Pharmacodynamics of Tigecycline against Phenotypically Diverse Staphylococcus aureus Isolates in a Murine Thigh Model

Abstract

Tigecycline is a currently marketed antimicrobial agent with activity against resistant gram-positive cocci, including methicillin-resistant Staphylococcus aureus (MRSA). Despite the proven efficacy of tigecycline in the treatment of infections caused by these pathogens, questions remain as to the exposure-response relationship best associated with its efficacy. The purpose of this study was to define this relationship against seven distinct S. aureus isolates by using a neutropenic murine thigh model. Single-dose pharmacokinetics were evaluated, and free drug exposures were calculated after determination of protein binding. Doses of 1.56 to 400 mg/kg of body weight divided 1 to 8 times daily were administered against two methicillin-susceptible S. aureus isolates, two hospital-associated MRSA (HA-MRSA) isolates, and three community-associated (CA-MRSA) isolates. Tigecycline pharmacokinetics were best described by a two-compartment model, with a mean half-life of 9.9 h. Protein binding was dose dependent (range, 92.9 to 81.2%). MICs were 0.25 μg/ml for all isolates, except for HA-MRSA 56 (MIC, 0.5 μg/ml) and CA-MRSA 156 (MIC, 0.125 μg/ml). Tigecycline displayed efficacy against all isolates, producing maximum decreases in log₁₀ numbers of CFU/ml of 1.8 to 2.3 from 0-h controls. Mean correlation coefficients for free-drug (f) concentration exposures derived from the parameters fT>MIC (the percentage of time during which the concentration of f remains above the MIC), fC_max/MIC (the ratio of the maximum concentration of f to the MIC), and fAUC/MIC (the ratio of the area under the concentration-time curve of f to the MIC) were 0.622, 0.812, and 0.958, respectively. Values for the mean effective exposure index at 80% (EI₈₀) and 50% (EI₅₀) for fAUC/MIC were 5.4 μg/ml (range, 2.8 to 13 μg/ml) and 2.6 μg/ml (range, 0.6 to 5.1 μg/ml), respectively. Experiments with nonneutropenic mice infected with CA-MRSA 156 resulted in maximum kill at all fAUC/MIC exposures tested (1.8 to 8.8 μg/ml). The fAUC/MIC ratio is the pharmacodynamic parameter most predictive of tigecycline efficacy. Furthermore, the presence of a functioning immune system markedly reduces the required exposure.

Tigecycline is a first-in-class glycylcycline antimicrobial agent with activity against many multidrug-resistant gram-positive bacteria, including community-associated methicillin-resistant Staphylococcus aureus (CA-MRSA) (13) and hospital-associated MRSA (HA-MRSA) (22). Currently, tigecycline is FDA approved for the treatment of complicated skin and skin structure infections (cSSSI), in addition to complicated intra-abdominal infections (cIAI). Despite these approvals and the drug's proven efficacy, many questions remain as to the innate pharmacodynamics of the drug.

Tigecycline is structurally related to the tetracycline class of antimicrobials (20). Historically, the exposure-response relationship of the tetracyclines is vastly understudied relative to that of other classes of antimicrobials, likely because the introduction of the most recent agents predates the conception of current pharmacodynamic theory (1). Given its relatively recent introduction to the antimicrobial armamentarium, the pharmacodynamics of tigecycline have been investigated in a few studies both in mice and in humans (12, 16, 21). A neutropenic murine thigh model study concluded that the percentage of time (T) at which the concentration of free drug (f) remained above the MIC (fT>MIC) was the exposure-response relationship that was most predictive of efficacy (21). Separate studies of patients treated for cSSSI and cIAI showed an association between the ratio of the area under the concentration-time curve to the MIC (AUC/MIC) and positive outcomes (12, 16).

Given the high propensity of S. aureus to cause disease and the organism-dependent results seen in previous evaluations (12, 16, 21), this study was designed to fully examine the exposure-response relationship of tigecycline against S. aureus through the use of the neutropenic murine thigh model.

MATERIALS AND METHODS

Antimicrobial test agents.

Standard analytical-grade tigecycline (lot RB5603; expiration date, October 2008; Wyeth, Madison, NJ) was used for all in vitro experiments and a greater proportion of the in vivo experiments. The remaining in vivo analysis was conducted using commercially available tigecycline (Tygacil) obtainable from Wyeth Pharmaceuticals. Comparison studies of pharmacokinetics and pharmacodynamics were conducted between standard and commercial tigecycline to ensure congruency. For all studies, the tigecycline powder was weighed at a quantity sufficient to produce the required dose in mg/kg of body weight and was reconstituted with normal saline (NS) immediately prior to its utilization. The resulting solution was used within 30 min of reconstitution.

Bacterial isolates.

In total, seven isolates of S. aureus were examined in these studies, consisting of two methicillin-susceptible S. aureus isolates (ATCC 29213 and ATCC 25923), two HA-MRSA isolates (56 and 152), and three CA-MRSA isolates (CA-MRSA 144, 147, and 156). Each of the MRSA isolates was collected clinically, and all have been genotypically and phenotypically described previously (4, 6-8, 11). 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) prior to use in the experiments. Tigecycline MICs were determined in triplicate by broth microdilution by following Clinical and Laboratory Standards Institute (CLSI) guidelines (2), and the modal MIC was reported.

Animal infection model.

Pathogen-free, female ICR mice weighing approximately 25 g were acquired from Harlan Sprague Dawley, Inc. (Indianapolis, IN), and utilized throughout these experiments. The animals were maintained and used in accordance with National Research Council recommendations and provided food and water ad libitum. Mice were rendered neutropenic with 150 and 100 mg/kg of cyclophosphamide (Cytoxan; Bristol-Myers Squibb, Princeton, NJ) injected intraperitoneally 4 days and 1 day prior to inoculation, respectively. Two hours prior to the initiation of antimicrobial therapy, each thigh of the neutropenic mice was inoculated intramuscularly with a 0.1-ml solution containing approximately 10⁷ CFU/ml of the test isolate prepared from a fresh subculture that had been incubated overnight.

Pharmacokinetic studies.

Mice were prepared as described in “Animal infection model” above, with S. aureus ATCC 29213 as the infecting isolate. Single doses of tigecycline were administered as 0.2-ml subcutaneous injections at concentrations of 50, 25, 12.5, and 6.25 mg/kg. Mice were euthanized by exposure to CO₂, and blood samples were extracted via cardiac puncture from groups of six mice each at 8 to 12 time points ranging from 0.5 to 24 h after tigecycline administration. Blood samples were separated following centrifugation, and sera were stored at −80°C and analyzed within 1 month of collection.

Tigecycline concentrations in murine sera were determined by high-performance liquid chromatography (9). The assay was linear over a range of 0.05 to 5 μg/ml (R² = 0.998). Intraday coefficients of variation for the low (0.2 μg/ml) and high (4 μg/ml) quality control samples were 3.7% and 1.9%, respectively. Interday coefficients of variation were 2.9% and 2.6%, respectively.

Pharmacokinetic parameters for single doses of tigecycline were calculated using first-order input and elimination, by using a nonlinear least-squares technique (WinNonlin version 5.0.1 software; Pharsight, Mountain View, CA). Compartment model selection and weighting schemes were based on visual inspection of the fit and use of the correlation between the observed and calculated concentrations. The pharmacokinetic parameters derived from the 6.25-mg/kg studies were used to simulate the concentration-time profiles for 3.13- and 1.56-mg/kg doses of tigecycline. The AUC for all dosing regimens was calculated by using the trapezoidal rule.

Ex vivo protein binding.

Protein binding studies were conducted in triplicate using Amicon Centrifree micropartition devices (Millipore, Bedford, MA) with 30,000 molecular weight cutoff filters according to the manufacturer's package insert. Briefly, a master stock solution containing 1 mg/ml of tigecycline was prepared in NS. From this solution, secondary solutions were produced at 100 times the final test concentrations, also in NS. Final solutions were prepared using a 1:100 dilution in mouse serum, resulting in test concentrations of 0.75, 1.5, 6, 12, and 25 μg/ml. Serum collection was performed on the day of the protein binding experiment from ICR mice euthanized by CO₂ exposure; blood was extracted via cardiac puncture and centrifuged prior to serum separation. Each serum-drug test solution was heated at 37^οC in a shaking water bath for 10 min. Next, 0.9 ml of each solution was transferred into three ultrafiltration devices and centrifuged for 45 min at 10^οC at 1,000 × g to generate a serum ultrafiltrate. Nonspecific binding of the drug to the filter device was determined by following the same procedure, with NS at a concentration of 10 μg/ml. The initial solutions and ultrafiltrates were stored at −80^οC prior to analysis.

Bacterial-density studies.

At 2 h postinfection, groups of three mice were given a single tigecycline regimen of 1.56 to 400 mg/kg/day, administered as single or multiple doses of 1.56, 3.13, 6.25, 12.5, 25, or 50 mg/kg. All tigecycline doses were given as 0.2-ml subcutaneous injections and were calculated from the mean weight determined 1 day prior to inoculation. At the initiation of dosing, one group of three mice was sacrificed and harvested to serve as the 0-h control group. All other mice were harvested at 24 h after the start of dosing. In addition to the mice that received the study medication, one group of mice received sterile NS injections at the same frequency, route, and volume as the most frequent treatment regimen. The harvesting procedure for all study mice commenced with euthanasia by CO₂ exposure, followed by cervical dislocation. After sacrifice, thighs were removed and individually homogenized in normal saline. Serial dilutions of the thigh homogenate were plated on Trypticase soy agar with 5% sheep blood for CFU determination. Efficacy, defined as the change in bacterial density, was calculated as the log₁₀ change in numbers of bacterial CFU/ml obtained for tigecycline-treated mice after 24 h from the preantibiotic counts measured from the 0-h control animals.

Pharmacodynamic index determination.

Using the concentration-time profiles derived from the pharmacokinetic studies in conjunction with protein binding data, the MICs of the test isolates, and the bacterial density data, graphs of the change in log₁₀ numbers of CFU at 24 h versus the pharmacodynamic parameters (fT>MIC, fC_max/MIC [the ratio of the maximum concentration of f to the MIC], and AUC/MIC) were constructed using f (which was calculated using two different methodologies [see “Protein binding” below]) and were plotted using the sigmoidal maximum effect (E_max) model. The values for the effective exposure index at 80% (EI₈₀) and 50% (EI₅₀) of E_max, in addition to stasis values, were calculated from the individual curves for each S. aureus isolate and from a composite curve for all seven isolates.

Immunocompetent-mouse infection model.

For these studies, groups of ICR mice were infected, dosed, processed, and evaluated as described for the neutropenic thigh infection studies, except that mice were not given the cyclophosphamide injections prior to being infected with an increased inoculum of approximately 10⁸ CFU/ml. CA-MRSA 156 was used for these studies, and tigecycline was administered once daily at doses of 1.56 to 12.5 mg/kg.

RESULTS

In vitro susceptibility.

The MIC of tigecycline against all S. aureus isolates tested was 0.25 μg/ml, except for HA-MRSA 56 and CA-MRSA 156, which had MICs of 0.5 and 0.125 μg/ml, respectively.

Pharmacokinetic determination.

The pharmacokinetics of tigecycline displayed linearity across the dosing range and was best described by a two-compartment model. The pharmacokinetic properties associated with each dose are listed in Table 1. Resulting AUC values for the doses used in the pharmacodynamic analysis ranged from 1.2 to 325 μg·h/ml.

TABLE 1.

Pharmacokinetic properties of single-dose tigecycline in neutropenic ICR mice infected in the thigh with S. aureus ATCC 29213

Tigecycline dose (mg/kg)	Cmax (μg/ml)	Tmax (h)	AUC0-24 (μg·h/ml)a	Elimination half-life (h)	Protein binding (%)
50	12.24	0.74	49.19	7.36	92.9
25	5.33	0.53	25.27	11.80	92.5
12.5	2.38	0.45	9.10	10.05	87.9
6.25	1.68	0.08	4.80	10.53	81.2

^aAUC_0-24, AUC from 0 to 24 h.

Protein binding.

The percentage of tigecycline bound to protein was determined to increase with increasing concentrations. Mean percentages of protein binding (±standard deviation) were 65.7% (2.26%), 82.7% (2.5%), 93.1% (0.9%), 92% (0.8%), and 93.6% (1.3%) at concentrations of 0.75, 1.5, 6, 12, and 25 μg/ml, respectively. Nonspecific protein binding was negligible. A sigmoidal E_max model explaining the relationship between protein binding and tigecycline concentrations was constructed, and the percent bound was determined based on the peak serum drug concentration (C_max) for each dose in the bacterial-density studies (Table 1). For any given dose, all absolute concentrations in the profile were corrected for binding, with the free value determined by the C_max. As a result, the fAUC values for the doses used in the bacterial-density studies ranged from 0.2 to 23.1 μg·h/ml.

An additional approach to free-drug determination was performed. Similar to the methods used by Scaglione and colleagues (18), free-drug exposures were determined by applying the percentage of tigecycline protein bound, not at the C_max value, but instead for each observed concentration determined from the individual sampling time points during the pharmacokinetic analysis (i.e., 1 h, 2 h, … 24 h). This method as much as tripled the half-lives of the single doses and resulted in a near doubling of the fAUC exposures displayed by the various regimens (range, 0.4 to 34.5 μg·h/ml).

Bacterial-density studies.

The mean initial bacterial load of S. aureus in infected 0-h control animals was 5.78 log₁₀ CFU/ml (range, 5.7 to 5.9). After 24 h, bacterial densities of untreated control mice displayed a mean increase of 2.2 ± 0.1 logs. The maximal CFU reduction seen with tigecycline-treated animals after 24 h ranged from 1.5 to 2.3, depending on the infecting S. aureus isolate.

Pharmacodynamic index determination.

The pharmacodynamics of tigecycline were seemingly similar for all S. aureus isolates tested (Table 2). Correction for protein binding according to the C_max resulted in mean correlation coefficient values (r²) for fT>MIC, fC_max/MIC, and fAUC/MIC of 0.62, 0.81, and 0.96, respectively, for the individual isolates. When these data were pooled to construct a composite curve of all seven isolates, the respective r² values were 0.55, 0.70, and 0.80. In all instances, as shown in Fig. 1 by HA-MRSA isolate 56 and the composite curve in Fig. 2, the fAUC/MIC ratio was the exposure index most highly predictive of efficacy against these S. aureus isolates. For studies in which the dose-isolate relationship produced free-tigecycline concentrations below the MIC for the entire interval (i.e., 0% fT>MIC), CFU reductions were still observed and, in some cases, approached the E_max value.

TABLE 2.

Correlation coefficient values for each exposure index^a

Isolate	fT>MIC	fCmax/MIC	fAUC/MIC
MSSA ATCC 29213	0.699	0.974	0.983
MSSA ATCC 25923	0.872	0.508	0.877
HA-MRSA 56	0.341	0.704	0.957
HA-MRSA 152	0.511	0.953	0.997
CA-MRSA 144	0.552	0.770	0.954
CA-MRSA 147	0.483	0.892	0.966
CA-MRSA 156	0.894	0.880	0.970

^aData show correlation coefficient (r²) values for each exposure index relative to efficacy as defined by a sigmoid E_max model.

FIG. 1.

fAUC/MIC exposure response curve for tigecycline against HA-MRSA 56 in thigh-infected neutropenic mice.

EI₈₀, EI₅₀, and stasis values for fAUC/MIC are shown in Table 3, in addition to the maximum reduction in the number of CFU predicted by the E_max model. Although the target indexes varied by isolate, the genotypic profile seemed uncorrelated with these differences. The mean values for EI₈₀, EI₅₀, and stasis were 5.4, 2.6, and 2.8, respectively. Similar results were noted when these data were determined from the aggregate of S. aureus isolates (Fig. 2). Based on a calculated maximal change in the log₁₀ number of CFU/ml of 1.95, the EI₈₀, EI₅₀, and stasis values were 5.7, 2.4, and 2.6, respectively.

TABLE 3.

fAUC/MIC associated with EI₈₀, EI₅₀, and stasis values for tigecycline against all S. aureus isolates in the neutropenic thigh model

Isolate	fAUC/MIC			Maximum change in log10 CFU/ml
	EI80	EI50	Stasis
MSSA ATCC 29213	4.1	0.6	0.5	2.3
MSSA ATCC 25923	7.0	5.0	5.3	1.4
HA-MRSA 56	4.2	2.5	2.5	2.0
HA-MRSA 152	2.8	1.5	1.6	1.8
CA-MRSA 144	3.2	2.2	2.6	1.4
CA-MRSA 147	3.5	1.5	1.7	1.8
CA-MRSA 156	13.0	5.1	5.3	2.1

FIG. 2.

Composite E_max model and 95% confidence interval of fAUC/MIC as a function of change in bacterial density for diverse S. aureus isolates.

Use of the methods described by Scaglione et al. (18) for free-drug determination resulted in substantial increases in fAUC/MIC targets. Increases observed for individual isolates ranged from 1.8 to 2.9 times those discerned by the C_max free-drug correction method and resulted in a doubling of mean exposure indices.

Immunocompetent-mouse infection model.

Secondarily to the increase in the infecting inoculum, the thighs of 0-h control animals contained 6.8 ± 0.1 log₁₀ CFU/ml. In the presence of a competent immune system, these numbers increased by 0.2 ± 0.8 log₁₀ CFU/ml in control mice after 24 h. In total, four once-daily tigecycline doses were administered: 1.56, 3.13, 6.25, and 12.5 mg/kg, with respective fAUC/MIC values of 1.8, 3.6, 7.2, and 8.8. After 24 h, each dose displayed maximal efficacy, with mean CFU decreases of 1.6, 1.9, 1.8, and 1.6 logs, respectively. Accordingly, the construction of an explanatory E_max model was not possible. In comparison with CA-MRSA 156-infected neutropenic mice given the same tigecycline exposures, the resulting changes in log₁₀ numbers of CFU/ml were +1.4, +0.5, +0.1, and −1.1, respectively.

DISCUSSION

To date, the sparse data available for tetracyclines as a class seem to suggest that the AUC/MIC is the pharmacodynamic parameter most associated with antibacterial activity (3). Through the use of the neutropenic murine thigh infection model, the results of the current assessment clearly validate this observation for the structurally related glycylcycline antibiotic tigecycline against diverse S. aureus isolates. Relative to this association, we identified mean fAUC/MIC targets with slight isolate-dependent variations. Other publications have associated the presence of the Panton-Valentine leukocidin (PVL) gene as a major virulence factor associated with cSSSI and necrotizing pneumonia, secondary to CA-MRSA infection (4, 5, 10). However, despite having the highest EI₈₀ value of all the isolates, CA-MRSA 156 was the only CA-MRSA isolate that did not display the PVL factor. Moreover, ATCC 25923 displayed the next highest EI₈₀ value in spite of being methicillin susceptible. Tetracycline resistance was displayed by isolates CA-MRSA 147 and HA-MRSA 56 and 152, the EI₈₀ values of which fell below the mean value. These observations seem to suggest that there is no association between the observed target variations and the non-tigecycline susceptibility patterns or known virulence factors. Moreover, the MICs for the isolates used in this evaluation (0.125 to 0.5 μg/ml) were consistent with those seen against S. aureus in a recently published, large, global surveillance study (17) and should support a broad application of the contained results.

In this analysis, we determined that the protein binding of tigecycline ranges from 81.2 to 92.9% and increases with increasing concentration; similar observations were made during protein binding studies with humans (14). An unpublished analysis of tigecycline protein binding in mouse plasma found comparable results, with the percent protein bound ranging from 74.9 to 86.2%, again, increasing with increasing concentration (data on file at Wyeth Research). This observation, however, was not made in another published murine neutropenic thigh analysis (21). In that study van Ogtrop and colleagues reported tigecycline to be 59% protein bound; the test concentration(s) is uncertain as the experimental methods were not reported (21).

The aforementioned study by van Ogtrop et al. (21) was a pharmacodynamic appraisal of tigecycline, using a neutropenic thigh model similar to that used in our study. That evaluation was conducted with four S. aureus isolates, two of which (ATCC 29213 and ATCC 25923) were used in this assessment. Additional studies were conducted with various Escherichia coli, Klebsiella pneumoniae, and Streptococcus pneumoniae isolates. Although results for exposure-response relationships were not available for S. aureus, based on results with the other three organisms, the authors concluded that fT>MIC was the parameter most associated with efficacy, followed closely by fAUC/MIC. Although our results too showed a close association between fAUC/MIC and efficacy, we found fT>MIC to be the parameter least predictive of tigecycline efficacy. The low predictive value was highlighted by the fact that despite doses possessing 0% fT>MIC, relatively large reductions in numbers of CFU were attained. These results are similar to the findings of an in vitro time-kill analysis assessing tigecycline against Acinetobacter baumannii. In that assessment, tigecycline at concentrations of 0.6 times the MIC produced a CFU decrease of approximately 1.5 logs after 24 h. When the concentrations were increased to 0.7 and 0.8 times the MIC, efficacy rivaled that of doses 4 times the MIC after 24 h (approximately 2-log decrease) and 48 h (approximately 3.5-log decrease), respectively (19).

Certain differences between the pharmacokinetics of our analysis and those of the previous analysis (21) may help explain the discordance observed in the determined optimal exposure response relationship. For example, the pharmacokinetics of tigecycline in our study were best described by a two-compartment model and produced a mean half-life of 9.9 h, compared with the nonlinear half-lives (1.5 to 2.34 h) seen by using a one-compartment model in the previous assessment. Although this increase in half-life would suggest that our model predicted higher concentrations and thus increases in AUC and T>MIC across the dosing range, the variation seen between the protein binding inverted these differences with regard to free-drug evaluations. As a result, we calculated three- to fourfold-lower fAUC and fC_max values, comparatively. Accordingly, exposures that we found to be below the MIC for the entire interval (0% fT>MIC) may have been deemed supraMIC for at least a portion of the interval in the previously conducted study (21).

Two recent pharmacodynamic analyses with humans have been completed and also show a relationship between AUC/MIC outcomes (12, 16). Each analysis defined a classification and regression tree-derived AUC/MIC breakpoint for positive clinical and microbiological outcomes, one for patients with cIAI and the other for patients with cSSSI; both analyses also divided patients into separate cohorts based on the identified pathogen(s), to more precisely define breakpoints. With respect to cIAI, an AUC/MIC breakpoint of 6.96 was discerned, although gram-positive organisms were not present in the included cohorts (16). The study of patients with cSSSI, however, identified a relationship in a cohort containing patients infected with S. aureus and/or Streptococcus spp. The total-drug AUC/MIC breakpoint defined in this population was 17.9 (12). While new information suggests that albumin concentration and pH have the potential to alter the free fraction (15), the reported protein binding of tigecycline in humans ranges from 71 to 87% (14). If 21% of tigecycline exposure is assumed to be free drug, the fAUC/MIC target drops to 3.76 for positive microbiological and clinical outcomes seen in this study. This corresponds with the EI₈₀ value of 5.4 derived from our assessment, a value that represents the exposure required for an approximately 1- to 2-log kill and the mean fAUC/MIC value required for stasis of 2.78. These data suggest that for cSSSI, an exposure close to stasis may have been all that was required for positive outcomes.

One important factor to consider when evaluating our derived fAUC/MIC targets is that these studies were conducted in the purest representation of the bug-drug relationship, the immunocompromised model. This concept was underscored by the results of our studies with immunocompetent mice. The EI₈₀ value derived from the neutropenic mice infected with CA-MRSA 156 was 13, the highest of all the isolates tested. Nevertheless, when CA-MRSA 156 infections were undertaken in nonneutropenic mice, minimal doses of tigecycline produced maximum CFU reductions. This associates the presence of a functioning immune system to a minimum sevenfold reduction in target exposure for fAUC/MIC.

Since we appreciate that the calculation of protein binding according to a fixed free-drug concentration based on C_max may not be optimal due to the rapid equilibrium of the protein-drug complex, we also determined free-drug exposures based on variable protein binding, using the novel methods described by Scaglione and colleagues (18). Given the unique relationship observed between tigecycline concentrations and protein binding, application of this method produced a transient decrease in slope during the beta elimination phase of the pharmacokinetic curve, using compartmental analysis. Namely, as the total drug concentrations decreased, free drug subsequently increased, elevating the calculated free AUCs. Similar observations were noted using noncompartmental calculations. This resulted in fAUC/MIC targets that would be unreachable in humans and render the drug seemingly ineffective, a fact which is obviously not the case given the efficacy described in the human trials (12, 16, 18). This additional assessment using an alternative method to account for free concentrations further highlights the difficulty in assessing what is believed to be the bioactive “free”-drug component.

While we acknowledge the difficulties in determining the absolute free-drug exposures for the above-noted reasons, our proposed target appears to be consistent with the proven clinical efficacy of this compound in humans. Furthermore, these mouse-derived data also suggest that a functioning immune system will further reduce the required efficacious exposure.