Addition of PNU-100480 to First-Line Drugs Shortensthe Time Needed to Cure Murine Tuberculosis

Kathy N. Williams1, Steven J. Brickner2*, Charles K. Stover3†, Tong Zhu2, Adam Ogden2, Rokeya Tasneen1,Sandeep Tyagi1, Jacques H. Grosset1, and Eric L. Nuermberger1

1Center for Tuberculosis Research, Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland;2Pfizer Inc., Groton, Connecticut; and 3Pfizer Inc., Kalamazoo, Michigan

Rationale: We recently reported strong bactericidal activity of theoxazolidinone PNU-100480 and its ability to increase the initialbactericidal effect of various combinations of first-line tuberculosisdrugs and moxifloxacin in a murine model.Objectives: To investigate whether the addition ofPNU-100480to thestandard first-line regimen of rifampin, isoniazid, and pyrazina-mide could shorten the duration of treatment necessary to preventrelapse after treatment discontinuation.Methods: Following aerosol infection with Mycobacterium tubercu-losis H37Rv and a 13-day incubation period, control mice weretreated with the first-line regimen while test mice received the sameregimen with PNU-100480 or linezolid added for the first 2 or4 months. Efficacy was assessed on the basis of quantitative culturesof lung homogenates performed monthly during treatment and3 months after completion of 3, 4, 5, or 6 months of treatment todetermine the relapse rate.Measurements and Main Results: After 2 months of treatment, micereceiving PNU-100480 in addition to the first-line regimen had lungCFU counts two orders of magnitude lower than control micereceiving the first-line regimen alone. Relapse rates after 4 monthsof treatment were 90, 35, and 5% when PNU-100480 was added tothe first-line regimen for 0, 2, and 4 months, respectively. When thetotal treatment duration was 3 months, relapse rates were 85 and35 to 45% when mice received PNU-100480 for 2 and 3 months,respectively; all control mice remained culture positive at the timeof treatment completion with 17 to 72 CFU per lung. Addition oflinezolid to the first-line regimen had an antagonistic effect result-ing in higher CFU counts and failure to render mice culture-negativein 4 months of treatment.Conclusions: Together with previous findings, these results confirmthat PNU-100480, which is now in Phase I clinical testing, hassterilizing activity in the murine model and suggest that it may becapable of shortening treatment duration for drug-susceptible aswell as drug-resistant tuberculosis in humans.

Keywords: tuberculosis treatment; oxazolidinone; pharmacokinetics;

linezolid; antagonism

Bactericidal activity against actively multiplying bacilli is a de-sirable characteristic of an antituberculosis drug. A rapidreduction in the bacterial burden early in the course oftreatment speeds clinical improvement, reduces transmissibil-ity, and prevents the selection of drug-resistant mutants.However, the ultimate duration of tuberculosis (TB) treatment

is determined by the rate at which a small subpopulation ofslow- or non-multiplying bacteria, also known as ‘‘persisters,’’ iseliminated. Though lacking genotypic resistance, these drug-tolerant persisters remain viable for months in the face ofintensive combination chemotherapy. The ability of some drugsto eradicate such persisters and prevent relapse after treatmentcompletion is commonly referred to as ‘‘sterilizing activity.’’ Inparticular, it is the collective sterilizing activity of rifampin(RIF) and pyrazinamide (PZA) that made the modern 6-month, short-course regimen possible (1, 2). Without thesedrugs, as is the case for many patients with multidrug-resistant(MDR) TB, treatment is 18 to 24 months in duration and is stillassociated with lower cure rates than drug-susceptible TB (3–5).

The development of new drugs with sterilizing activitycapable of further shortening the treatment of drug-susceptibleTB and/or MDR-TB is the major objective of current drugdevelopment efforts (6). Fortunately, three new drugs from twonew classes, each of which has shown evidence of sterilizingactivity in the mouse model, are under evaluation in Phase IIclinical trials for treatment of MDR-TB (7–10). But historysuggests some or all will fail to reach the market (11).Additional new-drug candidates are needed to bolster the TBdrug pipeline.

Linezolid (LZD), the only marketed oxazolidinone antibi-otic, has activity against Mycobacterium tuberculosis in vitro,including MDR- and extensively drug-resistant (XDR)-TBstrains (12–14). Although it is not approved for use in TB, ithas been used with some success in difficult-to-treat casesresistant to most other drugs (15–20). Still, it has largelybacteriostatic activity in the mouse model and in patients withsmear-positive pulmonary TB (13, 14, 21, 22). We recentlyfound that another oxazolidinone, PNU-100480 (PNU) (23), hasmuch more potent bactericidal activity than LZD in mice (22).

AT A GLANCE COMMENTARY

Scientific Knowledge on the Subject

Recent findings show that PNU-100480, an oxazolidinoneantibiotic now in Phase I trials for tuberculosis, has strongbactericidal activity and is capable of increasing the bacte-ricidal activity of first-line drugs in a murine model.

What This Study Adds to the Field

This study demonstrates that PNU-100480 has activityagainst persistent tubercle bacilli in mice such that theaddition of PNU-100480 to the standard first-line tubercu-losis regimen shortens the duration of treatment necessaryfor cure. These results suggest PNU-100480 may have thepotential to improve the treatment of both drug-susceptibleand drug-resistant tuberculosis.

(Received in original form April 26, 2009; accepted in final form June 9, 2009)

Supported by Pfizer.

* Present address: S.J. Brickner Consulting, LLC, 9 Fargo Drive, Ledyard, CT 06339.

† Present address: MedImmune, LLC, Gaithersburg, MD 20878.

Correspondence and requests for reprints should be addressed to Eric L.

Nuermberger, M.D., 1550 Orleans Street, Baltimore, MD 21231. E-mail:

enuermb@jhmi.edu

Am J Respir Crit Care Med Vol 180. pp 371–376, 2009

Originally Published in Press as DOI: 10.1164/rccm.200904-0611OC on June 11, 2009

Internet address: www.atsjournals.org

Moreover, PNU significantly improves the initial bactericidalactivity of several combinations of existing first-line drugs andmoxifloxacin. Based on these promising results, we sought todetermine whether PNU has sufficient sterilizing activity toshorten the duration of therapy needed to prevent relapse inmurine TB when added to the standard first-line regimen ofRIF-isoniazid (INH)-PZA.

Some of the results of these studies have been previouslyreported in the form of an abstract (24).

METHODS

Bacterial Strain

Mycobacterium tuberculosis H37Rv was passaged in mice, frozen in1-ml aliquots, and stored at 2808C before use. At the time of infection,an aliquot was thawed and subcultured in Middlebrook 7H9 brothsupplemented with 10% oleic acid-albumin-dextrose-catalase complex(Difco, Detroit, MI) and 0.05% Tween 80 (Sigma, St. Louis MO).

Antimicrobials

PNU and LZD were provided by Pfizer (Groton, CT). For adminis-tration to mice, both drugs were suspended in a vehicle composed of5% polyethylene glycol-200 (Sigma) and 95% methylcellulose (0.5%)(Fisher, Suwanee, GA) in distilled water. PZA was purchased fromFisher. INH and RIF were purchased from Sigma. Stock solutions wereprepared weekly using distilled water and stored at 48C, as previouslydescribed (25).

Aerosol Infection

Female BALB/c mice (Charles River, Wilmington, MA), aged 5 to6 weeks, were infected by the aerosol route using the InhalationExposure System (Glas-col, Terre Haute, IN) and a log phase brothculture with an optical density of approximately 1.0 at 600 nm. Micewere infected in one of four consecutive aerosol runs and then blockrandomized by run into treatment groups. Untreated mice were killed(1) on the day after infection to determine the number of (CFU)implanted in the lungs and (2) on the day of treatment initiation(D0), 13 days after infection, to determine the baseline CFU count.Quantitative cultures of lung homogenates were performed on selec-tive 7H11 agar plates (Becton-Dickinson, Sparks, MD) as previouslydescribed (25). All procedures involving animals were approved by theJohns Hopkins University Animal Care and Use Committee.

Drug Treatment

All antibiotics were administered once daily, five days per week, in0.2 ml by oral gavage. Both oxazolidinone suspensions were sonicatedbriefly before use and shaken between doses. Rifampin was given1 hour before administration of other drugs to avoid an adversepharmacokinetic interaction (26–28).

The scheme of the experiment is presented in Table 1. Negativecontrols went untreated (group 1). Positive controls (group 2) received

the standard 6-month short-course regimen of RIF-INH-PZA for2 months (initial phase) followed by RIF-INH for 4 months (contin-uation phase). Test mice received the same 2-month initial phaseregimen with the addition of either PNU (groups 3, 4, and 5) or LZD(groups 6 and 7). Mice that received RIF-INH-PZA-PNU during theinitial phase received only 2 months of treatment with RIF-INH-PNU(group 3), RIF-PNU (group 4) or RIF-INH (group 5) during thecontinuation phase. Mice that received RIF-INH-PZA-LZD duringthe initial phase received only 2 months of treatment with RIF-INH-LZD (group 6) or RIF-INH (group 7) during the continuation phase.The doses of RIF (10 mg/kg), INH (25 mg/kg) and PZA (150 mg/kg)were described previously (25). LZD and PNU were compared at dosesselected to provide equivalent area under the serum concentration-time curve (AUC) values in mice. The dose of LZD (65 mg/kg) waschosen, based on projections from preexisting pharmacokinetic (PK)data in uninfected mice (Pfizer, data on file), to produce an AUC inmice matching the AUC observed in humans after the standard 600-mgdose administered once daily (91–122 mg-h/ml) (29). However, PKstudy in mice infected with M. tuberculosis later suggested this doseproduces an AUC (190 mg-h/ml) (22) closer to the AUC values ob-served when 600 mg is administered twice daily to healthy volunteers(215–294 mg-h/ml) (30, 31) and patients with TB (233 mg-h/ml) (21).The PNU dose (160 mg/kg) was chosen to produce an AUC similar tothat observed with LZD at 65 mg/kg in mice (22).

Assessment of Treatment Efficacy

Five mice per group were killed monthly, beginning after 2 months oftreatment, for assessment of lung CFU counts (Table 1). Another 20mice per group were held for 3 additional months beyond completionof 3, 4, 5, or 6 months of treatment before being killed to determine theproportion with culture-positive relapse, as defined by isolation of 1 orgreater CFU after plating the entire lung homogenate.

Pharmacokinetics of First-Line Drugs Administered with

and without Linezolid

When an unexpected antagonistic effect of LZD on the activity of theRIF-INH-PZA combination became apparent at the 2-month time-point, the single dose serum PK profiles of RIF, INH and PZA weredetermined in uninfected 6-week-old female BALB/c mice afteradministration of the combination with or without LZD. Mice receivedone of the following three regimens: RIF-INH-PZA alone (in whichRIF was administered 1 hour before INH-PZA, as in the efficacyexperiment), RIF-INH-PZA-LZD (in which LZD was administeredimmediately after the INH-PZA component, as in the efficacy exper-iment), and LZD 1 RIF-INH-PZA (in which LZD was administered1 hour before the RIF component). Serum RIF concentrations wereassayed in samples obtained 1, 2, 4, 6, 8, and 16 hours after RIF dosing.INH and PZA concentrations were assayed in samples obtained at 0.5,1, 3, 5, and 7 hours after INH-PZA dosing. Three or four mice fromeach group were killed at each time point. Mice were anesthetized byisoflurane inhalation and exsanguinated by cardiac puncture. Serumwas harvested and frozen at 2808C. Concentrations of RIF, INH, andPZA were determined in the Infectious Diseases Pharmacokinetics

TABLE 1. SCHEME OF THE EXPERIMENT

No. of Mice for Lung CFU Counts (No. for Relapse Assessment) at the Indicated Time Point

Treatment groups* D–12 D0 Month 2 Month 3 Month 4 Month 5 Month 6

Controls

1) No Treatment 8 12 5

2) 2RHZ 1 4RH 5 5 5 (20) 5 (20) 5 (20)

Tests

3) 2RHZU 1 2RHU 5 5 (20) 5 (20)

4) 2RHZU 1 2RU 5 (20) 5 (20)

5) 2RHZU 1 2RH 5 (20) 5 (20)

6) 2RHZL 1 2RHL 5 5 (20) 5 (20)

7) 2RHZL 1 2RH 5 (20) 5 (20)

Definition of abbreviations: D–12 5 day after infection, 12 days before treatment initiation; D0 5 day of treatment initiation;

H 5 isoniazid; L 5 linezolid; R 5 rifampin; U 5 PNU-100480; Z 5 pyrazinamide.

* Drug doses: R, 10mg/kg; H, 25 mg/kg; Z, 150 mg/kg; U, 160 mg/kg; L, 65 mg/kg.

372 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 180 2009

Laboratory at the National Jewish Medical and Research Center(Dr. Charles Peloquin) using validated high-performance liquid chro-matography assays (25).

Data Analysis

Lung CFU counts (x) were log-transformed as log10(x 1 1) beforeanalysis. Group mean CFU counts after 2 months of treatment werecompared using one-way analysis of variance with Bonferroni’s post-test (GraphPad Prism v.4, GraphPad Software, San Diego, CA) toadjust for multiple comparisons, as appropriate. Relapse proportionswere compared using Fisher’s Exact Test (STATA 8.2, STATA Corp.,College Station, TX) with adjustment for multiple comparisons, asappropriate.

RESULTS

Change in CFU Counts during the Initial Phase

One day after aerosol infection, the mean lung CFU counts(6 SD) were 4.45 6 0.03, 4.42 6 0.02, 4.48 6 0.1, and 4.44 6

0.05 log10 (mean, 4.44 6 0.05 log10) in mice infected in aerosolruns 1, 2, 3, and 4, respectively. The mean lung CFU count onD0 was 7.92 6 0.15 log10. All untreated mice became moribundwithin 21 days of infection. Two months of treatment withRIF-INH-PZA reduced the mean lung CFU count to 3.17 6

0.27 log10, for a log kill of 4.75 compared with the baseline value(Figure 1). The addition of PNU resulted in an even greaterreduction in the mean lung CFU count to 0.71 6 0.26 log10, fora log kill of 7.21 log10 (P , 0.001). Surprisingly, mice treatedwith RIF-INH-PZA-LZD for 2 months had a mean lung CFUcount of 4.28 6 0.24, more than 1 log higher than mice treatedwith RIF-INH-PZA, indicating that LZD had a significantantagonistic effect (P , 0.001).

Change in CFU Counts during the Continuation Phase

After 3 months of treatment, all positive controls receiving RIF-INH remained culture-positive with a mean lung CFU countof 1.47 6 0.24 log10 (Figure 1). Among mice treated withRIF-INH-PZA-PNU in the initial phase, all five receiving RIF-INH-PNU for the third month were culture-negative, whereasonly three of five mice receiving RIF-PNU or RIF-INH wereculture-negative. The remaining two mice in each group hadonly 1 or 2 CFU isolated from the entire lung homogenate.Among mice treated with RIF-INH-PZA-LZD in the initialphase, mean lung CFU counts were 2.59 6 0.47 log10 and 2.53 6

0.23 log10 in those receiving RIF-INH-LZD and RIF-INH,respectively, for the third month.

After 4 months of treatment, all positive controls and allmice receiving PNU-containing regimens were culture-negativewith the exception of one of five mice in group 5 (receiving RIF-INH in the continuation phase) with 1 CFU isolated. All micereceiving LZD-containing regimens remained culture-positivewith mean lung CFU counts of 1.52 6 0.17 log10 and 0.74 6 0.53log10 in those receiving RIF-INH-LZD and RIF-INH, respec-tively. After 5 and 6 months of treatment, all positive controlswere culture-negative except one of five mice with 1 CFUisolated at the 5-month time-point.

Proportion of Mice Relapsing with Positive Cultures after

Completion of Treatment

Among positive control mice treated with 2 months of RIF-INH-PZA followed by RIF-INH, 90% relapsed if treatmentwas limited to 4 months; all were cured when treatment wasextended to 6 months (Table 2). The addition of PNU for thefirst 2 months only, as in group 5, reduced the relapse rate after4 months of treatment to 35%, indicating that PNU contributessterilizing activity during the initial phase of treatment. In-

corporation of PNU into both the initial and continuationphases of treatment, as in groups 3 and 4, contributed additionalsterilizing activity. Among mice receiving RIF-INH-PZA-PNUfor the first 2 months, the relapse rates were 45 and 35% inthose receiving RIF-INH-PNU and RIF-PNU, respectively,compared with 85% in those receiving RIF-INH for the thirdmonth. When treatment was extended to 4 months, 5% ofmice receiving PNU-containing continuation phase regimenshad positive cultures compared with 35% of mice receiving onlyRIF-INH during the continuation phase. Mice receiving LZD-containing regimens still had positive lung cultures when theycompleted 3 and 4 months of treatment. Only those completing4 months of treatment were assessed after a 3-month follow-upperiod. As expected, all 20 mice in each group had positive lungcultures.

Pharmacokinetics of First-Line Drugs Administered with

and without Linezolid

Due to the antagonism observed when LZD was added to RIF-INH-PZA, a single dose oral PK experiment was performed todetermine whether an adverse drug–drug interaction existsbetween LZD and one or more of the drugs in the RIF-INH-PZA combination. Administration of LZD just after the INH-PZA component, as per the efficacy experiment, reduced themaximum serum drug concentration (Cmax) and AUC of INHand PZA in a similarly significant way (Table 3, Figures 2A and2B), but neither the T1/2 of INH nor PZA was shortened upon

Figure 1. Change in lung CFU counts during treatment.

TABLE 2. RESULTS OF RELAPSE ASSESSMENTS*

Proportion (%) of Mice with Relapse after Treatment

Treatment group 3 mo 4 mo 5 mo 6 mo

2) 2RHZ 1 4RH n.d. 18 of 20 (90) 1 of 20 (5) 0 of 20 (0)

3) 2RHZU 1 2RHU 9 of 20 (45)† 1 of 20 (5)‡ n.d. n.d.

4) 2RHZU 1 2RU 7 of 20 (35)† 1 of 20 (5)‡ n.d. n.d.

5) 2RHZU 1 2RH 17 of 20 (85) 7 of 20 (35)‡ n.d. n.d.

6) 2RHZL 1 2RHL n.d. 20 of 20 (100) n.d. n.d.

7) 2RHZL 1 2RH n.d. 20 of 20 (100) n.d. n.d.

Definition of abbreviations: H 5 isoniazid; L 5 linezolid; n.d. 5 not done; R 5

rifampin; U 5 PNU-100480; Z 5 pyrazinamide.

* Relapse defined as positive lung culture 3 months after treatment cessation.† P 5 0.0187 for RHU vs. RH and P 5 0.0063 for RU vs. RH in continuation

phase.‡ P , 0.001 for RHU and RU vs. RHZ/RH; P 5 0.004 for RHZU/RH vs. RHZ/RH.

Williams, Brickner, Stover, et al.: Sterilizing Activity of PNU-100480 373

coadministration of LZD. Administration of LZD in this wayhad no effect on RIF concentrations (Figure 2C). However,when LZD was administered 1 hour before the RIF component,RIF concentrations were reduced but INH and PZA concen-trations were not significantly affected.

DISCUSSION

We recently discovered the strong bactericidal activity of PNUagainst established murine TB and found that adding PNU toRIF-INH-PZA significantly increases the activity of the first-line regimen (22). Here we extend those findings by demon-strating that PNU also lends additional sterilizing activity to thefirst-line regimen that is capable of shortening the duration oftreatment necessary for cure. This sterilizing effect of PNU isevident when use of PNU is limited to the initial phase oftreatment, but is even greater when PNU is given throughoutthe treatment, when it enables a 1- to 2-month reduction in thetreatment duration without sacrificing efficacy. PNU is moreeffective than INH in augmenting the sterilizing activity ofRIF during the continuation phase. In fact, the combination ofRIF-PNU is just as active as RIF-INH-PNU, indicating thatinclusion of INH is unnecessary when PNU is used during thisphase. Together these results strongly support the future clinicaldevelopment of PNU for treatment of drug-susceptible TB.Although combinations relevant for MDR-TB and XDR-TBwere not evaluated in this study, the ability of PNU to increasethe sterilizing activity of the first-line regimen suggests it mayalso be capable of shortening the treatment of MDR-TB andXDR-TB.

Unlike the addition of PNU, the addition of LZD didnot improve the activity of RIF-INH-PZA. On the contrary,a surprising antagonism was observed. In general, antagonismmay be caused by (1) a pharmacokinetic interaction, such asreduced absorption or enhanced clearance of one or more drugs,(2) by competition for drug uptake, activation or target-siteinteraction, or (3) by a mechanistic interaction in which the-antibacterial effect of one drug reduces the susceptibility of theorganism to another drug. The antagonism observed in thisexperiment can be explained, at least in part, by reduced serumconcentrations of INH and PZA when LZD is coadministeredwithout sufficient separation in time. This drug–drug interac-tion is due to reduced absorption, and not more rapid clear-

ance, of INH and PZA, as evidenced by the results of thesingle-dose PK study in which (1) INH and PZA exposureswere not significantly affected when LZD was administered2 hours before INH and PZA, but (2) INH and PZA Cmax andAUC, but not T1/2, were reduced when LZD was coadminis-tered. Assuming there is no change in volume of distribution ofthese drugs, the T1/2 is solely determined by the clearancevalues. RIF concentrations were not affected under the con-ditions used in the efficacy experiment, but an adverse in-teraction was observed in the PK study when LZD was given1 hour before RIF. In this case, the mean Cmax was not clearlyaffected but the AUC was reduced by 35% with previousadministration of LZD. Although this is only a single-dose PKstudy, induction of major isoforms of cytochrome P450 bylinezolid is generally not expected after repeated dosing inrodents (Pfizer, data on file). Because additional efficacy studieswith greater separation of dosing of LZD and companion drugswere not performed, we cannot exclude the possibility of

TABLE 3. SUMMARY OF SINGLE DOSE PK INUNINFECTED MICE

Cmax Tmax AUC Extrap T1/2

Drug Name 1Treatment mg/ml h mg*h/ml h

Isoniazid RHZ 19.4 6 2.0 0.5 29.4 6 8.0 1.6 6 0.2

RHZL 9.08 0.5 18.9 2.2

L1RHZ 15.6 6 0.5 0.5 22.5 6 2.3 2.4 6 0.5

Pyrazinamide RHZ 153 0.5 350 1.3

RHZL 90.8 0.5 299 1.6

L1RHZ 131 0.5 338 2.0 6 0.2

Rifampin RHZ 16.2 6 3.5 1-4 165 6 37.0 5.2 6 3.2

RHZL 16.9 6 2.5 1-2 141 6 15.9 4.5 6 1.4

L1RHZ 15.3 1 108 7.3

Data are reported as mean (n 5 2) or mean 6 SD when n . 2.1 Drug doses: rifampin (R), 10mg/kg; isoniazid (H), 25 mg/kg; pyrazinamide

(Z), 150 mg/kg; linezolid (L) 65 mg/kg, as in the efficacy experiment.

RHZ: R was administered 1 h before H and Z, as in the efficacy experiment.

RHZL: R was administered 1 h before H and Z, and L was administered

immediately after H and Z, as in the efficacy experiment.

L 1 RHZ: L was administered 1 h before R, and R was administered 1 h before

H and Z.

Figure 2. (A, B, and C) Serum pharmacokinetics of INH, PZA and RIF incombination with LZD.

374 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 180 2009

a second form of antagonism that is not related to absorption.One in vitro study described occasional synergism and noevidence of antagonism when LZD was combined with INHor RIF (32). We are aware of no other in vitro or in vivo studiesof LZD in combination with first-line TB drugs. Therefore,whether these results demonstrating antagonism have anysignificance for humans or are simply an artifact of drugadministration to mice is unknown. However, they do suggestcaution when evaluating LZD in combination with other drugsin animal models.

Although PNU was administered at a higher dose and in thesame formulation and time frame as LZD, no antagonisticeffect on CFU counts was observed. It is possible that similarreductions in the absorption of INH and PZA occurred withPNU coadministration, but the effect was obscured by thestrong bactericidal effect of PNU. If such drug–drug interac-tions did occur undetected, this would only serve to furtheremphasize the bactericidal and sterilizing properties of PNU.

There is a great need for new drugs with sterilizing activity toshorten the treatment duration for MDR- and XDR-TB andimprove treatment success rates. As PNU is now in phase Iclinical testing, it is encouraging that its sterilizing activity isstrong enough to improve the efficacy of the first-line regimen.Because PNU does not exhibit cross-resistance with existing TBdrugs (23), it may be capable of improving the treatment ofMDR- and XDR-TB in an even more dramatic fashion. Furtherexperiments are necessary to explore the treatment-shorteningpotential of PNU in combination with second-line TB drugsand, ideally, the new drugs in clinical development. Insofar assignificant bactericidal activity has been observed with PNUdoses as low was 50 mg/kg daily (13, 22), it will also beimportant to examine whether reductions in dosage will con-tinue to yield sterilizing activity while limiting the potential foradverse effects. Finally, given the increased sterilizing effectobserved when PNU was added to RIF-containing regimens, itwill be important to determine if the addition of PNU can alsoimprove upon experimental rifapentine-containing regimensthat are capable of curing mice in 3 months or less and arecurrently under evaluation in phase II clinical trials (33, 34).

Conflict of Interest Statement: K.N.W. received $1,001 to $5,000 in consultancyfees from Harper Labs. S.J.B. received more than $100,001 in salary as anemployee of Pfizer; received $5,001 to $10,000 for consulting on various aspectsusing one-atmosphere room temperature plasma generation for sterilization fromFlight Safety Technology Inc.; holds a patent for novel antibiotics (but receives nodirect financial benefit beyond salary) from Pfizer; has $5,001 to $10,000 companystock granted as restricted stock grant from Pfizer; received $10,001 to $50,000 forconsulting on TB drug development from Weill Medical College of CornellUniversity; and received $10,001 to $50,000 for consulting on TB drug develop-ment from the Global Alliance for TB; he was Upjohn Oxazolidinone Working Groupleader and Pfizer Research Project Leader for two teams that discovered anddeveloped PNU-100480 for TB, but received no financial compensation of any kindbeyond his salary for the work with this compound. As of March 31, 2009 S.J.B. is nolonger a Pfizer employee and started his own consulting business, which has theabove-stated two nonprofit clients with whom he is or will be working in the area ofTB drug discovery and development. C.K.S. received more than $100,001 as the Sr.Director Infectious Disease Research for MedImmune and has $1,001 to $5,000 inPfizer stock ownership or options; some mutual funds may own Pfizer stock. T.Z. isan employee of Pfizer, Inc., and has $10,001 to $50,000 from Pfizer, Inc. in stockownership or options. A.O. is a full-time employee of Pfizer, Inc. and has $10,001 to$50,000 in Pfizer stock, stock options, restricted stock units. R.T. does not havea financial relationship with a commercial entity that has an interest in the subject ofthis manuscript. S.T. does not have a financial relationship with a commercial entitythat has an interest in the subject of this manuscript. J.H.G. does not have a financialrelationship with a commercial entity that has an interest in the subject of thismanuscript. E.L.N. received $5,001 to $10,000 for promotional and nonpromo-tional lectures from Pfizer; $5,001 to $10,000 for promotional lectures from Wyeth;and $5,001 to $10,000 for promotional and nonpromotional lectures from Abbott;received more than $100,001 from Otsuka Pharmaceuticals in grants, and morethan $100,001 from Pfizer in grants; has a pending patent for combinationtreatment for tuberculosis from Pfizer; has $5,001 to $10,000 in stock ownershipor options from Wyeth; $1,001 to $5,000 in stock ownership or options fromViropharma; $1,001 to $5,000 in stock ownership or options from Mylan; and$1,001 to $5,000 in stock ownership or options from Amgen.

Acknowledgment: The authors gratefully acknowledge drugs provided by Pfizer(PNU-100480, linezolid).

References

1. Grosset J. Bacteriologic basis of short-course chemotherapy for tuber-

culosis. Clin Chest Med 1980;1:231–241.2. Mitchison DA. Role of individual drugs in the chemotherapy of

tuberculosis. Int J Tuberc Lung Dis 2000;4:796–806.3. Chan ED, Laurel V, Strand MJ, Chan JF, Huynh ML, Goble M, Iseman

MD. Treatment and outcome analysis of 205 patients with multidrug-resistant tuberculosis. Am J Respir Crit Care Med 2004;169:1103–1109.

4. Park SK, Kim CT, Song SD. Outcome of chemotherapy in 107 patients

with pulmonary tuberculosis resistant to isoniazid and rifampin. Int JTuberc Lung Dis 1998;2:877–884.

5. Yew WW, Chan CK, Chau CH, Tam CM, Leung CC, Wong PC, Lee J.

Outcomes of patients with multidrug-resistant pulmonary tuberculo-sis treated with ofloxacin/levofloxacin-containing regimens. Chest2000;117:744–751.

6. O’Brien RJ, Nunn PP. The need for new drugs against tuberculosis.

Obstacles, opportunities, and next steps. Am J Respir Crit Care Med2001;163:1055–1058.

7. Andries K, Verhasselt P, Guillemont J, Gohlmann HW, Neefs JM,

Winkler H, Van Gestel J, Timmerman P, Zhu M, Lee E, et al. Adiarylquinoline drug active on the ATP synthase of Mycobacteriumtuberculosis. Science 2005;307:223–227.

8. Matsumoto M, Hashizume H, Tomishige T, Kawasaki M, Tsubouchi H,

Sasaki H, Shimokawa Y, Komatsu M. OPC-67683, a nitro-dihydro-imidazooxazole derivative with promising action against tuberculosisin vitro and in mice. PLoS Med 2006;3:e466.

9. Nuermberger E, Tyagi S, Tasneen R, Williams KN, Almeida D,

Rosenthal I, Grosset JH. Powerful bactericidal and sterilizing activityof a regimen containing PA-824, moxifloxacin, and pyrazinamide ina murine model of tuberculosis. Antimicrob Agents Chemother 2008;52:1522–1524.

10. Stover CK, Warrener P, VanDevanter DR, Sherman DR, Arain TM,

Langhorne MH, Anderson SW, Towell JA, Yuan Y, McMurray DN,et al. A small-molecule nitroimidazopyran drug candidate for thetreatment of tuberculosis. Nature 2000;405:962–966.

11. Glickman SW, Rasiel EB, Hamilton CD, Kubataev A, Schulman KA.

Medicine. A portfolio model of drug development for tuberculosis.Science 2006;311:1246–1247.

12. Alcala L, Ruiz-Serrano MJ, Perez-Fernandez TC, Garcia DV, Diaz-

Infantes M, Marin-Arriaza M, Bouza E. In vitro activities of linezolidagainst clinical isolates of Mycobacterium tuberculosis that aresusceptible or resistant to first-line antituberculous drugs. AntimicrobAgents Chemother 2003;47:416–417.

13. Cynamon MH, Klemens SP, Sharpe CA, Chase S. Activities of several

novel oxazolidinones against Mycobacterium tuberculosis in a murinemodel. Antimicrob Agents Chemother 1999;43:1189–1191.

14. Fattorini L, Tan D, Iona E, Mattei M, Giannoni F, Brunori L, Recchia S,

Orefici G. Activities of moxifloxacin alone and in combination withother antimicrobial agents against multidrug-resistant Mycobacteriumtuberculosis infection in BALB/c mice. Antimicrob Agents Chemother2003;47:360–362.

15. Nam HS, Koh WJ, Kwon OJ, Cho SN, Shim TS. Daily half-dose

linezolid for the treatment of intractable multidrug-resistant tubercu-losis. Int J Antimicrob Agents 2009;33:92–93.

16. Condos R, Hadgiangelis N, Leibert E, Jacquette G, Harkin T, Rom WN.

Case series report of a linezolid-containing regimen for extensivelydrug-resistant tuberculosis. Chest 2008;134:187–192.

17. Fortun J, Martin-Davila P, Navas E, Perez-Elias MJ, Cobo J, Tato M,

De la Pedrosa EG, Gomez-Mampaso E, Moreno S. Linezolid for thetreatment of multidrug-resistant tuberculosis. J Antimicrob Chemo-ther 2005;56:180–185.

18. Park IN, Hong SB, Oh YM, Kim MN, Lim CM, Lee SD, Koh Y, Kim

WS, Kim DS, Kim WD, et al. Efficacy and tolerability of daily-halfdose linezolid in patients with intractable multidrug-resistant tuber-culosis. J Antimicrob Chemother 2006;58:701–704.

19. von der LB, Sandven P, Brubakk O. Efficacy and safety of linezolid in

multidrug resistant tuberculosis (MDR-TB): a report of ten cases.J Infect 2006;52:92–96.

20. Yew WW, Chau CH, Wen KH. Linezolid in the treatment of ‘difficult’

multidrug-resistant tuberculosis. Int J Tuberc Lung Dis 2008;12:345–346.

21. Dietze R, Hadad DJ, McGee B, Molino LP, Maciel EL, Peloquin CA,

Johnson DF, Debanne SM, Eisenach K, Boom WH, et al. Early and

Williams, Brickner, Stover, et al.: Sterilizing Activity of PNU-100480 375

extended early bactericidal activity of linezolid in pulmonary tuber-culosis. Am J Respir Crit Care Med 2008;178:1180–1185.

22. Williams KN, Stover CK, Zhu T, Tasneen R, Tyagi S, Grosset JH,Nuermberger E. Promising antituberculosis activity of the oxazolidi-none PNU-100480 relative to that of linezolid in a murine model.Antimicrob Agents Chemother 2009;53:1314–1319.

23. Barbachyn MR, Hutchinson DK, Brickner SJ, Cynamon MH,Kilburn JO, Klemens SP, Glickman SE, Grega KC, Hendges SK,Toops DS, et al. Identification of a novel oxazolidinone (U-100480)with potent antimycobacterial activity. J Med Chem 1996;39:680–685.

24. Williams K, Stover C, Brickner S, Zhu T, Tasneen R, Tyagi S, Grosset J,Nuermberger E. The oxazolidinone PNU-100480 has strong bacteri-cidal and sterilizing activity in the murine model of tuberculosis[abstract F1–1163a]. Final Program of the 48th Annual InterscienceConference on Antimicrobial Agents and Chemotherapy (ICAAC)and the Infectious Diseases Society of America (IDSA) 46th AnnualMeeting 2008:318.

25. Nuermberger E, Rosenthal I, Tyagi S, Williams KN, Almeida D,Peloquin CA, Bishai WR, Grosset JH. Combination chemotherapywith the nitroimidazopyran PA-824 and first-line drugs in a murinemodel of tuberculosis. Antimicrob Agents Chemother 2006;50:2621–2625.

26. Dhillon J, Dickinson JM, Sole K, Mitchison DA. Preventive chemo-therapy of tuberculosis in Cornell model mice with combinations ofrifampin, isoniazid, and pyrazinamide. Antimicrob Agents Chemother1996;40:552–555.

27. Dickinson J, Guy A, Mitchison DA. Bioavailability of rifampin inexperimental murine tuberculosis. Antimicrob Agents Chemother1992;36:2066–2067.

28. Grosset J, Truffot-Pernot C, Lacroix C, Ji B. Antagonism betweenisoniazid and the combination pyrazinamide-rifampin against tuber-culosis infection in mice. Antimicrob Agents Chemother 1992;36:548–551.

29. Stalker DJ, Jungbluth GL. Clinical pharmacokinetics of linezolid, a noveloxazolidinone antibacterial. Clin Pharmacokinet 2003;42:1129–1140.

30. Gee T, Ellis R, Marshall G, Andrews J, Ashby J, Wise R. Pharmaco-kinetics and tissue penetration of linezolid following multiple oraldoses. Antimicrob Agents Chemother 2001;45:1843–1846.

31. Stalker DJ, Jungbluth GL, Hopkins NK, Batts DH. Pharmacokineticsand tolerance of single- and multiple-dose oral or intravenous line-zolid, an oxazolidinone antibiotic, in healthy volunteers. J AntimicrobChemother 2003;51:1239–1246.

32. Rodriguez Diaz JC, Ruiz M, Lopez M, Royo G. Synergic activity offluoroquinolones and linezolid against Mycobacterium tuberculosis.Int J Antimicrob Agents 2003;21:354–356.

33. Rosenthal IM, Zhang M, Almeida D, Grosset JH, Nuermberger EL.Isoniazid or moxifloxacin in rifapentine-based regimens for experi-mental tuberculosis? Am J Respir Crit Care Med 2008;178:989–993.

34. Rosenthal IM, Zhang M, Williams KN, Peloquin CA, Tyagi S, VernonAA, Bishai WR, Chaisson RE, Grosset JH, Nuermberger EL. Dailydosing of rifapentine cures tuberculosis in three months or less in themurine model. PLoS Med 2007;4:e344.

376 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 180 2009