Intrapulmonary pharmacokinetics and pharmacodynamics of meropenem

International Journal of Antimicrobial Agents 26 (2005) 449–456

Intrapulmonary pharmacokinetics and pharmacodynamics of meropenem John E. Conte Jr a,b,c,∗ , Jeffrey A. Golden b , Mary Grace Kelley a , Elisabeth Zurlinden a a

c

Department of Epidemiology & Biostatistics, Infectious Diseases Research Group, University of California at San Francisco, San Francisco, CA 94143-0919, USA b Department of Medicine, Infectious Diseases Research Group, University of California at San Francisco, San Francisco, CA 94143-0919, USA Department of Microbiology and Immunology, Infectious Diseases Research Group, University of California at San Francisco, San Francisco, CA 94143-0919, USA Received 4 May 2005; accepted 20 August 2005

Abstract The objective of this study was to determine the plasma and intrapulmonary pharmacokinetic parameters of intravenously administered meropenem in healthy volunteers. Four doses of 0.5 g, 1.0 g or 2.0 g meropenem were administered intravenously to 20, 20 and 8 healthy adult subjects, respectively. Standardised bronchoscopy and timed bronchoalveolar lavage (BAL) were performed following administration of the last dose. Blood was obtained for drug assay prior to drug administration and at the time of BAL. Meropenem was measured in plasma, BAL fluid and alveolar cells (ACs) using a combined high pressure liquid chromatographic–mass spectrometric technique. Plasma, epithelial lining fluid (ELF) and AC pharmacokinetics were derived using non-compartmental methods. Cmax /MIC90 (where Cmax is the maximum plasma concentration and MIC90 is the minimum inhibitory concentration required to inhibit 90% of the pathogen), AUC/MIC90 (where AUC is the area under the curve for the mean concentration–time data), intrapulmonary drug exposure ratios and percent time above MIC90 during the dosing interval (%T > MIC90 ) were calculated for common respiratory pathogens with MIC90 values of 0.12–4 ␮g/mL. In the 0.5 g dose group, the Cmax (mean ± S.D.), AUC0–8 h and half-life for plasma were, respectively, 25.8 ± 5.8 ␮g/mL, 28.57 ␮g h/mL and 0.77 h; for ELF the values were 5.3 ± 2.5 ␮g/mL, 12.27 ␮g h/mL and 1.51 h; and for ACs the values were 1.0 ± 0.5 ␮g/mL, 4.30 ␮g h/mL and 2.61 h. In the 1.0 g dose group, the Cmax , AUC0–8 h and half-life for plasma were, respectively, 53.5 ± 19.7 ␮g/mL, 55.49 ␮g h/mL and 1.31 h; for ELF the values were 7.7 ± 3.1 ␮g/mL, 15.34 ␮g h/mL and 0.95 h; and for ACs the values were 5.0 ± 3.4 ␮g/mL, 14.07 ␮g h/mL and 2.17 h. In the 2.0 g dose group, the Cmax , AUC0–8 h and half-life for plasma were, respectively 131.7 ± 18.2 ␮g/mL, 156.7 ␮g h/mL and 0.89 h. The time above MIC in plasma ranged between 28% and 78% for the 0.5 g dose and between 45% and 100% for the 1.0 g and 2.0 g doses. In ELF, the time above MIC ranged from 18% to 100% for the 0.5 g dose and from 25% to 88% for the 1.0 g dose. In ACs, the time above MIC ranged from 0% to 100% for the 0.5 g dose and from 24% to 100% for the 1.0 g dose. Time above MIC in ELF and ACs for the 2.0 g dose was not calculated because of sample degradation. The prolonged T > MIC90 and high intrapulmonary drug concentrations following every 8 h administration of 0.5–2.0 g doses of meropenem are favourable for the treatment of common respiratory pathogens. © 2005 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved. Keywords: Meropenem; Intrapulmonary; Pharmacokinetics; Pharmacodynamics

1. Introduction

∗ Corresponding author at: University of California at San Francisco, MU 420 West, Box 0560, San Francisco, CA 94143, USA. Tel.: +1 415 476 1312. E-mail address: [email protected] (J.E. Conte Jr).

Meropenem is an approved carbapenem antibiotic that is active against common respiratory pathogens such as penicillin-sensitive Streptococcus pneumoniae, methicillinsensitive Staphylococcus aureus, Haemophilus influenzae, Pseudomonas aeruginosa, Klebsiella pneumoniae,

0924-8579/$ – see front matter © 2005 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved. doi:10.1016/j.ijantimicag.2005.08.015

450

J.E. Conte Jr et al. / International Journal of Antimicrobial Agents 26 (2005) 449–456

Enterobacter aerogenes, Acinetobacter baumannii and anaerobes [1]. Whilst the recommended dose in adults is 1.0 g given intravenously every 8 h, doses ranging from 1.5 g/day (0.5 g every 8 h) to 6.0 g/day (2.0 g every 8 h) have been reported in clinical trials [2–5]. The plasma elimination half-life, maximum plasma concentration (Cmax ), volume of distribution at steady state (Vdss ) and plasma clearance in healthy subjects receiving 1.0 g intravenously have been reported to be ca. 1.0 h, 60 ␮g/mL, 10.5 L and 15 L/h respectively [6]. Protein binding is low and concentration dependent, i.e. 13.8%, 10.4% and 22.3%, respectively, at 5 ␮g/mL, 50 ␮g/mL and 100 ␮g/mL [7]. The intrapulmonary concentrations of meropenem in patients following a single intravenous dose of 1.0 g have been reported previously [8,9]. We have previously decribed our techniques for the in vivo measurement of the concentration of antibiotics in pulmonary epithelial lining fluid (ELF) and alveolar cells (ACs) [10–16]. The purpose of this investigation was to determine the plasma and intrapulmonary pharmacodynamic parameters of three different multiple dose regimens administered to healthy volunteers.

according to the time of bronchoscopy of 1, 2, 3, 5 or 8 h following the last dose. Eight subjects in the 2.0 g arm were assigned to one of two groups of four subjects each according to the time of bronchoscopy of 1 h or 3 h following the last dose. Meropenem was administered intravenously over 30 min in a dose of 0.5 g, 1.0 g or 2.0 g every 8 h for a total of four doses. The four doses of study medication were administered in the General Clinical Research Center (GCRC) at the University of California at San Francisco. 2.2. Bronchoscopy and bronchoalveolar lavage Standardised bronchoscopy, bronchoalveolar lavage (BAL) and clinical monitoring were performed in the GCRC as previously reported [10–16]. 2.3. Blood samples Blood was obtained for meropenem assay just prior to administration of the first dose (baseline), at completion of the last dose (Cmax ) and at the time of BAL at 1, 2, 3, 5 or 8 h following initiation of the last dose.

2. Methods 2.4. Specimen handling 2.1. Study design and subjects This was a prospective, non-blinded study of the plasma and intrapulmonary concentrations of meropenem in healthy adults. All subjects gave written informed consent and were required to be between 21 and 45 years of age and have a body mass index of 18–29 [17]. The screening evaluation included: a medical history; physical examination; vital signs; height and weight; and baseline laboratory testing, including complete blood count with differential, platelet count, blood urea nitrogen, serum creatinine, aspartate aminotransferase, alanine aminotransferase, human chorionic gonadotropin (if female), alkaline phosphatase, total bilirubin, prothrombin time, partial thromboplastin time, albumin, urinalysis with microscopy and a urine drug screening (drugs of abuse, including cotanine). The evaluation was repeated prior to administration of meropenem and, except for the urine drug and alcohol screen, following bronchoscopy. Subjects were excluded who had: a history of clinically significant disease, seizure disorder, or major surgery within the previous 6 months; clinically significant abnormal findings at the screening physical examination (including laboratory tests); intolerance to meropenem, carbapenems, cephalosporins, penicillins or lidocaine; substance abuse or smoking within the past year; positive drug screen; pregnancy or lactation; those required to take chronic medications other than self-prescribed vitamins, birth control pills or hormone replacement therapy; and those receiving any investigational drug within 30 days prior to the study. Twenty subjects in each of the 0.5 g and 1.0 g groups were assigned to one of five groups of four subjects each

Blood samples were kept on ice until centrifugation. The plasma was separated and frozen at −70 ◦ C until assay. 2.5. Meropenem assay The meropenem assays were performed at AstraZeneca Pharmaceuticals (Wilmington, DE). Meropenem was determined in human plasma, bronchoalveolar lavage fluid (BALF) and alveolar cells (ACs) by reverse phase liquid chromatography and electrospray ionisation (ESI) tandem mass spectrometry (MS). Human plasma was extracted by protein precipitation with 50:50 acetonitrile:methanol. Aliquots of BALF were vortexed, centrifuged at 10 000 rpm at 4 ◦ C for 5 min and analysed directly. Owing to the lack of sufficient blank lavage matrix, this method has been validated for analysis of BALF and AC samples using a surrogate lavage solution (sBALF) composed of 10% blank lavage/90% normal saline (0.9% saline). ACs were diluted with 1 mL of sBALF, suspended by vortexing, lysed by ultrasonication and extracted by a protein precipitation procedure similar to plasma. All samples, supernatants and reagents were kept in an ice-bath whenever possible during sample preparation for all three matrices. The samples were injected onto a high pressure liquid chromatographic column using a switching valve to first divert the non-retained solvent front to waste, then to switch the flow to the mass spectrometer where retained analytes were detected by multiple reaction monitoring via positive ion ESI/MS using a PE/SCIEX API 365 mass spectrometer (PerkinElmer, Boston, MA).

J.E. Conte Jr et al. / International Journal of Antimicrobial Agents 26 (2005) 449–456

The calibration range for plasma was validated from a lower limit of 0.5 ␮g/mL to an upper limit of 50 ␮g/mL using a 20 ␮L aliquot of human plasma with heparin as the anticoagulant. Dilutions of up to 1:20 with blank human plasma extended the upper quantitation limit to 1000 ␮g/mL. The method is specific for acetylsalicylic acid, salicylic acid, acetaminophen and ibuprofen. The method has been shown to be precise and accurate in human plasma. Calibration standards prepared at 0.5, 1.0, 2.0, 5.0, 10.0, 25.0 and 50.0 ␮g/mL in human plasma containing heparin were assayed in duplicate in each of the four validation assays. The overall inter-assay precision of the human plasma calibration standards across the linear range was 3.8% relative standard deviations (RSD). The lower limit of quantitation (LLOQ) samples in human plasma were assayed in quadruplicate in each of the four validation assays. The intra-assay precision for the LLOQ for meropenem in human plasma was 7.0% RSD and the inter-assay precision was 9.1% RSD, with 0.2% absolute difference from theory. Human plasma quality controls (QCs), prepared at 1.5, 15.0 and 40.0 ␮g/mL, were assayed in quadruplicate in each of the four validation runs. The overall intra-assay precision for meropenem for the low, middle and high QCs was 7.7%, 5.1% and 6.8% RSD, respectively. The overall inter-assay precision for the low, middle and high QCs was 4.0%, 2.5% and 5.1% RSD, with an absolute difference from theory of 6.7%, 3.3% and 7.5%, respectively. No interference was observed in the plasma from six lots of human plasma. Meropenem stability has been demonstrated in human plasma for 20 months when stored at or below −70 ◦ C and at room temperature for 3 h. Stability of meropenem was demonstrated for four freeze–thaw cycles in human plasma. The calibration range for BALF and ACs was validated from a lower limit of 1.0 ng/mL to an upper limit of 200 ng/mL using a 150 ␮L aliquot of BALF or a 200 ␮L aliquot from an AC suspension. Dilutions of up to 1:5 with blank sBALF or blank cell suspensions extended the upper quantitation limit to 1000 ng/mL. The method was shown to be precise and accurate in sBALF. Limited availability of blank lavage permitted only two validation assays. Calibration standards prepared at 1.0, 2.0, 5.0, 10.0, 20.0, 50.0, 100 and 200 ng/mL in sBALF were assayed in duplicate in each of the two validation assays. The overall inter-assay precision of the sBALF calibration standards across the linear range was 8.0% RSD. The LLOQ samples in sBALF were assayed in quadruplicate in each of the two validation assays. The intra-assay precision for the LLOQ for meropenem in sBALF was 7.8% RSD. The interassay precision was 1.6% RSD, with 2.7% absolute difference from theory. sBALF QCs prepared at 4.0, 80.0 and 160 ng/mL were assayed in quadruplicate in each of the two validation runs. The overall intra-assay precision for meropenem in the low, middle and high QCs was 2.8%, 2.7% and 1.8% RSD, respectively. The overall inter-assay precision for the low, middle and high QCs was 3.5%, 1.8% and 3.4% RSD, with an absolute difference from theory of 1.8%, 4.5% and

451

2.5%, respectively. Meropenem stability was demonstrated in sBALF for 319 days when stored at or below −70 ◦ C. 2.6. Quantitation of the volume of ELF and concentration of antibiotics in ELF and ACs The amount of ELF recovered was calculated by the urea dilution method as described by Rennard et al. [18] and as reported in our previous pulmonary pharmacokinetic studies [10–16]. The concentration of urea in serum was analysed using previously reported methods [10–16,19]. The volume of ELF and ACs collected in BALF, the concentration of antibiotic in the ELF (ABXELF ) and the concentration of antibiotic in alveolar cells (ABXAC ) were derived using methods and calculations published previously [10–16]. 2.7. Statistical, pharmacokinetic and pharmacodynamic analyses Descriptive statistics, graphic representations, database management and the kinetic analysis were performed using Kinetica 2000 (version 4.3; InnaPhase Corporation, Philadelphia, PA), SPSS (version 11.0.1; SPSS, Inc., Chicago, IL) and PROPHET (version 6.0; AbTech, Charlottesville, VA). The mixed log-linear rule was used to compute the area under the curve (AUC) for the mean concentration–time data in plasma, ELF and ACs for 0–8 h after the last dose. If drug was not detectable at the 8 h time point, the AUC was calculated using the last observed value. The means of the plasma data at 1, 2, 3, 5 and 8 h for the 0.5 g and 1.0 g dose groups and at 1.0 h and 3.0 h for the 2.0 g dose group were fit to a monoexponential equation, y = dose/volume × exp(−Lz x) to calculate Lz , the elimination rate constant, and to calculate a projected 8 h plasma concentration in the 2.0 g dose group. The plasma, ELF and AC half-lives were calculated using the relationship T1/2 = 0.693/Lz . Fitting was performed using a weighting function (1/Y2 ) where 1/Y was the reciprocal of the observed concentration. These fitted curves were also used to estimate the time above minimum inhibitory concentration (MIC) for each dosage regimen and MIC combination. Analysis of variance (ANOVA) was used to compare the meropenem concentrations in plasma, ACs and ELF, as well as ELF and AC recovery at the different time periods. Linear regression was performed using the method of least squares estimation. P < 0.05 was regarded as significant. For the pharmacodynamic calculations, the concentrations of meropenem required to inhibit 90% (MIC90 ) of the following respiratory pathogens were used: H. influenzae and E. aerogenes, MIC90 = 0.12 ␮g/mL; S. pneumoniae, S. aureus and Enterobacter cloacae, MIC90 = 0.25 ␮g/mL; A. baumannii, MIC90 = 2 ␮g/mL; and P. aeruginosa, MIC90 = 4.0 ␮g/mL [1]. Pharmacokinetic/pharmacodynamic (PK/PD) indices, as suggested by Mouton et al. [20], were used to calculate the Cmax /MIC90 and AUC/MIC90 ratios and the percent time above the MIC90 (%T > MIC90 ).

452

J.E. Conte Jr et al. / International Journal of Antimicrobial Agents 26 (2005) 449–456

3. Results Forty-eight subjects, 20 in the 0.5 g group, 20 in the 1.0 g group and 8 in the 2.0 g group, completed the study. All of the results are presented, although the values obtained for concentrations of meropenem in ACs and ELF in the 2.0 g were implausibly low. After extensive evaluation, we concluded that the BAL specimens from these eight subjects had undergone unexplained degradation; thus, the PK/PD calculations for these specimens were not performed. In the 0.5 g group, the age (mean ± standard deviation (S.D.) of the 20 subjects was 33 ± 7.0 years. Eleven were men and nine were women. Twelve were Caucasian, four were Asian, two were African–American and two were Hispanic. The weight (mean ± S.D.) and serum creatinine were 68.2 ± 10.5 kg and 0.9 ± 0.2 mg/dL, respectively. The remaining screening laboratory tests were within normal limits. In the 1.0 g group, the age of the 20 subjects was 29.0 ± 5.7 years. Eleven were men and nine were women. Sixteen were White, two were Asian, one was African–American and one was indeterminate. The weight and serum creatinine were 71.5 ± 13.0 kg and 0.9 ± 0.2 mg/dL, respectively. The remaining screening laboratory tests were within normal limits. In the 2.0 g group, the age of the eight subjects was 33 ± 7.0 years. Five were men and three were women. Six were Caucasian, one was Asian and one was African–American. The weight and serum creatinine were 71.8 ± 12.5 kg and 1.0 ± 0.1 mg/dL, respectively. The remaining screening laboratory tests were within normal limits. None of the differences in age, weight or serum creatinine within or between the three dose groups were significant (P > 0.05). For the three dose groups, there were no serious adverse events and the subjects returned to their normal duties following bronchoscopy and BAL. Following the procedure, 3 subjects experienced chest discomfort, 1 had a transiently elevated temperature, 1 had shortness of breath, 7 had brief duration cough and 20 described self-limited lightheadedness. Transient rales or ronchi were present in eight subjects

following the procedure. This is an expected finding following instillation of fluid for the purpose of BAL. On repeat laboratory testing, two subjects had elevated liver function tests, none had an elevated serum creatinine, two had a slightly decreased haemoglobin concentration and one had a borderline elevated white blood cell count. The number or cells (mean ± S.D.) recovered from the three dose groups was not significantly different among the time groups (P > 0.05) (Table 1). AC recovery was not correlated with concentration of meropenem in ACs for the 0.5 g (R2 = 0.11; P = 0.15), 1.0 g (R2 = 0.06; P = 0.29) or 2.0 g (R2 = 0.02; P = 0.74) dose groups. The volume (mean ± S.D.) of ELF recovered from the three dose groups was not significantly different (P > 0.05) (Table 1). ELF recovery was not correlated with concentration of meropenem in ELF for the 0.5 g (R2 = 0.06; P = 0.31), 1.0 g (R2 = 0.00; P = 0.97) or 2.0 g (R2 = 0.17; P = 0.32) dose groups. 3.1. Plasma For the 8 h dosing interval, the mean (± S.D.) plasma concentrations (Cmax ) and the AUC0–8 h after the fourth dose for the 0.5 g (N = 20), 1.0 g (N = 20) and 2.0 g (N = 8) dose groups were, respectively, 25.8 ± 5.8 ␮g/mL and 28.57 ␮g h/mL, 53.5 ± 19.7 ␮g/mL and 55.49 ␮g h/mL, and 131.7 ± 18.2 ␮g/mL and 156.7 ␮g h/mL. Thus, the Cmax and AUC were proportional to dose, i.e. the kinetics were linear, as previously reported. Doubling of the dose resulted in an approximate two-fold increase in Cmax and AUC during the dosing interval. Comparison of the plasma concentrations determined at the completion of infusion between the time groups within each dose revealed no significant differences (P > 0.05) (Tables 2–4). There was no correlation between the weight of the subjects and the Cmax concentrations of meropenem following the fourth dose for the 0.5 g (R2 = 0.04; P = 0.38), 1.0 g (R2 = 0.00; P = 0.81) or 2.0 g (R2 = 0.02; P = 0.70) dose groups. Thus, although these doses were not weight-corrected, the plasma concentrations were not affected by the size of the subjects within the limits of criteria used for enrolment.

Table 1 Recovery of cells and epithelial lining fluid (ELF) from bronchoalveolar lavage in the 0.5 g (N = 20), 1.0 g (N = 20) and 2.0 g (N = 8) dose groupsa Dose (g) 0.5 Mean (cells/L) PMNs (%) Lymphocytes (%) Monocytes/macrophages (%) Eosinophils (%) Degenerated cells (%) ELF volume (mL)

1.4 × 108

1.0 (±1.5 × 108 )

1.4 ± 1.7 9.6 ± 8.5 81.5 ± 13.1 0.6 ± 1.1 7.0 ± 13.4 0.7 ± 0.3

0.82 × 108

2.0 (±0.44 × 108 )

1.7 ± 4.8 8.7 ± 7.1 82.7 ± 14.2 0.2 ± 0.4 6.8 ± 12.1 1.2 ± 0.7

0.1 × 108 (±0.2 × 108 ) 1.3 ± 1.4 13.3 ± 14.0 81.9 ± 13.4 1.1 ± 1.6 1.5 ± 2.3 1.0 ± 0.6

PMNs, polymorphonuclear leukocytes. a All data given as mean ± standard deviation. No significant differences among the groups for cell recovery, differential cell count or volume of ELF (P > 0.05).

J.E. Conte Jr et al. / International Journal of Antimicrobial Agents 26 (2005) 449–456

453

Table 2 Meropenem concentrations in plasma at the completion of infusion (Cmax ) and in plasma, epithelial lining fluid (ELF) and alveolar cells (ACs) at the time of bronchoalveolar lavage (BAL) in the 0.5 g dose group (N = 20)a BAL time (h)

Plasma Cmax (␮g/mL)b

Concentration at the indicated BAL time (␮g/mL)c Plasma

1 2 3 5 8

23.8 29.8 23.4 24.9 27.2

± ± ± ± ±

2.1 5.8 7.7 2.1 8.8

10.9 5.2 2.4 0.3 0.0

± ± ± ± ±

ACs 1.3 1.6 0.9 0.4 0.0

ELF

± ± ± ± ±

1.0 1.0 0.5 0.6 0.2

0.5 0.5 0.3 0.5 0.3

5.3 2.7 1.9 0.7 0.2

± ± ± ± ±

2.5 1.8 0.9 0.4 0.1

Data are given as mean ± 1 standard deviation. Blood drawn at the completion of infusion. There were no significant differences in the Cmax plasma concentrations among the time groups (P > 0.05). c Plasma concentrations were significantly greater than AC concentrations at 1, 2 and 3 h (P < 0.05) but not at 5 h or 8 h (P > 0.05); plasma concentration was significantly greater than ELF concentration only at 1 h (P < 0.01); ELF concentration was significantly greater than AC concentration only at 1 h (P < 0.01). No other differences were significant. a

b

Table 3 Meropenem concentrations in plasma at the completion of infusion (Cmax ) and in plasma, epithelial lining fluid (ELF) and alveolar cells (ACs) at the time of bronchoalveolar lavage (BAL) in the 1.0 g dose group (N = 20)a BAL time (h)

Plasma Cmax (␮g/mL)b

Concentration at the indicated time of BAL (␮g/mL)c Plasma

1 2 3 5 8

41.6 58.4 51.0 48.0 68.6

± ± ± ± ±

8.7 8.8 24.9 25.2 21.9

19.0 7.5 5.3 2.0 0.0

± ± ± ± ±

ACs 7.6 1.3 1.5 1.3 0.0

5.0 4.9 1.8 1.6 0.0

ELF

± ± ± ± ±

3.4 1.7 0.6 0.7 0.0

7.7 4.0 1.7 0.8 0.03

± ± ± ± ±

3.1 1.1 1.4 0.4 0.05

Data are given as mean ± 1 standard deviation. Blood drawn at the completion of infusion. There were no significant differences in the Cmax plasma concentrations among the time groups (P > 0.05). c Plasma concentrations were significantly greater than AC concentrations at 1 h and 3 h (P < 0.05); plasma concentrations were significantly greater than ELF concentrations at 1, 2 and 3 h (P < 0.05); ELF and AC concentrations were not significantly different at any time point (P > 0.05). a

b

Table 4 Meropenem concentrations in plasma at the completion of infusion (Cmax ) and in plasma, epithelial lining fluid (ELF) and alveolar cells (ACs) at the time of bronchoalveolar lavage (BAL) in the 2.0 g dose group (N = 8)a BAL time (h)

Plasma Cmax (␮g/mL)b

1 3

129.7 ± 21.1 133.8 ± 17.7

Concentration at the indicated time of BAL (␮g/mL)c Plasma

ACs

ELF

60.9 ± 8.0 12.8 ± 2.7

0.5 ± 0.4 0.0 ± 0.0

2.9 ± 1.0 2.8 ± 1.5

Data are given as mean ± 1 standard deviation (S.D.). Blood drawn at the completion of infusion. There were no significant differences in the Cmax plasma concentrations between the two time groups (P > 0.05). c AC and ELF concentrations in the 2 g dose group were greater than 20 S.D. from the expected means and were not included in the statistical analysis (see Section 2). a

b

Table 5 Plasma, epithelial lining fluid (ELF) and alveolar cell (AC) pharmacokinetic parameters in the 0.5 g, 1.0 g and 2.0 g dose groups Dose group

Cmax (␮g/mL) (N)

Tmax (h)

AUC0–8 h (␮g h/mL) (N)

T1/2 (h)

0.5 g (N = 20) Plasma ELF AC

25.8 ± 5.8 (20) 5.3 ± 2.5 (4) 1.0 ± 0.5 (4)

0.5 1.0 2.0

28.57 (20) 12.27 (20) 4.30 (20)

0.77 1.51 2.61

1.0 g (N = 20) Plasma ELF AC

53.5 ± 19.7 (20) 7.7 ± 3.1 (4) 5.0 ± 3.4 (4)

0.5 1.0 1.0

55.49 (20) 15.34 (20) 14.07 (20)

1.31 0.95 2.17

2.0 g (N = 8)a Plasma

131.7 ± 18.2 (8)

0.5

156.7 (8)

Cmax , maximum plasma concentration; Tmax , time to Cmax ; AUC, area under the curve for the mean concentration–time data; T1/2 , half-life. a Bronchoalveolar lavage specimens were degraded in the 2.0 g group, therefore ELF and AC concentrations were not calculated.

0.89

J.E. Conte Jr et al. / International Journal of Antimicrobial Agents 26 (2005) 449–456

454

Table 6 Plasma pharmacodynamic parameters by dose group MIC90 (␮g/mL)

0.12 0.25 2.0 4.0

Cmax /MIC90 ratio

AUC/MIC90 ratio

% Dosing interval plasma drug concentration above MIC90

0.5 g

1.0 g

2.0 g

0.5 g

1.0 g

2.0 g

0.5 g

1.0 g

2.0 g

215 103 13 6.5

446 214 27 13

1098 527 66 33

219 105 13 6.6

462 222 27.8 13.9

768 369 46 23

78 66 36 28

100 100 64 45

100 98 68 55

MIC90 , minimum inhibitory concentration required to inhibit 90% of the pathogen; Cmax , maximum plasma concentration; AUC, area under the curve for the mean concentration–time data.

The half-life in plasma (T1/2 ) was 0.77 h in the 0.5 g dose group, 1.31 h in the 1.0 g dose group and 0.89 h in the 2.0 g dose group (Table 5). The plasma Cmax /MIC90 and AUC/MIC90 ratios and %T > MIC90 are summarised in Table 6. 3.2. ELF For the 8 h dosing interval, the ELF concentrations (mean ± S.D.) determined at the time of bronchoscopy ranged from 5.3 ± 2.5 ␮g/mL at 1 h (Cmax , Tmax ) to 0.2 ± 0.1 ␮g/mL at 8 h (Cmin , Tmin ) for the 0.5 g dose group; from 7.7 ± 3.1 ␮g/mL at 1 h (Cmax, Tmax ) to 0.03 ± 0.05 ␮g/mL at 8 h (Cmin , Tmin ) for the 1.0 g dose group; and from 2.9 ± 1.0 ␮g/mL at 1 h (Cmax, Tmax ) to 2.8 ± 1.5 ␮g/mL at 3 h for the 2.0 g dose group (Tables 2–4). The half-life (T1/2 ) and AUC0–8 h , respectively, in ELF were 1.51 h and 12.27 ␮g h/mL in the 0.5 g dose group and 0.95 h and 15.34 ␮g h/mL in 1.0 g dose group (Table 5). In contrast to the concentrations in plasma, penetration of meropenem into the ELF was non-linear. Doubling of the dose from 0.5 g to 1.0 g resulted in a 44% increase in Cmax and a 25% increase in the AUC0–8 h . The ELF Cmax /MIC90

and AUC/MIC90 ratios and %T > MIC90 are summarised in Table 7. 3.3. ACs For the 8 h dosing interval, the AC concentrations (mean ± S.D.) determined at the time of bronchoscopy ranged from 1.0 ± 0.5 ␮g/mL at 1 h (Cmax , Tmax ) to 0.2 ± 0.3 ␮g/mL at 8 h (Cmin , Tmin ) for the 0.5 g dose group; from 5.0 ± 3.4 ␮g/mL at 1 h (Cmax , Tmax ) to 1.6 ± 0.7 ␮g/mL at 5 h (Cmin , Tmin ) for the 1.0 g dose group; and from 0.5 ± 0.4 ␮g/mL at 1 h to 0 at 3.0 h for the 2.0 g dose group (Tables 2–4). The half-life and AUC0–8 h , respectively, in ACs were 2.61 h and 4.30 ␮g h/mL in the 0.5 g dose group and 2.17 h and 14.07 ␮g h/mL in the 1.0 g dose group (Table 5). The half-life and AUC in ACs for the 2.0 g dose were not calculated because of the BAL sample degradation in this group (see Section 2). The kinetics in ACs were non-linear. Doubling of the dose from 0.5 g to 1.0 g resulted in an approximate five-fold increase in Cmax and a three-fold increase in the AUC. The AC Cmax /MIC90 and AUC/MIC90 ratios and %T > MIC90 are summarised in Table 8.

Table 7 Epithelial lining fluid (ELF) pharmacodynamic parameters by dose group MIC90 (␮g/mL)

0.12 0.25 2.0 4.0

Cmax /MIC90 ratio

AUC/MIC90 ratio

% Dosing interval ELF drug concentration above MIC90

0.5 g

1.0 g

0.5 g

1.0 g

0.5 g

1.0 g

45 21 2.7 1.3

64 31 3.9 1.9

102 49 6 3

128 61 7.7 3.8

100 90 34 18

88 75 38 25

MIC90 , minimum inhibitory concentration required to inhibit 90% of the pathogen; Cmax , maximum plasma concentration; AUC, area under the curve for the mean concentration–time data. Table 8 Alveolar cell (AC) pharmacodynamic parameters MIC90 (␮g/mL)

0.12 0.25 2.0 4.0

Cmax /MIC90 ratio

AUC/MIC90 ratio

% Dosing interval AC drug concentration above MIC90

0.5 g

1.0 g

0.5 g

1.0 g

0.5 g

1.0 g

8.5 4 0.51 0.26

42.1 20.2 2.5 1.3

35.8 17.2 2.2 1.1

117 56.3 7.0 3.5

100 85 0 0

100 100 50 24

MIC90 , minimum inhibitory concentration required to inhibit 90% of the pathogen; Cmax , maximum plasma concentration; AUC, area under the curve for the mean concentration–time data.

J.E. Conte Jr et al. / International Journal of Antimicrobial Agents 26 (2005) 449–456

4. Discussion The linearity of the plasma pharmacokinetics of meropenem has previously been reported in studies in human subjects and this study confirms those observations. Doubling of the dose from 0.5 g to 1.0 g resulted in a two-fold increase and doubling from 1.0 g to 2.0 g resulted in a 2.4-fold increase in Cmax . Whilst the latter is somewhat more than an exact doubling, the number or observations was limited (N = 8 in the 2.0 g group) and the difference is within the limits of the methodology. However, this study has also demonstrated a less than proportional penetration of meropenem into ELF and greater than proportional penetration into ACs. Doubling of the dose from 0.5 g to 1.0 g resulted in a 44% and 25% increase in Cmax and AUC, respectively, in ELF, but a five- and threefold increase in Cmax and AUC in ACs. The disproportionate increased penetration of meropenem into ACs may be explained in part by in vitro experiments performed by Cuffini et al., who demonstrated concentration of meropenem into human macrophages [21]. At clinically relevant extracellular meropenem concentrations from 0.125–1 ␮g/mL, meropenem was concentrated 3–12-fold within human macrophages. This process was not temperature-, environment- or energy-dependent and appeared to be a passive process not requiring cell viability. These authors also demonstrated that intracellular meropenem remained active against intracellular S. aureus. The reason for the non-linear penetration of meropenem into the ELF is unknown and is not elucidated in this study. The ELF/plasma penetration ratios that we observed of 49–80% in the 0.5 g group and 32–53% in the 1.0 g group are similar to the ELF/plasma ratios previously reported [9; Drusano et al., Abstracts of the 44th Interscience Conference on Antimicrobial Agents and Chemotherapy, 2004]. The ELF/plasma ratios that we observed in the 2.0 g group were unexpectedly low (4.8% at 1 h and 21.9% at 3 h). In our study, the AC/plasma penetration ratios ranged from 9% at 1 h and 20.8% at 3 h in the 0.5 g dose group, to 26% at 1 h and 34% at 3 h in the 1.0 g dose group. In the 2.0 g dose group, the ratios at 1 h and 3.0 h were implausibly low, i.e. 0.8% and 0%, respectively, providing additional evidence that the BAL samples in the 2.0 g dose group had undergone unexplained degradation. The half-lives, Cmax values and AUCs (Table 5) that we observed in plasma were comparable with those previously reported. The plasma elimination half-lives in the 0.5 g and 1.0 g groups were 0.77 h and 1.31 h, respectively. The ELF half-lives were 1.51 h and 0.95 h and the AC half-lives were 2.61 h and 2.17 h in the 0.5 g and 1.0 g dose groups, respectively. Thus, there appears to be an approximate doubling of the half-life in ACs. The biological reason for and clinical implication of this finding are unknown. The MIC90 values used for the PK/PD calculations in this study were those reported in the literature. The calculated PK/PD indices are clinically relevant and are those

455

that are likely to be achieved in the treatment of nosocomial or community-acquired respiratory infection; however, these indices apply only for those organisms whose MIC90 values fall within the concentrations used in this study. The time above MIC for ␤-lactams and carbapenems is an important correlate of favourable outcome [22–25]. In a murine Pseudomonas thigh infection model treated with meropenem, a time above MIC of ≥23% was best correlated with favourable outcome [26]. Maximal bacterial cell killing is obtained when time above MIC exceeds 3.2 h, i.e. 40% of the 8 h dosing interval [27]. In our study, the time above MIC in plasma ranged between 45% (at MIC = 4.0 ␮g/mL) and 100% (at MIC = 0.12 ␮g/mL) for the 1.0 g and 2.0 g doses and between 28% (at MIC90 = 4.0 ␮g/mL) and 78% (at MIC90 = 0.12 ␮g/mL) for the 0.5 g dose. Thus, except for the 0.5 g dose versus the highest MICs (2 and 4 ␮g/mL), all of the times above MIC in plasma were greater than 40%. In ELF, the time above MIC ranged from 18% (at MIC = 4 ␮g/mL) to 100% (at MIC = 0.12 ␮g/mL) in the 0.5 g dose and from 25% (at MIC = 4.0 ␮g/mL) to 88% (at MIC = 0.12 ␮g/mL) in the 1.0 g dose. We did not calculate time above MIC in ELF in the 2.0 g dose because we believed that the values observed were not valid, as previously discussed. Thus, for the 0.5 g and 1.0 g doses, time above MIC was greater than 30% for all MIC values except 4 ␮g/mL. The clinical significance of these findings in ELF is unknown and requires further investigation. In ACs, the time above MIC ranged from 0% (at MIC = 4 ␮g/mL) to 100% (at MIC = 0.12 ␮g/mL) for the 0.5 g dose and from 24% (at MIC = 4.0 ␮g/mL) to 100% (at MIC = 0.12 ␮g/mL) in the 1.0 g dose. Again, we did not calculate time above MIC in ACs in the 2.0 g dose because we believed that the values observed were not valid. These data suggest that the 0.5 g dose might not be adequate for the treatment of respiratory infection due to organisms with MICs of ≥2.0 ␮g/mL and that the 1.0 g dose given every 8 h would not be adequate for organisms with MICs of ≥4.0 ␮g/mL. The significance of the intrapulmonary (ACs and ELF) Cmax /MIC and AUC/MIC ratios is less well understood. For the 0.5 g, 1.0 g and 2.0 g doses, the above ratios were all above unity for all MICs between 0.12 ␮g/mL and 4.0 ␮g/mL, indicating that the maximum meropenem concentrations were greater than the MICs of potential pathogens. For the 0.5 g dose, the values were above unity for all MICs in plasma and ELF. In ACs the values were below unity only for organisms with MICs greater than 2.0 ␮g/mL. The contribution of the Cmax /MIC and AUC/MIC ratios to clinical efficacy for meropenem is unknown. The prolonged time above MIC observed in this study supports an every 8 h dosing regimen of 1.0 g meropenem for the treatment of respiratory infections due to pathogens with MICs ranging from 0.12–2.0 ␮g/mL. For organisms with a MIC of ≥4 ␮g/mL, it is likely that a dose of 2 g every 8 h would yield pharmacodynamic parameters that would be better correlated with a successful outcome. Further investigation is warranted in this area.

456

J.E. Conte Jr et al. / International Journal of Antimicrobial Agents 26 (2005) 449–456

At all time periods, plasma meropenem concentrations were greater than those observed in ELF and ACs. This was true for the 0.5 g and 1.0 g dose groups. This suggests that meropenem was partially restricted from entry into the fluid (ELF) and cellular (ACs) compartments of the lung. The physiological basis for this differential penetration into lung is unknown. This study was not designed to measure the effect of protein binding on meropenem concentrations or the calculated pharmacodynamics. Our assay measured total (free and protein-bound) antibiotic concentrations in plasma, ELF and ACs. The fraction of free drug in these compartments was not determined. Since protein binding is concentrationdependent and the total drug concentrations varied widely between the compartments and with time, it is likely that free drug concentrations and the pharmacodynamic ratios were less than those that we have reported and that this effect would be greatest in serum. Further investigation is warranted to determine the effect of protein binding on the pharmacodynamics of meropenem in these compartments. Acknowledgments This work was carried out with funds provided by AstraZeneca Pharmaceuticals and by NIH grant #MO1RR00079 (General Clinical Research Center) of the University of California at San Francisco. The authors thank Patty C. Davis at AstraZeneca for performing the meropenem analyses, and Sinead Noonan for manuscript preparation.

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17] [18]

[19] [20]

References [1] Pfaller MA, Jones RN. A review of the in vitro activity of meropenem and comparative antimicrobial agents tested against 30,254 aerobic and anaerobic pathogens isolated world wide. Diagn Microbiol Infect Dis 1997;28:157–63. [2] Bui KQ, Ambrose PG, Nicolau DP, Lapin CD, Nightingale CH, Quintiliani R. Pharmacokinetics of high-dose meropenem in adult cystic fibrosis patients. Chemotherapy 2001;47:153–6. [3] Dandekar PK, Maglio D, Sutherland CA, Nightingale CH, Nicolau DP. Pharmacokinetics of meropenem 0.5 and 2 g every 8 hours as a 3-hour infusion. Pharmacotherapy 2003;23:988–91. [4] Hurst M, Lamb HM. Meropenem: a review of its use in patients in intensive care. Drugs 2000;59:653–80. [5] Thalhammer F, Traunmuller F, El Menyawi I, et al. Continuous infusion versus intermittent administration of meropenem in critically ill patients. J Antimicrob Chemother 1999;43:523–7. [6] Drusano GL, Hutchison M. The pharmacokinetics of meropenem. Scand J Infect Dis Suppl 1995;96:11–6. [7] Catchpole CR, Wise R, Thornber D, Andrews JM. In vitro activity of L-627, a new carbapenem. Antimicrob Agents Chemother 1992;36:1928–34. [8] Tomaselli F, Maier A, Matzi V, Smolle-Juttner FM, Dittrich P. Penetration of meropenem into pneumonic human lung tissue as

[21]

[22]

[23]

[24]

[25]

[26]

[27]

measured by in vivo microdialysis. Antimicrob Agents Chemother 2004;48:2228–32. Allegranzi B, Cazzadori A, Di Perri G, et al. Concentrations of single-dose meropenem (1 g iv) in bronchoalveolar lavage and epithelial lining fluid. J Antimicrob Chemother 2000;46:319–22. Conte Jr JE, Golden JA, McQuitty M, et al. Effects of gender, AIDS, and acetylator status on intrapulmonary concentrations of isoniazid. Antimicrob Agents Chemother 2002;46:2358–64. Conte Jr JE, Golden JA, Kipps J, Zurlinden E. Intrapulmonary pharmacokinetics of linezolid. Antimicrob Agents Chemother 2002;46:1475–80. Conte Jr JE, Golden JA, Kipps J, Lin ET, Zurlinden E. Effects of AIDS and gender on steady-state plasma and intrapulmonary ethambutol concentrations. Antimicrob Agents Chemother 2001;45: 2891–6. Conte Jr JE, Golden JA, McQuitty M, Kipps J, Lin ET, Zurlinden E. Single-dose intrapulmonary pharmacokinetics of rifapentine in normal subjects. Antimicrob Agents Chemother 2000;44:985– 90. Conte Jr JE, Golden J, Duncan S, McKenna E, Lin E, Zurlinden E. Single-dose intrapulmonary pharmacokinetics of azithromycin, clarithromycin, ciprofloxacin, and cefuroxime in volunteer subjects. Antimicrob Agents Chemother 1996;40:1617–22. Conte Jr JE, Golden JA, Duncan S, McKenna E, Zurlinden E. Intrapulmonary pharmacokinetics of clarithromycin and of erythromycin. Antimicrob Agents Chemother 1995;39:334–8. Conte Jr JE, Golden JA, Kipps J, Zurlinden E. Steady-state plasma and intrapulmonary pharmacokinetics and pharmacodynamics of cethromycin. Antimicrob Agents Chemother 2004;48:3508–15. Anonymous. Dietary Guidelines for Americans, 2000. United States Department of Agriculture; 2000. p. 1–44. Rennard SI, Basset G, Lecossier D, et al. Estimation of volume of epithelial lining fluid recovered by lavage using urea as marker of dilution. J Appl Physiol 1986;60:532–8. Talke HSGE. Enzymatische harnstoffbestimmung im blut und serum im optischem test nach warburg. Klin Wochenschr 1965;43:174–5. Mouton JW, Dudley MN, Cars O, Derendorf H, Drusano GL. Standardization of pharmacokinetic/pharmacodynamic (PK/PD) terminology for anti-infective drugs: an update. J Antimicrob Chemother 2005;55:601–7. Cuffini AM, Tullio V, Allocco A, Giachino F, Fazari S, Carlone NA. The entry of meropenem into human macrophages and its immunomodulating activity. J Antimicrob Chemother 1993;32: 695–703. Mattoes HM, Kuti JL, Drusano GL, Nicolau DP. Optimizing antimicrobial pharmacodynamics: dosage strategies for meropenem. Clin Ther 2004;26:1187–98. Drusano GL, Craig WA. Relevance of pharmacokinetics and pharmacodynamics in the selection of antibiotics for respiratory tract infections. J Chemother 1997;9(Suppl. 3):38–44. Kuti JL, Dandekar PK, Nightingale CH, Nicolau DP. Use of Monte Carlo simulation to design an optimized pharmacodynamic dosing strategy for meropenem. J Clin Pharmacol 2003;43:1116–23. Mouton JW, Touzw DJ, Horrevorts AM, Vinks AA. Comparative pharmacokinetics of the carbapenems: clinical implications. Clin Pharmacokinet 2000;39:185–201. Takata T, Aizawa K, Shimizu A, Sakakibara S, Watabe H, Totsuka K. Optimization of dose and dose regimen of biapenem based on pharmacokinetic and pharmacodynamic analysis. J Infect Chemother 2004;10:76–85. Drusano GL. Prevention of resistance: a goal for dose selection for antimicrobial agents. Clin Infect Dis 2003;36(Suppl. 1):S42–50.