Pharmacodynamics of Daptomycin against Enterococcus faecium and Enterococcus faecalis in the Murine Thigh Infection Model

The Clinical and Laboratory Standards Institute (CLSI) daptomycin MIC susceptibility breakpoint for the treatment of enterococcal infections is ≤4 μg/ml. However, patients receiving daptomycin for the treatment of infections caused by enterococci with MICs of ≤4 μg/ml may experience treatment failures.

ABSTRACT

The Clinical and Laboratory Standards Institute (CLSI) daptomycin MIC susceptibility breakpoint for the treatment of enterococcal infections is ≤4 μg/ml. However, patients receiving daptomycin for the treatment of infections caused by enterococci with MICs of ≤4 μg/ml may experience treatment failures. We assessed the pharmacodynamics of daptomycin against enterococci in a neutropenic murine thigh infection model and determined the exposures necessary for bacteriostasis and a 1-log₁₀-CFU reduction of Enterococcus faecalis and Enterococcus faecium. We further characterized daptomycin efficacy at clinically achievable exposures. Six E. faecium and 6 E. faecalis isolates (daptomycin MICs, 0.5 to 32 μg/ml) were studied. Daptomycin was administered at various doses over 24 h to achieve area under the free drug concentration-time curve-to-MIC ratios (fAUC_0–24/MIC) ranging from 1 to 148. Daptomycin regimens that simulate mean human exposures following doses of 6, 8, and 10 mg/kg of body weight/day were also studied. Efficacy was assessed by the differences in the number of log₁₀ CFU per thigh at 24 h. The Hill equation was used to estimate the fAUC_0–24/MIC required to achieve bacteriostasis and a 1-log₁₀-CFU reduction. For E. faecium, a 1-log₁₀-CFU reduction required an fAUC_0–24/MIC of 12.9 (R² = 0.71). For E. faecalis, a 1-log₁₀-CFU reduction was not achieved, while the fAUC_0–24/MIC required for stasis was 7.2 (R² = 0.8). With a human-simulated regimen of 6 mg/kg/day, a 1-log₁₀-CFU reduction was observed in 3/3 E. faecium isolates with MICs of <4 μg/ml and 0/3 E. faecium isolates with MICs of ≥4 μg/ml; however, a 1-log₁₀-CFU reduction was not achieved for any of the 6 E. faecalis isolates. These results, alongside clinical data, prompt a reevaluation of the current breakpoint.

INTRODUCTION

As the sole lipopeptide antibiotic available, daptomycin has a unique mechanism of action, causing calcium-dependent depolarization of the bacterial cell membrane, which conveys robust in vitro activity against enterococci, staphylococci, and streptococci (1, 2). As such, daptomycin is one of the few antibiotic options available to patients with infections caused by vancomycin-resistant enterococci (VRE). Although the only enterococcal infections for which daptomycin has received FDA approval are skin and soft tissue infections caused by vancomycin-susceptible Enterococcus faecalis, it is frequently used off-label for complicated bacteremia, infective endocarditis, and other recalcitrant infections caused by VRE (3). For these indications, higher doses up to 10 mg/kg of body weight/day are associated with decreased mortality (4–7). The Clinical and Laboratory Standards Institute (CLSI) has defined the daptomycin MIC breakpoint for Enterococcus spp. to be ≤4 μg/ml (8). However, increased treatment failures have been observed in patients with infections caused by enterococci with daptomycin MICs of 3 to 4 μg/ml, even in patients who have received doses of ≥8 mg/kg/day (9, 10). While treatment failures might not be attributed to suboptimal dosing alone, dosing remains a possible contributor; therefore, the daptomycin MIC breakpoint might need to be lower. The objectives of this study were to define the pharmacodynamic (PD) profile of daptomycin against E. faecalis and E. faecium and to determine the efficacy of daptomycin at clinically achievable exposures in the neutropenic murine thigh infection model.

RESULTS

Bacterial isolates.

A total of 6 E. faecalis and 6 E. faecium isolates were utilized in the bacterial density studies. The daptomycin modal MICs for these isolates are reported in Table 1.

TABLE 1.

Isolates used for efficacy studies

Isolate (CAIRDa no.)	Daptomycin MIC (μg/ml)	Resistance mutation(s)
E. faecalis 4	1	NDb
E. faecalis 18	2	ND
E. faecalis 63	1	ND
E. faecalis 65	1	ND
E. faecalis 70	1	ND
E. faecalis 71	0.5	ND
E. faecium 64	2	ND
E. faecium 91	1.5	ND
E. faecium 100	32	ClsB
E. faecium 101	3	None
E. faecium 103	4	LiaF, LiaS, LiaR
E. faecium 104	8	LiaS

^aCAIRD, Center for Anti-Infective Research and Development.

^bND, not determined.

Pharmacokinetics of daptomycin.

The average murine pharmacokinetic parameters of daptomycin were as follows: a volume of distribution of 0.15 liters · kg⁻¹, an absorption rate constant of 6.8 h⁻¹, an elimination rate constant of 0.18 h⁻¹, a half-life of 3.85 h, and a total clearance of 0.027 liter · h⁻¹ · kg⁻¹. The area under the free drug concentration-time curve (AUC) over 24 h (fAUC_0–24) increased proportionately with the dose across the range of doses utilized in the dose-ranging studies (Fig. 1). Human pharmacokinetic data from healthy volunteers were obtained from the daptomycin package insert (11) and used to simulate daptomycin concentration-versus-time profiles at steady state for doses of 6, 8, and 10 mg/kg/day. Based on these simulations, it was determined that a four-dose murine regimen of 9.5 mg/kg, 5 mg/kg, 3 mg/kg, and 1.5 mg/kg at 0 h, 4 h, 10 h, and 17 h, respectively, would provide an exposure similar to that achieved in humans after a dose of 6 mg/kg/day at steady state. Each of the murine doses was increased linearly to simulate human daptomycin doses of 8 mg/kg and 10 mg/kg (Table 2). The predicted fAUC_0–24 of each human-simulated regimen was comparable to the corresponding fAUC_0–24 observed in healthy volunteers (Table 3). Confirmatory pharmacokinetic studies validated the 6-mg/kg and 10-mg/kg human-simulated regimens against the predicted human and murine concentrations (Fig. 2 and 3).

FIG 1.

Dose versus fAUC_0–24 in the murine thigh infection model after single doses of daptomycin.

TABLE 2.

Doses and times of administration for each human-simulated regimen

Human-simulated dose (mg/kg/day)	Murine dose (mg/kg)
	0 h	4 h	10 h	17 h
6	9.5	5	3	1.5
8	12.7	6.7	4	2
10	15.8	8.3	5	2.5

TABLE 3.

Comparison of observed human exposures and predicted murine exposures of human-simulated regimens^a

Human dose (mg/kg q24h)	fAUC0–24 observed (mg · h/liter)	Murine regimen (mg/kg HSR)	fAUC0–24 predicted (mg · h/liter)
6	53.7 ± 6.6	6	60.8
8	72.9 ± 18.1	8	75.7
10	88.3 ± 15.1	10	94.6

^aq24h, every 24 h; HSR, human-simulated regimen.

FIG 2.

Daptomycin human-simulated-regimen (HSR) free plasma concentration-time profile in the murine thigh infection model compared with a simulated human profile of 6 mg/kg/day at steady state. Data are means ± standard deviations.

FIG 3.

Daptomycin human-simulated-regimen (HSR) free plasma concentration-time profile in the murine thigh infection model compared with a simulated human profile of 10 mg/kg/day at steady state. Data are means ± standard deviations.

The high-performance liquid chromatography (HPLC) assay of daptomycin in murine plasma was linear over a range of 1.0 to 100 μg/ml (R² = 0.997). Intraday (n = 10) coefficients of variation for the low (2 μg/ml) and high (75 μg/ml) quality control samples were 4.8% and 3.6%, respectively. Interday (n = 14) coefficients of variation for the low (2 μg/ml) and high (75 μg/ml) quality control samples were 5.1% and 4.9%, respectively.

Dose-ranging studies.

For mice inoculated with E. faecalis, the average bacterial burden across all isolates at 0 h was 6.52 ± 0.16 log₁₀ CFU/thigh. Over 24 h, the bacterial burden increased by 1.14 ± 0.39 log₁₀ CFU/thigh in untreated control mice. The relationship between fAUC_0–24/MIC and the change in the bacterial density at 24 h is presented in Fig. 4, and the inhibitory sigmoid maximum-effect (E_max) model fit of these data is presented in Fig. 5. For the composite of tested isolates, the relationship between exposure and response was relatively strong, with R² being equal to 0.80. An fAUC_0–24/MIC of 7.23 was required to achieve bacterial stasis, while a 1-log₁₀-CFU reduction was not achieved.

FIG 4.

Response of E. faecalis (EFC) to total daily doses of daptomycin (DAP) from 1 to 25 mg/kg at 24 h compared to the 0-h controls.

FIG 5.

Curve of best fit to the fAUC_0–24/MIC ratios and the change in the number of log₁₀ CFU at 24 h for the composite of E. faecalis (EFC) isolates in neutropenic mice.

Human-simulated regimen studies.

The results of the human-simulated-regimen studies are reported in Fig. 6 and Fig. 7. The average reductions in bacterial density at 24 h achieved with human-simulated regimens of 6, 8, and 10 mg/kg against E. faecalis were 0.72 ± 0.36, 0.82 ± 0.44, and 0.84 ± 0.33 log₁₀ CFU/thigh, respectively. In the untreated control groups, the average change in bacterial density at 24 h for all E. faecalis isolates was 0.83 ± 0.28 log₁₀ CFU/thigh. Despite the difference in the starting inoculum between the two sets of studies, the efficacy achieved with the 8- or 10-mg/kg/day human-simulated regimens was similar to the maximum predicted efficacy from the sigmoid E_max model constructed from the dose-ranging data.

FIG 6.

Response of E. faecalis (EFC) to three daptomycin human-simulated regimens at 24 h compared to the 0-h controls.

FIG 7.

Response of E. faecium (EFM) to three daptomycin human-simulated regimens at 24 h compared to the 0-h controls.

Against E. faecium isolates with an MIC ≤4 μg/ml, the average reductions in bacterial density at 24 h achieved with human-simulated regimens of 6, 8, and 10 mg/kg/day were 1.23 ± 0.45, 1.25 ± 0.40, and 1.24 ± 0.47 log₁₀ CFU/thigh, respectively. Against E. faecium isolates with an MIC of >4 μg/ml, the average reductions in bacterial density at 24 h achieved with human-simulated regimens of 6, 8, and 10 mg/kg/day were 0.23 ± 0.29, 0.67 ± 0.71, and 0.72 ± 0.70 log₁₀ CFU/thigh, respectively. In the untreated control groups, the average change in bacterial density at 24 h for all six E. faecium isolates was 0.02 ± 0.24 log₁₀ CFU/thigh. For these six isolates, net stasis or growth was achieved in the control groups. Six additional isolates tested in this model demonstrated a statistically significant decline in bacterial density in control mice at 24 h and were not included in the final analysis (data not shown). The inhibitory sigmoid E_max model fit of fAUC_0–24/MIC versus the change in bacterial density at 24 h for E. faecium is presented in Fig. 8. A moderate relationship between the exposure and the response was seen (R² = 0.71). Bacterial stasis and a 1-log₁₀-CFU reduction were achieved with fAUC_0–24/MIC ratios of 0.85 and 12.9, respectively. For isolates with an MIC of 4 μg/ml, mean daptomycin exposure from a dose of 6 mg/kg/day yielded an fAUC_0–24/MIC of 13.4, which is sufficient to meet PD targets for both E. faecalis and E. faecium.

FIG 8.

Curve of best fit to the fAUC_0–24/MIC ratios and change in the number of log₁₀ CFU at 24 h for the composite of all E. faecium (EFM) isolates in neutropenic mice.

DISCUSSION

While the evidence remains inconclusive, data have been published suggesting that patients who have infections with enterococci with MICs of 3 to 4 μg/ml have a higher rate of clinical failure regardless of the daptomycin dose (9, 10). Although treatment failures are multifactorial, other data suggest that patients who receive <8 mg/kg/day of daptomycin for enterococcal bacteremia have a higher likelihood of mortality (4–7). Together, these observations have prompted a reevaluation of the current CLSI daptomycin MIC breakpoints for Enterococcus spp.; at the time of writing, the present susceptible category is defined by an MIC of ≤4 μg/ml, with no intermediate or resistant categories being defined to date (8).

This breakpoint was set based on limited clinical and preclinical data for E. faecalis only. In daptomycin phase 2 and 3 clinical trials conducted prior to the establishment of the breakpoints, a very limited number of enterococci with daptomycin MICs of ≥4 μg/ml were isolated from the patients (12). Three nonclinical in vivo studies had been conducted: two in a murine thigh infection model and one in a murine renal infection model. Dandekar et al. studied two isolates, one E. faecalis isolate and one E. faecium isolate, in the thigh infection model and predicted a maximum effect of 2.09-log₁₀-CFU killing and 1.53-log₁₀-CFU killing for each organism, respectively, with a 99% effective dose (ED₉₉) at an fAUC_0–24/MIC of 157 and 38, respectively (13). Safdar et al. studied two E. faecium isolates in a thigh infection model, showing a maximum effect approaching, but not achieving, 2-log₁₀-CFU killing, yet 1-log₁₀-CFU killing was achieved against the isolates at fAUC_0–24/MIC values of approximately 0.4 and 2.9 (14). A bactericidal effect higher than that observed in the thigh was achieved by Alder et al., who studied 10 isolates in a renal infection model, achieving up to 4-log₁₀-CFU killing for some isolates, with the ED₉₉ reached at an fAUC_0–24/MIC of 4 (15). However, because of the differences between the two models, direct comparisons are difficult. The renal infection model used 3 days of treatment, as opposed to the 24 h of treatment given in the thigh infection models. Additionally, as the authors note (15), the renal infection model is limited by the fact that daptomycin is renally eliminated, and daptomycin concentrations were measured only in the plasma, potentially underestimating the actual exposures achieved in renal tissue.

From these three nonclinical in vivo studies, a pharmacodynamic target AUC/MIC of 48 (equivalent to an fAUC_0–24/MIC of 4) was chosen to be used in a Monte Carlo simulation, which calculated the probability of target attainment (PTA) to be 96.2% against enterococci with daptomycin MICs of ≤8 μg/ml at a daptomycin dose of 4 mg/kg/day (12). However, the association between doses of <8 mg/kg/day and increased mortality in enterococcal bacteremia casts doubt on the current susceptibility breakpoint and suggests that the pharmacodynamic target of an fAUC_0–24/MIC of 4 should be reevaluated. In this study, we determined the pharmacodynamic profile of daptomycin against E. faecium and E. faecalis in the murine thigh infection model and estimated the magnitude of the pharmacodynamic target necessary to achieve various efficacy targets. We also characterized the bactericidal activity achievable with clinically relevant daptomycin exposures in this model.

The PD profile of E. faecalis isolates built from the dose-ranging studies demonstrates that a 1-log₁₀-CFU reduction was not routinely achievable against this species in this model, while bacteriostasis was achieved at an fAUC_0–24/MIC of 7.23. Thus, the target for stasis for E. faecalis is expected to be attainable with the mean daptomycin exposure achieved with a dose of 6 mg/kg/day (fAUC_0–24 = 53.7 mg · h/liter) when the MIC is ≤4 μg/ml. This is corroborated by results from human-simulated exposures: at a dose of 6 mg/kg/day, bacterial stasis was achieved against E. faecalis isolates, yet none of the isolates achieved a 1-log₁₀-CFU reduction. Only one isolate (E. faecalis 4) was able to achieve 1-log₁₀-CFU killing with a higher human-simulated regimen of 8 mg/kg/day. These results are consistent with those observed by Dandekar et al. in a similar thigh infection model (13). Notably, we studied the same isolate (E. faecalis 63) used by these authors and observed a similar PD profile. We demonstrated a magnitude of killing lower than that seen in the 72-hour renal infection model (>2 log₁₀ CFU), which can be explained by the shorter duration of therapy that we used and the potentially higher daptomycin exposures in renal tissue.

In contrast, E. faecium displayed a distinct PD profile. Compared with E. faecalis, greater killing of E. faecium was achieved at 24 h. In human-simulated-regimen studies, 6 mg/kg/day was successful in achieving 1-log₁₀-CFU bacterial killing of all E. faecium isolates with daptomycin MICs of <4 μg/ml, and higher exposures were not associated with considerably greater bacterial killing. For isolates with MICs ranging from 4 to 8 μg/ml, an exposure higher than 6 mg/kg/day was needed to achieve comparable reductions in bacterial densities.

Our PD model constructed from the human-simulated-regimen studies predicted that a 1-log₁₀-CFU reduction of E. faecium was attainable at an fAUC_0–24/MIC of 12.9. This species responded differently in the murine thigh infection model than E. faecalis, with E. faecium-infected control groups achieving stasis, while the organisms in control groups infected with E. faecalis grew approximately 1 log₁₀ CFU/thigh. Establishing infection with E. faecium in this model was a challenge; 6 out of 12 E. faecium isolates studied failed to achieve growth or stasis in 24-h control groups. These isolates were excluded from the final analysis because a decline in bacterial density in the control groups may bias the results in favor of the drug: the killing could be due to the drug plus host and not simply the drug, as intended.

Although it would be ideal to consider the same efficacy threshold (i.e., bacteriostasis or 1-log₁₀-CFU killing) for both species, the response of E. faecalis relative to that of E. faecium was subdued. Bacteriostasis for E. faecalis and 1-log₁₀-CFU killing for E. faecium represent the maximum achievable thresholds for each organism in the present model. Both stasis and 1-log₁₀-CFU killing targets have been used as surrogates for clinical efficacy in neutropenic murine thigh infection models (16, 17). In the case of E. faecalis, it is possible that greater killing could be achieved with daptomycin exposures greater than those associated with a 10-mg/kg/day dose in humans. However, if 1-log₁₀-CFU killing were possible, it would require an fAUC_0–24/MIC higher than the current estimate of 7.23 for bacteriostasis. While the clinical usage of daptomycin against E. faecalis is limited since β-lactams are effective against this species and the majority of VRE isolates are E. faecium, the MIC breakpoint has traditionally been defined to be the same for both organisms. Additional data will need to be considered to determine whether a distinct breakpoint for each species is warranted.

The target exposures for 1-log₁₀-CFU bacterial killing of E. faecium (12.9) and bacteriostasis in E. faecalis (7.23) estimated from the current study are greater than the previously reported fAUC_0–24/MIC of 3.6 for a 2-log₁₀-CFU reduction achieved in the renal infection model (15) as well as exposures predicted for 1-log₁₀-CFU killing in two isolates in the murine thigh infection model (14). Based on these PD indices, a dose of 6 mg/kg/day should theoretically provide enough exposure to derive efficacy for E. faecalis and E. faecium infections when the MIC is ≤4 μg/ml. However, it is critical to recognize that the systemic exposure reported in the package insert with the 6-mg/kg/day dose is an average value. Due to interpatient variability, a significant proportion of patients might fail to meet the target with this dose. Furthermore, our PD target of 12.9 for 1-log₁₀-CFU killing in E. faecium is more than three times greater than the target fAUC_0–24/MIC of 4 used in the Monte Carlo analysis which supported the current MIC breakpoint. Because the AUC of daptomycin increases linearly with dose in humans (10, 18), a tripling of the fAUC_0–24 will require a tripling of the dose. This implies that daptomycin doses of 12 mg/kg/day would be required to achieve a PTA similar to that seen in the Monte Carlo analysis if our PD target had been utilized. Although doses this high are currently used to treat enterococcal endocarditis (19) and bacteremia (7), the practice is not universal, and our work demonstrates that organisms which may currently be deemed susceptible may not be maximally treated except with doses of daptomycin greater than 6 mg/kg/day. Since the current maximum approved dose of daptomycin is 6 mg/kg/day, our data suggest that the current breakpoint of ≤4 μg/ml should be lowered.

This study of 12 enterococcal isolates in the neutropenic murine thigh infection model is the largest such study to date. The PD targets for efficacy that we identified for both organisms were found to be higher than the PD target previously utilized for breakpoint determination. Our data suggest that current breakpoints may overestimate the susceptibility of some isolates to daptomycin exposures achieved with doses of 6 mg/kg/day. Finally, although we demonstrated that the maximum bactericidal activity of daptomycin was achieved with human-simulated dosing regimens, these data represent mean responses and should not be construed to imply that a dose of 6 mg/kg/day will achieve the maximum bactericidal effect in all patients. Translational application of these data will require further study of the entire range of achievable exposures in various patient populations and the ability to link these exposures to meaningful clinical outcomes.

MATERIALS AND METHODS

Antimicrobial test agents.

Daptomycin vials (lot 928333; Teva Pharmaceuticals USA, North Wales, PA) were purchased from Cardinal Health Inc., Dublin, OH. The vials were stored under refrigeration at 2 to 8°C. Daptomycin vials were reconstituted with 0.9% sodium chloride (Hospira, Lake Forest, IL) according to manufacturer recommendations and were further diluted with 0.9% sodium chloride to obtain final dosing concentrations based on the mean weight of the study mouse population. All doses were administered as subcutaneous injections of 0.2 ml.

Bacterial isolates.

Daptomycin MICs for isolates were determined by broth microdilution in triplicate using broth supplemented with CaCl₂ (20). Selection of isolates was based on the susceptibility to daptomycin to include isolates with a broad range of MICs. Isolates were screened for growth in the murine model without antimicrobial treatment, and isolates with the best growth for each MIC were selected by examination of the mean ± standard deviation change in bacterial density at 24 h.

Animal infection model. (i) Laboratory animals.

Specific-pathogen-free ICR mice weighing 20 to 22 g were obtained from Envigo RMS, Inc. (Indianapolis, IN). The animals were allowed to acclimate for a minimum of 48 h before the beginning of any experiments and were provided food and water ad libitum. The protocol was reviewed and approved by the Institutional Animal Care and Use Committee at Hartford Hospital. The mice were rendered transiently neutropenic via intraperitoneal (i.p.) injections of cyclophosphamide at a dose of 150 mg/kg of body weight 4 days before inoculation and 100 mg/kg 1 day before inoculation. To induce a controlled degree of renal impairment, 5 mg/kg uranyl nitrate was administered via i.p. injections 3 days before inoculation.

(ii) Neutropenic thigh infection model.

All isolates were previously frozen at −80°C in skim milk (BD Biosciences, Sparks, MD). Each isolate was transferred twice sequentially onto Trypticase soy agar plates with 5% sheep blood (TSA II; Becton, Dickinson, & Co., Sparks, MD). Each transfer was incubated at 37°C for approximately 24 h. After an 18- to 24-h incubation period, the second transfer was used to make a bacterial suspension of approximately 10⁷ CFU/ml (pharmacokinetic and dose-ranging studies) for inoculation. It was determined after the completion of the dose-ranging studies that a higher inoculum of 10⁸ CFU/ml was needed to induce infection with E. faecium; thus, for consistency, this inoculum (10⁸ CFU/ml) was used for both E. faecalis and E. faecium during the assessment of the in vivo efficacy of human-simulated-regimen studies. For all studies, the final inoculum concentrations were determined by serial dilution and plating techniques. The thigh infection was produced by intramuscular injection of 0.1 ml of the inoculum into each thigh of the mouse 2 h before the initiation of antimicrobial therapy.

(iii) Pharmacokinetic studies.

Pharmacokinetic parameters of daptomycin in the murine model were determined by administering single doses of daptomycin of 1 mg/kg, 5 mg/kg, 20 mg/kg, 50 mg/kg, and 100 mg/kg to mice prepared as described above. Using these parameters, pharmacokinetic simulations (Phoenix, version 6.3; Pharsight Corp., Mountain View, CA) were performed to generate dosing regimens in mice targeting fAUC_0–24 values similar to those achieved with doses of 6 mg/kg, 8 mg/kg, and 10 mg/kg per day at steady state in healthy volunteers (10). The pharmacokinetic parameters of daptomycin in healthy volunteers (10) were used to simulate the concentration-versus-time profile at steady state. Confirmatory pharmacokinetic studies were conducted with these human-simulated regimens to ensure that the observed murine exposures were similar to the predicted murine and observed healthy volunteer exposures.

Infected animals received 0.2-ml subcutaneous injections as required based on the dosing regimens. Groups of 6 mice were euthanized at 8 predefined time points for all studies except the 10-mg/kg human-simulated-regimen confirmatory pharmacokinetic study, which included 4 predefined time points. Terminal blood samples from CO₂-asphyxiated mice were collected via cardiac puncture and placed in K₂EDTA Vacutainer tubes (BD, Franklin Lakes, NJ). Plasma was separated by centrifugation for 10 min at 4°C at 10,000 relative centrifugal force before transfer into polypropylene tubes. The tubes were stored at −80°C until they were analyzed. Daptomycin concentrations were determined using a validated high-performance liquid chromatography (HPLC) assay (11). Free (unbound) daptomycin plasma concentrations were calculated by multiplying total daptomycin plasma concentrations by 8.5%, the mean value of the range reported in the daptomycin package insert (10). The same value was also used to calculate free daptomycin concentrations in murine plasma, which range from 7.5% to 10% (21).

(iv) Dose-ranging studies.

The purpose of the dose-ranging studies was to describe the full range of daptomycin activity achievable against E. faecalis isolates having a range of susceptibilities to daptomycin. Mice were prepared and inoculated as described above using a starting inoculum of approximately 10⁷ CFU/ml, and treatment was initiated 2 h later. All doses were studied in groups of 3 mice over a 24-h treatment period. Control animals were dosed with the diluent vehicle (normal saline) at the same volume, by the same route, and on the same schedule as the most frequently dosed drug regimen. For each isolate tested, 3 untreated mice were used as the 0-h controls and 3 additional mice (receiving normal saline) were used as the 24-h controls. Five escalating doses were utilized in each study; therefore, 21 mice were needed for each isolate studied. In total, 6 E. faecalis isolates were utilized in these studies. Mice received doses ranging from 1 to 25 mg/kg/day as single doses. After the 24-h treatment period, all animals were euthanized by CO₂ asphyxiation followed by cervical dislocation. After sacrifice, the thighs were removed and individually homogenized in normal saline. Serial dilutions were plated on appropriate agar medium for determination of the number of CFU. Reductions in CFU numbers at 24 h relative to the 0-h starting inoculum were evaluated for each isolate to define the total reduction in the number of log₁₀ CFU per thigh. Antibacterial efficacy was defined as bacterial stasis or a ≥1-log₁₀-CFU reduction at 24 h compared with the count in the 0-h control.

Using a similar methodology, daptomycin dose-ranging studies were also conducted against 12 E. faecium isolates using a starting inoculum of approximately 10⁷ CFU/ml. However, adequate infection was not achieved in the murine model at this starting inoculum; a decline in the bacterial burden was observed at 24 h in the absence of antimicrobial treatment (24-h control groups) (data not shown). Other authors have also encountered this difficulty (11, 15). Given that the bacterial killing by daptomycin was confounded by the intrinsic eradication of the infection by the host, the data from the dose-ranging studies against E. faecium were not utilized in subsequent data analyses.

(v) Human-simulated exposure studies.

The purpose of the human-simulated exposure studies was to determine the range of daptomycin activity against enterococci achievable at exposures comparable to those achieved in humans receiving clinically relevant doses of daptomycin: 6, 8, and 10 mg/kg/day. Mice were prepared and inoculated as described above using a starting inoculum of approximately 10⁸ CFU/ml, and treatment was initiated 2 h later. The higher starting inoculum was necessary to establish bacterial infection with E. faecium isolates. All dosing regimens were studied in groups of 3 mice over a 24-h treatment period. Control animals were dosed with the diluent vehicle (normal saline) at the same volume, by the same route, and on the same schedule as the most frequently dosed drug regimen. For each isolate tested, 3 untreated mice were used as the 0-h controls and 3 additional mice (receiving normal saline) were used as 24-h controls. Three human-simulated dosing regimens were utilized in each study; therefore, 15 mice were needed for each isolate studied. In total, 12 isolates (6 E. faecalis and 6 E. faecium) were utilized in these studies. For each isolate, mice were treated with all three human-simulated regimens described above (Table 3). After the 24-h treatment period, all animals were euthanized by CO₂ asphyxiation followed by cervical dislocation. After sacrifice, the thighs were removed and individually homogenized in normal saline. Serial dilutions were plated on appropriate agar medium for determination of the number of CFU. Reductions in CFU numbers at 24 h relative to the 0-h starting inoculum were evaluated for each isolate to define the total reduction in the number of log₁₀ CFU per thigh. Antibacterial efficacy was defined as bacterial stasis or a ≥1-log₁₀-CFU reduction at 24 h compared with the count in the 0-h control. Stasis, growth, and decline were defined in terms of the 95% confidence interval around the mean change in the number of log₁₀ CFU per thigh at 24 h compared with the count in the 0-h controls. If this interval included 0, the isolate was deemed static, while growth and decline were defined when the entire interval was greater than or less than 0, respectively. Data for isolates which failed to achieve stasis or growth in the 24-h controls were excluded from the final data analysis.

Pharmacodynamic modeling.

From the bacterial density studies, two composite data sets of changes in the number of log₁₀ CFU at 24 h were compiled. The first data set included all 6 E. faecalis isolates from the dose-ranging studies, and the second set included all 6 E. faecium isolates from the human-simulated-regimen studies.

The free daptomycin exposure-to-MIC ratio (fAUC_0–24/MIC) was calculated for each dosing regimen and organism used; these values were plotted against the corresponding changes in the number of log₁₀ CFU for each thigh. An inhibitory sigmoid E_max model using the Hill equation was fit to each data set and used to determine the fAUC_0–24/MIC necessary for bacterial stasis and a 1-log₁₀-CFU reduction (Phoenix, version 6.3; Pharsight Corp., Mountain View, CA).