Acquired blaCfxA-3 carried by a conjugative transposon or duplicated intrinsic blaCME-3 mediates cefiderocol resistance in Elizabethkingia anophelis clinical isolates.

Objectives: Elizabethkingia spp. are resistant to multiple antibiotics. This study aimed to determine in vitro and in vivo activities of cefiderocol against Elizabethkingia spp. and to investigate resistance mechanisms.

Methods: Bloodstream isolates were collected from four hospitals. In vitro and in vivo activities were determined using broth microdilution and the wax moth model, respectively. Genome comparison and gene editing were used to confirm the contribution of target genes. Conjugation experiments and serial passage were used to determine transferability and stability, respectively. A MIC of ≤ 4 mg/L was designated as the susceptibility breakpoint.

Results: Among 228 non-duplicated isolates, 226 exhibited a MIC of ≤ 4 mg/L with MIC_50/90 of 1/2 mg/L. Two isolates had a MIC of 128 mg/L; both source patients had multiple comorbidities, were ventilator-dependent, and had not received cefiderocol previously. Resistance was attributable to acquisition of bla_CfxA-3, carried by a conjugative transposon from Prevotella jejuni, and duplication of intrinsic bla_CME-3, which led to its overexpression. tetQ coexisted with bla_CfxA-3 in this conjugative transposon and minocycline facilitated its transfer among E. anophelis. Antibiotics prescribed for source patients did not induce bla_CME-3 duplication. The stabilities of bla_CfxA-3 and double bla_CME-3 were 100% and > 90%, respectively, after 10-day serial passage. Cefiderocol failed to rescue moth larvae infected with resistant strains, but removal of resistance mechanisms restored in vivo efficacy.

Conclusions: Cefiderocol was in vitro and in vivo active against Elizabethkingia spp. but resistance may emerge due to the availability, transferability, and/or stability of resistance mechanisms.