Eravacycline as a last resort for difficult-to-treat resistant Acinetobacter baumannii infections in critically ill patients: three case reports with pharmacokinetic insights

Abstract

Objectives

To describe eravacycline use as a salvage treatment for ventilator-associated pneumonia (VAP) caused by difficult-to-treat resistant (DTR) Acinetobacter baumannii in critically ill patients.

Methods

We reported three cases of DTR A. baumannii VAP with multiple organ failure treated with eravacycline. Patients were critically ill with confirmed VAP by distal pulmonary cultures. Eravacycline was administered at 1 mg/kg q12h in combination with IV colistin or as primary therapy. Clinical and microbiological outcomes were assessed.

Results

Eravacycline MICs ranged from 0.25 to 0.75 mg/L. Microbiological success was observed in the three cases, including one patient who was successfully weaned and discharged alive with no further samples submitted for microbiological culture, and two other patients who were repeatedly sampled and remained negative for A. baumannii. Clinical success could not be confirmed in one case. No adverse effects were observed. Pharmacokinetic analysis of concentrations from a single patient revealed a maximal concentration (C_max) of 1.47 mg/L at 1 h and an AUC_0–6 of 2.88 mg·h/L. The epithelial lining fluid/plasma concentration ratio was 0.1.

Conclusions

Eravacycline showed promise as a salvage therapy for DTR A. baumannii VAP in critically ill patients. Further studies are needed to confirm its efficacy and optimal dosing in this setting.

Introduction

Acinetobacter baumannii is a non-fermentative, Gram-negative opportunistic bacterium that has emerged as a major burden in hospitals, particularly in ICUs.^1,2 Furthermore, strains with multiple resistant mechanisms including MBLs and siderophore receptor mutations reducing susceptibility to sulbactam/durlobactam and cefiderocol have been described recently.^3,4

Eravacycline is a novel synthetic fluorocycline approved by the US FDA and the EMA in complicated intra-abdominal infections in 2018^5,6 with MICs 2–8-fold lower than other cyclins against Gram-negative bacteria.⁷

To date, few cohorts have described eravacycline efficacy against carbapenem-resistant A. baumannii (CRAB) in hospital-acquired infections.^8,9 None has focused on eravacycline as a salvage treatment of ventilator-associated pneumonia (VAP) caused by difficult-to-treat resistant (DTR) A. baumannii.

We report three cases of DTR bla_NDM-1 and bla_OXA-23  A. baumannii VAP with multiple organ failure treated with eravacycline as last-line treatment.

Methods

Ethics

This retrospective study was set in compliance with the 1975 Declaration of Helsinki and falls into the framework of MR-004 methodology, where written informed consent is not required according to the French public health laws. However, patients were informed that their medical records might be used for research in accordance with the privacy rule.

Patients

Patients included in our study were critically ill patients with multiple organ dysfunctions (see details in the Table 1). They were all treated for an A. baumannii-associated VAP, defined as an infection occurring at least 48 h after endotracheal intubation with new or progressive infiltrates on chest imaging, fever or leucocytosis and at least purulent respiratory secretions or worsening oxygenation parameters associated with an A. baumannii-positive culture [bronchoalveolar lavage (BAL) or plugged telescoping catheter (PTC)]. A secondary bloodstream infection (BSI) was diagnosed in one case. The detection of carbapenemase producers was first confirmed using the Coris^® BioConcept Resist Acineto immunochromatographic test, which detects the enzymes NDM, OXA-23 and OXA-40/58 directly from the colonies. These resistance genes were confirmed by sequencing. All strains were then genotyped. In addition, for cases 1 and 3, NDM had been detected by the BioFire^® FilmArray^® Pneumonia plus Panel as soon as the sample was taken. The strains were the same for cases 2 and 3, which was ST1 with NDM-1 and OXA-23 but different from case 1, which was ST2 with only NDM-1.

Table 1.

Main features of the three cases of DTR A. baumannii infections treated with eravacycline

Case	Indication of transfer in ICU	Organ failures	Antibiotic treatment	Type of AB infections and diagnostic tests	Susceptibility [MICs if available (mg/L)]	Genotyping	Type and duration of treatment	Clinical/microbiological success/mortality
Case 1
42y woman Mixed connective tissue disease BMI = 27 kg/m2 SAPS II = 48	Pneumonia and diarrhoea on lupus enteropathy leading to multiple organ failure	ARDS eGFR (CKD-EPI) > 90 mL/min/1.73 m2 Norepinephrine (maximum): 0.6 mg/h	Meropenem prior to AB infection Initial therapy with cefiderocol, tigecycline and nebulized colistin with clinical failure and cytolysis	VAP (D13) PTC: 107 cfu/mL AB	Ampicillin/sulbactam: Rc Tigecyclinea: R MIC = 3 Minocyclinea: ND Eravacyclineb: S MIC = 0.25 Cefiderocolb: R MIC = 4 Colistinb: S MIC = 1 Amikacin: Sc	bla NDM-1	Rescue therapy: Eravacycline 1 mg/kg q12h + IV Colistin 3MUI q8h 7 days	Clinical: ICU discharge on D32 No SAE Microbiological: No additional sampling
Case 2
52y woman COPD BMI = 43 kg/m2 SAPS II = 39	ARDS (influenza)	Severe ARDS: VV-ECMO Renal failure: haemodialysis Norepinephrine (maximum): 3.8 mg/h	Multiple VAP episodes and VV-ECMO cannula infections treated with: Amoxicillin/clavulanic acid cefotaxime meropenem Initial therapy: tigecycline, tobramycin and ampicillin/sulbactam	VAP (D44) BAL: 107 cfu/mL AB	Ampicillin/sulbactama: R MIC > 256 Tigecyclinea: R MIC = 6 Minocyclinea: Rc Eravacyclinea: S MIC = 0.75 Cefiderocolb: R MIC = 8 Colistinb: S MIC = 1 Amikacin: Rc	bla NDM-1 bla OXA-23 tet(B)	Rescue therapy: Eravacycline 1 mg/kg q12h + IV Colistin 1.5MUI q12h 8 days	Clinical: Death on D55 from worsening ARDS Microbiological: AB eradication but DTR P. aeruginosa blaVIM treated with Cefiderocol Negative BAL at D54
Case 3
59y male Aortic aneurysm BMI = 18 kg/m2 SAPS II = 69	Cardiac failure after cardiac surgery Previous colonization with blaNDM AB	ARDS Cardiac failure: intra-aortic balloon and VA-ECMO Renal failure: haemodialysis Norepinephrine (maximum): 5 mg/h	No previous antibiotics	VAP (D10) and BSI (D12) BAL: 107 cfu/mL AB Blood cultures (×4): 11 h, 9 h, 14 h, 12 h	Ampicillin/sulbactama: R MIC > 256 Tigecyclinea: R MIC = 4 Minocyclinea: R MIC = 48 Eravacyclinea: S MIC = 0.75 Cefiderocolb: R MIC = 4 Colistinb: S MIC = 1 Amikacin: Rc	bla NDM-1 bla OXA-23 tet(B)	Initial therapy: Eravacycline 1 mg/kg q12h + IV Colistin 1.5MUI q12h 17 days	Clinical: VA-ECMO weaned on D14 Norepinephrine weaned on D32 Death from subarachnoid haemorrhage on D36 Microbiological: Negative blood culture on D14 Negative BAL at D25

AB, A. baumannii; eGFR, estimated glomerular filtration rate; D, days from ICU hospitalization; ND, not determined; SAE, serious adverse event; SAPS, Simplified Acute Physiology Score; y, years old.

^aMIC determined by diffusion-gradient method.

^bMIC determined by microdilution.

^cSome MICs were not systematically generated and were not available.

Results

MIC of eravacycline was 0.75 mg/L for two strains and 0.25 mg/L for one strain. Strains were resistant to cefiderocol and had a tigecycline MIC of >2 mg/L. Eravacycline was introduced in combination with IV colistin as rescue therapy in two cases and as the primary therapy for one patient previously colonized with DTR A. baumannii. Microbiological success was observed in the three cases including one patient who was successfully weaned and discharged alive with no further samples submitted for microbiological culture, and two other patients who were repeatedly sampled and remained negative for A. baumannii. Clinical success could not be confirmed in case 2, as it is difficult to exclude a contribution of VAP to the exacerbation of acute respiratory distress syndrome (ARDS), with death occurring on Day 55 despite optimal treatment and microbiological success with a negative BAL on Day 54. For case 3, we considered the clinical success based on the weaning of ECMO and normalization of respiratory parameters even after discontinuation of antimicrobials. Subarachnoid haemorrhage, although probably favoured by several factors including curative anticoagulation, prolonged hospitalization, ECMO and surgery, did not appear to be directly related to A. baumannii-associated VAP.

In all three cases, we used 1 mg/kg q12h and no side effects were observed, with no digestive or liver intolerance during therapy. For one single patient (case 3), serial plasma concentrations for drug monitoring after the first administration were done. The pharmacokinetic profiles revealed maximal concentration (C_max) of 1.47 mg/L at 1 h and a plasma AUC_0–6 for eravacycline (using non-compartmental analysis with PKNCA package in R, version 4.3.3) of 2.88 mg·h/L. Eravacycline concentration was measured in the pulmonary epithelial lining fluid (ELF) at 2 h after injection with an ELF/plasma concentration ratio of 0.1 (10%). Eravacycline and urea concentrations were determined in blood plasma and ELF using LC-tandem MS, Waters Xevo TQD^® (Milford, MA, USA), according to published data.¹⁰

Discussion

We have reported three cases of eravacycline use for the treatment of A. baumannii in the absence of therapeutic alternatives. In 2024, IDSA guidelines recommended the use of combination therapy with ampicillin/sulbactam or sulbactam/durlobactam and another key antimicrobial (cefiderocol, polymyxin B, minocycline) for severe infections due to CRAB.¹¹ Sulbactam/durlobactam was not available in France and all patients included in our study had A. baumannii resistant to ampicillin/sulbactam and cefiderocol. In the case of eravacycline, these recommendations underline the paucity of data limiting its use to the absence of therapeutic alternatives.

Two out the three cases reported had a proven infection with A. baumannii that was considered cured by the combination eravacycline/colistin. Therefore, our study suggested that in cases of VAP or BSIs and in the presence of DTR A. baumannii, eravacycline should be considered as a therapeutic option in combination therapy.

Numerous studies have highlighted the in vitro activity of eravacycline alone and in combination on CRAB.^12,13 However, only a few studies have focused on the use of eravacycline in real-life settings as a treatment of CRAB. Alosaimy et al.⁸ have reported the treatment with eravacycline of 46 patients with infections due to A. baumannii. In this study, 32 patients had CRAB. However, many of these patients had a therapeutic alternative as 35.7% of the CRAB were considered minocycline susceptible, 14.8% were ampicillin/sulbactam susceptible and susceptibility to cefiderocol was not provided. In another study, nearly all patients received combination therapy of ampicillin/sulbactam and eravacycline as a treatment of CRAB infections in COVID-19 patients.⁹ All strains in our cohort were considered cefiderocol resistant, with MICs of >2 mg/L according to the EUCAST 2023 guidelines. Nevertheless, it is important to note that the cut-offs between EUCAST, CLSI and FDA differ, with ≤2,  ≤ 4 and ≤1 mg/L to define susceptibility of A. baumannii.¹⁴ Furthermore, no resistance mechanism was detailed in the latter cohorts, especially the association of bla_NDM-1 and bla_OXA-23. However, the rarity of NDM-1 positivity in CRAB isolates and the similar mutational profiles of the two strains from cases 2 and 3 could limit the generalizability of our results.

The MIC distribution of eravacycline against CRAB is correlated and approximately 2–8 times lower than that of tigecycline in in vitro studies.^11,12 However, the overall clinical relevance of this difference remains uncertain given that MIC levels were determined by different methods, particularly due to the lack of standardization of technique during the early stages of eravacycline commercialization, which may limit the consistency of antimicrobial susceptibility testing (AST) results. However, our cohort suggested that tigecycline and minocycline microbiologically considered resistance does not preclude the use of eravacycline.

Data concerning eravacycline penetration into the ELF and alveolar cells (ACs) are scarce. Only one Phase I study showed that eravacycline concentrations in the ELF and ACs of healthy adults achieved greater levels than plasma by 6- and 50-fold, respectively.¹⁵ By analogy, more data concerning tigecycline lung penetration showed contrasting data in critically ill patients with the median ELF/plasma concentration ratio ranging from 0.03 to 1.5.^16,17 Our results are in concordance with the study by Burkhardt et al.¹⁶, which showed a low ELF/plasma ratio of tigecycline in patients with underlying pulmonary pathology. Further pharmacokinetic studies are needed to assess the possibility of underdosing eravacycline in the treatment of pneumonia. Finally one pharmacokinetic study suggested that eravacycline AUC_0–12 was not reduced by veno-venous extracorporeal membrane oxygenation (VV-ECMO) and that it would not impact antimicrobial efficacy.¹⁸ To our knowledge, this cohort was the first to include a patient treated by eravacycline under venoarterial (VA)-ECMO and microbiological success was observed, underlining the possible use of this key agent in this setting.

We have only reported three cases of DTR A. baumannii treated with eravacycline in combination with colistin in a retrospective design. Further prospective studies are needed to support the use of eravacycline alone or in combination with another antimicrobial in the treatment of DTR A. baumannii infections.

Conclusions

This study described the use of eravacycline as a last resort in three patients with DTR A. baumannii infections. As two patients had microbiological and clinical cure, our cohort highlighted the importance of considering eravacycline as a salvage therapy in the absence of other therapeutic alternatives. More data are needed to confirm the use of eravacycline in this indication and its optimal dosing in critically ill patients.

Funding

This study was carried out as part of our routine work. No funding.

Transparency declarations

The authors declare that they have no conflicts of interest.

Author contributions

M.T., L.M. and J.F.T. conceived the study. E.R. performed the microbiological analysis. M.P.L. performed the pharmacological analysis including the dosages. M.T., L.M. and J.F.T. drafted the manuscript and critically revised the manuscript. All authors critically revised the manuscript and consent for publication.

No other assistance.

Data availability

All data generated or analysed during this study are included in this study or its supplementary material files. Further inquiries can be directed to the corresponding author.