KPC Presence in Pseudomonas aeruginosa Has Minimal Impact on the In Vivo Efficacy of Carbapenem Therapy

Abstract

While reports of Klebsiella pneumoniae carbapenemase (KPC) production among Pseudomonas aeruginosa strains have emerged from a number of countries worldwide, outcome data are lacking. This is the first report evaluating how KPC production in P. aeruginosa impacts the efficacy of carbapenems by using the murine thigh infection model. Our findings suggest that the impact of KPC-2 in vivo is less pronounced than would be anticipated based on the in vitro potency.

TEXT

Carbapenemases represent the most versatile family of β-lactamases, with a wide spectrum of β-lactam-hydrolyzing activity compared with that of other enzymes. The molecular class A carbapenemases of the Klebsiella pneumoniae carbapenemase (KPC) family (KPC-1 to KPC-12) are a potent group of carbapenemases recognized in numerous pathogens. Although KPC-type carbapenemases have been predominantly found in Klebsiella pneumoniae, reports of these enzymes in other pathogens are commonplace (1–4); of interest, reports of KPC production among Pseudomonas aeruginosa have appeared, albeit uncommonly, since the first detection of KPC-2 on a plasmid in 2007 (4, 5).

P. aeruginosa is a problematic nosocomial pathogen due to intrinsic and acquired antibiotic resistance factors, as well as high rates of associated mortality and morbidity. Since carbapenems have been extensively used for the treatment of this organism, acquisition of the KPC enzyme in P. aeruginosa has the potential to further threaten the utility of its use. Carbapenem resistance in P. aeruginosa can be mediated by several mechanisms, including both enzyme- and non-enzyme-mediated processes; therefore, it is important to understand how KPC production impacts the antimicrobial activity of the carbapenems. Previous studies conducted by our group evaluating combination therapy with doripenem and ertapenem revealed that the combination demonstrated enhanced antimicrobial activity against K. pneumoniae isolates harboring KPC enzymes compared with that of either agent alone (6). Herein, we evaluated the antimicrobial activity of doripenem and ertapenem alone and in combination against carbapenem-resistant P. aeruginosa with and without KPC enzyme production, using the murine thigh infection model with human-simulated exposures of these drugs.

Commercially available doripenem (Ortho-McNeil-Janssen Pharmaceuticals, Inc., Raritan, NJ) and ertapenem (Merck & Co., Inc., Whitehouse Station, NJ) were used for in vivo studies. Immediately prior to experimentation, each drug was reconstituted and diluted to achieve the desired concentration with normal saline. Three doripenem-resistant P. aeruginosa isolates, one KPC-2 producing and two non-KPC producing, were utilized (Table 1). Doripenem MIC values were determined via Etest according to the manufacturer's specifications (bioMérieux North America) or the broth microdilution method as defined by Clinical and Laboratory Standards Institute (CLSI) guidelines (7). The study was reviewed and approved by the Hartford Hospital Institutional Animal Care and Use Committee. Animals were maintained and used in accordance with National Research Council recommendations and were provided food and water ad libitum. For in vivo studies, the standard immunocompetent and neutropenic mouse thigh infection models were conducted as previously described (8). Human-simulated doses of 2 g doripenem every 8 h as a 4-h infusion and 1 g ertapenem every 24 h as a 30-min infusion, as determined in prior analyses and confirmed in multiple studies over several years, were utilized (6, 9–11). Of note, the doripenem regimen evaluated in this study, while outside FDA approvals, is consistent with strategies employed to optimize carbapenem pharmacodynamics (i.e., high dose, prolonged infusion) against multidrug-resistant organisms. To compare the antimicrobial efficacy between regimens, a Student t test or a Mann-Whitney U test, if the data were not normally distributed, was used. A P value of ≤0.05 was defined as statistically significant.

Table 1.

MICs and mechanisms of carbapenem resistance in P. aeruginosa isolates

Isolate	Doripenem MIC (μg/ml)	KPC producing
PSA JJ1-29	32	No
PSA AZ11-18	64	No
PSA 1463	>64	Yes

The results of the efficacy studies in immunocompetent animals are shown in Fig. 1, and the results of neutropenic studies are shown in Fig. 2. In the immunocompetent model, the numbers of organisms recovered from the thighs of infected animals serving as 0-h controls (2 h postinoculation) ranged from 6.42 to 7.09 log₁₀ CFU. The bacterial densities of control mice after 24 h ranged from 7.04 to 9.19 log₁₀ CFU. While 50.0% of mice in the control group and 50.0% of ertapenem treatment mice expired before the end of the 24-h study period, animals that received either doripenem or combination therapy survived to the completion of the study. As anticipated, ertapenem monotherapy yielded bacterial densities similar to those of untreated control animals for all 3 isolates evaluated. Regardless of the in vitro doripenem MIC, antimicrobial activity for both doripenem monotherapy and combination therapy were statistically greater against the KPC-2-producing P. aeruginosa isolate than each of the 2 non-KPC-producing P. aeruginosa isolates (P < 0.01). There was no statistical difference in antimicrobial activities between doripenem monotherapy and combination therapy for any of the isolates.

Fig 1.

Change in log₁₀ CFU (mean ± SD) for human-simulated regimens of doripenem and ertapenem, alone and in combination, for KPC-2-producing and non-KPC-producing Pseudomonas aeruginosa isolates in the immunocompetent murine thigh infection model.

Fig 2.

Change in log₁₀ CFU (mean ± SD) for human-simulated regimens of doripenem and ertapenem, alone and in combination, for KPC-2-producing and non-KPC-producing Pseudomonas aeruginosa isolates in the neutropenic murine thigh infection model.

In the neutropenic studies, the numbers of organisms recovered from the thighs of infected animals serving as 0-h controls ranged from 4.82 to 5.54 log₁₀ CFU, and bacterial densities of control mice at 24 h ranged from 8.67 to 9.27 log₁₀ CFU. The mortalities of the control and ertapenem groups were 66.7% and 58.3%, respectively. All animals that received either doripenem or combination therapy survived to the completion of the study. While ertapenem monotherapy yielded bacterial densities similar to untreated control animals for all 3 isolates evaluated, statistically better activity was noted for doripenem monotherapy and combination therapy against the KPC-2-producing isolate (PSA 1463) than that of the non-KPC producer PSA AZ11-18 (P < 0.001). Similarly to the immunocompetent model, there was no difference in antibacterial activity between doripenem monotherapy and combination therapy for any of the isolates.

When comparing the activity of doripenem treatment between KPC and non-KPC-producing P. aeruginosa isolates, we found that despite having reduced in vitro potency, greater activity was observed against the KPC-2-producing P. aeruginosa isolate. Previous studies conducted by our group with doripenem monotherapy found maximal activity against P. aeruginosa isolates with MICs of ≤16 μg/ml and variable activity against isolates with MICs of 32 μg/ml; organisms with MICs above 32 g/ml were not evaluated (9). As such, the results of our current studies with the non-KPC-producing isolates, demonstrating variable activity against JJ1-29 and no activity against AZ11-19, are consistent with our previous conclusions and further emphasize the discordance between the in vitro and in vivo findings of the KPC-producing strain studied herein.

We saw no difference in activity between doripenem monotherapy and combination therapy. Given the proposed mechanism of action, namely, ertapenem acting as a suicide inhibitor of the KPC enzyme (12), this observation was not unexpected for the non-KPC-producing strains. For the KPC-2-producing strain, the noted activity of doripenem monotherapy suggests that the KPC enzyme played little role in vivo; thus, the addition of ertapenem would not be expected to change the overall outcome.

It is important to note that given the relative rarity of KPC production in P. aeruginosa, compared with that of other Gram-negative organisms, this initial study evaluated only a single KPC-producing strain. Moreover, while we can say for certain that these clinical strains either produced KPC enzymes or did not, the full genetic backgrounds and relative contribution of each resistance mechanism that resulted in the observed carbapenem MICs have not been elucidated. As such, future studies evaluating a larger number strains would be useful and warranted to validate these initial findings. Moreover, the inclusion of genetically described and/or isogenic strains in future studies would be helpful in extricating KPC production as an efficacy determinate.

In conclusion, while the presence of KPC-2 has the potential to enhance the phenotypic profile of P. aeruginosa, our data, as determined from a single isolate, suggest that the contribution of KPC-2 in vivo is less pronounced in the background of multiple mechanisms of carbapenem resistance in pseudomonal isolates, and accurate identification of specific carbapenem resistance mechanisms may help determine appropriate therapy for such strains.