|Home | About | Journals | Submit | Contact Us | Français|
Doripenem is a new carbapenem antimicrobial with activity against a range of gram-negative organisms, including Pseudomonas aeruginosa. Previous animal studies have shown efficacy of a 500-mg dose of doripenem given as a 4-h infusion against P. aeruginosa with MICs of ≤4 μg/ml. The purpose of this study is to evaluate the efficacy of 1- and 2-g-dose prolonged infusions of doripenem against a wide range of P. aeruginosa isolates in the neutropenic murine thigh model. Eighteen clinical P. aeruginosa isolates (MIC range, 2 to 32 μg/ml) were used; 15 of these were multidrug resistant. After infection, groups of mice were administered doripenem doses designed to simulate the free time above the MIC (fT>MIC) observed in humans given 1 or 2 g of doripenem every 8 h as a 4-h infusion. Efficacy correlated well with published fT>MIC bactericidal targets of 40%. After 24 h, 1- and 2-g doses achieved approximately ≥2 log decreases in CFU against isolates with MICs of ≤8 and 16 μg/ml, respectively (fT>MIC range, 52.5 to 95%). Results with organisms with higher MICs, where fT>MIC was 0%, were variable, including both increases and decreases in CFU. Compared with 1-g doses, statistically greater efficacy was noted for 2-g doses against three of the eight isolates with MICs of ≥16 μg/ml. While MIC distributions of P. aeruginosa at present necessitate increased exposures for only the most-resistant isolates, the ability of increased doses to achieve pharmacodynamic targets and the efficacy observed when these targets were attained could prove useful when these resistant isolates are encountered.
Doripenem is the newest Food and Drug Administration (FDA)-approved addition to the carbapenem class of antimicrobials. At present, doripenem is approved in the United States for the treatment of complicated intra-abdominal infections and complicated urinary tract infections at a dose of 500 mg every 8 h for patients with normal renal function (15). As with all β-lactam antibiotics, the efficacy of doripenem is determined by the percentage of time the free drug concentrations exceed the MIC of the infecting organism (fT>MIC) (10, 11). A simple strategy to optimize this exposure-response relationship is to increase the length of drug infusion (5, 11). Thus, while approved doripenem doses utilize standard 1-h infusions, a recent clinical trial used an increased infusion time of 4 h and found favorable results for the treatment of ventilator-associated pneumonia (3).
One drawback to increasing infusion times is that it results in a reduction in the maximal concentrations attainable by a given dose. For new medications, such as doripenem, resistance rates are often low, and simply prolonging the infusion provides ample exposure against the majority of organisms in the MIC distribution. Given present doripenem MIC distributions against most gram-negative organisms, this is certainly the case (16). However, for a portion of gram-negative pathogens, there are small groups of doripenem-resistant isolates that fall outside the range readily achievable by standard prolonged infusions. One such example is Pseudomonas aeruginosa, for which United States clinical isolate MICs range from 0.03 to >32 μg/ml (16). In an in vivo murine study using human simulated exposures, 500 mg of doripenem, despite 4-h infusion times, was unable to achieve appreciable efficacy against P. aeruginosa isolates with MICs above 4 μg/ml (11). Similarly, based solely on pharmacodynamic target attainments, Monte Carlo analyses have highlighted the inability of standard doripenem doses to achieve targets at MICs above 4 μg/ml, despite prolonged infusion times (2, 9). To combat these higher-MIC organisms, these analyses have proposed the use of doripenem at increased doses coupled with prolonged infusions. Given these observations and the high propensity for P. aeruginosa to be pathogenic in infections commonly treated with doripenem (3, 13, 17), we evaluated the efficacy of high-dose prolonged infusions of doripenem against a wide distribution of P. aeruginosa isolates by simulating human concentration-time profiles in mice infected via the neutropenic thigh model.
Analytical-grade doripenem (Johnson & Johnson Pharmaceutical Research & Development, LLC) was utilized for all in vitro and in vivo studies. Based on the sponsor's supplied potency, doripenem powder was weighed in a quantity sufficient to achieve the required concentration and reconstituted immediately prior to use. Doripenem solutions were stored at room temperature, protected from light, and discarded after 8 h.
A total of 18 clinical isolates of P. aeruginosa collected from patients at Hartford Hospital in Hartford, CT, were used in this analysis. The MIC of each isolate was determined, in triplicate, by broth microdilution, using methods outlined by the Clinical and Laboratory Standards Institute (4), and the modal MIC was reported. Isolates were maintained in double-strength skim milk (BD Biosciences, Sparks, MD) at −80°C. Each isolate was subcultured twice on Trypticase soy agar with 5% sheep blood (BD Biosciences) and grown at 35°C prior to use in the experiments.
Pathogen-free, female ICR mice weighing approximately 25 g were acquired from Harlan Sprague Dawley, Inc. (Indianapolis, IN) and utilized throughout these experiments. The study was reviewed and approved by the Hartford Hospital Institutional Animal Care and Use Committee. Animals were maintained and used in accordance to National Research Council recommendations and provided food and water ad libitum. Mice were rendered neutropenic with 100- and 150-mg/kg intraperitoneal injections of cyclophosphamide (Cytoxan; Bristol-Myers Squibb, Princeton, NJ) given 1 and 4 days prior to inoculation, respectively. Three days prior to inoculation, mice were also given a single 5-mg/kg intraperitoneal injection of uranyl nitrate. This produces a predictable degree of renal impairment to slow drug clearance (1). Two hours prior to the initiation of antimicrobial therapy, each thigh was inoculated intramuscularly with 0.1 ml solution containing approximately 106 CFU/ml of the test isolate.
Utilizing the pharmacokinetic parameters described in a previous murine pharmacokinetic analysis (11), we determined a dosing regimen, in mice, that simulated the fT>MIC observed in humans given 1- and 2-g doses of doripenem every 8 h as a 4-h prolonged infusion across a range of MICs (WinNonlin version 5.0.1; Pharsight, Mountain View, CA). Exposures in humans were derived from the extrapolation of data generated in healthy volunteers dosed with 500 mg doripenem every 8 h as a 4-h infusion (Johnson & Johnson Pharmaceutical Research & Development, LLC, data on file). Previously described protein binding values of 25.2 and 8.5% were used to calculate free drug concentrations in mice and humans, respectively (2, 8).
Confirmatory pharmacokinetics studies were undertaken in infected mice prior to the use of these regimens in the pharmacodynamic analyses. For these studies, mice were dosed with the above-described calculated regimens, and groups of six mice were euthanized at eight time points throughout the dosing interval. Blood samples were taken via cardiac puncture, and serum samples were stored at −80°C until analysis. Doripenem concentrations were analyzed using a validated high-performance liquid chromatography assay (18). Interday and intraday coefficients of variation for high and low standards were less than 5%.
For each of the 18 P. aeruginosa isolates, groups of three mice were administered 1- or 2-g-dose human simulated regimens of doripenem beginning 2 hours after the initiation of infection. All doses were administered as 0.2-ml subcutaneous injections and consisted of three 8-h dosing intervals (i.e., 24 h). To serve as control animals, an additional group of mice were administered normal saline at the same volume, route, and frequency as the treatment regimens. All animals were harvested 24 h after the initiation of therapy. Mice that failed to survive for 24 h were harvested at the time of expiration (11). The harvesting procedure for all study mice began with euthanization by CO2 exposure followed by cervical dislocation. After sacrifice, thighs were removed and individually homogenized in 5 ml of normal saline. Serial dilutions of the thigh homogenate were plated onto Trypticase soy agar with 5% sheep blood for CFU determination. In addition to the above-mentioned treatment and control groups, another group of three infected, untreated mice were harvested at the initiation of dosing (i.e., 0 h control). Efficacy, designated as the change in bacterial density, was calculated as the change in log10 bacterial CFU/ml obtained for doripenem-treated mice after 24 h from the 0 h control animals. A comparison of efficacies between 1- and 2-g doses against each isolate was made using a Student t test or Mann-Whitney U test if data were not normally distributed. A P value of <0.05 was defined a priori as statistically significant.
The phenotypic profiles for each of the 18 P. aeruginosa isolates utilized in this study are listed in Table Table1.1. As evidenced by the fact that 83% of the isolates are multidrug resistant (nonsusceptibility to ≥3 antimicrobials commonly used to treat P. aeruginosa infections), this conglomeration of isolates represented a clinically challenging population with very few treatment options. For doripenem, MICs ranged from 2 to 32 μg/ml.
To reproduce human exposures in mice, each 8-h dosing interval required eight individual doses. For the 1-g-dose regimen, doses of 11, 4.5, 9, 9, 9, 9, 1.5, and 0.75 mg/kg were given at 0, 0.5, 1.5, 2.5, 3.5, 4.5, 6, and 7.5 h, respectively. The same schedule was used for the 2-g-dose simulation, but the respective doses were increased to 22, 9, 18, 18, 18, 18, 3, and 1.5 mg/kg. The calculated free drug pharmacokinetic profiles for the simulated regimens and respective human exposures are displayed in Fig. Fig.1,1, as are the confirmatory results of the in vivo pharmacokinetic studies. In mice, the maximum concentration of free drug in serum (fCmax; μg/ml) and free area under the concentration-time curve (fAUC; μg·h/ml) attained for 1-g-dose (15.9 and 68.6, respectively) and 2-g-dose (31.9 and 137.1, respectively) simulations were similar to those observed in humans given 1 (15.6 and 64.7)- and 2 (31.2 and 129.3)-gram doses. More importantly, the pharmacodynamic parameter of interest, fT>MIC, was in close association between mice and humans across a range of MICs (Table (Table22).
At the start of dosing, 0 h control mice displayed a mean bacterial burden of 5.03 ± 0.31 log10 CFU/ml per thigh. This number increased by an average of 3.73 ± 0.62 logs in untreated mice after 24 h. While all doripenem-treated mice survived to the 24 h sampling point, a number of control mice died prior to that time. The results of the efficacy studies are shown in Fig. Fig.2.2. Compared with 0 h control animals, 1-g-dose simulations produced an approximately ≥2 log CFU decrease for isolates with MICs of 2 to 8 μg/ml. Similar results were noted with 2-g-dose simulations for MICs of 2 to 16 μg/ml. Moreover, compared with the 1-g-dose regimen, the 2-g-dose simulation resulted in statistically greater efficacy for three of the eight isolates with MICs of 16 and 32 μg/ml (P < 0.05).
Doripenem is one of the first antibiotics to be utilized in pharmacodynamically enhancing prolonged infusions during a drug registration trial. In that trial, 500 mg of doripenem was given as a 4-h infusion to patients infected with ventilator-associated pneumonia (3). Extending the infusion time is not a new concept for β-lactam antibiotics; in fact, this practice was first described over 50 years ago (6). The principle behind such a concept is centered on the time-dependent nature of the efficacy of this class of antimicrobials. Namely, the efficacy of β-lactams is determined by the fT>MIC of the infecting organism (5). While extending the infusion is often useful for optimizing the fT>MIC for organisms on the lower end of the MIC range, use of increased doses in conjunction with prolonged infusions is an approach to combat organisms with higher MICs (12). In the present analysis, through use of the murine thigh infection model, we evaluated the efficacy of human doses two and four times those used in the above-described clinical trial against a diverse population of P. aeruginosa isolates.
Within our results, there is a clear delineation between the results of the 1-g-dose simulation at MICs of ≤8 μg/ml and those at MICs of ≥16 μg/ml. For the isolates with MICs at or below 8 μg/ml, maximal efficacy was attained against all isolates, while above this MIC, the results were variable, including both increases and decreases in CFU. Many studies have identified an fT>MIC target of 40% for carbapenems to achieve bactericidal activity and 20% for bacteriostatic activity (20). As such, the findings of our analysis are not unexpected, considering the fT>MIC of the 1-g-dose regimen at MICs of 8 and 16 μg/ml are 52.5 and 0%, respectively. Similar results were noted for the 2-g-dose simulations but were 1 MIC dilution higher (i.e., bactericidal activity at MICs of ≤16 μg/ml and variable efficacy at an MIC of 32 μg/ml) and again correlated well with pharmacodynamic target attainments. While our data are consistent with these targets, a similar study previously conducted by our group found somewhat different results (11). In that study, 4-h prolonged infusions of 500 mg doripenem every 8 h produced inconsistent results against P. aeruginosa with MICs of 4 μg/ml, despite an fT>MIC of 50%. Two of the isolates with MICs of 4 μg/ml used in that study were repeated in the present study to assess the effects of higher exposures. In the previous study, isolate 944 resulted in a near 1 log growth after 24 h, and isolate 1050, while decreasing 1.5 logs based on mean data, displayed 1 log of variability (11). With the higher exposures used in the current study, both isolates exhibited maximal decreases in CFU, and variability decreased to less than 0.2 and 0.5 logs for isolates 944 and 1050, respectively. Moreover, in the previous study, four of the five isolates with MICs of 8 and 16 μg/ml resulted in growth at 24 h. By doubling or quadrupling the doripenem exposures, the CFU counts of these isolates decreased by at least 2 logs.
Contrary to what might be expected, various degrees of efficacy were noted for both 1- and 2-g doses against isolates when doripenem concentrations never exceeded the MIC of the infecting organism (i.e., 0% fT>MIC). A number of factors may help explain this observation. First, while every effort was made to correctly identify the true MIC of each P. aeruginosa isolate, determination via the broth microdilution method is limited by its doubling of dilutions. That is to say, an isolate with an MIC of 16 μg/ml merely identifies the isolate's ability to survive in a drug concentration of 8 μg/ml and its inability to live at concentrations of 16 μg/ml. Accordingly, the “true” MIC of the isolate could be anywhere between 8 and 16 μg/ml. This, taken in light of the pharmacokinetic data displayed in Fig. Fig.1,1, clearly shows that if an isolate with a reported MIC of 16 μg/ml had a “true” MIC of 9 μg/ml, the fT>MIC of a 1-g-dose doripenem simulation could very well exceed 40%. A similar example could be drawn for the 2-g dose when evaluating isolates with MICs of 32 μg/ml. Second, pharmacokinetic simulations are a very complex undertaking with many moving parts. During our confirmatory pharmacokinetic studies, groups of six mice were infected just as they were in the pharmacodynamic analyses and sampled over time. Groups of mice were used in hopes of accounting for interanimal variability. While this is an excellent approach, it is impossible to ensure that each study animal is achieving exactly the same exposure. Not unlike human pharmacokinetics analyses, if an animal exhibited, for example, infection-related renal impairment impeding clearance above that displayed during the pharmacokinetic analyses, the free doripenem concentrations could have certainly surpassed 16 or 32 μg/ml for 1- and 2-g doses, respectively.
When evaluating the efficacy of the simulated regimens used in our analysis against the higher MIC isolates, it is important to consider these results in the context of current MIC distributions. A recent U.S. surveillance study (16) of 875 clinical P. aeruginosa isolates revealed 4 μg/ml as the MIC90. Furthermore, 97.4 and 99.3% of the isolates had MICs of 8 and 16 μg/ml or lower, respectively. Similarly, during doripenem clinical trials, all pretreatment MICs of P. aeruginosa isolated from patients treated with doripenem for hospital- or ventilator-associated pneumonia (n = 48) were <4 μg/ml (3, 17). Given these distributions, coupled with our results, doripenem doses of 1 or 2 g should produce sufficient pharmacokinetic exposures against nearly all P. aeruginosa isolates.
The human pharmacokinetic data used in these analyses were derived from data obtained from healthy volunteers given 500 mg every 8 h as a 4-h infusion (Johnson & Johnson Pharmaceutical Research & Development, LLC, data on file). Extrapolation to 1- and 2-g doses based on the mean data obtained from these subjects was possible given the linearity of doripenem described in early pharmacokinetic studies (14, 19). Moreover, our chosen exposure targets for 1-g doses were validated by Monte Carlo simulation using a population pharmacokinetic model derived from different phase 1 data than those used in our analysis (2). In this study, Bhavnani and colleagues showed a 90% probability that a patient given 1 g of doripenem as a 4-h infusion would achieve an ≥45% fT>MIC at an MIC of 8 μg/ml. Similar results were noted in another Monte Carlo simulation based upon a population pharmacokinetic model derived from Japanese patients undergoing abdominal surgery (9). In this study, in which patients were given 1 g as a 4-h infusion, all of the simulated patient profiles achieved ≥40% fT>MIC at an MIC of 8 μg/ml and only 37% did at an MIC of 16 μg/ml. Compared with these data, our healthy-volunteer-derived target of 0% fT>MIC at an MIC of 16 μg/ml represents a more conservative estimate of pharmacokinetic exposures at this MIC. In order to validate our 2-g-dose simulations, we conducted a 5,000-patient Monte Carlo simulation (Crystal Ball 2000; Decisioneering Inc., Denver, CO), using the population pharmacokinetic model described by Bhavnani et al. (2); the results of this investigation further confirmed our exposure targets (data not shown). It should be noted that while dose-ranging studies have found no association between adverse events and dose, of the doses we examined, only the 1-g dose has been evaluated and proved to be generally safe, albeit in a relatively small sample of approximately 200 patients (7, 14; Johnson & Johnson Pharmaceutical Research & Development, LLC, data on file). Moreover, only doses of 500 mg are approved by regulatory agencies worldwide.
The ability to simulate human exposures within an animal model afforded us the opportunity to assess the efficacy of doripenem against a variety of P. aeruginosa isolates at doses above those tested in clinical trails to date. While MIC distributions at present suggest that these increased dosages would be needed for only the upper 5% of the population, clearly data herein suggesting activity against isolates with MICs of ≤16 μg/ml could prove invaluable when the need arises. With a lack of human data, advocacy for these increased doses is not entirely possible; however, based on the results we have presented, future clinical studies should consider the incorporation of these pharmacodynamically enhanced regimens.
We thank Henry Christensen, Lindsay Tuttle, Debora Santini, Jennifer Hull, and Christina Sutherland for their assistance with the conduct of the animal experimentation and the analytical determinations of doripenem.
This work was supported by a grant from Johnson & Johnson Pharmaceutical Research & Development, LLC., Raritan, NJ.
Published ahead of print on 21 September 2009.