Cefiderocol for the Treatment of Adult and Pediatric Patients With Cystic Fibrosis and Achromobacter xylosoxidans Infections

Abstract

Treatment options for Achromobacter xylosoxidans are limited. Eight cystic fibrosis patients with A. xylosoxidans were treated with 12 cefiderocol courses. Pretreatment in vitro resistance was seen in 3 of 8 cases. Clinical response occurred after 11 of 12 treatment courses. However, microbiologic relapse was observed after 11 of 12 treatment courses, notably without emergence of resistance.

Antimicrobial resistance is a major threat to the world’s population. Achromobacter xylosoxidans, formerly known as Alcaligenes denitrificans subsp xylosoxidans, is a motile, aerobic, glucose-nonfermenting gram-negative bacillus found in soil and water. Achromobacter xylosoxidans is intrinsically resistant to most antibiotics and often acquires in vitro resistance to additional antibiotics after antibiotic exposure [1, 2]. An estimated 3%–8% of patients with cystic fibrosis (CF) are colonized with A. xylosoxidans [1]. Non-CF patients at risk for A. xylosoxidans infections include patients with cancer and patients with non-CF bronchiectasis [2]. Cefiderocol (previously S-649266) is a novel siderophore cephalosporin antibacterial agent with broad-spectrum activity against multidrug-resistant gram-negative bacilli, including A. xylosoxidans [3].

Here we report 8 patients with CF who were infected with extensively multidrug-resistant A. xylosoxidans and treated on a compassionate use basis with cefiderocol.

METHODS

A multicenter, retrospective case series of patients with A. xylosoxidans infections who were treated with cefiderocol was performed. Patients were identified through evaluation of the United States (US) cefiderocol compassionate use program. Centers that had treated patients for A. xylosoxidans infection with cefiderocol were approached for participation. For adult patients, a standard dose of 2 g intravenously (IV) every 8 hours was used; 1 patient with an estimated creatinine clearance >120 mL/minute received 2 g IV every 6 hours. For pediatric patients, a dose of 60 mg/kg every 8 hours was used. All doses were adjusted for renal function according to the manufacturer’s recommendations. Cefiderocol was obtained through compassionate use from its manufacturer, with approval from local institutional review boards. The cefiderocol minimum inhibitory concentration (MIC) for pretreatment isolates was determined through broth microdilution utilizing Clinical and Laboratory Standards Institute–approved methodology in iron-depleted media, performed at International Health Management Associates (Schaumburg, Illinois). Breakpoints for Pseudomonas aeruginosa were used to define susceptibility. Microbiologic relapse was defined as isolation of A. xylosoxidans within 180 days of completion of treatment with cefiderocol. Clinical condition at the 30-day timepoint was defined by treating physicians and graded as follows: complete resolution of all symptoms associated with infection (resolution), improvement of symptoms without resolution (improvement), stable symptoms, worsening symptoms, and death. Resolution and improvement were grouped as clinical response. Stable and worse symptoms were grouped with death as clinical failures.

RESULTS

Overall, 31 A. xylosoxidans isolates were tested for cefiderocol susceptibility in the context of compassionate use in the US; 5 (16%) of these isolates were in vitro nonsusceptible to cefiderocol. Here, we describe 8 of these patients—6 adult and 2 pediatric CF patients from 4 centers who were treated with cefiderocol for A. xylosoxidans infections (Table 1). All patients had A. xylosoxidans infections in the respiratory tract. Pneumonia was complicated by bacteremia (patients 1 and 4) and empyema (patient 7). Patient 6 was also infected with Achromobacter ruhlandii. A total of 12 separate cefiderocol treatment courses were given; 4 patients received 2 courses of cefiderocol. The duration for each course ranged from 2 days to 6 weeks. In 11 of 12 courses, additional antibiotics with anti-gram-negative activity were given. Three of 8 initial pretreatment A. xylosoxidans isolates were not susceptible to cefiderocol in vitro. These testing results generally became available after treatment was started. Overall, clinical response was observed after 11 of 12 treatment courses. Patient 3 was stable without clinical improvement or worsening after 19 days of cefiderocol treatment; treatment was stopped when the index isolate was reported to be in vitro resistant to cefiderocol. Two adverse events were reported; patient 1 developed unexplained chylothorax. Patient 5 reported mild tingling and numbness of lower extremities during cefiderocol treatment, which was ameliorated with premedication with diphenhydramine. Microbiologic relapse occurred after 11 of 12 treatment courses. In patient 6, no further A. xylosoxidans was isolated in sputum cultures obtained 8, 16, and 28 weeks after the second course. The course of this pediatric patient, who also received phage therapy, was recently reported separately [4]. No treatment-emergent resistance was observed. In 5 instances where the pretreatment isolate was susceptible to cefiderocol, the posttreatment isolate was tested and remained susceptible.

Table 1.

Treatment Characteristics

Patient	Age, y	Treatment Course	Duration, d	Additional Antibioticsa	CFDC MIC Pretreatment	Syndrome	Lung Transplantation	Clinical Outcome	Microbiologic Relapse	CFDC Susceptibility Posttreatment
1	28	1a	42	TZP	0.12 mg/L (S)	Bacteremic pneumonia	Post	Resolved	Yes	Susceptible
		1b	42	None	≤0.03 mg/L (S)	Pneumonia	Post	Resolved	Yes	Not tested
2	17	2a	42	MEM, SXT, iTOB	1 mg/L (S)	Post-BOLT regimen	Post	Improved	Yes	Susceptible
3	29	3a	19	CZA, SXT	>64 mg/L (R)	Post-BOLT regimen	Post	Stable	Yes	Not tested
4	41	4a	21	ERV, DLX, iCST	0.06 mg/L (S)	Pneumonia	Pre	Resolved	Yes	Susceptible
		4b	21	ERV, iCST	17 mmb (I)	Bacteremia pneumonia	Post	Resolved	Yes	Susceptible
5	25	5a	21	ERV, IPM, iAMK	0.06 mg/L (S)	Pneumonia	NA	Resolved	Yes	Susceptible
6c	10	6a	14	MVB, phaged	32 mg/L (R)	Pneumonia	NA	Improved	Yes	Resistant
		6b	14	MVB, phaged	Achromobacter xylosoxidans 16 mg/L (R) Achromobacter ruhlandii 1 mg/L (S)	Pneumonia	NA	Resolved	No	NA
7e	56	7a	2	CST	Strain 1: 1 mg/L (S) Strain 2: 64 mg/L (R)	Pneumonia	Pre	Improved	Yes	Not tested
		7b	14	SXT	Not tested	Empyema	Post	Resolved	Yes	Not tested
8	28	8a	14	SXT, iAMK	20 mmb (S)	Pneumonia	Post	Resolved	Yes	Not tested

Abbreviations: BOLT, bilateral orthotopic lung transplantation; CFDC, cefiderocol; CST, colistin; CZA, ceftazidime-avibactam; DLX, delafloxacin; ERV, eravacycline; I, intermediate; iAMK, inhaled amikacin; iCST, inhaled colistin; IPM, imipenem-cilastatin; iTOB, inhaled tobramycin; MEM, meropenem; MIC, minimum inhibitory concentration; MVB, meropenem-vaborbactam; NA, not applicable; R, resistant; S, susceptible; SXT, trimethoprim-sulfamethoxazole; TZP, piperacillin-tazobactam.

^aAntibiotics with anti-gram-negative bacterial activity are listed.

^bAs determined by disk diffusion.

^cThis case was previously reported [4].

^dBacteriophage therapy (Ax2CJ45ϕ2).

^ePatient 7 underwent 2 courses of cefiderocol: a 2-day course prior to transplantation and 14 days after transplantation (see text for details).

Cefiderocol, A. xylosoxidans, and Lung Transplantation

In 6 patients, cefiderocol treatment was given peri–lung transplantation (Table 1). Two patients (2 and 3) received cefiderocol as a planned part of a combination peri–lung transplant antibacterial regimen. In patient 3, treatment was stopped after 19 days, when results of susceptibility testing revealed in vitro resistance to cefiderocol. Both patients were discharged in good condition after lung transplant hospitalization, but A. xylosoxidans was again isolated in both patients after treatment was stopped. Patient 1 was treated posttransplantation with two 6-week courses for A. xylosoxidans bacteremic pneumonia and nonbacteremic pneumonia. Patient 7 received 4 doses of cefiderocol for A. xylosoxidans pneumonia pretransplantation and was then transferred to another institution for lung transplantation. The posttransplant course was complicated by A. xylosoxidans empyema, treated successfully with 2 weeks of cefiderocol. Patient 8 was treated with a 14-day course of cefiderocol for A. xylosoxidans pneumonia diagnosed 12 days after lung transplantation with symptom resolution. Achromobacter xylosoxidans was again isolated from the respiratory tract after treatment completion for both patient 7 and 8.

DISCUSSION

In our series of 8 patients with CF who were treated with cefiderocol for A. xylosoxidans infections, we observed a number of important findings. First, baseline resistance was higher than previously reported [5]. In 3 of 8 patients, the first isolate—prior to any cefiderocol exposure—was not susceptible to cefiderocol, using P. aeruginosa breakpoints. In general, susceptibility data for cefiderocol show broad spectrum of activity toward highly resistant gram-negative bacteria, including Enterobacterales, P. aeruginosa, Acinetobacter baumannii, and Stenotrophomonas maltophilia [6–9]. A recent study by Rolston et al, investigating in vitro activity of cefiderocol against gram-negative isolates from patients with cancer, showed marked activity toward Achromobacter spp, with a 50% MIC of <0.03 μg/mL and a 90% MIC of 0.125 μg/mL [5].

Second, the apparent clinical efficacy of cefiderocol was good, with a clinical response noted after 11 of 12 treatment courses. Determination of treatment response can be difficult in respiratory infections in patients with CF and we were unable to include a control group for comparison. In a recent small case series of patients infected with A. baumannii, S. maltophilia, or Klebsiella pneumoniae, a 70% clinical success rate and 10% 30-day mortality were reported [10]. In a phase 3 randomized clinical trial in patients with nosocomial pneumonia, cefiderocol (n = 148) was found to be noninferior to meropenem (n = 150); all-cause mortality at 14 days was 12% in both groups [11]. Data from the CREDIBLE-CR study on efficacy and safety of cefiderocol or best available therapy for the treatment of serious infections caused by carbapenem-resistant Gram-negative bacteria (ClinicalTrials.gov identifier NCT02714595) was recently published [12]. The efficacy of cefiderocol (n = 101) was compared to best available therapy (BAT; n = 49) in severe infections due to carbapenem-resistant bacteria. While the primary clinical and microbial outcomes were equivalent between cefiderocol and BAT, a significantly higher mortality was observed in patients treated with cefiderocol (34% vs 18%). The reason for this difference in mortality was unclear. Achromobacter xylosoxidans was not a pathogen in the CREDIBLE-CR trial. In our small case series, all patients were alive 180 days after infection. While CF patients have not been included in any cefiderocol clinical trials to date, they are often colonized with extensively resistant gram-negative bacteria. Therefore, observational data such as those reported here will help define the role of cefiderocol in this setting.

The third observation was that microbiologic relapse was near-universal. This is similar to the frequently observed persistently colonized state seen after other gram-negative pneumonias in CF patients. Except for a pediatric patient who received combination therapy with bacteriophages, all patients had subsequent isolation of A. xylosoxidans after treatment. Importantly, microbiologic relapse was not associated with apparent treatment-emergent resistance. A possible contributing factor to microbiologic relapse is the concentration of cefiderocol in epithelial lining fluid, which is estimated at approximately one-tenth of plasma concentrations in healthy adults [13].

Cefiderocol was reasonably well tolerated in our patients. Tolerability of cefiderocol is generally equivalent to other broad-spectrum cephalosporins. In a phase 2 noninferiority trial comparing imipenem-cilastatin to cefiderocol for complicated urinary tract infections, serious adverse events in the cefiderocol arm occurred similarly or slightly less often than the imipenem-cilastatin arm (5% vs 8%, respectively) [14]. Similarly, in the nosocomial pneumonia trial, the rate of drug-related treatment-emergent adverse events was 10% vs 11% in patients treated with cefiderocol vs meropenem [11].

In several of our patients, cefiderocol was used in the setting of lung transplantation. In general, cefiderocol is an attractive option for a peritransplant regimen. Cefiderocol has no important drug–drug interactions with medications commonly used after transplantation, is well tolerated, and lacks the substantial anaerobic activity of comparator agents like piperacillin-tazobactam. The availability of an agent with in vitro activity against pretransplant pathogens is of crucial importance when evaluating peritransplant risk of poor outcomes. The optimal duration of posttransplant prophylaxis remains uncertain.

In summary, the role of cefiderocol in severe A. xylosoxidans infections is still being defined. While the clinical outcomes reported here are encouraging, the microbiologic relapse and high baseline resistance rate are concerning. Further studies are needed to clarify how we should use cefiderocol in CF patients with A. xylosoxidans infections.

Note

Potential conflicts of interest. A. M. L. has performed consulting for Shionogi and MicrogenDx. J. A. has served on speakers’ bureaus and advisory boards for Shionogi and Merck, and has served as a speaker for bioMérieux. K. A. has served on advisory boards for Shionogi and Becton Dickinson. M. J. B. holds equity in Adaptive Phage Therapeutics. R. K. S. has served on advisory boards for Shionogi, Merck, Entasis, Utility, Venatorx, and Summit; and has received grants from Shionogi, Merck, Venatorx, Allergan, and Tetraphase. D. v. D. has served on advisory boards for Actavis, Tetraphase, Sanofi-Pasteur, MedImmune, Astellas, Merck, Allergan, T2Biosystems, Roche, Achaogen, Neumedicine, Shionogi, Karius, Wellspring, QPex, Pfizer, Entasis, and Utility; and has received grants from the National Institutes of Health. All other authors report no potential conflicts.

All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.