Impact of bacterial species and baseline resistance on fosfomycin efficacy in urinary tract infections

• Published in: Journal of antimicrobial chemotherapy 2020-04, Vol.75 (4), p.988-996
• Main Authors: Abbott, Iain J, Dekker, Jordy, van Gorp, Elke, Wijma, Rixt A, Raaphorst, Merel N, Klaassen, Corné H W, Meletiadis, Joseph, Mouton, Johan W, Peleg, Anton Y
• Format: Article
• Language: English
• ISSN: 0305-7453; 1460-2091
• DOI: 10.1093/jac/dkz519

Abstract Objectives To assess the antibacterial effects of a single 3 g oral fosfomycin dose on Escherichia coli and Klebsiella pneumoniae clinical isolates within a dynamic bladder infection model. Methods An in vitro model simulating dynamic urinary fosfomycin concentrations was used. Target fosfomycin exposure (Cmax = 1984 mg/L and Tmax = 7.5 h) was validated by LC-MS/MS. Pharmacodynamic responses of 24 E. coli and 20 K. pneumoniae clinical isolates were examined (fosfomycin MIC ≤0.25–128 mg/L). Mutant prevention concentration (MPC), fosfomycin heteroresistance, fosfomycin resistance genes and fosA expression were examined. Pathogen kill and emergence of high-level resistance (HLR; MIC >1024 mg/L) were quantified. Results Following fosfomycin exposure, 20 of 24 E. coli exhibited reductions in bacterial counts below the lower limit of quantification without regrowth, despite baseline fosfomycin MICs up to 128 mg/L. Four E. coli regrew (MIC = 4–32 mg/L) with HLR population replacement. At baseline, these isolates had detectable HLR subpopulations and MPC >1024 mg/L. All E. coli isolates were fosA negative. In contrast, 17 of 20 K. pneumoniae regrew post exposure, 6 with emergence of HLR (proportion = 0.01%–100%). The three isolates without regrowth did not have a detectable HLR subpopulation after dynamic drug-free incubation. All K. pneumoniae had MPC >1024 mg/L and were fosA positive. WGS analysis and fosA expression failed to predict fosfomycin efficacy. Conclusions E. coli and K. pneumoniae isolates demonstrate discrepant responses to a single fosfomycin dose in a dynamic bladder infection in vitro model. Treatment failure against E. coli was related to an HLR subpopulation, not identified by standard MIC testing. Activity against K. pneumoniae appeared limited, regardless of MIC testing, due to universal baseline heteroresistance.