Variability in Bacteriophage and Antibiotic Sensitivity in Serial Pseudomonas aeruginosa Isolates from Cystic Fibrosis Airway Cultures over 12 Months

Abstract

Antibiotic treatment for Pseudomonas aeruginosa (Pa) in cystic fibrosis is limited in efficacy and may lead to multi-drug resistance (MDR). Alternatives such as bacteriophages are being explored but well designed, and controlled trials are crucial. The rational selection of patients with bacteriophage susceptible infections is required for both safety and efficacy monitoring. We questioned whether bacteriophage susceptibility profiles were constant or variable over time, variability having been reported with antibiotics. Serial Pa isolates (n = 102) from 24 chronically infected cystic fibrosis (CF) patients over one year were investigated with plaque and antibiotic disc diffusion assays. Variable number tandem repeat (VNTR) analysis identified those patients with >1 isolate. A median (range) of 4 (3-6) isolates/patient were studied. Twenty-one (87.5%) individuals had a single VNTR type; three (12.5%) had two VNTR types at different times. Seventy-five percent of isolates were sensitive to bacteriophage at ≥ 1 concentration; 50% of isolates were antibiotic multidrug resistant. Serial isolates, even when representing a single VNTR type, varied in sensitivity to both bacteriophages and antibiotics. The rates of sensitivity to bacteriophage supports the development of this therapy; however, the variability in response has implications for the selection of patients in future trials which must be on the basis of current, not past, isolate testing.

Keywords: Pseudomonas aeruginosa; adjunctive therapy; antimicrobial resistance; bacteriophage; cystic fibrosis; novel antimicrobials; pulmonary infection.

Figure 2

Antibiograms and bacteriophage sensitivity profiles for 4 patients with the Liverpool Epidemic Strain. Isolates are graphically represented by different symbols chronologically throughout the year. The first isolate is represented by , with subsequent isolates , , , , if there were 5 isolates through the course of 2016. Open symbols represent different strains, identified by VNTR analysis. Abbreviations—AMK: amikacin; AZM: aztreonam; CEF: ceftazidime; CIP: ciprofloxacin; GEN: gentamicin; MER: meropenem; PTZ: piperacillin-tazobactam; SXT: co-trimoxazole (no EUCAST breakpoint); TOB: tobramycin. Disc diffusion assay for antibiotic sensitivity testing was performed once at each time point and EUCAST sensitivity cutoffs are indicated by the bars. Biological triplicates of phage assay are represented individually and bars indicate both neat phage (>2 × 10² PFU/mL) and dilute phage (>2 × 10⁵ PFU/mL) cutoffs. (a–d) Patients G, K, M and N were colonized with the LES. All strains qualified as MDR based on antibiograms and 14/18 (78%) LES isolates showed some lytic activity to phage using the lower sensitivity cutoff. Using the higher sensitivity cutoff of 10³ serial dilutions, 9/18 (50%) showed sensitivity to phage. (a) Patient G had isolates that are broadly resistant to antibiotics-sensitive to tobramycin at 2 and to meropenem at 1 time points, respectively, but broadly sensitive to phage at all time points; (b) Patient K was colonized by a different strain at the first time point, but all subsequent isolates were LES with lytic activity to phage; (c) Patient M’s LES strain was sensitive only to piperacillin-tazobactam but sensitive to phage at all time points; and (d) Patient N was colonized with LES resistant to all antibiotics tested. Only 1 of this patient’s 5 isolates showed lytic activity to the neat phage cocktail and no isolates were deemed sensitive using the higher cutoff.

Figure 1

Example antibiograms and bacteriophage sensitivity profiles for 4 different patients chosen to illustrate variability in results. Isolates are graphically represented by different symbols chronologically throughout the year. The first isolate is represented by , with subsequent isolates , , , , if there were 5 isolates through the course of 2016. Open symbols represent different strains, identified by VNTR analysis. Abbreviations: AMK: amikacin; AZM: aztreonam; CEF: ceftazidime; CIP: ciprofloxacin; GEN: gentamicin; MER: meropenem; PTZ: piperacillin-tazobactam; SXT: co-trimoxazole (no EUCAST breakpoint); TOB: tobramycin. Disc diffusion assay for antibiotic sensitivity testing was performed once at each time point and EUCAST sensitivity cutoffs are indicated by the bars. Biological triplicates of phage assay are represented individually and bars indicate both neat phage (>2 × 10² PFU/mL) and dilute phage (>2 × 10⁵ PFU/mL) cutoffs. (a) Patient H is an example of a patient colonized with MDR Pa, sensitive only to tobramycin and phage at all 5 time points throughout 2016. (b) Patient B represents a patient harboring a Pa strain showing high variability to all antibiotics as well as the phage cocktail. Of note, this patient’s first isolate was resistant to the phage cocktail, while subsequent isolates throughout the year showed susceptibility. (c) Patient D harbored a Pa strain resistant to the phage cocktail at the first two timepoints in the year, but sensitive throughout the rest of 2016. (d) Patient F was one of three from whom two distinct Pa strains were identified. The first two isolates in the year are one strain, while the subsequent two are another. There are distinct antibiograms for the two strains, as well as a distinct pattern of susceptibility/resistance to the phage cocktail.