Comparative Pharmacodynamics of the New Oxazolidinone Tedizolid Phosphate and Linezolid in a Neutropenic Murine Staphylococcus aureus Pneumonia Model

Abstract

Tedizolid phosphate (TR-701) is a novel oxazolidinone prodrug (converted to the active form tedizolid [TR-700]) with potent Staphylococcus aureus activity. The current studies characterized and compared the in vivo pharmacokinetic/pharmacodynamic (PD) characteristics of TR-701/TR-700 and linezolid against methicillin-susceptible S. aureus (MSSA) and methicillin-resistant S. aureus (MRSA) in the neutropenic murine pneumonia model. The pharmacokinetic properties of both drugs were linear over a dose range of 0.625 to 40 mg/kg of body weight. Protein binding was 30% for linezolid and 85% for TR-700. Mice were infected with one of 11 isolates of S. aureus, including MSSA and community- and hospital-acquired MRSA strains. Each drug was administered by oral-gastric gavage every 12 h (q12h). The dosing regimens ranged from 1.25 to 80 mg/kg/12 h for linezolid and 0.625 to 160 mg/kg/12 h for TR-701. At the start of therapy, mice had 6.24 ± 0.40 log₁₀ CFU/lungs, which increased to 7.92 ± 1.02 log₁₀ CFU/lungs in untreated animals over a 24-h period. A sigmoid maximum-effect (E_max) model was used to determine the antimicrobial exposure associated with net stasis (static dose [SD]) and 1-log-unit reduction in organism relative to the burden at the start of therapy. The static dose pharmacodynamic targets for linezolid and TR-700 were nearly identical, at a free drug (non-protein-bound) area under the concentration-time curve over 24 h in the steady state divided by the MIC (AUC/MIC ratio) of 19 and 20, respectively. The 1-log-unit kill endpoints were also similar, at 46.1 for linezolid and 34.6 for TR-700. The exposure targets were also comparable for both MSSA and MRSA isolates. These dosing goals support further clinical trial examination of TR-701 in MSSA and MRSA pneumonia.

INTRODUCTION

Staphylococcus aureus has become a leading cause of community- and hospital-acquired pneumonia. Furthermore, infection with community- and health care-acquired strains of methicillin-resistant S. aureus (MRSA) is increasingly common and limits antimicrobial options (13, 25, 29, 31). Tedizolid phosphate (TR-701) is a novel oxazolidinone prodrug. The active moiety, tedizolid (TR-700), exhibits potent activity against Gram-positive bacteria, including MRSA (48). The goals of our studies were to identify the pharmacodynamic (PD) target (AUC/MIC ratio) for TR-700 against methicillin-susceptible S. aureus (MSSA) and MRSA strains in a neutropenic murine pneumonia model and compare the PD targets of TR-700 to that of the only FDA-approved oxazolidinone antibiotic, linezolid.

MATERIALS AND METHODS

Bacteria, media, and antibiotics.

Eleven isolates of S. aureus were utilized for these experiments. The strains included four MSSA strains (ATCC 6538P, ATCC 25923, ATCC 29213, and ATCC Smith), one hospital-acquired MRSA strain (ATCC 33591), and six community-acquired MRSA strains (MW2, R2527, 04-045, 04-154, 05-051, and UW 307109). Organisms were grown, subcultured, and quantified in Mueller-Hinton broth (Difco Laboratories, Detroit, MI) and Mueller-Hinton agar (Difco Laboratories, Detroit, MI). Linezolid was obtained from the University of Wisconsin pharmacy. Tedizolid (TR-700) and tedizolid phosphate (TR-701) were supplied by the manufacturer (Trius Therapeutics, San Diego, CA). Prior to in vivo treatment studies, TR-701 was dissolved in sterile water at the appropriate drug concentrations. TR-700 was dissolved in sterile dimethyl sulfoxide (DMSO) and diluted with medium to the appropriate concentration for in vitro susceptibility testing.

In vitro susceptibility studies.

The MICs of linezolid and TR-700 were determined three times in duplicate using Clinical and Laboratory Standards Institute (CLSI) microdilution methods M07-A8 (15). The duplicates were identical, and the final results were reported as the median value of the three MIC tests (see Table 1).

Table 1.

In vitro activity of linezolid and TR-700 against select S. aureus isolates

Organism	MIC (mg/liter)a			Lineage
	Linezolid	TR-700	Methicillin
MSSA
ATCC 25923	2	0.06	0.25
ATCC 29213	2	0.25	0.125
ATCC 6538P		0.125	0.5
ATCC Smith		0.5	0.125
MRSA
04-045	2	0.125	>16
04-154	2	0.25	>16
MW2	2	0.5	>16	USA400
307109		0.25	>16
ATCC 33591		0.25	>16	USA200
R2527		0.125	>16	USA300
05-051		0.25	>16

^aThe MICs are reported as the median values of three MIC tests.

Murine infection model.

All animal studies were approved by the Animal Research Committees of the University of Wisconsin and William S. Middleton Memorial VA Hospital. Six-week-old, specific-pathogen-free, female ICR/Swiss mice weighing between 24 and 27 g (Harlan Sprague-Dawley, Indianapolis, IN) were used for all studies. Three mice were utilized for each treatment and control group. Mice were rendered neutropenic (absolute neutrophil count of <100/mm³) by injecting cyclophosphamide (Mead Johnson Pharmaceuticals, Evansville, IN) intraperitoneally 4 days (150 mg/kg of body weight) and 1 day (100 mg/kg) prior to experimental infection. Previous studies have shown that this regimen produces persistent neutropenia in this animal model for 5 days (3). Broth cultures of freshly plated bacteria were grown to logarithmic phase. The inoculum for each organism was diluted to a 0.3 absorbance at 580 nm. Viable plate count confirmation of the inoculum ranged from 10⁶ to 10⁷ CFU/ml. Animals were anesthetized by isoflurane inhalation. The mice were held upright, and 50 μl of the inoculum was administered to the nares and inhaled into the lung. Treatment commenced 2 h after challenge, at which time mice had 6.24 ± 0.40 log₁₀ CFU/lungs.

Drug pharmacokinetics.

Single-dose plasma pharmacokinetic (PK) studies were performed in infected mice given oral doses of linezolid (0.625, 2.5, 10, and 40 mg/kg) or TR-701 (0.625, 2.5, 10, and 40 mg/kg). Blood samples were obtained from groups of three mice at 0.25, 0.5, 1, 3, 6, 9, 12, and 24 h. The concentrations of compounds were analyzed by a validated liquid chromatography-tandem mass spectrometry (LC-MS-MS) by Midwest Bioresearch, Skokie, IL, as previously described (43). The assay lower limit of quantitation was 1 ng/ml, and the coefficient of variation ranged from 1 to 7%. A noncompartmental model was used for PK analyses. PK parameters, including elimination half-life and concentration at time zero (C₀), were calculated via nonlinear least-squares techniques. The area under the concentration-time curve (AUC) was calculated by the trapezoidal rule. For treatment doses in which no kinetics were determined, the PK index was estimated by linear extrapolation for higher and lower dose levels and interpolation for dose levels within the dose range studied. Protein binding was previously determined by the sponsor, Trius Pharmaceuticals (85%).

Treatment protocols.

Two hours after pulmonary challenge, mice were treated with either linezolid or TR-701. Linezolid was administered by oral gavage (OG) every 12 h over a dose range of 1.25 to 80 mg/kg. TR-701 was administered by OG every 12 h over a dose range of 0.625 to 160 mg/kg. Groups of three mice were utilized for each dose administered, 0-h and 24-h untreated controls. After 24 h of treatment the mice were euthanized by carbon dioxide asphyxiation. Untreated control mice were assessed at time of treatment initiation. Following euthanasia, the lungs were harvested using sterile technique, homogenized, diluted, plated, and incubated, and viable plate counts were assessed as previously described (18). Data are expressed as the mean change in log₁₀ CFU/lungs for three mice per treatment group compared to the organism burden at the start of therapy.

Data analysis.

The results were analyzed using a sigmoid E_max model: log₁₀ SD = {log₁₀ [E/(E_max − E)]/N } + log ED₅₀, where SD is the static dose or dose level necessary to maintain organism burden the same as at the start of treatment, E is the control growth, or in this case the log change in CFU/ml lung homogenate in untreated controls after 24 h, E_max is the maximum effect, ED₅₀ is the dose required to achieve 50% of E_max, and N is the slope of the dose-response curve. The indices E_max, ED₅₀, and N were calculated for each drug and strain using nonlinear least-squares regression. The dose needed to achieve net stasis (SD) and 1-log-unit kill was determined using the formula for each drug against each strain tested. The area under the concentration-time curve over 24 h in the steady state divided by the MIC (AUC/MIC ratio) has been shown in previous studies to be the pharmacodynamic (PD) parameter predictive of treatment efficacy for oxazolidinones (8, 35). Therefore, the total drug AUC/MIC (tAUC/MIC ratio) and free drug (non-protein-bound) AUC/MIC (fAUC/MIC ratio) for each drug-organism combination was also determined. The coefficient of determination (R²) was used to estimate the variance that could be due to regression with the drug exposure index. Analysis of variance (ANOVA) was used to determine whether the differences in PD target needed for stasis and killing were significant between linezolid and TR-700. The Mann-Whitney U test was used to compare TR-700 PD targets for MSSA and MRSA strain groups.

RESULTS

In vitro susceptibility testing.

The MICs of linezolid and tedizolid (TR-700) are listed in Table 1. The MICs for linezolid were the same for the five organisms tested, whereas the MICs for TR-700 varied by 8-fold (0.06 to 0.5 μg/ml). In vitro potency was 4- to 32-fold higher for TR-700. Methicillin potency did not influence the linezolid or TR-700 MICs.

Pharmacokinetics.

The time course plasma levels of linezolid and TR-700 in infected neutropenic mice after oral administration of 0.625, 2.5, 10, and 40 mg/kg are shown in Fig. 1. PK parameters, including elimination half-life, maximum plasma drug concentration (C_max), and area under the drug-concentration curve from 0 h to infinity (AUC_0–∞) are also shown; however, these were not determined for the lowest linezolid dose (0.625 mg/kg), as there were too few data points to accurately calculate the parameters. The plasma pharmacokinetics of both drugs were linear over the dose range (R² of >99% based upon linear regression of AUC). The elimination half-life of linezolid ranged from 0.4 to 0.9 h. The C_max increased from 0.6 to 30.0 mg/liter over the dose range, and the AUC ranged from 0.6 to 56.2 mg · h/liter. The elimination half-life of TR-700 ranged from 3.3 to 4.3 h. The C_max increased from 0.2 to 20.1 mg/liter over the dose range, and the AUC ranged from 1.4 to 131 mg · h/liter. Protein binding values were previously reported for linezolid at 30% (51). The binding value for TR-700 was based upon prior studies performed by the sponsor (85%).

Fig 1.

Concentrations for linezolid (A) and TR-700 (B) in plasma after administration of single oral doses of 0.625, 2.5, 10, and 40 mg/kg in neutropenic infected mice. Each symbol represents the geometric mean ± standard deviation (error bar) of the levels measured in three mice. Cmax, peak level in plasma; AUC, plasma AUC_0–∞; T1/2, plasma elimination half-life (in hours). PK parameters were not determined for the lowest dose of linezolid, as there were too few data points to estimate these parameters.

Dose response, 24-h static dose, 1-log-unit kill determination.

The dose-response curves following treatment with escalating doses of linezolid and tedizolid phosphate (TR-701) against 5 and 11 S. aureus isolates, respectively, are shown in Fig. 2. At the start of therapy, mice had 6.24 ± 0.40 log₁₀ CFU/lung. The organisms grew to 7.92 ± 1.02 log₁₀ CFU/lung in untreated control mice over 24 h. The maximal reductions of burden in S. aureus-infected mice treated with linezolid and TR-701 were 3.27 ± 0.56 log₁₀ CFU/lung and 3.93 ± 0.24 log₁₀ CFU/lung, respectively. The dose-response curves for linezolid were nearly identical, as were the in vitro MICs. The MIC variation for TR-700 did appear to impact the dose-response relationship. Higher TR-701 doses were needed for similar effects against organisms with higher MICs compared to those with lower MICs. Tables 2 and 3 show the 24-h total dose required to achieve a static effect and 1-log-unit kill for linezolid and TR-701 against each S. aureus isolate. The mean plasma 24-h tAUC/MIC ratios associated with a static endpoint were 27.2 and 133 for linezolid and TR-700, respectively. However, when protein binding was considered, the mean plasma fAUC/MIC ratios for linezolid and TR-700 were similar at 19 and 20, respectively. The mean plasma 24-h fAUC/MIC associated with 1-log-unit kill reduction was roughly 2-fold higher than that needed for stasis (46.1 for linezolid and 34.6 for TR-700) and were not significantly different (P = 0.334). Additionally, the plasma 24-h fAUC/MIC associated with net stasis and 1-log-unit kill was not significantly different in MSSA and MRSA isolates (P = 0.41) for TR-700. The relationships between the AUC/MIC ratio and effect for linezolid and TR-700 are shown in Fig. 3. The exposure-response relationships were relatively strong based upon the coefficient of determination (R² = 0.68 for linezolid and R² = 0.48 for TR-700).

Fig 2.

Dose-response relationships for linezolid (A) and TR-701 (B). Mice received one of a series of 2-fold increasing doses of linezolid or TR-701 every 12 h over a 24-hour treatment period. Each symbol represents the mean organism burden in the lungs of three mice. The error bars represent the standard deviations. The dashed horizontal line represents the organism burden at the start of therapy. Symbols below the line represent organism reduction compared to the start of therapy. Symbols above the line represent organism growth.

Table 2.

In vivo activity of linezolid against select S. aureus isolates^a

S. aureus isolate or parameter	MIC (mg/liter)	SD (mg/kg/24 h)	24-h SD tAUC/MIC	24-h SD fAUC/MIC	1-log-unit kill dose (mg/kg/24 h)	24-h 1-log-unit kill tAUC/MIC	24-h 1-log-unit kill fAUC/MIC
S. aureus isolates
ATCC 25923	2	18.8	21.6	15.1	50.1	70.3	49.2
ATCC 29213	2	35.1	48.2	33.7	61.6	86.5	60.5
MW2	2	30.3	40.4	28.3	52.6	73.8	51.7
04-154	2	13.6	13	9.1	39.8	55.8	39
04-045	2	13.4	12.7	8.9	31.8	42.7	28.9
Mean	2	22.2	27.2	19	47.2	65.8	46.1
Median	2	18.8	21.6	15.1	50.1	70.3	49.2
SD	0	9.9	16.3	11.4	11.6	16.9	11.9

^aSD, static dose; tAUC/MIC, total drug AUC/MIC; fAUC/MIC, free drug (non-protein-bound) AUC/MIC. The means, medians, and static doses are shown in boldface type.

Table 3.

In vivo activity of TR-701 or TR-700 against select S. aureus isolates^a

S. aureus isolate or parameter	MIC (mg/liter)	TR-701 SD (mg/kg/24 h)	TR-700 24-h SD tAUC/MIC	TR-700 24-h SD fAUC/MIC	TR-701 1-log-unit kill dose (mg/kg/24 h)	TR-700 24-h 1-log-unit kill tAUC/MIC	TR-700 24-h 1-log-unit kill fAUC/MIC
S. aureus isolates
ATCC Smith	0.5	9.1	86.8	13	14	148.3	22.2
UW 307109	0.25	6.4	124.6	18.7	9.5	181	27.3
MW2	0.5	11	108.1	16.2	15.9	173.2	26
ATCC 29213	0.25	14.3	303.7	45.5	20.3	461	69.2
ATCC 33591	0.25	7.5	144.2	21.6	15.3	331.1	49.7
ATCC 6538P	0.125	3.7	145.9	21.9	6.3	242.9	36.4
R2527	0.125	7.2	277.3	41.6	13.9	588.2	88.2
05-051	0.25	3.4	67.7	10.1	5.1	99.3	14.9
ATCC 25923	0.06	1.4	116.2	17.4	1.8	147.3	22.1
04-154	0.125	0.8	27.6	4.1	1.8	70.4	10.6
04-045	0.25	3.1	61.9	9.29	4.7	91.2	13.7
Mean	0.24	6.2	133.1	20	9.9	230.4	34.6
Median	0.25	6.4	116.2	17.4	9.5	173.2	26
SD	0.14	4.2	85.9	12.9	6.3	165.2	24.8

^aSD, static dose; tAUC/MIC, total drug AUC/MIC; fAUC/MIC, free drug (non-protein-bound) AUC/MIC. The means, medians, and static doses are shown in boldface type.

Fig 3.

Relationship between linezolid (A) and TR-700 (B) plasma 24-h fAUC/MIC ratios and in vivo efficacy against multiple strains of S. aureus (5 and 11 strains, respectively). Mice were treated with 2-fold increasing concentrations of either linezolid or TR-701 given every 12 h over a 24-h treatment period. Each symbol represents the geometric mean in the change of organism burden from the start of therapy in the lungs of three mice. The dashed horizontal line represents the burden in the lungs at the start of therapy. Symbols below the line represent organism reduction compared to the start of therapy. Symbols above the line represent organism growth. The sigmoid lines represent the best-fit curves. R² is the coefficient of determination.

DISCUSSION

Antimicrobial pharmacokinetic-pharmacodynamic (PK/PD) studies are useful to identify the therapeutic potential of a drug through the integration of the PK properties, in vitro potency (MIC), and outcome. These approaches have proven useful in guiding design of effective drug regimens in humans and the development of susceptibility breakpoints (1, 2, 17, 21, 40, 41, 45). The emergence of antibiotic resistance has further elevated the relevance of these investigations to optimize treatment strategies. Drug resistance in Gram-positive organisms, especially S. aureus, has become epidemic worldwide and is linked to escalating morbidity and mortality (19, 20, 52, 54). Among the infections produced by this pathogen, health care- and ventilator-associated pneumonia is associated with the most dismal outcomes, and S. aureus is now recognized as the most commonly implicated organism (28, 50). Recent experimental and clinical investigations suggest the potential advantage of treatment with the oxazolidinone linezolid for this disease state due to potency against MRSA and penetration into the infection site relative to vancomycin and daptomycin (12, 16, 24, 30, 36, 38, 56–58). The present studies were designed to characterize the PK/PD relationships of a new oxazolidinone, TR-701, for treatment of these infections. We further compared these PD relationships to those from a concurrent study of linezolid.

Prior PK/PD study with oxazolidinones, including TR-701, has demonstrated potency against both MSSA and MRSA and the relevance of the AUC/MIC index (2, 22, 23, 27, 34, 35, 37, 44, 53, 55). For example, Louie et al. (35) performed a dose fractionation PK/PD evaluation of TR-701 against MRSA and MSSA using the murine thigh infection model. The total daily dose ranged from 0 to 100 mg/kg and was fractionated into every 24 h (q24h), q12h, and q6h dosing regimens. The dose-response curves were nearly identical for each dosing interval, suggesting that the amount of drug rather than the dosing frequency was most important for optimizing efficacy. Analysis of the dose fractionation data demonstrated that the AUC/MIC ratio was the driving PD index with the strongest regression fit (R² = 0.98).

While the index associated with therapeutic success has been extensively examined, few studies have attempted to identify the AUC/MIC magnitude or target associated with efficacy in treatment against MSSA or MRSA. In an in vivo study of linezolid using a neutropenic murine thigh infection model, the PD index AUC/MIC provided the best correlation with efficacy and the static dose target was a fAUC/MIC of 58 (8). A single clinical PK/PD study examined linezolid efficacy in patients with MRSA cellulitis, bacteremia, and pneumonia (44). Clinical success was more common when the linezolid plasma fAUC/MIC value was near 30. More recently, the murine thigh infection model was used to explore the treatment efficacy of TR-701 against a single MRSA isolate (35). Similar to linezolid, the AUC/MIC ratio was the predictive index for efficacy, and net stasis in infectious burden was noted at a fAUC/MIC of 43 at the 24-h study endpoint.

The AUC/MIC stasis target identified in the current staphylococcal pneumonia studies was 2- to 4-fold lower than those reported in previous studies for both linezolid and TR-700 in soft tissue models. The stasis fAUC/MIC target in this study against nearly a dozen staphylococcal isolates was a value near 20. The similarity of the fAUC/MIC targets for both compounds when considering unbound drug (binding of 30% for linezolid versus 85% for TR-700) highlights the relevance of protein binding consideration when comparing PK/PD relationships across compounds within a similar mechanistic class (2, 17). The TR-700 fAUC/MIC magnitude associated with the 1-log-unit killing endpoint was only 34.6. The mechanistic explanation for the difference in PD targets in this lung model compared to previous soft tissue models is not entirely clear. The designs of each of these studies were similar in many ways, including the mouse species, drug exposures (PK), and the neutropenic state of immunosuppression. One small difference was the level of protein binding utilized for calculating free drug (non-protein-bound) exposures. The current study used 85% as the protein-bound fraction, whereas the previous study used 80% (35). Accounting for this difference alone would make the fAUC/MIC PD targets more congruent. Another key difference between the current and previous studies is the infection site. One intuitive theory to explain the discrepancy is the differential penetration of the oxazolidinones into the lung infection site (12, 16, 24, 26, 55). For example, at multiple time points after administration of linezolid (600 mg per os [p.o.]) twice a day (BID) for 2.5 days to humans, bronchoscopic examination of epithelial lining fluid (ELF) linezolid concentrations revealed 2.4- to 4.2-fold-higher concentrations of drug in ELF than in plasma (16). Similar results have been demonstrated with TR-700 with 40-fold-higher accumulation of active drug in ELF than in plasma (26). This PK difference at the infection site is a plausible explanation for the lower PD targets for oxazolidinones in pulmonary models compared to soft tissue models.

The majority of antimicrobial PD investigations have shown that the PD target is similar in treatment of organisms exhibiting drug resistance. We explored this tenet by study of multiple S. aureus strains with beta-lactam, macrolide, and lincosamide resistance mechanisms. The treatment fAUC/MIC targets were congruent across these strain genotypes as previously described for other drug classes (4–7, 47).

A valuable application of preclinical PD studies is the ability to translate the results from preclinical animal model studies into the clinical realm to predict whether certain drug exposures in patients are likely to be successful for a given organism, site of infection, and MIC distribution. PK studies of healthy volunteers have been recently reported with TR-701 (10, 11, 39, 42). Oral doses of 200 mg once a day produced an AUC between 22 and 26 mg · h/liter, whereas oral doses of 400 mg once daily produced an AUC between 54 and 56 mg · h/liter (10, 39, 42). If one considers the free drug AUC associated with these doses, the highest MIC for which a stasis fAUC/MIC target of 20 would be achieved is approximately 0.5 mg/liter for a 400-mg once-daily regimen. The MIC ceiling associated with the 1-log-unit kill PD targets would be an MIC of 0.25 mg/liter. In vitro surveillance study with TR-700 for S. aureus has identified an MIC₅₀ and MIC₉₀ of 0.25 and 0.5 mg/liter, respectively (9, 14, 48, 59).

In conclusion, the present study characterized the in vivo drug exposure response of a novel oxazolidinone, TR-701, in a neutropenic murine S. aureus pneumonia model. The results suggest that TR-701 would exhibit an efficacy similar to the efficacy of linezolid for S. aureus pneumonia. An unanswered question in the current study is whether TR-701 would be expected to have the same efficacy for a linezolid-resistant isolate, as has been suggested by in vitro results (32, 33, 46, 49). The results suggest further study of TR-701 in clinical trials of pneumonia due to S. aureus is warranted and may be especially useful given the increased burden of MSSA and MRSA pneumonia in the health care setting.