Age-Related Effects on Nelfinavir and M8 Pharmacokinetics: a Population Study with 182 Children

As a relationship between nelfinavir antiretroviral efficacyand plasma concentrations has been previously established, nelfinavirpharmacokinetics was investigated in order to optimize the individualtreatment schedule in a pediatric population. A population pharmacokineticmodel was developed to describe the concentration-time courseof nelfinavir and its active metabolite M8. Individual characteristicswere used to explain the large interindividual variability inchildren. Data from therapeutic drug monitoring in 182 childrentreated with nelfinavir were analyzed with NONMEM. Then Foodand Drug Administration (FDA) current recommendations were evaluatedestimating the percentage of children who reached the targetminimum plasma concentration (0.8 mg/liter) by using Bayesianestimates. Nelfinavir pharmacokinetics was described by a one-compartment model with linear absorption and elimination. Pharmacokineticestimates and the corresponding intersubject variabilities forthe model were as follows: nelfinavir total clearance, 0.93liters/h/kg (39%); volume of distribution, 6.9 liters/kg (109%);absorption rate, 0.5 h^–1; formation clearance fractionto hydroxy-tert-butylamide (M8), 0.025; M8 elimination rate,1.88 h^–1 (49%). Apparent nelfinavir total clearance andvolume of distribution decreased as a function of age. M8 eliminationrate was increased by concomitant administration of nevirapineor efavirenz. Our data confirm that the FDA recommendationsfor children from 2 to 13 years are optimal and that the doserecommended for children younger than 2 years is adequate forthe children from 2 months to 2 years old. However, in childrenyounger than 2 months, the proposed nelfinavir newborn doseof 40 mg/kg of body weight twice daily is inadequate and wesuggest increasing the dose to 50 to 60 mg/kg administered thricedaily. This assumption should be further evaluated.

INTRODUCTION

Top

Introduction

Nelfinavir is a protease inhibitor commonly used in a highlyantiretroviral therapy to treat human immunodeficiency virustype 1 (HIV-1) infection. In children, the antiretroviral efficacyobtained with this drug was disappointing (18). However, thisdrug is still recommended as first-line therapy because it iswell tolerated by children and few alternatives are available.Optimization of administration schedule, due to a better understandingof nelfinavir pharmacokinetics in children, could help at improvingthe efficacy of this drug.

Nelfinavir is metabolized via CYP2C19 (cytochrome P450 isoenzyme)in the liver into an active metabolite, hydroxy-tert-butylamide(M8), which is in turn metabolized via CYP3A4 (15). It has beenreported that 2- to 13-year-old children demonstrate a two-to threefold increase in apparent oral clearance of nelfinavircompared to adult values (13) and that children younger than2 years have lower nelfinavir concentrations (and higher apparentclearance) than older children (7). However, most studies wereperformed in small and heterogeneous groups of children andshowed a very large variability in drug pharmacokinetic parameters(5, 8, 25).

The aim of the present study was to characterize the nelfinavirand M8 pharmacokinetics in children. This was achieved by (i)developing an integrated pharmacokinetic model to simultaneouslydescribe the nelfinavir and M8 pharmacokinetics and (ii) usinga pharmacostatistic model to identify the patient characteristicsthat can influence nelfinavir and M8 pharmacokinetics. Suchresults should be useful to optimize nelfinavir treatment sincea significant relationship was demonstrated between nelfinavirantiretroviral efficacy/safety and minimum plasma concentration(6, 21, 24).

MATERIALS AND METHODS

Top

Materials and Methods

Patients. The population included children receiving oral nelfinavir fortreatment of HIV infection and whose antiretroviral drug plasmaconcentrations were monitored on a routine basis. Nelfinavirwas administered according to body weight (BW), only as 250-mgtablets (never as powder), and tablets were crumbled in a smallvolume of water and added to milk or food when children couldnot swallow them. Nelfinavir powder was not used because ofthe large volume to administer, the unpleasant consistency,and the difficulties in dissolving it in milk or food (18).

Except newborns, children were outpatients. For outpatients,the dosing information was obtained by the clinician from thepatient or the parents. For each child, if the time elapsedbetween drug administration and blood sampling times was lessthan 13 h, gender, BW, height, body mass index (BMI), body surfacearea (BSA), and age were carefully recorded, as well as combinedtreatments, particularly of other antiretroviral drugs. Whenlow compliance was suspected by the clinician or by the pharmacologist(undetectable plasma concentrations of nelfinavir and M8), thedata were not included in the analysis. When sample time wasgreater than 13 h and also nonadherence or error on the lastadministration time was highly probable, the records were notincluded in the data set. This included only 16 plasma samples(<2%). Ethics committee approval and patient consent arenot compulsory in France in order to use therapeutic drug monitoringdata, and thus they were not obtained.

Analytical method. Nelfinavir and M8 plasma concentrations were measured by high-performanceliquid chromatography. Briefly, the method involved the extractionof the drugs and the internal standard (clazepam SL 72469) from200 µl of plasma with a 6-ml mixture of ethyl acetate-hexane(vol/vol) in alkaline medium (0.5 ml of 0.2 M sodium carbonate).After evaporation, the residue was dissolved in eluent consistingof acetonitrile-perchlorate tetramethyl ammonium (0.01 M) intrifluoroacetic acid (0.01%) (37:63, vol/vol). Chromatographywas performed using a reverse-phase-phase C₈ analytical column(Nucleosil C8 125*, 4.6 mm by 3 µm; Macherey-Nagel) andgradient elution with an increase of acetonitrile from 37 to45%. UV detection at 205 nm was used. Linearity of the methodwas obtained in the concentration range of 0.2 to 20 mg/literand 0.05 to 8 mg/liter for nelfinavir and M8, respectively.Based on standard samples, interday accuracy for the two analytesranged from 92.9 to 97.6%, and based on quality control samples,interday precision expressed as percent coefficient of variationwas less than 10%.

Population pharmacokinetic modeling of nelfinavir and M8. Concentrations that were beyond the limit of quantificationwere set to half the limit of quantification (i.e., 0.1 µg/ml)(3).

Data were analyzed using the nonlinear mixed-effect modelingsoftware program NONMEM (version V, level 1.1, double precision)with the DIGITAL FORTRAN compiler (4). The first-order conditionalestimation with the Interaction option was used. The pharmacokineticsof nelfinavir and M8 were studied sequentially. Nelfinavir datawere first analyzed according to a one-compartment open model.Nelfinavir concentrations versus time were fitted using theNONMEM subroutine ADVAN2 TRANS2. Parameters of the model werethe absorption rate constant (K_a), distribution volume (V),and elimination clearance (CL_T). The mean population parametersof the parent compound were then used to produce the input functioninto the metabolite compartment, allowing a first estimationof the M8 parameters. Then, all parameters of the nelfinavir-M8system were estimated. Parameters of the nelfinavir-M8 modelwere K_a, V of nelfinavir, nelfinavir total clearance (CL_T),M8 apparent formation clearance fraction (F_MT), and M8 eliminationrate constant (K_M0) (Fig. 1). The M8 distribution volume wasnot identifiable. The equation for the metabolite pharmacokineticsis derived in the Appendix. All clearance and volume terms wereapparent parameters, i.e., V/F, CL_T/F etc., where F is the bioavailabilityfraction.

View larger version (11K):
[in this window]
[in a new window]
FIG. 1. Schematic representation of the pharmacokinetic compartment model for the simultaneous prediction of nelfinavir and plasma M8 concentration after nelfinavir oral administration. Nelfinavir (in compartment 1) underwent irreversible biotransformation into M8 (in compartment 2).

Several error models (i.e., exponential, proportional, and additiveerror models) to describe intersubject (ISV) and residual variabilitieswere investigated.

The influence of each patient covariate was systematically testedvia a generalized additive model according to the followingequation, using CL_T, for example:

Formula 1 (1)
where TV(CL_T) is the typical value of clearancefor a patient with the median covariate value and ${theta}$ _AGE is theestimated influential factor for age.

Categorical covariates, including gender and combined antiretroviraldrugs, were tested using an inducing drug effect, for example:

Formula 2 2
or, in the case of an inhibitorydrug effect,

Formula 3 (3)
where ${theta}$ _DRUG is the estimated influential factor for the drug effect.

Covariates were selected in the final population model if (i)their effect was biologically plausible, (ii) they produceda minimum reduction of 7 (P < 0.01, one degree of freedom)in the objective function value (OFV), and (iii) they produceda reduction in the variability of the pharmacokinetic parameter,assessed by the associated intersubject variability. An intermediatemultivariate model was then obtained including all significantcovariates. In order to keep only those covariates with thelargest contribution in the final multivariate model, a changeof 11 (P < 0.001, one degree of freedom) of the objectivefunction was required for the retention of a single parameterduring backward stepwise multiple regression analysis.

For evaluation of the goodness of fit, graphs of observed concentrationsversus predictions (PRED), weighted residuals versus time, andweighted residuals versus PRED, as well as the correspondinggraphs using individual predictions, were compared. Diagnosticgraphics and distribution statistics were obtained using theR program (12).

Bootstrap validation. The accuracy and robustness of the final population model wereassessed using a bootstrap method, as previously described indetail (19). Briefly, this included the following steps: (i)from the original data set of n individuals, B bootstrap sets(B = 1,000) of n individuals were drawn with replacement (resampling);(ii) for each of the B bootstrap sets, the population pharmacokineticparameters were estimated; (iii) with the B estimates of eachpopulation pharmacokinetic parameter, the corresponding meanand standard deviation were estimated; and (iv) to validatethe model, the parameters estimated from the bootstrap neededto be close to estimates obtained from the original populationset.

The entire procedure was performed in an automated fashion usingWings for NONMEM (10). This procedure also provided nonparametricstatistics (median and 2.5th and 97.5th percentiles) of thepopulation parameters.

Individual minimum plasma concentrations. Individual pharmacokinetic parameters using the Posthoc optionof NONMEM were used to calculate the daily dosage to obtaina minimum plasma concentration of 0.8 mg/liter (6). Three homogeneousgroups of children were distinguished by age: younger than 2months, 2 months to 2 years, and 2 to 13 years. Then the dailydosage to obtain a minimum plasma concentration of 0.8 mg/literwas calculated in the three age groups as if the daily dosewas given both ways: every 12 h (i.e., twice daily [BID]) andevery 8 h (i.e., thrice daily [TID]). For each category of ageand regimen, a cumulated curve was drawn to show immediatelyfor a given daily dose regimen which percentage of childrenwould have a minimum plasma concentration above 0.8 mg/liter.Current Food and Drug Administration (FDA) nelfinavir dose recommendations(11), depending on drug regimen (every 8 or 12 h), were evaluatedin each of the three age groups.

RESULTS

Top

Results

Demographic data. A total of 182 children (95 boys, 87 girls) were available forpharmacokinetic evaluation. Table 1 summarizes the dosage regimens.A total of 742 nelfinavir concentrations (a median of threesamples per patient) and 557 M8 concentrations (a median oftwo per patient) were collected. Sampling times were as follows:median was 3.5 h, 50% were between 2.8 and 4.9 h, 2.5% were<1.2 h, and 2.5% were between 12 and 13 h (Fig. 2). Medianage was 8.2 years (from 3 days to 17 years), 25 children wereyounger than 2 months, 36 were 2 months to 2 years old, and121 were older than 2 years. Median body weight was 21 kg (from1.7 to 70 kg). Coadministered antiretroviral drugs known toinfluence plasma nelfinavir concentrations were efavirenz (n= 10, 53 samples), nevirapine (n = 33, 133 samples), ritonavir(n = 3, 11 samples), and saquinavir (n = 10, 48 samples). Thesubjects who took ritonavir were included in a first analysis,but the number of subjects was too small, producing an unstablemodel, so they were excluded in the final analysis. Plasma sampleswere generally collected at steady state, which means afterat least 10 daily administrations of nelfinavir. Eighteen plasmasamples were not at steady state, since they were obtained inchildren less than 10 days old.

View this table:
[in this window]
[in a new window]
TABLE 1. Dosage regimens

View larger version (23K):
[in this window]
[in a new window]
FIG. 2. Graphic of distribution of sample collection times (frequency versus time after dose).

Population pharmacokinetics. (i) Nelfinavir pharmacokinetic model building. A one-compartment pharmacokinetic model adequately describedthe data. Intersubject and residual variabilities were bestdescribed by exponential and additive error models, respectively.The available data were not sufficient to estimate an intersubjectvariability for K_a, and exclusion of this random effect hadno influence on the objective function value. Body size descriptors,BW, BSA, and BMI, had significant effects on CL_T, resultingin an at least 44-U decrease of OFV. BW had the most significanteffect, with a 63-U decrease of OFV. Because CL_T was proportionalto BW and nelfinavir dosing is based upon BW, this effect wasthen fixed in the following step of the analysis.

Formula 4 (4)

Formula 5 (5)

The use of allometric principles suggests an exponent of 0.75for the clearance, 1 for volume of distribution, and –0.25for elimination constant rate (1, 22). With our data set, therewere no significant differences in terms of OFV and goodnessof fit when using an exponent of 1 for clearance and 1 for volume.

Age had a significant effect on CL_T and V, resulting, respectively,in 93- and 13-U decreases in the OFV. This effect could be observedfrom the plot of nelfinavir apparent clearance (Fig. 3) andvolume (data not shown) using maximum posteriori Bayes estimatesfrom the base model versus age. Adding the same age effect onboth V and CL_T resulted in a 101-U decrease in OFV. The followingequations describe the final covariate model for nelfinavir:

Formula 6 (6)

Formula 7 (7)

Formula 8 (8)

View larger version (30K):
[in this window]
[in a new window]
FIG. 3. Changes in apparent elimination clearance of nelfinavir as a function of age.
${circ}$ , Individual clearance estimates from the base model;
—, clearance predicted by the model = 0.92 x (AGE/8.2)^–0.29.

(ii) M8 pharmacokinetic model building. The M8 pharmacokinetics was modeled as a metabolite compartment.The nelfinavir pharmacokinetic parameters, including the effectof age on CL_T and V, were fixed, and M8 parameters were estimatedseparately. The intersubject variability for F_MT was not significant,and exclusion of this random effect had no influence on theOFV. The covariate submodeling was then established for theM8 elimination rate constant K_M0. Efavirenz and nevirapine arenonnucleosidic inhibitors (NNI); both of them are inductors.Estimating the specific effect for each drug on K_M0, we foundthat the inducer effects of efavirenz and nevirapine were notsignificantly different. We defined ${theta}$ _NNI, which is the commoninfluential factor used when one of these drugs is administered(the two drugs are never administered simultaneously).

At this step, the following equation described the final covariatemodel:

Formula 9 (9)
Adding thiscovariate modeling resulted in a 42-U decrease in OFV.

(iii) Nelfinavir-M8 pharmacokinetic model building. Nelfinavir and M8 were simultaneously fitted to the parent-metabolitemodel with first-order administration and elimination, includingthe covariate submodelings, in order to verify and refine theparameter estimates. This step led to minor changes in the previousestimates. The addition of a covariance term between total clearanceand M8 elimination rate led to a significant 11-U decrease inOFV. Then covariate deletion was performed to verify the nelfinavir-M8pharmacokinetic model. Exclusion of each covariate of the modelled to deterioration of the fit with OFV and ISV increase: respectively,106 U and 15% on CL_T, 7% on V for age, and 40 U and 5% on theelimination rate constant for NNI coadministration. At thisfinal step, the following equations described the covariatemodel:

Formula 10 (10)

Formula 11 (11)

Formula 12 (12)
Table 2 summarizes the final populationpharmacokinetic estimates.

View this table:
[in this window]
[in a new window]
TABLE 2. Population pharmacokinetic parameters of nelfinavir and metabolite M8 and bootstrap validation

(iv) Model performance. Final model performance could be appreciated by comparing thepopulation predicted and observed plasma concentrations (Fig.4). There was a correlation of r = 0.74 (P < 0.001) and r= 0.79 (P < 0.001) between individual predicted and observedconcentrations for nelfinavir and M8, respectively, and thebias and precision were, respectively, –0.02 (±1.5)and 2.2 (±3.8) for nelfinavir and –0.03 (±0.6)and 0.3 (±1.2) mg/liter for M8.

View larger version (41K):
[in this window]
[in a new window]
FIG. 4. Population predicted versus observed nelfinavir and plasma M8 concentrations from the final model. Solid line, identity line.

Bootstrap assessment of the final population model. The final model obtained with the original data set was subjectedto bootstrap analysis. As shown in Table 2, the mean, standarderror, and variability of parameter estimates obtained fromthe bootstrap process, which entailed 1,000 runs, were verysimilar to the estimates previously obtained with the originaldata set.

Relevance of FDA recommendations. In children from 2 to 16 years (n = 121), using the minimaldoses currently recommended, 25 mg/kg of body weight TID or50 mg/kg BID, the predicted concentration was above 0.8 mg/literin 96% of children with a 25-mg/kg administration every 8 hand 91% of children with a 50-mg/kg administration every 12h (Fig. 5). This large group was again split into two subgroups,2 to 7 and 8 to 16 years, to refine the analysis: for the recommendeddoses, the percentages of children that had a trough concentrationabove 0.8 mg/liter were not significantly different betweenthe two groups.

View larger version (18K):
[in this window]
[in a new window]
FIG. 5. Percentage of the 121 children from 2 to 13 years with a minimum plasma concentration above 0.8 mg/liter as a function of daily dose and frequency of administration. Solid line, administration every 8 h; dotted line, administration every 12 h; vertical line, minimal FDA-recommended doses of 25 mg/kg TID (solid line) and 50 mg/kg BID (dotted line).

In children from 2 months to 2 years (n = 36), using the minimaldoses proposed, 40 mg/kg TID or 60 mg/kg BID, the predictedconcentration was above 0.8 mg/liter in 100% of children witha 40-mg/kg administration every 8 h and in 89% of children witha 60-mg/kg administration every 12 h (Fig. 6). Thus, the dailydosage for children from 2 to 16 years should be 75 to 100 mg/kg,whereas the children from 2 months to 2 years should receivemore, 120 mg/kg.

View larger version (17K):
[in this window]
[in a new window]
FIG. 6. Percentage of the 36 children from 2 months to 2 years with a minimum plasma concentration above 0.8 mg/liter as a function of daily dose and frequency of administration. Solid line, administration every 8 h; dotted line, administration every 12 h; vertical line, minimal FDA-recommended doses of 40 mg/kg TID (solid line) and 60 mg/kg TID (dotted line).

In children younger than 2 months (n = 25), using 40 mg/kg every12 h, the predicted concentration was above 0.8 mg/liter in4% of children. For 50 and 60 mg/kg every 8 h, 88% and 100%of children, respectively, had a minimum plasma concentrationabove 0.8 mg/liter (Fig. 7) (between 0.7 and 2.1 mg/liter for50 mg/kg every 8 h and between 0.8 and 2.5 mg/liter for 60 mg/kgevery 8 h).

View larger version (15K):
[in this window]
[in a new window]
FIG. 7. Percentage of the 25 children younger than 2 months with a minimum plasma concentration above 0.8 mg/liter as a function of daily dose and frequency of administration. Solid line, administration every 8 h; dotted line, administration every 12 h; vertical dotted line, newborn's FDA-recommended dose of 40 mg/kg BID.

Whichever age, the percentage of children with a minimum plasmaconcentration above 0.8 mg/liter was higher with an administrationevery 8 h than with an administration every 12 h, correspondingto better TID than BID FDA-recommended doses (11).

DISCUSSION

Top

Discussion

The nelfinavir-M8 pharmacokinetics was satisfactorily describedby the proposed compartmental model. The present study showeda great consistency in the final nelfinavir-M8 population modelderived from sequential analyses of nelfinavir and M8, confirmingthe robustness of the process. The basic one-compartment modelwas already used in adults for nelfinavir (21). The pharmacokineticsof the metabolite produced from the parent compound should bedescribed by a two-exponential equation, but the sparse dataset (a median of only two M8 concentration-time samples perchild) did not allow the identification of two exponential components.So only an integrated modeling of parent-metabolite pharmacokineticscould provide a reliable estimate of M8 elimination, since theinformation for the fast exponential decay is provided by nelfinavirdata. Indeed, in this approach, data on the metabolite may addinformation to the observations on the parent and vice versa.

The following observations support the use of the proposed pharmacokineticmodel. (i) Nelfinavir mean plasma clearance (CL_T/F = 0.92 ±0.04 liter/h/kg) was consistent with previously reported values:1.0 to 1.3 liters/h/kg in 18 children (2.1 to 10.8 years) (25)and 1.57 liters/h/kg in 26 children (0.6 to 16 years) (8). (ii)Nelfinavir apparent plasma clearance and volume (CL_T/F, V/F)decreased with age, being much higher in children younger than2 years, in agreement with previous studies. Bergshoeff et al.(5) showed that the clearance in children aged <2 years was1.5 times higher than in older children (2 to 18 years). Veryhigh clearance in infants was reported by Litalien et al. (4.2liters/h/kg for children from 2.3 to 8.5 months) (16), Capparelliet al. (2.7 liters/h/kg in infants between 15 days and 2 years)(7), Payen et al. (2.13 liters/h/kg for children younger than2 years) (20), and Mirochnick et al. (2.1 liters/h/kg at weeks1 and 6 of life) (17). (iii) Finally, the goodness of fit, depictedin Fig. 4, was also a factor.

The residual error, 1.65 µg/ml, was probably overestimatedbecause it included some part of interoccasion variability thatcould not be estimated here, since only one sample was availableat each occasion. If there was a significant interoccasion variabilityin the nelfinavir pharmacokinetics, it could result in an underestimationof the ISVs, including CL_T intersubject variability.

A major aim of population pharmacokinetics is to determine whichmeasurable pathophysiological factor can cause changes in thedose-concentration relationship and to estimate the degree towhich they do so, such that an appropriate dose adjustment canbe made. This is particularly relevant for drugs that exhibitan appreciable degree of intersubject variability, such as nelfinavir,in children.

In the present study, age and the NNI drugs influenced the nelfinavir-M8pharmacokinetics.

As shown, CL_T and V normalized to BW decreased as an inversefunction of age. CL_T and V were apparent parameters (CL_T/F andV/F). In our model, the same age effect was added to increaseboth V and CL_T in the younger children, as we supposed thatthe age effect was due to a decrease in bioavailability (F).

A number of factors may explain the decrease of the bioavailabilityin infants. For instance, a diet which differs in content andcalories from that of older children may play a role, as theinfluence of diet on the bioavailability of nelfinavir is wellestablished (16). Moreover, newborns have an alkaline gastricpH (pH 6 to 8) and gastric acid production increases over thenext 24 to 48 h before declining and remains relatively lowin the first months of life. This high gastric pH in the newbornand young infant may reduce the bioavailability of weakly acidiccompounds such as nelfinavir. Also, a smaller absorption areaand binding of nelfinavir to a baby's inner side bottle mayalso be suggested. In these young children, an increase in themetabolism did not seem relevant. There is no argument in favorof an overexpression of the P glycoprotein in infants. Moreover,CYP2C19, which metabolizes nelfinavir to M8 (16), has a lowactivity during the first year of life (30% of the adult activity)(23).

The plasma M8 concentrations were 1.9-fold lower in patientstreated with efavirenz or nevirapine, consistent with an inductionof CYP3A4 by these drugs, with M8 being metabolized via CYP3A4(2). Furthermore, very high plasma M8 concentrations were observedin all samples (n = 11) from three children who received ritonavir,a known CYP3A4 inhibitor (14), but these data were too scantyto reach statistical significance.

It was previously shown that the antiretroviral response wasimproved in children with a minimum plasma concentration above0.8 mg/liter (6). Using a Bayesian approach, we showed thatthis target concentration was reached more often with an administrationevery 8 h than an administration every 12 h (Fig. 5 to 7). Thisis in agreement with a previous study which showed that a significantlyhigher percentage of children in the twice-daily group had subinhibitoryminimum plasma concentrations of nelfinavir than the thrice-dailyregimen (9). Nelfinavir has a short half-life (5.5 h), whichexplains why an administration every 8 h maintained a highertrough concentration than an administration every 12 h. Thisdifference in minimum plasma concentration between administrationsevery 8 or 12 h is more important in the youngest children.The youngest children, who have a smaller absorption area, mayhave a better bioavailability with a small dose administeredthrice daily than with a higher dose administered twice daily.Therefore, the thrice-daily regimen should be preferred to thetwice daily regimen, especially in this group. FDA-recommendeddoses for nelfinavir were then simulated as doses given withuniform intervals (every 8 h TID and every 12 h BID); however,nelfinavir is taken during a meal to increase bioavailabilityand children do not eat exactly every 8 or 12 h.

For children from 2 to 13 years, we showed that the new currentFDA recommendations, 25 to 35 mg/kg TID or 50 to 60 mg/kg BID,were optimal. We confirmed also that the proposed nelfinavirdoses for children younger than 2 years, 40 to 50 mg/kg TIDor 60 to 75 mg/kg BID, are optimal for children from 2 monthsto 2 years. However, more children had a trough concentrationabove 0.8 mg/liter with the TID than with the BID recommendedregimen. For children younger than 2 months, a 40-mg/kg doseof nelfinavir administered twice daily is inadequate. We recommenda nelfinavir dose between 50 and 60 mg/kg administered thricedaily. Mirochnick et al. also proposed further investigationsof larger doses, such as 75 mg/kg twice a day for infants youngerthan 6 weeks (17). As stated above, the predicted trough concentrationsthat served to determine this drug dosage recommendation wereobtained from a Bayesian approach, and because the residualvariability was rather high, these predictions are likely tobe close to the mean population trough values. However, themain consequence, i.e., the need to increase dosage in the youngestchildren, is also supported by direct examination of their observedconcentrations at various times after administration that weremainly lower than 0.8 µg/ml. These data confirm the FDAdosage recommendations for children older than 2 months. However,in younger children, it is suggested to increase the dosageand to give it thrice daily. Nevertheless, the results of thispopulation pharmacokinetic analysis should be confirmed by aprospective analysis.

Top

The differential system connected with the model depicted inFig. 1 is:

Formula A1 A1

Formula A2 A2

Formula A3 A3
where G corresponds to the gut compartment, (nelfinavir)to nelfinavir amount and (M8) to the metabolite's amount, K= CL_T/V is the total nelfinavir constant rate, FMT is the nelfinavir-to-M8formation clearance fraction (fraction between 0 and 1), andK_M0 is the M8 elimination rate constant (K_M0 = CL_m0/V_m, withV_m = 1).

The solution giving the profile of the metabolite (m = M8) compartmentis:

Formula A4 A4
where t isthe time elapsed between drug administration and blood samplingand T is the time interval between two administrations.

ACKNOWLEDGMENTS

This work was supported by a CRES (contrat de recherche stratégieINSERM) from INSERM.

FOOTNOTES

* Corresponding author. Mailing address: Pharmacologie Clinique, Hôpital Saint Vincent de Paul, 74-82 Avenue Denfert-Rochereau, 75014 Paris, France. Phone: 33140488611. Fax: 33140488223. E-mail: deborah.hirt{at}univ-paris5.fr.

REFERENCES

Top