Ceftolozane/tazobactam heteroresistance in cystic fibrosis-related Pseudomonas aeruginosa infections

Abstract

Objectives: Cystic fibrosis (CF) patients are often colonized with Pseudomonas aeruginosa. During treatment, P. aeruginosa can develop subpopulations exhibiting variable in vitro antimicrobial (ABX) susceptibility patterns. Heteroresistance (HR) may underlie reported discrepancies between in vitro susceptibility results and clinical responses to various ABXs. Here, we sought to examine the presence and nature of P. aeruginosa polyclonal HR (PHR) and monoclonal HR (MHR) to ceftolozane/tazobactam in isolates originating from CF pulmonary exacerbations.

Methods: This was a single-centre, non-controlled study. Two hundred and forty-six P. aeruginosa isolates from 26 adult CF patients were included. PHR was defined as the presence of different ceftolozane/tazobactam minimum inhibitory concentration (MIC) values among P. aeruginosa isolates originating from a single patient specimen. Population analysis profiles (PAPs) were performed to assess the presence of MHR, defined as ≥4-fold change in the ceftolozane/tazobactam MIC from a single P. aeruginosa colony.

Results: Sixteen of 26 patient specimens (62%) contained PHR P. aeruginosa populations. Of these 16 patients, 6 (23%) had specimens in which PHR P. aeruginosa isolates exhibited ceftolozane/tazobactam MICs with categorical differences (i.e. susceptible versus resistant) compared to results reported as part of routine care. One isolate, PSA 1311, demonstrated MHR. Canonical ceftolozane/tazobactam resistance genes were not found in the MHR isolates (MHR PSA 1311 or PHR PSA 6130).

Conclusions: Ceftolozane/tazobactam PHR exists among P. aeruginosa isolates in this work, and approximately a quarter of these populations contained isolates with ceftolozane/tazobactam susceptibiilty interpretations different from what was reported clinically, supporting concerns surrounding the utility of traditional susceptibility testing methodology in the setting of CF specimens. Genome sequencing of isolates with acquired MHR to ceftolozane/tazobactam revealed variants of unknown significance. Future work will be centred on determining the significance of these mutations to better understand these data in clinical context.

Figure 1.

Ceftolozane/tazobactam, ceftazidime/avibactam and imipenem/relebactam MIC distributions against P. aeruginosa isolates from 26 adult CF patients.

Figure 2.

Examples of isolates demonstrating PHR. Isolates 1940 (MIC 0.5 mg/L) and 1943 (MIC 0.125 mg/L) demonstrate 2-fold change in ceftolozane/tazobactam MIC (Patient 8). Isolates 2746 (MIC 2 mg/L) and 2751 (MIC <0.064 mg/L) demonstrate >2-fold change in ceftolozane/tazobactam MIC (Patient 12). Isolates 3358 (MIC >256 mg/L) and 3366 (MIC 2 mg/L) demonstrate a change in susceptibility interpretation. S, susceptible; R, resistant.

Figure 3.

Results from PSA 1311 (ceftolozane/tazobactam MIC 1/4 mg/L) PAPs at 2-, 4-, 6- and 8-fold the ceftolozane/tazobactam MIC. PSA 1311 demonstrated growth on PAPs up to 4-fold the MIC (16/4 mg/L). No growth was seen at 6- or 8-fold the ceftolozane/tazobactam MIC. The isolate recovered at 4-fold ceftolozane/tazobactam MIC was designated PSA 6130 and saved for additional analyses.

Figure 4.

Time–kill experiments displaying the activity of antimicrobials alone and in combination at trough concentrations against (a) PSA 1311 and (b) PSA 6130 (MIC). Against PSA 1311 (ceftolozane/tazobactam MIC 1 mg/L), bacterial regrowth from baseline was observed by 24 h. No reduction in bacterial density over time was observed with ciprofloxacin (MIC 2 mg/L) and amikacin (MIC 32 mg/L) monotherapy. At 6 h, the combinations of amikacin plus ceftolozane/tazobactam and ciprofloxacin plus ceftolozane/tazobactam reduced the bacterial density, but regrowth above baseline was observed by 24 h. Against PSA 6130 (ceftolozane/tazobactam MIC 32 mg/L), there were no reductions in bacterial density in the ceftolozane/tazobactam, ciprofloxacin (MIC 1 mg/L) and amikacin (MIC 32 mg/L) arms.

Figure 5.

Phylogenetic relationships between isolates (labelled ‘patient number.sample number’) displaying ceftolozane/tazobactam HR. HR P. aeruginosa isolates from CF respiratory specimens exhibit a high level of clonality.