Published Ahead of Print 24 March 2014. 10.1128/AAC.02564-13.

2014, 58(6):3191. DOI:Antimicrob. Agents Chemother. Evelyne Jacqz-Aigrain

andFrédérique Duquesne, Thomas Storme, André Baruchel Wei Zhao, Daolun Zhang, May Fakhoury, Mony Fahd, with Malignant Hematological DiseaseOptimization of Vancomycin in Children Population Pharmacokinetics and Dosing

http://aac.asm.org/content/58/6/3191Updated information and services can be found at:

These include:REFERENCES

http://aac.asm.org/content/58/6/3191#ref-list-1This article cites 23 articles, 8 of which can be accessed free at:

CONTENT ALERTS more»articles cite this article),

Receive: RSS Feeds, eTOCs, free email alerts (when new

http://journals.asm.org/site/misc/reprints.xhtmlInformation about commercial reprint orders: http://journals.asm.org/site/subscriptions/To subscribe to to another ASM Journal go to:

on May 15, 2014 by guest

http://aac.asm.org/

Downloaded from

on May 15, 2014 by guest

http://aac.asm.org/

Downloaded from

Population Pharmacokinetics and Dosing Optimization ofVancomycin in Children with Malignant Hematological Disease

Wei Zhao,a,b Daolun Zhang,a May Fakhoury,a Mony Fahd,c Frédérique Duquesne,c Thomas Storme,d André Baruchel,c,e

Evelyne Jacqz-Aigraina,b,e

Department of Pediatric Pharmacology and Pharmacogenetics, Hôpital Robert Debré, APHP, Paris, Francea; Clinical Investigation Center CIC1426, INSERM, Paris, Franceb;Department of Pediatric Hemato-Oncology, Hôpital Robert Debré, APHP, Paris, Francec; Department of Pharmacy, Hôpital Robert Debré, APHP, Paris, Franced; UniversitéParis Diderot, Sorbonne Paris Cité, Paris, Francee

An increase in vancomycin dose has been proposed in adults with malignant hematological disease. As pediatric data are limited,our aim was to evaluate the population pharmacokinetics of vancomycin in order to define the appropriate dosing regimen inchildren with malignant hematological disease. Vancomycin concentrations were collected prospectively during therapeuticmonitoring. Population pharmacokinetic analysis was performed using NONMEM software. Seventy children (age range, 0.3 to17.7 years) were included. With the current recommended dosing regimen of 40 to 60 mg/kg/day, 53 children (76%) had sub-therapeutic steady-state trough concentrations (Css/min of <10 mg/liter). A one-compartment model with first-order eliminationwas developed. Systematic covariate analysis identified that weight significantly influenced clearance (CL) and volume of distri-bution (V) with power functions of 0.677 for CL and 0.838 for V. Vancomycin CL also significantly increased with increases increatinine clearance and seemed to be higher in children with malignant hematological disease than in the general pediatric pop-ulation. The model was validated internally. Its predictive performance was further confirmed in an external validation byBayesian estimation. A patient-tailored dosing regimen was developed based on the final pharmacokinetic model and showedthat a higher proportion of patients reached the target Css/min than with the traditional mg/kg-basis dose (60% versus 49%) andthat the risks associated with underdosing or overdosing were reduced. This is the first population pharmacokinetic study ofvancomycin in children with malignant hematological disease. An optimized dosing regimen, taking into account patientweight, creatinine clearance, and susceptibility of the pathogens involved, could routinely be used to individualize vancomycintherapy in this vulnerable population.

Vancomycin is primarily effective against Gram-positive cocci.Staphylococcus aureus and Staphylococcus epidermidis, includ-

ing both methicillin-susceptible (MSSA and MSSE, respectively,for S. aureus and S. epidermidis) or methicillin-resistant (MRSAand MRSE, respectively, S. aureus and S. epidermidis) species, areusually sensitive to vancomycin. It is often included in the initialempirical antibiotic therapy in patients with malignant hemato-logical disease (1).

Vancomycin is administered intravenously. It is 25 to 50%protein bound, mainly to albumin and IgA, and is almost exclu-sively eliminated by the renal route. The pharmacokinetic-phar-macodynamic breakpoint of vancomycin was defined as the ratioof the area under the concentration-time curve from 0 to 24 h(AUC0 –24) to the MIC and is at least 400 h in adults with Staphy-lococcus aureus pneumonia (2).

The pharmacokinetics of vancomycin has shown large interin-dividual variability, primarily linked to a patient’s age, clinicalcondition, and disease (3, 4). Patients in oncology represent acritical population in whom inadequate empirical antibacterialtherapy may result in infection-related morbidity and increasedmortality. In addition, the pharmacokinetic parameters oftenpresent different characteristics than in nononcology patients(5–7), making optimization of the dosing regimen essential.Vancomycin was reported to have an increased clearance inadults with malignant hematological disease compared withadults without cancer (8, 9). In children with cancer, pharma-cokinetic data are limited (10, 11), and the optimal dosingregimen remains undefined. Innovative strategies, such asmodeling and simulation approaches, were initiated to indi-

vidualize dosing of vancomycin in neonates and adults basedon identified covariates (12). Thus, the objectives of the presentwork were to evaluate the population pharmacokinetics of van-comycin in children with malignant hematological disease andto optimize vancomycin therapy based on a defined pharma-cokinetic-pharmacodynamic breakpoint.

MATERIALS AND METHODSStudy population. Children with malignant hematological disease receiv-ing vancomycin as an intravenous infusion were included in the Depart-ment of Pediatric Hematolo-Oncology at Robert Debré Hospital between2010 and 2011 if at least one vancomycin serum concentration was as-sayed for therapeutic drug monitoring (TDM). The following data werecollected prospectively by a trained research assistant: age, weight, serumcreatinine concentration, and details of vancomycin administration (doseand infusion start and stop times), type of hematological disease, andbone marrow transplantation. Patients with incomplete dosing informa-tion were excluded. This study was designed in accordance with legalrequirements and the Declaration of Helsinki, registered at the Commis-sion Nationale Informatique et Liberté (CNIL), and approved by the local

Received 25 November 2013 Returned for modification 10 January 2014Accepted 16 March 2014

Published ahead of print 24 March 2014

Address correspondence to Wei Zhao, wei.zhao@rdb.aphp.fr, orEvelyne Jacqz-Aigrain, evelyne.jacqz-aigrain@rdb.aphp.fr.

Copyright © 2014, American Society for Microbiology. All Rights Reserved.

doi:10.1128/AAC.02564-13

June 2014 Volume 58 Number 6 Antimicrobial Agents and Chemotherapy p. 3191–3199 aac.asm.org 3191

on May 15, 2014 by guest

http://aac.asm.org/

Downloaded from

research ethics committee (Comité de l’Evaluation de l’Ethique des Pro-jets de Recherche Biomédicale [CEERB], Robert Debré Hospital, Paris,France).

Dosing regimen and sampling. Vancomycin (Sandoz, Levallois-Per-ret, France) was administered as an intravenous infusion over 60 min. Theempirically initial dosing regimen is 40 to 60 mg/kg/day in four divideddoses. Monitoring of vancomycin concentrations was performed in orderto maintain a steady-state trough concentration per minute (Css/min) ofbetween 10 and 20 mg/liter.

Assay of serum vancomycin and creatinine. The serum vancomycinconcentrations were determined by a fluorescence polarization immuno-assay method (FPIA) using the Cobas Integra 400 plus system (RocheDiagnostics, Meylan, France). The calibration curve ranged from 0.74 to80 mg/liter. The accuracy and coefficients of variation (CVs) of intralabo-ratory controls (7.2, 21.4, and 35.6 mg/liter) were 90% to 110% and!3.9%, respectively. The lower limit of quantification (associated CV)was 0.74 mg/liter (CV of !3.3%). Serum creatinine concentrations weremeasured by an enzymatic method using an Advia 1800 chemistry system(Siemens Medical Solutions Diagnostics, Puteaux, France).

Pharmacokinetic modeling. Pharmacokinetic analysis was carriedout using the nonlinear mixed-effects modeling program NONMEM,version 7.2.0 (Icon Development Solutions, Ellicott City, MD, USA). Thefirst-order conditional estimation (FOCE) method with the interactionoption was used to estimate the pharmacokinetic parameters and theirvariability.

Structure model. One- or two-compartment open models with first-order elimination were compared. The basic model was evaluatedthrough visual inspection of routine diagnostic plots. Interindividualvariability of the pharmacokinetic parameters was estimated using an ex-ponential model and was expressed as follows:

!i " !TV # e$i (1)

where "i represents the parameter value of the ith subject, "TV is the typicalvalue of the parameter in the population, and #i is the variability between

subjects, which is assumed to follow a normal distribution with a mean ofzero and variance of one.

A residual-variability (additive, proportional, exponential, or mixed)model was selected according to improvement of the objective functionvalue (OFV) and visual inspection of routine diagnostic plots.

Covariate analysis. The effects of age, weight, serum creatinine con-centration, creatinine clearance, and type of hematological disease (leu-kemia or lymphoma) were investigated as potential variables on pharma-cokinetic parameters. The stepwise covariate modeling and likelihoodratio test were used to test the effect of each variable.

The selection of covariates was based on a forward-backward processand biological plausibility. During forward selection, a covariate was se-lected if a significant (P ! 0.05, $2 distribution with 1 degree of freedom)decrease (reduction of %3.84) in the OFV from the basic model was ob-tained. At the end, all the significant covariates were added simultaneouslyinto a “full” model. The importance of each covariate was reassessed bybackward selection, and a covariate was independently removed from thefull model if the increase in the OFV was less than 7.88 (P ! 0.005, $2

distribution). The resulting model was considered the “final” populationpharmacokinetic model.

Model validation. Model validation was based on graphical and sta-tistical criteria. Goodness-of-fit plots, including observed (DV) versusindividual prediction (IPRED), DV versus population prediction(PRED), conditional weighted residuals (CWRES) versus time, andCWRES versus PRED, were used initially for diagnostic purposes (13).The stability and performance of the final model were also assessed bymeans of a nonparametric bootstrap with resampling and replacement.Resampling was repeated 500 times, and the values of estimated parame-ters from the bootstrap procedure were compared with those estimatedfrom the original data set. The entire procedure was performed in anautomated fashion, using the PsN module (14). The final model was alsoevaluated graphically and statistically by visual predictive checks (VPCs)and normalized prediction distribution errors (NPDE) (15). One thou-sand data sets were simulated using the final population model parame-ters. For the VPCs, the 5th, 50th, and 95th percentiles of the simulatedconcentrations were processed using the R platform, plotted againstelapsed time, and compared to observed concentrations. For a model inwhich random effects are well estimated, approximately 90% of the ob-served data are expected to be within the 5th to 95th prediction interval.NPDE results were summarized graphically by default as provided by the

TABLE 1 Baseline characteristics of 70 children

CharacteristicMean value(SD)a

Medianvalue Range

No. of patients 70

Gender (no.)Female 29Male 41

Bone marrow transplantation(no. of patients)

25

Samples (no.) 98Vancomycin dose (mg) 327 (203) 250 50–950Vancomycin dose (mg/kg) 13.0 (3.4) 13.6 7.0–31.5Vancomycin concn (mg/liter) 9.4 (5.3) 8.1 1.8–27.3Weight (kg) 25.7 (15.5) 20.2 5.6–71.0Age (yr) 6.8 (4.8) 5.6 0.3–17.7Serum creatinine (&mol/liter) 32 (17) 30 10–141Creatinine clearance (ml/min)b 199.8 (63.3) 191.0 48.7–457.0

Hematology disease (no. of patients)Acute lymphoblastic leukemia 40Acute myeloblastic leukemia 17Juvenile myelomonocytic

leukemia5

Lymphoma 5Other 3

a Values are means (SD) except where indicated as number (no.).b Creatinine clearance was calculated by the Schwartz formula.

TABLE 2 Covariate analysisa

Model or variable(s)PKparameter OFV 'OFV

Structure model 398.6Body wt CL 352.3 (46.3Age CL 368.8 (29.8Creatinine clearance CL 377.2 (21.4Serum creatinine CL 397.0 (1.6Type of hematological disease

(leukemia or lymphoma)CL 396.6 (2.0

Body wt, age, and creatinineclearance

CL 323.6 (75.0

Age and creatinine clearance CL 351.3 (47.3Body wt and creatinine clearance CL 326.0 (72.6Body wt V 363.2 (35.5Age V 364.4 (34.2Body wt and age V 357.9 (40.7Allometric modelb 314.4 (84.2Final modelc 312.8 (85.8a Body weight and creatinine clearance were identified as significant covariates (valuesin boldface).b Body weight on CL and V with allometric scaling and creatinine clearance on CL.c Body weight on CL and V and creatinine clearance on CL.

Zhao et al.

3192 aac.asm.org Antimicrobial Agents and Chemotherapy

on May 15, 2014 by guest

http://aac.asm.org/

Downloaded from

NPDE R package (version 1.2) (16) using (i) a Q-Q plot (where Q isquantile) of the NPDE and (ii) a histogram of the NPDE.

Given that the objective of the analysis was to use the final model forprediction purposes, the predictive performance of the developed model wasfurther evaluated in an independent group of children with malignanthematological disease. The individual concentrations were predictedby Bayesian estimation (MAXEVAL ) 0 in the estimation step, whereMAXEVAL is the maximum number of model evaluations that can beused) with NONMEM using the population pharmacokinetic parame-ters. The predictive performance was evaluated by calculating the predic-tion error (PE) and absolute prediction error (APE) using the followingequations (ABS is the absolute function):

PE (%) " (Bayesian estimated concentration% measured concentration) ⁄ measured concentration (2)

APE (%) " ABS !Bayesian estimated concentration% measured concentration"/measured concentration (3)

Dosing optimization based on a pharmacokinetic model. MonteCarlo simulations were performed using the parameter estimates ob-tained from the final model in order to define the optimal dosing regimenable to attain the target AUC/MIC of 400 h in about 50% of patients,under the assumption of a comparable safety profile.

Traditional pediatric dose (mg/kg basis) simulation approach. Inthe traditional approach, the pediatric dose of vancomycin was simulatedon a mg/kg basis according to different age groups. Thus, various mg/kgdosing regimens (40, 50, 60, 70, 80, 90, and 100 mg/kg) were simulated ineach pediatric group: infants (28 days to 23 months), children (2 to 11years), and adolescents (12 to 18 years). One thousand simulations wereperformed using the original data set, and AUC0 –24 and Css/min valueswere calculated for each patient simulation. The target attainment ratewas then calculated for each dosing regimen to define the optimal doseregimen in each pediatric group.

Patient-tailored dose. A patient-tailored dose was assessed to evaluatethe advantage of personalized therapy. In this simulation scenario, theindividual dose was calculated based on population pharmacokinetic pa-rameters and covariates in each patient as follows:

Optimized daily dosing i (mg ⁄ day) " target AUC0%24 # CLi (4)

where CLi is calculated using the equation developed from the model,and i stands for individual. The target AUC0 –24 was defined accordingto the pharmacokinetic-pharmacodynamic breakpoint of vancomy-cin: AUC0 –24/MIC of &400 h. For example, if the MIC was 1 mg/liter,the target AUC0 –24 should be at least 400 mg · h/liter.

The simulation process was similar to that described above for dosesimulation on a mg/kg basis. At the end, we compared the variability ofAUC0 –24 and Css/min values between the mg/kg basis and patient-tailoreddose.

RESULTSSerum concentrations of intravenous vancomycin infusion over60 min were monitored in 70 children (41 boys) with a mean ageof 6.8 years (standard deviation [SD], 4.8 years; range, 0.3 to 17.7years) and a mean weight of 25.7 kg (SD, 15.5 kg; range, 5.6 to 71.0kg). A total of 98 vancomycin concentrations were analyzed. Asummary of patient demographic and clinical characteristics ispresented in Table 1.

Blood samples were drawn at the median of 54 h after initiationof treatment. The concentrations ranged from 1.8 to 27.3 mg/liter.The number of patients in the concentration ranges (!5, 5 to 10,10 to 15, and %15 mg/liter) were 12 (17%), 41 (59%), 11 (16%),and 6 (9%), respectively.

Population pharmacokinetic modeling. A total of 98 vanco-mycin concentrations were available for population modeling.Data fitted a one-compartment model with first-order elimina-tion. Interindividual variability was best described by an exponen-tial model and was then estimated for V and CL. Residual variabil-ity was best described by a combined proportional and additivemodel.

The systematic covariate analysis identified body weight as themost important covariate implemented on V, which caused a sig-nificant drop in the OFV of 35.5 points. For clearance, body

TABLE 3 Population pharmacokinetic parameters of vancomycin and bootstrap results

ParameteraPK parametervalue RSE (%)b

Bootstrap value (n ) 500)

Median2.5th

percentile97.5th

percentile

Volume of distribution, V (liters)V ) "1 * (body weight/20.2)"2

"1 119 13.4 116 81 157"2 0.838 25.1 0.803 0.401 1.32

Clearance, CL (liters/h)CL ) "3 * (body weight/20.2)"4 * RF

"3 4.37 4.8 4.33 4.01 4.76"4 0.677 12.2 0.668 0.505 0.878

RF ) (CLcr/191)"5

"5 1.03 21.2 1.03 0.585 1.55

Interindividual variability (%)V 77.0 35.9 69.3 43.4 96.2CL 34.8 22.9 32.7 22.2 39.8

Residual variabilityProportional (%) 5.3 79.2 6.0 2.0 16.0Additive (mg/liter) 1.17 26.6 1.15 0.68 1.54

a RF, renal function; CLcr, creatinine clearance (ml/min) calculated by the Schwartz formula.b RSE, relative standard error.

Pharmacokinetics of Vancomycin in Children

June 2014 Volume 58 Number 6 aac.asm.org 3193

on May 15, 2014 by guest

http://aac.asm.org/

Downloaded from

Zhao et al.

3194 aac.asm.org Antimicrobial Agents and Chemotherapy

on May 15, 2014 by guest

http://aac.asm.org/

Downloaded from

weight was also identified as a significant covariate, causing a sig-nificant drop in the OFV of 46.3 points. The model was furthersignificantly improved by introducing creatinine clearance('OFV of 26.3 points) on CL (Table 2). Therefore, the influenceof covariates on CL and V was retained in the model as follows:

CLi " CLref # !WTi ⁄ WTref"!1 # !CLCRi ⁄ CLCRref"!2 (5)

Vi " Vref # !WTi ⁄ WTref"!3 (6)

where CLi and Vi are, respectively, the CL and V of the ithindividual, WTi and CLCRi are the weight and creatinine clear-ance of the ith individual, and WTref and CLCRref are the ref-erence weight and creatinine clearance values, respectively. Inour study, the reference weight and creatinine clearance werethe median values of our population, 20.2 kg and 191 ml/min.The exponents were estimated by the model.

After incorporation of all significant covariates, interindi-vidual variability decreased from 61.6 to 34.8% for CL and from117.9 to 77.0% for V. The shrinkage was 21% for CL and 23% forV. The final population pharmacokinetic parameters are given inTable 3.

Model diagnostics showed acceptable goodness-of-fit criteriafor the final model. As shown in Fig. 1A to D, population andindividual predictions are acceptable. In addition, the mean pa-rameter estimates resulting from the bootstrap procedure veryclosely agreed with the respective values from the final populationmodel, indicating that the final model is stable and can redeter-mine the estimates of population pharmacokinetic parameters.The results of 500 bootstrap replicates are summarized in Table 3.The NPDE distribution and histogram indicate that the assump-tion of normal distribution of the differences between individualpredictions and observed data is acceptable (P ) 0.45, Shapiro-Wilk test of normality) (Fig. 1E and F). No trends were observedon the diagnostic plots of NPDE versus time or predicted concen-trations. The VPC (Fig. 1G) shows that observed concentrationswere well predicted by the model (exact binomial test, 11.2% outof limits observed; 95% confidence interval, 5.7% to 19.2%). Boxplots of the distributions of observed and predicted Css/min from1,000 simulations derived from the final model are presented inFig. 1H. The median values of Css/min from the observed data andthe simulation as well as the interquartile range were similar, in-dicating acceptable predictive capability of the final model. Figure2 shows the relationship between individual vancomycin clear-ance and covariates (body weight and creatinine clearance). Thetypical CLs of patients weighting 20, 40, and 60 kg were 4.3, 6.9,and 9.1 liters/h, respectively.

The performance of the developed model was further evalu-ated in an independent group of 20 children with malignant he-matological disease with a mean age of 8.1 years (SD, 4.1 years;range, 2.4 to 17.2 years), a mean weight of 28.8 kg (SD, 11.6 kg;range, 13.9 to 62.0 kg), and a mean creatinine clearance of 191.3ml/min (SD, 61.2 ml/min; range, 84.0 to 350.7 ml/min; Schwartz

formula). Twenty-five concentrations consisting of peak, trough,and scavenged samples were available and ranged from 4.1 to 67.2mg/liter. The Bayesian estimated concentrations were highly cor-related with measured concentrations (r2 ) 0.99). The mean PEand APE were 1.0% (5th to 95th percentile, (4.7% to 8.7%) and4.7%, respectively, indicating a good predictive performance ofthe developed model on new patients. The estimates of populationpharmacokinetic parameters did not change obviously after inclu-sion of the validation data set into the model.

Dosing optimization based on pharmacokinetic model. (i)Traditional pediatric dose (mg/kg basis). The target attainmentrates as a function of dose and age groups for a standard MICsusceptibility breakpoint of 1 mg/liter are shown in Fig. 3. Thecurrent recommended dose of 60 mg/kg/day results in only 15%of infants and 24% of children and adolescents achieving the tar-

FIG 1 Model evaluation. (A and B) Routine diagnostic goodness-of-fit plots: population predicted (PRED) versus observed concentrations (DV) and individualpredicted (IPRED) versus observed concentrations (DV). (C and D) Conditional weighted residuals (CWRES) versus time and conditional weighted residuals(CWRES) versus population predicted concentrations (PRED). (E and F) Normalized prediction distribution errors (NPDE): Q-Q plot of the distribution of theNPDE versus the theoretical N(0, 1) distribution and a histogram of the distribution of the NPDE, with the density of the standard Gaussian distribution overlaid.(G) Visual predictive check (VPC). Circles represent the observed, while dashed lines depict the 5th and 95th percentiles of the simulated data. The solid linesindicate the median obtained from the simulated data. (H) Box plots of the distributions of observed (Obs) and predicted (Sim) Css/min values from 1,000simulations. The bold horizontal bars in the middle show the median values of vancomycin Css/min values, while the outer boundaries of the boxes represent theranges of the 25th and the 75th percentiles (interquartile ranges).

FIG 2 The relationship between vancomycin clearance and covariates. (A)Vancomycin clearance versus weight. (B) Vancomycin clearance (normalizedby size) versus creatinine clearance. Size is defined as (body weight/20.2)0.677.Dashed lines depict the typical covariate-CL relationship.

Pharmacokinetics of Vancomycin in Children

June 2014 Volume 58 Number 6 aac.asm.org 3195

on May 15, 2014 by guest

http://aac.asm.org/

Downloaded from

get AUC/MIC. These simulated values were in agreement with theobserved values in the present study. In fact, 76% of our patientshad a risk of underdosing (Css/min of !10 mg/liter). To reach thetarget AUC/MIC of 400 h in about 50% of patients, 90 mg/kg/daywas required for infants, and 80 mg/kg/day was required for chil-dren and adolescents (Fig. 3). The proportion of patients withrisks of overdosing (Css/min of %20 mg/liter) was 26% for the newproposed dosing regimen (Fig. 3).

(ii) Patient-tailored dose. A patient-tailored dose was calcu-lated for each patient based on equations 4 and 7:

CL " 4.37 # !body weight ⁄ 20.2"0.677

# !creatinine clearance ⁄ 191"1.03 (7)

where body weight is in kg and creatinine clearance is in ml/min(Schwartz formula).

The expected AUC0 –24 and Css/min values in simulated trials forinfants receiving 90 mg/kg/day and in children and adolescentsreceiving 80 mg/kg/day in comparison to a patient-tailored dose

are shown in Fig. 4. The proportion of patients achieving the targetCss/min (10 to 20 mg/liter) is 60% using the patient-tailored dose,which is higher than the percentage for the traditional mg/kg-basis dose (49%). The proportions of patients with risks of under-dosing (Css/min of !10 mg/liter) or overdosing (Css/min of %20mg/liter) are 20% and 20%, respectively, using the patient-tai-lored dose, which is lower than values for the traditional mg/kg-basis dose (24% for underdosing and 26% for overdosing).

DISCUSSIONIn the present study, the population pharmacokinetics of vanco-mycin was evaluated for the first time in children with malignanthematological disease. The high variability of vancomycin con-centrations emphasizes the need for dosing optimization andTDM-based dosage adaptation in this high-risk population.

The population pharmacokinetic model offered a practicaltool to optimize vancomycin dosing in children with malignanthematological disease. The rational dosing of antimicrobials

FIG 3 Target attainment rates and fraction of patients above maximum recommended Css/min. The target attainment rates and fraction of patients above themaximum recommended C0 for the 100 simulated trials are presented as a function of dose and age group. The AUC/MIC target is 400 h; the MIC is 1 mg/liter.The maximum recommended Css/min is 20 mg/liter.

Zhao et al.

3196 aac.asm.org Antimicrobial Agents and Chemotherapy

on May 15, 2014 by guest

http://aac.asm.org/

Downloaded from

should take into account pharmacokinetics in the target popula-tion, the relationship between drug exposure and outcome, andthe susceptibility of the pathogens involved. In children, the stan-dard dose evaluation studies of antimicrobials are usually basedon a “nonselected” pediatric population and do not take into ac-count the potential impact of the disease and disease state, whichare the main factors that ultimately influence drug exposure in theclinical setting (17). For the purposes of comparison, the demo-graphics and pharmacokinetic parameters of vancomycin ob-tained from different pediatric studies (10, 18, 19) are summa-rized in Table 4. The magnitude of the differences provides strongargument for studying pharmacokinetics in selected subgroups of

patients. Indeed, the estimated vancomycin CL in children withmalignant hematological disease was higher than that reported ingeneral nonselected pediatric populations. Such an impact of he-matological malignancies was previously reported with vancomy-cin (8, 9) and additional antimicrobials in adult patients (5–7).

According to regulatory guidelines (20, 21), vancomycin is agood example of a drug for which the modeling and simulationapproach can be used to establish optimal dosage recommen-dations in children. The pharmacokinetic-pharmacodynamicbreakpoint of vancomycin was defined as an AUC0 –24/MIC ratioof at least 400 h in adults with Staphylococcus aureus pneumonia(2), in which the MIC for of methicillin-resistant Staphylococcus

FIG 4 AUC and Css/min distribution. Simulated vancomycin AUC and Css/min distribution in infants and in children and adolescents receiving the mg/kg-basisdose and patient-tailored dose.

TABLE 4 Pharmacokinetics of vancomycin in children

Parameter

Value for the group (reference or source)a

Pediatric patients with vancomycin TDM according to:Children withcancer (10)

Children with malignanthematologic disease(this study)Reference 18 Reference 19 Reference 10

No. of patients 78 6 31 33 70Age (yr) 7b 6.9 6.8 (0.3–17.7)Wt (kg) 25b 21.2 25.7 (5.6–71.0)CL (liters/h/kg) 0.103 0.11 (0.08–0.14) 0.114 0.149 0.22 (0.04–0.73)a Values are expressed as mean (range) except where indicated.b Median value.

Pharmacokinetics of Vancomycin in Children

June 2014 Volume 58 Number 6 aac.asm.org 3197

on May 15, 2014 by guest

http://aac.asm.org/

Downloaded from

aureus was 0.5 mg/liter in 76% (n ) 28) of infected patients and 1mg/liter in 24% (n ) 9). The pharmacokinetic-pharmacody-namic breakpoint is postulated to be similar in children and istherefore used for dosing optimization in children.

The current recommended pediatric dose of vancomycin is 40to 60 mg/kg/day (i.e., on a classical mg/kg-basis dose) and resultedin a high risk of underdosing in our target population, as 76%patients had Css/min of !10 mg/liter. The dose simulation was inagreement with this observation and showed that the dose of 60mg/kg/day resulted in 15% of infants and 24% of children andadolescents achieving the target AUC/MIC. Therefore, in order toincrease the likelihood of treatment, the vancomycin dose needsto be increased in children with malignant hematological disease.In the present study, the simulation approach demonstrated thatthe doses of 90 mg/kg/day for infants and of 80 mg/kg for childrenand adolescents allow nearly 50% of patients to achieve the AUC/MIC target, with the standard MIC susceptibility breakpoint of 1mg/liter which was clinically validated primarily in adult pneumo-nia studies in MRSA. However, according to recent data, there is atrend to a decrease in vancomycin efficacy, linked to the increasein the vancomycin MIC during MRSA infection (22, 23). Indeed,to obtain similar efficacy, treatment of MSRA with a higher MICwould require a higher AUC. An increase in vancomycin CL to-gether with creep in MIC values will require an increase in thevancomycin daily dosage to overcome the risk of therapeutic fail-ure. However, until now, the pharmacokinetic parameters andsafety of increased vancomycin dosing have not been available.The dosing simulation was based on the assumption of liner phar-macokinetics. Extrapolation of the dosing regimen outside theobserved range cannot preclude the possibility of nonlinear phar-macokinetics and should be evaluated in further study. In addi-tion, when increased vancomycin doses are administered, the in-fusion-related adverse events (i.e., “red man syndrome”) need tobe evaluated, and slowing the rate of infusion so that it exceeds 2 hshould be recommended.

One of the main advantages of modeling and simulation ap-proaches is to optimize personalized therapy. Therefore, in thesecond simulation scenario, we evaluated the patient-tailoreddose. Traditionally, the pediatric dose is defined on a mg/kg basisaccording to the different age groups, as we did in the first simu-lation scenario. This approach assumes an “average child” with an“average weight” in each age group (neonates, infants, children,and adolescents) and a standard mg/kg dose is calculated accord-ingly. However, both developmental factors and clinical condi-tions have a profound impact on pharmacokinetics. Many demo-graphic, biological, and clinical covariates are known to influencevancomycin pharmacokinetics. In our data, vancomycin clear-ance increased with body weight and creatinine clearance, show-ing that in children the standard weight-based dosing (mg/kg) isnot adapted to the whole range of pediatric ages. This could beexplained by the nonlinear correlation between body weight anddevelopmental changes in vancomycin clearance (24). Further-more, the individual dose needs optimization according to abnor-mal renal function. Therefore, weight and renal function have themost important impact on vancomycin pharmacokinetics in chil-dren (3), providing the scientific basis for rational patient-tailoreddosing schemes. The simulation clearly supports the use of a pa-tient-tailored dose, which showed a narrow spread between AUCand Css/min values compared with the traditional mg/kg-basisdose. The patient-tailored dose resulted in a higher proportion of

patients within the target trough concentrations, associated withlower risk of underdosing or overdosing. Individual dosage pre-scription and adjustment based on these covariates will require acomputer-based tool. This practice has been set up in our neonatalintensive care unit (NICU) to individualize vancomycin dosing inneonates (25) and undoubtedly will be implemented for the treat-ment of pediatric patients with selected diseases.

A limitation of our study was that the population pharmaco-kinetic model was developed based on vancomycin TDM data; theprediction of maximum concentration at steady state will likelyshrink toward population mean values, reducing the ability toidentify a “special” patient. Taking scavenged pharmacokineticsamples should improve individual parameter estimation withoutincreasing the burden of pediatric clinical practice.

Conclusion. In the present study, we developed a populationpharmacokinetic model of vancomycin in children with malig-nant hematological disease. Vancomycin clearance was markedlyhigher than that in children without cancer, with weight and cre-atinine clearance being significant covariates. We have shown thatthe patient-tailored dose reduced variability in vancomycin AUCand Css/min values compared to the mg/kg-basis dose, making themodeling approach an important tool for dosing individualiza-tion. Explaining residual variability requires the identification ofadditional covariates. A prospective study is warranted to evaluatethe potential clinical benefits and safety of this optimized dosingregimen.

ACKNOWLEDGMENTSWe acknowledge the technicians (Christel Grondin, Michel Popon,Samira Benakouche, and Yves Médard) for technical support.

This work was supported by the project Global Research in Pediat-rics—Network of Excellence (European Union-funded FP7 project, grantagreement number 261060).

We declare that we have no conflicts of interest related to this work.

REFERENCES1. Hughes WT, Armstrong D, Bodey GP, Bow EJ, Brown AE, Calandra T,

Feld R, Pizzo PA, Rolston KV, Shenep JL, Young LS. 2002. 2002guidelines for the use of antimicrobial agents in neutropenic patients withcancer. Clin. Infect. Dis. 34:730 –751. http://dx.doi.org/10.1086/339215.

2. Moise-Broder PA, Forrest A, Birmingham MC, Schentag JJ. 2004.Pharmacodynamics of vancomycin and other antimicrobials in patientswith Staphylococcus aureus lower respiratory tract infections. Clin. Phar-macokinet. 43:925–942. http://dx.doi.org/10.2165/00003088-200443130-00005.

3. Marsot A, Boulamery A, Bruguerolle B, Simon N. 2012. Vancomycin: areview of population pharmacokinetic analyses. Clin. Pharmacokinet. 51:1–13. http://dx.doi.org/10.2165/11596390-000000000-00000.

4. Zhao W, Kaguelidou F, Biran V, Zhang D, Allegaert K, Capparelli E,Holford N, Kimura T, Lo YL, Peris JE, Thomson A, van den Anker JN,Fakhoury M, Jacqz-Aigrain E. 2013. External evaluation of populationpharmacokinetic models of vancomycin in neonates: the transferability ofpublished models to different clinical settings. Br. J. Clin. Pharmacol. 75:1068 –1080. http://dx.doi.org/10.1111/j.1365-2125.2012.04406.x.

5. Bubalo JS, Munar MY, Cherala G, Hayes-Lattin B, Maziarz R. 2009.Daptomycin pharmacokinetics in adult oncology patients with neutro-penic fever. Antimicrob. Agents Chemother. 53:428 – 434. http://dx.doi.org/10.1128/AAC.00943-08.

6. Romano S, Fdez de Gatta MM, Calvo MV, Caballero D, Dominguez-GilA, Lanao JM. 1999. Population pharmacokinetics of amikacin in patientswith haematological malignancies. J. Antimicrob. Chemother. 44:235–242. http://dx.doi.org/10.1093/jac/44.2.235.

7. Zeitany RG, El Saghir NS, Santhosh-Kumar CR, Sigmon MA. 1990.Increased aminoglycoside dosage requirements in hematologic malig-nancy. Antimicrob. Agents Chemother. 34:702–708. http://dx.doi.org/10.1128/AAC.34.5.702.

Zhao et al.

3198 aac.asm.org Antimicrobial Agents and Chemotherapy

on May 15, 2014 by guest

http://aac.asm.org/

Downloaded from

8. Buelga DS, del Mar Fernandez de Gatta M, Herrera EV, Dominguez-GilA, García MJ. 2005. Population pharmacokinetic analysis of vancomycinin patients with hematological malignancies. Antimicrob. Agents Che-mother. 49:4934 – 4941. http://dx.doi.org/10.1128/AAC.49.12.4934-4941.2005.

9. Fernández de Gatta Mdel M, Santos Buelga D, Sánchez Navarro A,Dominguez-Gil A, García MJ. 2009. Vancomycin dosage optimization inpatients with malignant haematological disease by pharmacokinetic/pharmacodynamics analysis. Clin. Pharmacokinet. 48:273–280. http://dx.doi.org/10.2165/00003088-200948040-00005.

10. Chang D. 1995. Influence of malignancy on the pharmacokinetics ofvancomycin in infants and children. Pediatr. Infect. Dis. J. 14:667– 673.http://dx.doi.org/10.1097/00006454-199508000-00004.

11. Piro CC, Crossno CL, Collier A, Ho R, Koyama T, Frangoul H. 2009.Initial vancomycin dosing in pediatric oncology and stem cell transplantpatients. J. Pediatr. Hematol. Oncol. 31:3–7. http://dx.doi.org/10.1097/MPH.0b013e31818b3520.

12. Jacqz-Aigrain E, Zhao W, Sharland M, van den Anker JN. 2013. Use ofantibacterial agents in the neonate: 50 years of experience with vancomy-cin administration. Semin. Fetal. Neonatal. Med. 18:28 –34. http://dx.doi.org/10.1016/j.siny.2012.10.003.

13. Hooker AC, Staatz CE, Karlsson MO. 2007. Conditional weighted re-siduals (CWRES): a model diagnostic for the FOCE method. Pharm. Res.24:2187–2197. http://dx.doi.org/10.1007/s11095-007-9361-x.

14. Lindbom L, Ribbing J, Jonsson EN. 2004. Perl-speaks-NONMEM(PsN)-a Perl module for NONMEM related programming. Comput.Methods Programs Biomed. 75:85–94. http://dx.doi.org/10.1016/j.cmpb.2003.11.003.

15. Brendel K, Comets E, Laffont C, Laveille C, Mentré F. 2006. Metrics forexternal model evaluation with an application to the population pharma-cokinetics of gliclazide. Pharm. Res. 23:2036 –2049. http://dx.doi.org/10.1007/s11095-006-9067-5.

16. Comets E, Brendel K, Mentré F. 2008. Computing normalised predictiondistribution errors to evaluate nonlinear mixed-effect models: the npdeadd-on package for R. Comput. Methods Programs Biomed. 90:154 –166.http://dx.doi.org/10.1016/j.cmpb.2007.12.002.

17. Zhao W, Fakhoury M, Fila M, Baudouin V, Deschênes G, Jacqz-Aigrain

E. 2012. Individualization of valganciclovir prophylaxis for cytomegalo-virus infection in pediatric kidney transplant patients. Ther. Drug Monit.34:326 –330. http://dx.doi.org/10.1097/FTD.0b013e3182509e3a.

18. Lamarre P, Lebel D, Ducharme MP. 2000. A population pharmacoki-netic model for vancomycin in pediatric patients and its predictive valuein a naive population. Antimicrob. Agents Chemother. 44:278 –282. http://dx.doi.org/10.1128/AAC.44.2.278-282.2000.

19. Wrishko RE, Levine M, Khoo D, Abbott P, Hamilton D. 2000. Vancomycinpharmacokinetics and Bayesian estimation in pediatric patients. Ther. DrugMonit. 22:522–531. http://dx.doi.org/10.1097/00007691-200010000-00004.

20. European Agency for the Evaluation of Medicinal Products. 2000. Points toconsider on pharmacokinetics and pharmacodynamics in the development ofantibacterial medicinal products. Document CPMP/EWP/2655/99. Euro-pean Medicines Agency, London, United Kingdom. www.ema.europa.eu/ema/pages/includes/document/open_document.jsp?webContentId)WC500003420.

21. FDA. 2003. Guidance for industry: exposure-response relationships—studydesign, data analysis, and regulatory applications. Food and Drug Adminis-tration, Rockville, MD. http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/ucm072109.pd.

22. Ho PL, Lo PY, Chow KH, Lau EH, Lai EL, Cheng VC, Kao RY. 2010.Vancomycin MIC creep in MRSA isolates from 1997 to 2008 in a health-care region in Hong Kong. J. Infect. 60:140 –145. http://dx.doi.org/10.1016/j.jinf.2009.11.011.

23. Steinkraus G, White R, Fredrich L. 2007. Vancomycin MIC creep innon-vancomycin-intermediate Staphylococcus aureus (VISA), vancomy-cin-susceptible clinical methicillin-resistant S. aureus (MRSA) blood iso-lates from 2001–2005. J. Antimicrob. Chemother. 60:788 –794. http://dx.doi.org/10.1093/jac/dkm258.

24. Anderson BJ, Allegaert K, Van den Anker JN, Cossey V, Holford NH.2007. Vancomycin pharmacokinetics in preterm neonates and the predic-tion of adult clearance. Br. J. Clin. Pharmacol. 63:75– 84. http://dx.doi.org/10.1111/j.1365-2125.2006.02725.x.

25. Zhao W, Lopez E, Biran V, Durrmeyer X, Fakhoury M, Jacqz-AigrainE. 2013. Vancomycin continuous infusion in neonates: therapeutic drugmonitoring and dosing optimization. Arch. Dis. Child. 98:449 – 453. http://dx.doi.org/10.1136/archdischild-2012-302765.

Pharmacokinetics of Vancomycin in Children

June 2014 Volume 58 Number 6 aac.asm.org 3199

on May 15, 2014 by guest

http://aac.asm.org/

Downloaded from