFI-SITA study, which evaluated 84 immunocompromised patients and found no differences in treatment indications, use of combination therapy, clinical cure, or mortality compared with non-immunocompromised patients. Moreover, in multivariable analysis, immunosuppression was not associated with 30-day mortality, which was instead driven by other factors, including A. baumannii infection.⁸⁸

Taken together, these findings suggest that cefiderocol may achieve comparable outcomes in immunocompromised patients, supporting its use in this high-risk population despite their greater clinical complexity.

Use of Cefiderocol as Monotherapy or in Combination Regimens

The use of cefiderocol as monotherapy or in combination with a second antibiotic remains a matter of debate.^61,89 Clinical trials have demonstrated the efficacy of cefiderocol monotherapy in the treatment of complicated urinary tract infections and nosocomial pneumonia;^6,7 however, monotherapy may be insufficient in cases of severe or deep-seated infections.

With regard to the causative pathogen, current guidelines recommend cefiderocol monotherapy for infections due to MBL-producing Enterobacterales or P. aeruginosa. Conversely, for CRAB, cefiderocol, listed as an alternative to sulbactam/durlobactam, is advised as part of a combination regimen, typically with high-dose ampicillin/sulbactam (to achieve 9 g/day of sulbactam).²⁰

The use of combination therapy for CRAB infections remains particularly controversial and is likely influenced by the results of the CREDIBLE-CR trial, in which patients with A. baumannii infection treated with cefiderocol showed higher mortality (50% vs. 18% with BAT), although this finding may have been confounded by differences in disease severity and baseline characteristics.⁸

A recent prospective multicenter study published by Giacobbe et al further highlighted the lack of robust evidence on this topic.⁶¹ In this cohort of 200 patients, approximately half received combination therapy and half monotherapy. Prior colonization with CRAB was independently associated with the use of combination regimens as empirical therapy; however, isolation of A. baumannii was not associated with combination therapy in the targeted treatment setting. Moreover, no significant differences were observed in treatment strategy according to immunocompromised status.⁸⁸

The rationale for combination therapy includes broadening antimicrobial coverage, overcoming resistance mechanisms, preventing resistance emergence in high-inoculum infections, and achieving potential synergistic effects, as suggested for cefiderocol combined with tigecycline or colistin against A. baumannii. In addition, combination therapy may enhance tissue penetration and accelerate bacterial clearance in severe infections. However, it also carries potential drawbacks, including an increased risk of adverse effects and drug–drug interactions. In contrast, cefiderocol monotherapy offers an excellent safety profile, with minimal side effects and a reduced selective pressure that may help limit the emergence of antibiotic resistance.⁸⁹

In this context, several real-world studies have reported comparable outcomes with cefiderocol monotherapy compared with combination therapy.^{26,29,30,62,90–94}

In particular, a study by Giannella et al,⁹⁰ including 147 patients with CRAB infections, found that 66.7% received combination therapy. The most common companion agents were colistin (18.4%), tigecycline (15%), and fosfomycin (12.9%). Clinical success rates were higher in the monotherapy group (61.2%) compared with the combination therapy group (49.0%).

Similarly, in the aforementioned study by Mazzitelli et al,²⁹ comparing cefiderocol versus BAT for A. baumannii infections, a sub-analysis of 30 patients receiving cefiderocol monotherapy and 30 receiving combination therapy revealed no significant differences in mortality, clinical cure, or microbiological cure. In this study, tigecycline, meropenem, and fosfomycin were the most frequently used combination partners.

These findings were included in a meta-analysis by Onorato et al (2024),³² which also incorporated three smaller retrospective observational studies^26,92,93 and two case series^91,94 evaluating cefiderocol monotherapy versus combination therapy for A. baumannii infections. Across these seven studies, a significantly lower 30-day mortality rate was observed among patients treated with cefiderocol monotherapy (p = 0.024) and no significant differences were found in microbiological or clinical failure rates. The authors of the meta-analysis nevertheless advised cautious interpretation of these results, emphasizing that the observational, retrospective nature of the included studies could have introduced bias, as patients receiving monotherapy were often in better clinical condition, potentially influencing mortality outcomes.³²

An additional prospective observational study by Manesh et al,⁹⁵ including 161 patients with CRAB infection, published after the aforementioned meta-analysis, further confirmed the absence of significant difference in 30-day mortality between monotherapy and combination therapy, also when evaluated through propensity score analysis. Notably, in this study, baseline characteristics were comparable between groups (including age, type of infection or presentation with septic shock).

At present, there is no clear evidence either supporting or discouraging the addition of a second agent to cefiderocol for the treatment of CRAB infections. Furthermore, substantial variability exists in the choice of companion drug, and no data currently favor one option over another. Although observational studies have not consistently reported worse outcomes with cefiderocol monotherapy, it should be emphasized that these data largely derive from observational, often retrospective studies, which are inherently subject to selection bias, heterogeneity in study design, and frequently small sample sizes. Therefore, these findings should be interpreted with caution and require confirmation in randomized controlled trials.

In particular, in the setting of severe infections caused by CRAB, the use of cefiderocol in combination therapy may warrant careful consideration in order to increase the likelihood that at least one fully active agent is included in the treatment regimen, as also recommended by current IDSA guidelines.¹³

Conclusions

In conclusion, real-life data support the role of cefiderocol as a valuable therapeutic option against MDR Gram-negative pathogens, in some cases representing a potential last-resort agent capable of overcoming multiple resistance mechanisms. However, several questions remain unanswered and warrant further investigation. First, the emergence of resistance to cefiderocol has been reported and requires ongoing surveillance. Second, the role of cefiderocol in CRAB infections may need to be reconsidered in light of newly available targeted therapies, potentially positioning it as a second-line option in selected cases. Finally, although current real-world evidence does not demonstrate a clear advantage of combination therapy over monotherapy, this question has not yet been adequately addressed in randomized controlled trials.

Cefiderocol, together with other recently introduced antimicrobials, has contributed to improving the management of infections caused by MDR Gram-negative bacteria. Precisely because of its potential clinical value, its use, whether as empirically guided or targeted therapy, should be guided by careful clinical judgment and adherence to antimicrobial stewardship principles, in order to help preserve its efficacy from further resistance mechanisms.