Streptococcus pneumoniae

PHARMACOLOGY

crossm Comparative In Vivo Efficacies of Tedizolid in Neutropenic versus Immunocompetent Murine Streptococcus pneumoniae Lung Infection Models Kamilia Abdelraouf,a David P. Nicolaua,b Center for Anti-Infective Research and Development, Hartford Hospital, Hartford, Connecticut, USAa; Division of Infectious Diseases, Hartford Hospital, Hartford, Connecticut, USAb

ABSTRACT Given that tedizolid exhibits substantial lung penetration, we hypothe-

size that it could achieve good efficacy against Streptococcus pneumoniae lung infections. We evaluated the pharmacodynamics of tedizolid for treatment of S. pneumoniae lung infections and compared the efficacies of tedizolid human-simulated epithelial lining fluid (ELF) exposures in immunocompetent and neutropenic murine lung infection models. ICR mice were rendered neutropenic via intraperitoneal cyclophosphamide injections and then inoculated intranasally with S. pneumoniae suspensions. Immunocompetent CBA/J mice were inoculated similarly. Single daily tedizolid doses were administered 4 h postinoculation (termed 0 h). Changes in log10 CFU at 24 h compared with 0-h controls were estimated. Ratios of area under the free-drug concentration-time curve to MIC (fAUC0 –24/MIC) required to achieve various efficacy endpoints against each isolate were estimated using the Hill equation. Tedizolid doses in neutropenic and immunocompetent mice that mimic the human-simulated ELF exposure were examined. Stasis, 1-log reduction, and 2-log reduction were achieved at fAUC0 –24/MIC of 8.96, 24.62, and 48.34, respectively, in immunocompetent mice and 19.21, 48.29, and 103.95, respectively, in neutropenic mice. Tedizolid at 40 mg/kg of body weight/day and 55 mg/kg/day in immunocompetent and neutropenic mice, respectively, resulted in ELF AUC0 –24 comparable to that achieved in humans following a 200-mg once-daily clinical dose. These human-simulated ELF exposures were adequate to attain ⬎2-log reduction in bacterial burden at 24 h in 3 out of 4 isolates in both models and 1.58- and 0.74-log reductions with the fourth isolate in immunocompetent and neutropenic mice, respectively. Tedizolid showed potent in vivo efficacy against S. pneumoniae in both immunocompetent and neutropenic lung infection models, which support its consideration for S. pneumoniae lung infections.

Received 9 September 2016 Returned for modification 30 September 2016 Accepted 19 October 2016 Accepted manuscript posted online 31 October 2016 Citation Abdelraouf K, Nicolau DP. 2017. Comparative in vivo efficacies of tedizolid in neutropenic versus immunocompetent murine Streptococcus pneumoniae lung infection models. Antimicrob Agents Chemother 61:e01957-16. https://doi.org/ 10.1128/AAC.01957-16. Copyright © 2016 American Society for Microbiology. All Rights Reserved. Address correspondence to David P. Nicolau, [email protected].

KEYWORDS oxazolidinedione, pharmacodynamics, pneumonia

S

treptococcus pneumoniae represents a challenging pathogen for clinicians, as this organism has been responsible for a wide range of potentially life-threatening infections in both the community and nosocomial settings, including communityacquired pneumonia, sepsis, and meningitis (1). Pneumococcal pneumonia is a serious public health problem resulting in approximately 600,000 hospital admissions or visits to health care providers among adults annually (2). In 30% of severe S. pneumoniae infections, the bacteria are found to be resistant to at least one class of antibiotics, such as ß-lactams and macrolides, which complicates treatment and can result in as many as 7,000 deaths annually (2–4). Resistant pneumococcal pneumonia cases result in additional visits to health care providers and account for approximately $96 million in extra medical costs (2). Vancomycin tolerance, a precursor phenotype to resistance, has

January 2017 Volume 61 Issue 1 e01957-16

Antimicrobial Agents and Chemotherapy

aac.asm.org 1

Abdelraouf and Nicolau


FIG 1 Relationship between tedizolid fAUC0 –24 and the dose in the immunocompetent (A) and the neutropenic (B) lung infection models.

emerged among S. pneumoniae isolates and threatens the clinical utility of this lastresort agent (5–7). Additionally, S. pneumoniae isolates with reduced susceptibility to linezolid (MIC of ⱖ4 ␮g/ml in 0.3 to 1.4% of isolates) have been reported; thus, the emergence of resistance in the near future to this relatively new agent is unfortunately likely (8–10). As a result of limited therapeutic options available due to both the reduced susceptibility and the potential risk of drug toxicities, the development of new therapies with activity against S. pneumoniae infections is urgently warranted. Tedizolid, a novel oxazolidinedione, possesses potent in vitro activity against S. pneumoniae, including multidrug-resistant S. pneumoniae; thus, it may offer a novel approach to combating this pathogen (11). Similar to linezolid, tedizolid exerts its activity by binding to the 23S rRNA of the 50S subunit, thus preventing the formation of the bacterial 70S initiation complex and ultimately inhibiting bacterial protein synthesis (11). In 2014, tedizolid was approved for the treatment of acute bacterial skin and skin structure infections caused by susceptible Gram-positive bacteria, such as Staphylococcus aureus (including methicillin-resistant [MRSA] and methicillin-susceptible strains), various Streptococcus species, and Enterococcus faecalis (12). In light of its substantial pulmonary distribution in the epithelial lining fluid (ELF) observed in healthy adults (13), tedizolid is currently being studied in phase 3 for the treatment of life-threatening hospital-acquired bacterial pneumonia and ventilator-associated bacterial pneumonia. Knowledge of the pharmacodynamics of tedizolid for the treatment of lung infections caused by this relevant Grampositive organism will help to support these efforts. While several studies described tedizolid in vivo pharmacodynamics against S. aureus (14–16), limited data are available for the treatment of S. pneumoniae lung infections. The objective of this study was to evaluate and compare the pharmacodynamics of tedizolid for the treatment of lung infections caused by S. pneumoniae in the immunocompetent and the neutropenic murine lung infection models. Furthermore, we compared the in vivo efficacy achieved with tedizolid human-simulated ELF exposures in these two infection models. (This study was presented in part at IDWeek 2016, New Orleans, LA, 26 to 30 October 2016 [17, 18]). RESULTS Single-dose pharmacokinetics. Tedizolid was satisfactorily detected in mouse plasma for up to 24 h postadministration in both infection models. The observations were satisfactorily described by a 1-compartment linear model. Good model fits were achieved for all of the doses administered (R2 ⱖ 0.996). Based on the best-fit parameter estimates, the tedizolid area under the free-drug concentration-time curve from 0 to 24 January 2017 Volume 61 Issue 1 e01957-16

aac.asm.org 2

Tedizolid Efficacy in Murine Lung Infection Model


TABLE 1 Comparative ELF exposures of tedizolid in human and murine immunocompetent and neutropenic lung infection models Model Human Murine immunocompetent Murine neutropenic

Tedizolid daily dosing regimen 200 mg 40 mg/kg 55 mg/kg

ELF AUC0–24 (mg · h/liter) 109.30 110.22 116.90

ELF penetration ratio 41.24 3.75 3.68

h (fAUC0 –24) was estimated for each dose and appeared to increase nonlinearly with the dose in both infection models (Fig. 1). The relationships between the fAUC0 –24 and the dose were best described by the equations y ⫽ 2E ⫺ 07x4 ⫺ 3E ⫺ 06x3 ⫹ 0.0101x2 ⫹ 0.3129x ⫹ 0.2997 (R2 ⫽ 1) and y ⫽ ⫺4E ⫺ 06x4 ⫹ 0.0011x3 ⫺ 0.0854x2 ⫹ 2.9723x ⫺ 19.789 (R2 ⫽ 0.994) in the immunocompetent and the neutropenic infection models, respectively. Bronchopulmonary pharmacokinetics and human-simulated ELF exposures of tedizolid. Tedizolid was also satisfactorily detected in mice BAL fluid for up to 24 h postadministration in both infection models. A tedizolid dose of 40 mg/kg of body weight/day resulted in an ELF AUC0 –24 of 110.22 mg · h/liter in the immunocompetent mice, while a tedizolid dose of 55 mg/kg/day resulted in a mean ELF AUC0 –24 of 116.90 mg · h/liter in the neutropenic mice. These exposures were comparable to that achieved in humans (109.30 mg · h/liter) following the administration of a 200-mg once-daily (QD) clinical dose. However, due to the differences in the tedizolid penetration ratios in mice and humans, the mean plasma fAUC0 –24 achieved in mice with these doses were higher than that achieved in humans following a clinical dose (13). The ELF exposures of tedizolid in human and mouse models are summarized in Table 1. The profiles of tedizolid concentration in ELF following the administration of the identified human-simulated regimens in the mouse models utilized and that achieved in healthy volunteers following administration of the 200-mg QD dose are shown in Fig. 2. The calculated mean penetration ratios of tedizolid into the ELF were 3.75 and 3.68 in the immunocompetent and the neutropenic infection models, respectively. In vivo efficacy studies. At 0 h, the average bacterial burdens in the lungs were 8.01 ⫾ 0.27 and 7.13 ⫾ 0.06 log10 CFU/lung in the immunocompetent and neutropenic mice, respectively. The bacterial burdens increased over 24 h by magnitudes of 0.66 ⫾ 0.50 and 0.76 ⫾ 0.28 log10 CFU/lung in the untreated control immunocompetent and neutropenic mice, respectively. The relationships between the fAUC0 –24/MIC ratio and

FIG 2 Profiles of tedizolid concentration in ELF following the administration of single doses of tedizolid: 40 mg/kg in the immunocompetent infection model (dotted line), 55 mg/kg in the neutropenic infection model (dashed line), and 200 mg in healthy volunteers (solid line). Data are means ⫾ standard deviations. January 2017 Volume 61 Issue 1 e01957-16

aac.asm.org 3



FIG 3 Curve of best fit to the fAUC0 –24/MIC ratios and change in log10 CFU at 24 h for ATCC 700905 (A), ATCC 6303 (B), SPN 95 (C), SPN 102 (D), and the composite of S. pneumoniae isolates (E) in the immunocompetent mice (solid line) and the neutropenic mice (dotted line). The open circles represent the actual reductions in the bacterial burdens observed in the immunocompetent mice, while the closed circles represent the actual reductions in bacterial burdens observed in the neutropenic mice. Data are means ⫾ standard deviations.

change in log10 CFU at 24 h for each S. pneumoniae isolate in the immunocompetent and neutropenic mice are shown in Fig. 3. The exposure-response relationships for each isolate were relatively strong based upon the coefficient of determination (R2 of ⱖ0.93 in the immunocompetent model and R2 of ⱖ0.90 in the neutropenic model). The January 2017 Volume 61 Issue 1 e01957-16

aac.asm.org 4



TABLE 2 Tedizolid exposures required to achieve net stasis and 1-log and 2-log kills against S. pneumoniae isolates in immunocompetent and neutropenic murine lung infection models fAUC0–24/MIC ratio required to achieve: Isolate (tedizolid MIC, mg/liter) ATCC 700905 (0.25) ATCC 6303 (0.5) SPN 95 (0.25) SPN 102 (0.25) Composite (all isolates) aR,

Penicillin susceptibilitya R S S I

Stasis Immunocompetent 4.39 16.43 8.64 14.77 8.96

1-Log reduction Neutropenic 14.60 33.95 19.05 15.40 19.21

Immunocompetent 17.46 37.14 18.52 34.33 24.62

2-Log reduction Neutropenic 37.94 90.40 42.13 39.24 48.29

Immunocompetent 38.46 479.85 30.61 54.71 48.34

Neutropenic 72.42 395.81 81.59 79.40 103.95

resistant; S, susceptible; I, intermediate.

average maximal reductions of burden (Imax) at 24 h achieved in immunocompetent mice treated with tedizolid was higher than that achieved in the neutropenic mice: 5.25 ⫾ 0.64 log10 CFU/lung and 4.10 ⫾ 0.31 log10 CFU/lung, respectively. However, this apparent difference was not statistically significant (P ⫽ 0.1). Table 2 shows the tedizolid exposures required to achieve a static effect and 1-log and 2-log kills against each S. pneumoniae isolate in each infection model as well as the tedizolid MIC for each isolate. The plasma fAUC0 –24/MIC ratios associated with net stasis and 1-log kill in the neutropenic model were higher than those needed to achieve a similar effect in the immunocompetent model for 3 out of 4 isolates examined and were comparable for the fourth isolate. For the composite of tested isolates, the plasma fAUC0 –24/MIC ratios associated with a static endpoint was 8.96 and 19.21 in the immunocompetent and the neutropenic model, respectively. The plasma fAUC0 –24/MIC ratio associated with 1-log kill in the immunocompetent model was 2-fold lower than that needed to achieve a similar effect in the neutropenic model (24.62 versus 48.29). Similarly, the plasma fAUC0 –24/MIC ratio associated with 2-log kill in the immunocompetent model was roughly 2-fold lower than that needed to achieve a similar effect in the neutropenic model (48.34 versus 103.95). The comparative efficacy of the tedizolid human-simulated ELF exposures against each S. pneumoniae isolate in the two mouse models is shown in Table 3. At 24 h, the increase in bacterial burdens in the lungs of the untreated control immunocompetent and neutropenic mice was comparable for each S. pneumoniae isolate (P ⬎ 0.05). The average reduction in log10 CFU/lung at 24 h achieved in immunocompetent mice was greater than that achieved in the neutropenic mice receiving the same ELF exposure profiles. More important was the fact that these human-simulated exposures were adequate to attain a more than 2-log reduction in lung bacterial burden in 3 out of 4

TABLE 3 Comparative efficacy of human-simulated ELF exposures of tedizolid in the immunocompetent and the neutropenic infection models against S. pneumoniae isolates Change in log10CFU at 24 ha Isolate (MIC, mg/liter) ATCC 700905 (0.25) Control Treatment

Immunocompetent

Neutropenic

P value

0.37 ⫾ 0.34 ⫺3.88 ⫾ 1.02

0.41 ⫾ 0.46 ⫺2.90 ⫾ 0.44

0.85 0.06

ATCC 6303 (0.5) Control Treatment

1.38 ⫾ 0.31 ⫺1.58 ⫾ 0.41

1.10 ⫾ 0.33 ⫺0.74 ⫾ 0.57

0.17 0.01

SPN 95 (0.25) Control Treatment

0.59 ⫾ 0.27 ⫺4.46 ⫾ 0.61

0.76 ⫾ 0.19 ⫺2.22 ⫾ 0.56

0.18 ⬍0.0001

SPN 102 (0.25) Control Treatment

0.30 ⫾ 0.45 ⫺3.85 ⫾ 0.52

0.76 ⫾ 0.51 ⫺2.89 ⫾ 0.78

0.13 0.03

aData

are means ⫾ standard deviations.


aac.asm.org 5



isolates in both infection models and 1.58- and 0.74-log reduction in the fourth isolate in the immunocompetent and the neutropenic mice, respectively. DISCUSSION In this study, we examined the tedizolid exposure-response relationship in the neutropenic and immunocompetent murine S. pneumoniae lung infection model. Four S. pneumoniae isolates were selected for this examination. The selection of the isolates was based on the ability of the isolates to establish viable infection in the murine lungs as well as the tedizolid MIC for the isolates; the MIC for the isolates ranged between 0.25 and 0.5 mg/liter, which was consistent with the tedizolid MIC distributions reported in previously published studies (19–21). The results of the in vivo efficacy studies showed that the extent of tedizolid bacterial killing correlated well with the fAUC0 –24/MIC ratios. Substantial bacterial killing was achieved at higher tedizolid exposures for all isolates tested in the two infection models within 24 h. The exposures associated with net stasis and 1-log reduction in lung bacterial burden relative to that at the start of therapy identified in the neutropenic model were 19.21 and 48.29, respectively. These exposures were somewhat consistent with the tedizolid pharmacodynamics targets previously reported in a neutropenic murine S. aureus pneumonia model (fAUC/MIC ratios of 20 and 34.6 for net stasis and 1-log reduction in lung bacterial burden, respectively) (15). The similarity in the findings between the two studies implies that these exposure targets could be bacterial species independent. Overall, higher tedizolid exposures were needed for similar effects in the neutropenic model compared with the immunocompetent model. The difference between the two models is likely attributed to the enhanced tedizolid bacterial killing in the presence of granulocytes. However, this observed difference was only modest; the presence of neutropenia was associated with either no or only a 2- to 3-fold reduction in efficacy, which is comparable to the difference observed with ␤-lactams and quinolones (22, 23). Furthermore, a reduction in lung bacterial burden by up to 4 log10 CFU/lung within 24 h was achievable even in the presence of neutropenia with the dose range utilized. This indicated that the presence of granulocytes was not pertinent to the efficacy of tedizolid. This finding contradicts that previously reported by Drusano et al. (16). Using a murine thigh infection model, Drusano et al. compared the antistaphylococcal killing effect of tedizolid at doses equivalent to human exposures ranging from 200 to 3,200 mg/day in both immunocompetent and neutropenic mice. The authors observed a vast difference in tedizolid bacterial killing capability between the two mouse models; in the granulocytopenic mice, stasis was achieved at humanequivalent doses of slightly below 2,300 mg/day at 24 h, whereas a similar effect was achieved in immunocompetent animals at human-equivalent doses as low as 100 mg/day. Thus, the attenuation of efficacy was greater than 20-fold in the presence of neutropenia. The authors concluded that some tedizolid bacterial cell killing was due to the direct effect of the drug, but the majority of the drug killing capability was mediated through granulocytes. The discrepancy in the findings between our study and Drusano et al. could be attributed, at least in part, to the difference in the infection model utilized. Tedizolid shows substantial accumulation in the lung epithelium, as evidenced by its high ELF penetration ratio both in humans and in mice (13, 14), which could account for its enhanced activity against lung infections through the direct effect of the drug even if the host immune defenses were suppressed. We further compared the extent of bacterial killing achieved in the immunocompetent and neutropenic mice with doses that simulate the exposure achieved in humans following a clinical tedizolid dose of 200 mg daily. As previous studies by our group highlighted the relevance of antimicrobial efficacy and exposures at the site of infection (14), we examined the bronchopulmonary pharmacokinetics of tedizolid in infected immunocompetent and neutropenic mice to estimate the drug ELF exposures and determine a tedizolid dose in each mouse model that mimics the ELF exposure observed in humans (13). Although the plasma fAUC0 –24 achieved in immunocompetent and neutropenic mice was higher than that achieved in humans, the target site January 2017 Volume 61 Issue 1 e01957-16

aac.asm.org 6



exposures, ELF AUC0 –24, in the mice were comparable to the bronchopulmonary profile in humans. This is attributed to the enhanced penetration of tedizolid in human lungs compared with mice. This observation corroborates our previous conclusion on the importance of the assessment of drug exposure at the infection site when attempting to simulate human exposures in animal models. It is also worth mentioning that the tedizolid human-simulated exposures identified in this study were different in the two mouse models utilized and also different from that previously reported in the BALB/c mouse model of staphylococcal pneumonia (14). This alludes to the importance of the discrete assessment of the pharmacokinetics in each mouse model utilized, as the drug exposures at the site of infection may vary with the mouse strain as well as with the immune status (24). The tedizolid human-simulated ELF exposures were associated with considerable reduction in lung bacterial burden within 24 h for all of the tested isolates in both mouse models; these exposures were sufficient to achieve more than 3-log and 2-log reductions in lung bacterial burdens in the immunocompetent and neutropenic mice, respectively, in the majority of the isolates examined. Although the average reduction in bacterial burden was statistically significantly different between the two models in 3 out of 4 isolates, that difference is not deemed to be clinically significant. The extent of bacterial killing achieved with the human-simulated ELF exposures strongly suggests that human patients with S. pneumoniae lung infections who are treated with 200 mg/day of tedizolid should have a successful treatment outcome regardless of their immune status. A limitation of comparative immune function studies such as our current investigation is the fact that a higher bacterial inoculum generally is required to establish a viable infection in animals with fully functioning immune systems relative to their neutropenic counterparts (16). Given that the increase in lung bacterial burden in the untreated control mice at 24 h was consistent between the two models, the apparent absence of evidence of inoculum effect with the oxazolidinediones (11, 25, 26), and that enhanced kill was observed in the immunocompetent animals despite the higher starting inoculum, we believe that our comparative assessments are valid. While these initial starting inocula may have an influence on the overall magnitude of kill, our observation of enhanced tedizolid efficacy in immunocompetent hosts is consistent with previous in vivo studies with the compound (15). In summary, our findings suggest that tedizolid possesses potent in vivo efficacy against S. pneumoniae lung infections in both the immunocompetent and neutropenic hosts. The administration of human-simulated tedizolid ELF exposures resulted in a high degree of killing in both the immunocompetent and neutropenic lung infection models. While we are mindful of the importance of granulocytes in the defense against bacterial infections, the absence of granulocytes from the neutropenic host did not jeopardize the outcome of the treatment with tedizolid. These preclinical data utilizing achievable bronchopulmonary exposures supports the consideration of the currently utilized clinical dose of tedizolid for the treatment of life-threatening pneumonia due to S. pneumoniae in clinical trials. MATERIALS AND METHODS Antimicrobial test agents. For in vitro susceptibility testing, analytical-grade tedizolid phosphate (lot 33130169; Merck & Co., Inc., Kenilworth, NJ) was used. For the animal studies, tedizolid phosphate (Sivextro) for injection (lot number 14TE1A; Cubist Pharmaceuticals U.S., Lexington, MA) was used. Prior to each in vivo experiment, the drug was reconstituted with sterile water for injection and diluted to the desired concentrations with sterile saline. The reconstituted drug solutions were kept under refrigeration and used within 24 h. Isolates. Four S. pneumoniae isolates were used to assess the efficacy of tedizolid: penicillin-resistant ATCC 700905, penicillin-susceptible ATCC 6303 and SPN 95, and penicillin-intermediate SPN 102. Isolates were stored at ⫺80°C in double-strength skim milk (Remel, Lenexa, KS), subcultured twice onto Trypticase soy agar with 5% sheep blood (TSA II; Becton Dickinson and Co., Sparks, MD), and grown for 18 to 20 h at 37°C under 5% CO2 prior to use in the experiments. Susceptibility studies. The MICs of tedizolid were determined in triplicate for all test organisms using the broth microdilution methodology as outlined by the Clinical and Laboratory Standards Institute (CLSI) (27). The modal MIC was used to characterize the isolates for final pharmacodynamics analyses. January 2017 Volume 61 Issue 1 e01957-16

aac.asm.org 7



Animals. For the in vivo studies, specific-pathogen-free female CBA/J mice (7 to 9 weeks old) and female ICR mice (20 to 22 g) were purchased from Envigo RMS, Inc. (Indianapolis, IN). The mice received food and water ad libitum. A 48-h acclimation time was utilized before commencement of experimentation. All animals were cared for in accordance with the highest humane and ethical standards. The protocol was approved by the Institutional Animal Care and Use Committee of Hartford Hospital. Immunocompetent lung infection model. Given the low infectivity of S. pneumoniae in immunocompetent outbred mouse strains (such as ICR), inbred CBA/J mice were utilized throughout the immunocompetent lung infection experiments. A suspension of the S. pneumoniae isolate was freshly prepared from the second subculture of each organism that had been incubated for 18 to 20 h and diluted in 5% dextrose in normal saline to achieve an inoculum of 109 CFU/ml. For isolates ATCC 700905 and ATCC 6303, mice were inoculated with 0.05 ml of the bacterial suspension intranasally under isoflurane anesthesia as described previously (28), whereas for isolates SPN 95 and SPN 102 the inoculation volume was increased to 0.1 ml to achieve comparable baseline bacterial burden in the lungs. Neutropenic lung infection model. For neutropenic lung infection experiments, ICR mice were utilized. The mice were rendered transiently neutropenic by injecting cyclophosphamide (Baxter Healthcare Corp., Deerfield, IL) intraperitoneally at a dose of 250 mg/kg of body weight 4 days and 100 mg/kg 1 day prior to the inoculation. The mice were then inoculated as described above under the immunocompetent lung infection model. A bacterial suspension of 108 CFU/ml was used for the inoculation. For all of the isolates examined, an inoculation volume of 0.05 ml was utilized. Single-dose pharmacokinetic studies. Tedizolid pharmacokinetics were examined in infected CBA/J and ICR mice. Previously published dose range and dose fractionation studies for tedizolid in a neutropenic murine thigh model of S. aureus infection proved that the ratio of the area under the concentration-time curve to the MIC (AUC/MIC) was the pharmacodynamic index most closely linked to the bacterial cell killing rate (29). Therefore, the focus of the single-dose pharmacokinetic studies was to accurately estimate the area under the free-drug concentration-time curve over 24 h (fAUC0 –24) achieved in each mouse strain with the tedizolid doses utilized in the in vivo efficacy studies. Mice were infected as described above for the specific infection model. Four hours after S. pneumoniae bacterial inoculation, groups of 24 mice were administered a single dose of tedizolid intraperitoneally. At 4 time points ranging from 1 to 24 h after dosing, groups of 6 mice were euthanized by CO2 exposure, followed by blood collection via intracardiac puncture and ultimately cervical dislocation. Blood samples were collected in K⫹ EDTA tubes. The blood was centrifuged at 10,000 rpm for 10 min, and then the plasma was stored at ⫺80°C until being analyzed for tedizolid content. A total of six pharmacokinetic studies were undertaken to adequately characterize the plasma profile of tedizolid over the range of doses utilized in the immunocompetent mouse lung infection model (dose range, 1 to 140 mg/kg every 24 h [q24h]). Likewise, eight pharmacokinetics studies were undertaken in the neutropenic mouse model (dose range, 10 to 140 mg/kg q24h). The total systemic exposure measured as the AUC for each of the administered doses was calculated using the trapezoidal rule. Since the protein binding value for tedizolid in mouse plasma was previously estimated as 85% (30), the free-drug concentration was calculated as 0.15⫻ the total drug concentration at each given time point. The fAUC0 –24 for each administered dose was calculated using the trapezoidal rule. Bronchopulmonary pharmacokinetics and human-simulated ELF exposures of tedizolid. For studies of bronchopulmonary pharmacokinetics and human-simulated ELF exposures of tedizolid, immunocompetent CBA/J and neutropenic ICR mice were infected as previously described. For each infection model, cohorts of 24 mice each received single doses of tedizolid intraperitoneally. At 4 different time points, groups of 6 mice each were euthanized by CO2 exposure followed by blood collection via intracardiac puncture. The blood was centrifuged and the plasma was processed as described above. Following blood collection, bronchoalveolar (BAL) fluid was collected from the mice at the same 4 time points as described previously (31); a catheter was inserted into the trachea of the mice, and lungs were lavaged with 4 aliquots of 0.4 ml of normal saline. The BAL fluid was stored at ⫺80°C. The plasma and BAL fluid samples were assayed for tedizolid concentrations. Additionally, portions of the BAL fluid and the plasma were retained for urea determinations. The concentration of tedizolid in the BAL fluid combined with the urea concentration in the BAL fluid and the plasma were used to determine the ELF concentration for each mouse at each of the four time points according to the following formula: TZDELF ⫽ TZDBAL ⫻ (Ureaplasma/UreaBAL), where TZDBAL is the measured concentration of tedizolid in the BAL fluid sample and Ureaplasma and UreaBAL are the concentrations of urea in paired plasma and BAL fluid samples from each mouse, respectively. The resultant concentration of tedizolid in ELF was assumed to be free drug. For instances in which sample volume was insufficient to allow for urea analysis, the mean urea concentration in BAL fluid from other mice within that time point was utilized for the calculation of the tedizolid ELF concentrations. Once concentration data were compiled, the total ELF exposure for each dose (estimated as the area under the ELF concentration-time curve [ELF AUC]) was calculated using the trapezoidal rule. The tedizolid ELF AUC data were utilized to simulate a dosing regimen that approximates the target ELF profile observed in humans (109.30 mg · h/liter) following the administration of a 200-mg QD clinical dose (13) in each infection model. Once a regimen was mathematically identified, confirmatory bronchopulmonary studies were undertaken by following the same methodology as outlined above to ensure the appropriate exposure was achieved. In addition, the penetration ratio of tedizolid into the ELF was calculated by dividing the observed tedizolid ELF AUC0 –24 by the observed fAUC0 –24 in plasma for each infection model. Tedizolid concentration determination. Plasma and BAL samples were assayed for tedizolid concentrations at the Center for Anti-Infective Research and Development Laboratory at Hartford January 2017 Volume 61 Issue 1 e01957-16

aac.asm.org 8



Hospital using a validated high-performance liquid chromatography assay (32). The assay was linear, with a coefficient of determination (R2) of 0.999 for both plasma and BAL fluid tedizolid concentrations. The percent coefficient of variation of the quality control samples was less than 5% for intraday and interday assays. The lower limit of detection of the assay was 0.2 mg/liter in both plasma and BAL fluid. Urea concentration analysis. Aliquots of plasma and BAL samples were used for the determination of urea to derive drug levels in the ELF as described previously (13). Urea concentrations in BAL fluid and plasma were analyzed by colorimetric enzymatic assay (Teco Diagnostics, Anaheim, CA) via a spectrophotometer detection method (Cary 50 series; Varian, Walnut Creek, CA) at CAIRD. The urea assay was linear, with an R2 of ⱖ0.999 for both plasma and BAL fluid urea concentrations over the range of 0.1 to 2.0 mg/dl. Quality control samples of 0.15 and 1.5 mg/dl had intraday variability with coefficient of variation percentages of 6.52 and 1.36%, respectively, and interday variability with coefficient of variation percentages of 5.9 and 4.3%, respectively. In vivo efficacy studies. For these studies, cohorts of 60 immunocompetent CBA/J or neutropenic ICR mice were inoculated with one of the S. pneumoniae isolates utilized, as described above. Eight groups of six mice each received escalating once-daily doses of tedizolid intraperitoneally, including the dose required to achieve the human-simulated ELF exposure in each mouse infection model. Treatment was initiated 4 h postinoculation (designated 0 h). Untreated control mice (6 mice per group) were sacrificed 4 h postinoculation to serve as the 0-h control animals representing initial bacterial burden. Another untreated control group (6 mice per group) received 0.2 ml sterile normal saline intraperitoneally and was sacrificed 24 h later to serve as the 24-h control group. Treatment mice were also sacrificed at the end of the 24-h period. For harvesting, all animals were euthanized by CO2 exposure followed by cervical dislocation. Subsequently, all of the lobes of the lung were removed and homogenized in normal saline. Serial dilutions of the lung homogenates were plated on BD Columbia CNA agar with 5% sheep blood (Becton Dickinson and Co., Sparks, MD) for CFU determination. Efficacy was calculated as the change in bacterial burden obtained in treated mice after 24 h compared with that in the 0-h control animals. The change in bacterial density in tissues, expressed as change in log10 CFU at 24 h, for both treated and untreated animals was estimated and the plot of change in log10 CFU versus fAUC0 –24/MIC was constructed for each isolate in each infection model using the bioactive, free-drug exposures as determined using values obtained from the protein binding and pharmacokinetic studies. Data were fitted to a sigmoidal inhibitory Emax model using WinNonlin (version 5.0.1; Pharsight Corp., Mountain View, CA), and the effective index required to achieve net stasis and a 1- or 2-log bacterial cell kill for each isolate was estimated in each infection model utilized. For statistical analyses, the change in log10 CFU at 24 h was compared between the two infection models using Student’s t test. A P value of ⱕ0.05 was considered statistically significant.

ACKNOWLEDGMENTS This study was sponsored by a grant from Merck & Co., Inc., Kenilworth, New Jersey. We acknowledge the superior assistance of Deborah Santini, Christina Sutherland, Jennifer Tabor-Rennie, Elizabeth Cyr, Sara Robinson, Kimelyn Greenwood, Mordechai Grupper, Islam Ghazi, Marguerite Monogue, and Abrar Thabit in the performance of this study.

REFERENCES 1. Musher DM. 1992. Infections caused by Streptococcus pneumoniae: clinical spectrum, pathogenesis, immunity, and treatment. Clin Infect Dis 14:801– 807. https://doi.org/10.1093/clinids/14.4.801. 2. CDC. 2013. Antibiotic resistance threats in the United States. Centers for Disease Control and Prevention, U.S. Department of Health and Human Services, Atlanta, GA. 3. Jenkins SG, Brown SD, Farrell DJ. 2008. Trends in antibacterial resistance among Streptococcus pneumoniae isolated in the USA: update from PROTEKT US years 1-4. Ann Clin Microbiol Antimicrob 7:1. https:// doi.org/10.1186/1476-0711-7-1. 4. Van Bambeke F, Reinert RR, Appelbaum PC, Tulkens PM, Peetermans WE. 2007. Multidrug-resistant Streptococcus pneumoniae infections: current and future therapeutic options. Drugs 67:2355–2382. https://doi.org/ 10.2165/00003495-200767160-00005. 5. Novak R, Henriques B, Charpentier E, Normark S, Tuomanen E. 1999. Emergence of vancomycin tolerance in Streptococcus pneumoniae. Nature 399:590 –593. https://doi.org/10.1038/21202. 6. McCullers JA, English BK, Novak R. 2000. Isolation and characterization of vancomycin-tolerant Streptococcus pneumoniae from the cerebrospinal fluid of a patient who developed recrudescent meningitis. J Infect Dis 181:369 –373. https://doi.org/10.1086/315216. 7. Gillis LM, White HD, Whitehurst A, Sullivan DC. 2005. Vancomycintolerance among clinical isolates of Streptococcus pneumoniae in Mississippi during 1999-2001. Am J Med Sci 330:65– 68. https://doi.org/ 10.1097/00000441-200508000-00003. January 2017 Volume 61 Issue 1 e01957-16

8. Wolter N, Smith AM, Farrell DJ, Schaffner W, Moore M, Whitney CG, Jorgensen JH, Klugman KP. 2005. Novel mechanism of resistance to oxazolidinones, macrolides, and chloramphenicol in ribosomal protein L4 of the pneumococcus. Antimicrob Agents Chemother 49:3554 –3557. https://doi.org/10.1128/AAC.49.8.3554-3557.2005. 9. Mutnick AH, Enne V, Jones RN. 2003. Linezolid resistance since 2001: SENTRY Antimicrobial Surveillance Program. Ann Pharmacother 37: 769 –774. https://doi.org/10.1345/aph.1C437. 10. Decousser JW, Desroches M, Bourgeois-Nicolaos N, Potier J, Jehl F, Lina G, Cattoir V, Vandenesh F, Doucet-Populaire F, Microbs Study Group. 2015. Susceptibility trends including emergence of linezolid resistance among coagulase-negative staphylococci and meticillin-resistant Staphylococcus aureus from invasive infections. Int J Antimicrob Agents 46: 622– 630. https://doi.org/10.1016/j.ijantimicag.2015.07.022. 11. Locke JB, Zurenko GE, Shaw KJ, Bartizal K. 2014. Tedizolid for the management of human infections: in vitro characteristics. Clin Infect Dis 58(Suppl 1):S35–S42. https://doi.org/10.1093/cid/cit616. 12. Shorr AF, Lodise TP, Corey GR, De Anda C, Fang E, Das AF, Prokocimer P. 2015. Analysis of the phase 3 ESTABLISH trials of tedizolid versus linezolid in acute bacterial skin and skin structure infections. Antimicrob Agents Chemother 59:864 – 871. https://doi.org/10.1128/AAC .03688-14. 13. Housman ST, Pope JS, Russomanno J, Salerno E, Shore E, Kuti JL, Nicolau DP. 2012. Pulmonary disposition of tedizolid following administration of once-daily oral 200-milligram tedizolid phosphate in healthy adult volaac.asm.org 9


14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

unteers. Antimicrob Agents Chemother 56:2627–2634. https://doi.org/ 10.1128/AAC.05354-11. Tessier PR, Keel RA, Hagihara M, Crandon JL, Nicolau DP. 2012. Comparative in vivo efficacies of epithelial lining fluid exposures of tedizolid, linezolid, and vancomycin for methicillin-resistant Staphylococcus aureus in a mouse pneumonia model. Antimicrob Agents Chemother 56:2342–2346. https://doi.org/10.1128/AAC.06427-11. Lepak AJ, Marchillo K, Pichereau S, Craig WA, Andes DR. 2012. Comparative pharmacodynamics of the new oxazolidinone tedizolid phosphate and linezolid in a neutropenic murine Staphylococcus aureus pneumonia model. Antimicrob Agents Chemother 56:5916 –5922. https:// doi.org/10.1128/AAC.01303-12. Drusano GL, Liu W, Kulawy R, Louie A. 2011. Impact of granulocytes on the antimicrobial effect of tedizolid in a mouse thigh infection model. Antimicrob Agents Chemother 55:5300 –5305. https://doi.org/10.1128/ AAC.00502-11. Abdelraouf K, Nicolau DP. 2016. Human-simulated exposure and pharmacodynamic (PD) assessment of tedizolid (TZD) against Streptococcus pneumoniae in an immunocompetent murine lung infection model, presentation 1946. Abstr IDWeek 2016. IDWeek, New Orleans, LA. Abdelraouf K, Nicolau DP. 2016. Comparative in vivo efficacy of humansimulated exposures of tedizolid (TZD) in neutropenic (I⫺) versus immunocompetent (I⫹) murine Streptococcus pneumoniae lung infection model, presentation 1961. Abstr IDWeek 2016. IDWeek, New Orleans, LA. Zurenko G, Bien P, Bensaci M, Patel HN, Thorne G. 2014. Use of linezolid susceptibility test results as a surrogate for the susceptibility of Grampositive pathogens to tedizolid, a novel oxazolidinone. Ann Clin Microbiol Antimicrob 13:46. https://doi.org/10.1186/s12941-014-0046-0. Schaadt R, Sweeney D, Shinabarger D, Zurenko G. 2009. In vitro activity of TR-700, the active ingredient of the antibacterial prodrug TR-701, a novel oxazolidinone antibacterial agent. Antimicrob Agents Chemother 53:3236 –3239. https://doi.org/10.1128/AAC.00228-09. Brown SD, Traczewski MM. 2010. Comparative in vitro antimicrobial activities of torezolid (TR-700), the active moiety of a new oxazolidinone, torezolid phosphate (TR-701), determination of tentative disk diffusion interpretive criteria, and quality control ranges. Antimicrob Agents Chemother 54:2063–2069. https://doi.org/10.1128/AAC.01569-09. Andes D, Craig WA. 2003. Pharmacodynamics of the new des-f(6)quinolone garenoxacin in a murine thigh infection model. Antimicrob Agents Chemother 47:3935–3941. https://doi.org/10.1128/AAC.47 .12.3935-3941.2003. Craig WA, Andes DR. 2008. In vivo pharmacodynamics of ceftobiprole



24.

25.

26.

27.

28.

29.

30.

31.

32.

against multiple bacterial pathogens in murine thigh and lung infection models. Antimicrob Agents Chemother 52:3492–3496. https://doi.org/ 10.1128/AAC.01273-07. Keel RA, Crandon JL, Nicolau DP. 2012. Pharmacokinetics and pulmonary disposition of tedizolid and linezolid in a murine pneumonia model under variable conditions. Antimicrob Agents Chemother 56: 3420 –3422. https://doi.org/10.1128/AAC.06121-11. Bostic GD, Perri MB, Thal LA, Zervos MJ. 1998. Comparative in vitro and bactericidal activity of oxazolidinone antibiotics against multidrugresistant enterococci. Diagn Microbiol Infect Dis 30:109 –112. https:// doi.org/10.1016/S0732-8893(97)00210-1. Lee DG, Murakami Y, Andes DR, Craig WA. 2013. Inoculum effects of ceftobiprole, daptomycin, linezolid, and vancomycin with Staphylococcus aureus and Streptococcus pneumoniae at inocula of 10(5) and 10(7) CFU injected into opposite thighs of neutropenic mice. Antimicrob Agents Chemother 57:1434 –1441. https://doi.org/10.1128/ AAC.00362-12. CLSI. 2016. Performance standards for antimicrobial susceptibility testing. 26th ed CLSI supplement M100S. Clinical and Laboratory Standards Institute, Wayne, PA. Kerr AR, Paterson GK, Riboldi-Tunnicliffe A, Mitchell TJ. 2005. Innate immune defense against pneumococcal pneumonia requires pulmonary complement component C3. Infect Immun 73:4245– 4252. https:// doi.org/10.1128/IAI.73.7.4245-4252.2005. Louie A, Liu W, Kulawy R, Drusano GL. 2011. In vivo pharmacodynamics of torezolid phosphate (TR-701), a new oxazolidinone antibiotic, against methicillin-susceptible and methicillin-resistant Staphylococcus aureus strains in a mouse thigh infection model. Antimicrob Agents Chemother 55:3453–3460. https://doi.org/10.1128/AAC.01565-10. Keel RA, Tessier PR, Crandon JL, Nicolau DP. 2012. Comparative efficacies of human simulated exposures of tedizolid and linezolid against Staphylococcus aureus in the murine thigh infection model. Antimicrob Agents Chemother 56:4403– 4407. https://doi.org/ 10.1128/AAC.00122-12. Laohavaleeson S, Tessier PR, Nicolau DP. 2008. Pharmacodynamic characterization of ceftobiprole in experimental pneumonia caused by phenotypically diverse Staphylococcus aureus strains. Antimicrob Agents Chemother 52:2389 –2394. https://doi.org/10.1128/AAC.01422-07. Santini DA, Sutherland CA, Nicolau DP. 2015. Development of a high performance liquid chromatography method for the determination of tedizolid in human plasma, human serum, saline and mouse plasma. J Chromatogr Sep Tech 6:270.

aac.asm.org 10