A Pharmacokinetic Evaluationof Concomitant Administrationof Linezolid and Aztreonam

Theresa Lasher Sisson, BS, Gail L. Jungbluth, PhD,and Nancy K Hopkins, BS

Linezolid, a new oxazolidinone antimicrobial agent, has aspectrum of activity encom7passing a wide variety of Gram-positive bacteria. Thepurpose ofthis study was to evaluate thepharmacokinetics oflinezolid and aztreonam, an antimicro-bial agent with selective activity against Gram-negative bacte-ria, when given alone and in combination. Healthy subjectswere randomized to receive single, 30-minute intravenousinfusions of (1) linezolid 375 mg, (2) aztreonam 1000 mg,and (3) linezolid 375 mg plus aztreonam 1000 mg in anopen-label, crossover manner. The only statistically signifi-cant differences observed with combination treatment rela-tive to each drug alone were an increase in the maximum

The oxazolidinones are a novel synthetic class ofantibacterial agents that selectively inhibit bacte-

rial protein synthesis. Mechanism of action studiessuggest that they inhibit the initiation phase of transla-tion1 most likely by blocking formation of the 30Ssubunit initiation complex.4 The exact mechanism-ofaction of the oxazolidinones appears to be unique, asno cross resistance has been found in bacterial strainsresistant to other protein synthesis inhibitors or toother antimicrobial agents, including multi-drug-resistant strains.5` In general, oxazolidinones appearto be bacteriostatic; however, bactericidal activity hasbeen observed against Streptococcus pneumoniae.4

Linezolid, the lead candidate of a series of oxa-zolidinone compounds, is chemically identified as (S)-N-{ { 3-{ 3-Fluoro-4-(4-morpholinyl)phenyl}-2-oxo-5-oxazolidinyl}methyl}-acetamide (Figure 1). Linezoliddemonstrates a significant level of in vitro activity

From Pharmacia & Upjohn, Kalamazoo, Ml. Submitted for publicationDecember 30, 1 998; revised version accepted August 1 3, 1 999. Addressfor reprints: Theresa Lasher Sisson, BS, Pharmacia & Upjohn, ClinicalPharmacokinetics, 301 Henrietta St., Kalamazoo, Ml 49007.

plasma concentration of linezolid (-18%) and an approxi-mate 7% decrease in the apparent elimination rate of aztreo-nam, neither ofwhich are expected to be clinically significant.In healthy subjects, the combination oflinezolid and aztreo-nam was safe and well tolerated compared with each agentused alone. Pharmacokinetic data demonstrate that coad-ministration of linezolid and aztreonam does not alter thedisposition of either agent under single-dose conditions.Therefore, it is not expected that a dose alteration of eitheragent will be necessary in a clinical setting.

Journal of Clinical Pharmacology, 1999;39:1277-1282©D1999 the American College of Clinical Pharmacology

Figure 1. Linezolid chemical structure.

against staphylococci (including methicillin-resistantStaphylococcus aureus and methicillin-resistantStaphylococcus epidermidis), enterococci (includingvancomycin-resistant strains), and pneumococci(including penicillin-intermediate and penicillin-resistant strains).9'0 It is also effective in vitro againstCorynebacterium spp.,9'0 Bacillus spp.,910 Listeria mono-cytogenes,9 Staphylococcus spp.,9',0 Streptococcus spp.9 10

Moraxella catarrhalis,9"0 and mycobacteria.Clinical data show that linezolid is well tolerated in

single intravenous doses of 625 mg given twice daily

J Clin Pharmacol 1999;39:1277-1282 1 277

SISSON ETAL

for up to 7 days. Phase III clinical efficacy studies areevaluating linezolid doses of 400 mg and 600 mg giventwice daily for up to 28 days. The compound has anelimination half-life of 5 to 7 hours."-3 Approximately35% of an administered dose of linezolid appears inthe urine as the parent compound, and approximately50% appears as two major inactive metabolites.i'4Nonrenal clearance of linezolid is mediated by nonen-zymatic chemical oxidation, and approximately 31%of the administered drug is bound to proteins.

In a clinical setting, it is likely that an antimicrobialagent with Gram-negative activity will be sought foruse as a companion agent to linezolid. This would beuseful in the treatment of mixed (Gram-positive andGram-negative) infections and for situations in whichantimicrobial therapy must begin before pathogen sus-ceptibility information becomes available.

Aztreonam (Azactam®, Bristol-Myers Squibb Com-pany) is a currently marketed monolactam antibacte-rial agent with selective activity against Gram-negativeaerobic bacteria (e.g., Escherichia coli, Enterobacterspp., Pseudomonas aeruginosa). Approximately 60%to 80% of an administered dose of aztreonam isexcreted by the kidney, mainly by glomerular filtrationwith some tubular secretion."5-7 Nonrenal clearance isthrough hepatic excretion, and serum protein bindingis approximately 60%.15-17

Since the kidney is in part responsible for the elimi-nation of both drugs, and the exact mechanism for therenal elimination of linezolid is unknown, there ispotential for a pharmacokinetic interaction. They arevery likely to be coadministered in clinical practicedue to their complementary spectra of activity. Thissingle-dose drug interaction study was conducted todetermine the actual potential for a pharmacokineticinteraction when the two drugs are coadministered.

The linezolid dosing regimen in phase II efficacy tri-als ranged from 250 to 375 mg every 8 to 12 hours; asingle dose at the upper end of this range was selectedfor this interaction study. The usual starting dose ofaztreonam is 0.5 to 1 g every 8 hours. The doses chosenfor this drug interaction trial are clinically relevantand provide adequate plasma levels of each agent toassess and compare appropriate pharmacokineticparameters.

SUBJECTS AND METHODS

Study Protocol

This study of healthy male and female subjects wasconducted at the Pharmacia & Upjohn Clinical Re-search Unit, Brighton, England, using a randomized,

open-label, single-dose, three-way crossover design.Each subject received a single intravenous infusion onthree separate occasions, separated by a washout pe-riod of 7 days. Subjects received the three study treat-ments in randomized order:

Treatment A: intravenous linezolid (375 mg) as a30-minute infusion (187.5 mL);

Treatment B: intravenous aztreonam (1000 mg) as a30-minute infusion (100 mL);

Treatment C: intravenous linezolid (375 mg; 187.5mL) plus intravenous aztreonam (1000 mg; 100 mL),concomitantly infused over 30 minutes.

Subjects were healthy, nonsmoking men (n = 7) andwomen (n = 6). Females included in the trial were non-breastfeeding and surgically sterilized or postmeno-pausal and were required to have a negative serumpregnancy test. All subjects were in good health, as as-sessed by medical history, physical examination,echocardiogram, complete blood count, blood chemis-try, hepatitis B surface antigen test, and urinalysis.Subjects also underwent blood alcohol testing andurine drug screening. This study was performed in ac-cordance with the Brighton Upjohn Independent Re-search Ethics Committee, and all participants signedan informed consent.

Subjects were confined to the clinic from approxi-mately 12 hours prior to dosing until 48 hours post-dose (the time of the last blood draw). Following anovernight fast, subjects were dosed at 0.0 hours on day1 of each treatment period via a dedicated peripheralline using an intravenous infusion pump. For treat-ment C (linezolid/aztreonam combination), two intra-venous infusion pumps were used to deliver the drugssimultaneously into separate vascular sites within theantecubital area of one arm. Blood samples for thedetermination of drug concentrations were drawnfrom the antecubital area of the contralateral arm 10minutes before dosing, at 30 and 45 minutes, and at 1,1.5, 2, 3, 4, 6, 8, 10, 12, 16, 24, 30, 36, and 48 hourspostdose. Subjects continued to fast until 4 hours afterdosing.

Analytical Methods

Plasma was immediately harvested from blood sam-ples and frozen (-70°C at the clinic and < -30°C fromarrival at Pharmacia & Upjohn, Kalamazoo, MI) untilassayed for linezolid concentrations. Linezolid wasquantified in human plasma using a sensitive andselective high-performance liquid chromatographic(HPLC) method that was validated at AvTech

1278 * J Clin Pharmacol 1999;39:1277-1282

ADMINISTRATION OFLINEZOLID ANDAZTREONAM

Laboratories, Inc. (Kalamazoo, MI). Plasma specimens(0.500 mL) were spiked with an internal standard (IS)and extracted using prepared solid-phase extraction(SPE) cartridges (Isolute C2, Jones Chromatography,Lakewood, CO). Each sample was eluted from the SPEwith methanol. After evaporation of the organic mate-rial, the residue was reconstituted in acetonitrile/water and transferred to injection vials. An aliquot wasinjected onto the chromatography system (BeckmanGold) with separation accomplished on a reversed-phase analytical column (Zorbax RX-8, Dupont). Themobile phase was composed of trifluoroacetic acid/tet-rahydrofuran/methanol/water (0.1:1.2:25:73.7, v/v).Detection was by ultraviolet (UV) absorbance at 251nm. Retention times of linezolid and the IS wereapproximately 7.0 and 10.0 minutes, respectively.Mean recoveries for linezolid and the IS were approxi-mately 95.4% and 95.8%, respectively. Calibrationstandard responses were linear over the range of 0.010to 25.0 ,ug/mL, using a weighted (1/concentration) lin-ear least squares regression. The lower limit of quanti-tation was 0.010 gg/mL.

Coefficients of variation (CV%) were used toexpress the precision of the back-calculated calibra-tion standards. The calibration standards had CVs thatwere < 5.3%, with mean accuracy between 96.7% and104%. Interday accuracy and precision were furthermonitored by analysis of three linezolid quality con-trols prepared with target concentrations of 0.0400,4.00, and 15.0 gg/mL. Interday precision (CV) for thethree quality controls was < 3.9% with assay accuracyfrom 96.7% to 100%.

Pharmacia & Upjohn (Crawley, UK) developed andvalidated an HPLC procedure for the quantitation ofaztreonam in human plasma. Aliquots (0.250 mL) ofplasma were diluted with acetate buffer, pH 6.0, andproteins were precipitated with acetonitrile that con-tained-the IS. The samples were centrifuged and thesupernatant purified on an SPE (Isolute PH). Sampleswere washed and eluted with trifluoracetic acid inmethanol. The eluate was dried and reconstituted inmobile phase (methanol/0.1 M ammonium acetatebuffer, pH 6.0, 30:70, v/v). Detection was by UV absor-bance at 292 nm. The calibration range was linear from0.10 to 50.0 gg/mL. Quality controls were preparedwith target concentrations of 0.365, 15.0, and 40.1gg/mL. Interday precision for all QCs was less than11.6% with accuracy from 99.9% to 102%.

Both linezolid and aztreonam have shown good sta-bility under storage conditions similar to those used inthis study, and neither drug interfered with the chro-matographic determination of the other.

Pharmacokinetic Analysis

Linezolid and aztreonam plasma concentration datawere analyzed by noncompartmental methods. Themaximum plasma concentration (CmJ) and the time toCmax (tmax) were determined by inspection of the con-centration versus time curves for each subject. Theapparent terminal elimination rate constant (Xz) wasestimated using simple linear regression of the termi-nal segment of the natural log concentration versustime for each subject's data. The area under the con-centration versus time curve (AUCO-t) to the last mea-surable concentration (Ct) was estimated using thetrapezoidal rule, then extrapolated to infinity (AUCU J_)by adding the quotient of Ct/Xz. The apparent elimina-tion half-life (t1/2) was calculated as (ln 2)/Xz. Totalclearance (CL) was calculated as dose/AUC0,>. The vol-ume of distribution (Vss) was calculated as CL x (meanresidence time - [duration of infusion/2]).

Statistical Analysis

All data analyses were performed using SAS software(SAS Institute, Cary, NC). Each of the two treatmentcomparisons was made using parametric statisticalmethods. The linezolid and aztreonam pharmacoki-netic parameters were evaluated using a mixed-effectsanalysis of variance (ANOVA) model-with group,period, and treatment as fixed effects and subjectwithin group as a random effect. Separate analyseswere done for linezolid and aztreonam, and each analy-sis used a type IV sum of squares to allow for missingcells in the crossover design. Values are reported as themean ± standard deviation;p < 0.05 was considered astatistically significant difference.

RESULTS

Clinical Results

Of the 13 subjects enrolled, 12 completed the entirestudy. One female subject withdrew from the study forpersonal reasons after having received only a singledose of aztreonam. This subject was therefore excludedfrom the pharmacokinetic analysis.

Data for 5 females and 7 males were included in thepharmacokinetic analysis. Study subjects were between20 and 50 years of age (mean ± [SD]; 35 [8.7] years),with a mean weight of 73.3 (6.5) kg and mean height of171.2 (10.1) cm.A total of 15 adverse events were reported in 7 sub-

jects: 7 events followed a single dose of linezolid, 5

PHARMACOKINETICS AND PHARMACODYNAMICS 1 279

SISSON ETAL

Table I Linezolid Pharmacokinetic Parameters Resulting from a Single 30-MinuteInfusion of Linezolid 375 mg Administered to 12 Subjects Either Alone (treatment A)

or in Combination with Aztreonam 1000 mg (treatment C)

Parameter Treatment A (linezolid)

AUCO- (,ug x h/mL)Cl,,,,, (itg/mL)?kz (h-1)t112b (h)CL (mL/min)CL (mL/min/kg)Vss (L)Vss (L/kg)

56.0 ± 17.910.3 ± 1.9

0.158 ± 0.0594.39

121 ± 341.66 ± 0.4742.3 ± 6.70.58 ± 0.07

Treatment C (linezolid/aztreonam)

60.9 ± 21.612.2 ± 2.1

0.141 ± 0.0454.91

114 ± 391.58 ± 0.6042.5 ± 8.30.58 ± 0.10

Data are presented as the mean ± standard deviation. AUCn., area under the concentration-time curve extrapolated to infinity; Cn,;,x, maximumn plasma conccni-tration; Xz, terminal rate constant; t1/2 apparent elimination half-life; CL, total clearance; Vss, volume of distribution.a. Significant treatment effect at p < 0.05 by analysis of variance (ANOVA).b. Harmonic mean half-life.

events followed a single dose of aztreonam, and 3events followed the combination treatment. Therewere no serious adverse events, and most of theadverse events were mild and not considered studydrug related.

There were no clinically significant differencesamong the three treatments in vital sign measure-ments, echocardiograms, or laboratory safety tests.

Pharmacokinetics

LinezolidMean plasma linezolid pharmacokinetic parametersfollowing administration of treatment A (linezolidalone) or treatment C (linezolid plus aztreonam) areshown in Table I. The mean (± standard deviation)plasma concentration-time curves for linezolid areshown in Figure 2. There were no significant differ-ences between treatment groups for AUCe..., Xz, CL, orVss. Neither was there a significant difference betweentreatment groups in CL or Vss, when both wereadjusted for body weight. However, the Cm. of line-zolid was approximately 18% higher (p = 0.03) whenlinezolid was administered in combination withaztreonam, compared with linezolid alone. The powerto detect a 40% difference in linezolid clearancebetween treatments was 83.9%, given a coefficient ofvariation of 30%, with an alpha level of 5%.

Aztreonam

Mean plasma aztreonam pharmacokinetic parametersfollowing administration of treatment B (aztreonamalone) or treatment C (linezolid plus aztreonam) are

shown in Table II. The mean (± standard deviation)plasma concentration-time curves for aztreonam areshown in Figure 3. There were no significant differ-ences between treatment groups for AUC a-- C.,,ax, CL,or Vss. Neither was there a significant differencebetween treatment groups in CL or Vss, when bothwere adjusted for body weight. There was, however, asignificant difference between treatment groups in kz;coadministration resulted in a 6% greater apparentelimination half-life compared with administration ofaztreonam alone (1.7 hours vs. 1.6 hours). The powerto detect a 40% difference in aztreonam clearancebetween treatments was 99.9%, given a coefficient ofvariation of 6%, with an alpha level of 5%.

Figure 2. Linezolid plasma concentrations in 12 healthv subjects(mean + standard deviation).

1280 * J Clin Pharmacol 1999;39:1277-1282

p-Value'

0.43450.03060.2941

0.66140.71410.76980.7960

14-JE 122CD0

10 -0 Linezolid Alone0

88 Linezolid*Aztreonamo

8C 4

2 -

0 4 8 12 16 20 24

Time (hours)

ADMINISTRATION OF LINEZOLID AND AZTREONAM

Table II Aztreonam Pharmacokinetic Parameters Resulting from a Single 30-MinuteInfusion of Aztreonam 1000 mg Administered to 12 Subjects Either Alone (treatment B)

or in Combination with Linezolid 375 mg (treatment C)

Treatment B (aztreonam) Treatment C (linezolid/aztreonam)

AUC,, (g x h/mL) 121 21 128 ± 28 0.0579

C,,,,0 (g/mL) 62.0 10.4 60.0 ± 7.5 0.5813

kz (h-1) 0.432 0.029 0.403 ± 0.039 0.0100

t112h (h) 1.60 1.72CL (mU/min) 141 ± 24 136 ± 32 0.1415

CL (mL/min/kg) 1.93 ± 0.25 1.85 ± 0.33 0.1092

Vss (L) 18.6 ± 2.7 19.2 ± 3.3 0.4108

Vss (L/kg) 0.25 ± 0.02 0.26 ± 0.03 0.3887

Data are presented as the mean ± standard deviation. AUCG0-, area under the concentration-time curve extrapolated to infinity; C,l, maximum plasma con-

centration; Xz, terminal rate constant; t1,2, apparent elimination half-life; CL, total clearance; Vss, volume of distribution.a. Significant treatment effect at p < 0.05 by analysis of variance (ANOVA).b. Harmonic mean half-life.

DISCUSSION

Linezolid is currently under development for the treat-ment of infections caused by a wide variety of Gram-positive bacteria. In vitro studies have demonstratedthat linezolid is effective against a wide range ofGram-positive species.9'0 In clinical practice, aztreo-nam is a likely candidate for coadministration withlinezolid because of its effectiveness against Gram-negative organisms; the activity spectra of the twoagents are expected to complement one another in thetreatment of mixed infections.

In this interaction trial, the pharmacokinetics oflinezolid and aztreonam were not significantly altered

80-J

E

00E 60

c

Loc 400cJ8cd 20E

O

0+- Aztreonam Alone

-a-) Aztroonam/Linezolid

4 8 12 16 20 24

Time (hours)

Figure 3. Aztreonam plasma concentrationis in 12 healthy subjects(mean standard deviation).

0

by concomitant, single-dose administration of the twodrugs in healthy subjects. Although an 18% increasein Cmax of linezolid was observed with combinationtherapy, this difference is not expected to be clinicallysignificant, given the large intersubject variability inthe pharmacokinetics of this agent. The pharmacoki-netic parameters and associated variability reported inthis study are similar to those reported in other line-zolid studies."-"3 Similarly, the elimination rate con-

stant of aztreonam was lower when the drug was

given in combination than when subjects received italone and resulted in only a very small difference (<10%) in elimination half-life. It is probable that theobserved small difference in Xz was statistically sig-nificant because of the higher statistical power in thisstudy for aztreonam compared with linezolid, as theformer drug exhibits much lower intersubject variabil-ity. The pharmacokinetic parameters reported in thisstudy for aztreonam are similar to those reported in theliterature."-"59A component of elimination for each drug is renal

excretion. Aztreonam is in part filtered and secreted bythe kidney. It has not yet been determined whether therenal elimination of linezolid includes secretion by thekidney. Therefore, the potential for an interactionbetween linezolid and aztreonam is based on this pos-sible common pathway of elimination. The pharma-cokinetic data from this study demonstrate that coad-ministration of linezolid and aztreonam does not alterthe disposition of either agent under single-dose con-

ditions. Therefore, it is not-anticipated that dose altera-tions will be necessary in a clinical setting. Further-more, serious safety problems are not likely whenlinezolid and aztreonam are given together.

PHARMACOKINETICS AND PHARMACODYNAMICS

Parameter p-Value'

1281

SISSON ETAL

REFERENCES

1. Demyan WF, Swaney SM, Shinabarger DL: The oxazolidinone li-nezolid inhibits translation initiation in bacteria (abstract 5375). Pa-per presented at the 37th Interscience Conference on AntimicrobialAgents and Chemotherapy, 28 September-i October 1997, Toronto,Ontario.2. Swaney SM, Shinabarger DL: Oxazolidinone mechanism of ac-

tion: eperezolid inhibits the initiation event of protein synthesis inbacteria (abstract A-057). Paper presented at the 96th general meet-ing of the American Society for Microbiology, 19-23 May 1996, NewOrleans, LA.3. Shinabarger DL, Marotti KR, Murray RW, Lin AH, Melchior EP,Swaney SM, et al: Mechanism of action of oxazolidinones: effects oflinezolid and eperezolid on translation reactions. Antimicrob AgentsChemother 1997;41:2132-2136.4. Lin AH, Murray RW, Vidmar TJ, Marotti KR: The oxazolidinoneeperezolid binds to the 50S ribosomal subunit and competes withthe binding of chloramphenicol and lincomycin. Antimicrob AgentsChemnother 1997;41:2127-2131.5. Daly JS, Eliopoulos GM, Willey S, Moellering RC Jr: Mechanism ofaction and in vitro and in vivo activities of S-6123, a new oxazolidinonecompound. Antimicrob Agents Chen7other 1988;32(9):1341-1346.6. Daly JS, Eliopoulos GM, Reiszner E, Moellering RC Jr: Activity andmechanism of action of DuP 105 and DuP 721, new oxazolidinonecompounds. JAntimicrob Chemother 1988;21:721-730.7. Eustice DC, Feldman PA, Slee AM: The mechanism of action ofDuP 721, a new antibacterial agent: effects on macromolecular syn-

thesis. Biochem Biophys Res Comm 1988;150(3):965-971.8. Eustice DC, Feldman PA, Zajac 1, Slee AM: Mechanism of action ofDuP 721: inhibition of an early event during initiation of proteinsynthesis. Antimicrob Agents Chemother 1988;32(10):1218-1222.9. Zurenko GE, Yagi BH, Schaadt RD, Allison JW, Kilburn JO, Glick-man SE, et al: In vitro activities of U-100592 and U-100766, noveloxazolidinone antibacterial agents. Antimicrob Agents Chemother1996;40:839-845.

10. Jones RN, Johnson DM, Erwin ME: In vitro antimicrobial activityand spectrum of U-100592 and U-100766, two novel oxazolidi-nones. Antimicrob Agents Chemother 1996;40:720-726.11. Pawsey SD, Daley-Yates PT, Wajszczuk CP, Stalker DJ: U-100766safety, toleration and pharmacokinetics after oral and intravenousadministration. Abstracts of European Congress of AntimicrobialChemotherapy, 1996.12. Stalker DJ, Wajszczuk CP, Batts DH: Linezolid safety, tolerance,and pharmacokinetics following oral dosing twice daily for 14.5days. Paper presented at the 37th Interscience Conference on Anti-microbial Agents and Chemotherapy, 28 September-i October 1997,Toronto, Ontario.13. Stalker DJ, Wajszczuk CP, Batts DH: Linezolid safety, tolerance,and pharmacokinetics after intravenous dosing twice daily for 7.5days. Paper presented at the 37th Interscience Conference on Anti-microbial Agents and Chemotherapy, 28 September-i October 1997,Toronto, Ontario.14. Feenstra KL, Slatter JG, Stalker DJ, Welshman IR, Hauer KS, Cath-cart MT, et al: Metabolism and excretion of the oxazolidinone antibi-otic linezolid (PNU-100766) following oral dosing of ["4C1PNU-100766 to healthy volunteers. Paper presented at the 38th In-terscience Conference on Antimicrobial Agents and Chemotherapy,24-27 September 1998, San Diego, CA.15. Mattie H: Clinical pharmacokinetics of aztreonam. Clin Pharma-cokinet 1988;14:148-155.16. Mattie H: Clinical pharmacokientics of aztreonam: an update.Clin Pharmacokinet 1994;26:99-106.17. el Guanidine MA, Nawishy S, Abd el Bary M, Sabbour MS:Single-dose pharmacokinetics of aztreonam in healthy volunteersand renal failure patients. J Chemother 1989;1:164-169.18. Guay DR, Koskoletos C: Aztreonam, a new monolactam antimi-crobial. Clin Phorm 1985;4:516-526.19. Swabb EA, Sugerman AA, Platt TB, Pilkiewicz FG, Frantz M:Single-dose pharmacokinetics of the monobactam azthreonam (SQ26,776) in healthy subjects. Antimicrob Agents Cheinother 1992;21:944-949.

1282 * J Clin Pharmacol 1999;39:1277-1282