A pharmacokinetic and pharmacogenetic study of efavirenz in children: Dosing guidelines can result in subtherapeutic concentrations

Article (PDF Available)inAntiviral therapy 13(6):779-87 · January 2008with15 Reads
Source: PubMed
Abstract
Our main objectives were to study the population pharmacokinetics of efavirenz and to explore the adequacy of dosing guidelines. A total of 33 HIV-1-infected patients were recruited from the Emma Children's Hospital (Amsterdam, the Netherlands). Gender, age, drug formulation, the presence of the c.516G>T polymorphism in the CYP2B6 gene and the quantitation of liver enzymes alanine aminotransferase and aspartate aminotransferase at baseline were collected. A non-linear mixed effect pharmacokinetic model was developed. CYP2B6 genotype and drug formulation significantly influenced efavirenz pharmacokinetics. Clearance was 29.7% lower in children carrying the CYP2B6-516-G/T genotype compared with children carrying the G/G genotype. Relative bioavailiability of the oral liquid compared with tablets or capsules was 46.6%. Children carrying the CYP2B6-516-G/G genotype had a 50-70% probability of developing a subtherapeutic trough level of efavirenz and only 1-3% probability of developing a trough level >4 mg/l. To reduce the probability of developing a subtherapeutic trough concentration, we propose to give an adult efavirenz dose to children weighing > or =25 kg and to allometrically scale doses for other weight levels a priori. The dose of the oral solution should be twice the dose of capsules. Population pharmacokinetics of efavirenz in children were adequately described. Current dosing guidelines can result in subtherapeutic concentrations in children carrying the CYP2B6-516-G/G genotype and with the liquid formulation. A priori dose adaptations in the paediatric population seem feasible and need prospective validation.

Figures

© 2008 International Medical Press 1359-6535 779
Antiviral Therapy 13:779–787
Background: Our main objectives were to study the
population pharmacokinetics of efavirenz and to explore
the adequacy of dosing guidelines.
Methods: A total of 33 HIV-1-infected patients were
recruited from the Emma Children’s Hospital (Amster-
dam, the Netherlands). Gender, age, drug formulation, the
presence of the c.516G>T polymorphism in the CYP2B6
gene and the quantitation of liver enzymes alanine ami-
notransferase and aspartate aminotransferase at baseline
were collected. A non-linear mixed effect pharmacokinetic
model was developed.
Results: CYP2B6 genotype and drug formulation
significantly influenced efavirenz pharmacokinetics. Clear-
ance was 29.7% lower in children carrying the CYP2B6-
516-G/T genotype compared with children carrying the
G/G genotype. Relative bioavailiability of the oral liquid
compared with tablets or capsules was 46.6%. Children
carrying the CYP2B6-516-G/G genotype had a 50–70%
probability of developing a subtherapeutic trough level
of efavirenz and only 1–3% probability of developing a
trough level >4 mg/l. To reduce the probability of develop-
ing a subtherapeutic trough concentration, we propose to
give an adult efavirenz dose to children weighing 25 kg
and to allometrically scale doses for other weight levels
a priori. The dose of the oral solution should be twice the
dose of capsules.
Conclusions: Population pharmacokinetics of efavirenz in
children were adequately described. Current dosing guide-
lines can result in subtherapeutic concentrations in children
carrying the CYP2B6-516-G/G genotype and with the liq-
uid formulation. A priori dose adaptations in the paediatric
population seem feasible and need prospective validation.
Optimal treatment of children with an HIV infection is
a challenging task. Adherence to antiretroviral drugs is a
key factor for successful therapy. However, adherence is a
substantial problem in the paediatric population. A once
daily regimen containing efavirenz is useful to tackle the
problems of adherence and has been proven to be a safe,
convenient and potent treatment for HIV type-1 (HIV-1)
infection in children [1]. Apart from adherence, dosing
is another challenge in the treatment of paediatric HIV-1
infection. Drug absorption, interactions and metabolism
differ in children and change during growth. Pharmaco-
kinetic guiding of efavirenz dosing by means of thera-
peutic drug monitoring (TDM) in the paediatric popula-
tion is therefore considered important [2].
Efavirenz exposure has been well correlated with treat-
ment outcome and target trough concentrations (C
min
) in
plasma have been proposed to be between 1 and 4 mg/l
[3–5]. Current paediatric efavirenz dosing guidelines for
children >3 years are depicted in Table 1. Recommended
dosages vary with weight, formulation and age. No efa-
virenz dosing strategies have been investigated in chil-
dren <3 years of age. As it stands, there is only limited
data on the pharmacokinetics of efavirenz in children.
Recent findings of a high prevalence of subtherapeutic
efavirenz plasma concentrations in children indicate
that current paediatric dosing guidelines might not be
sufficient and that a large proportion of children tak-
ing efavirenz need a dose increase [6–8]. However, these
studies reporting the high prevalence of subtherapeutic
drug concentrations of efavirenz were of observational
nature and did not suggest alternative dosing regimes to
prevent subtherapeutic efavirenz concentrations.
Original article
A pharmacokinetic and pharmacogenetic study of
efavirenz in children: dosing guidelines can result in
subtherapeutic concentrations
Rob ter Heine
1
*,
Henriette J Scherpbier
2
, Kristel ML Crommentuyn
1
, Vincent Bekker
2
,
Jos H Beijnen
1
,
Taco W Kuijpers
2
and Alwin DR Huitema
1
1
Department of Pharmacy & Pharmacology, Slotervaart Hospital, Amsterdam, the Netherlands
2
Department of Pediatrics, Emma Children’s Hospital, Amsterdam, the Netherlands
*Corresponding author: E-mail: rob.terheine@slz.nl
Introduction
ter Heine.indd 779 5/9/08 16:59:25
© 2008 International Medical Press
780
R ter Heine et al.
Several factors are known to influence efavirenz
metabolism. Efavirenz is mainly metabolized by the
cytochrome P450 enzyme CYP2B6. Efavirenz expo-
sure is known to be influenced by polymorphisms in
the CYP2B6 gene, resulting in a higher efavirenz expo-
sure [9,10]. Unlike other CYP2B6*6 polymorphisms,
the c.516G>T polymorphism has been shown to be the
causal sequence variation for decreased expression and
function of CYP2B6 [11]. Also, drug formulation has
been shown to play a role; previous studies showed a
relative bioavailability of the oral solution of 62% or
83% when compared with efavirenz administered as a
tablet or capsule, respectively [3,12].
Because adequate exposure from the start of therapy
(that is, the correct dose administered the first time) is
desirable and because TDM might not always be avail-
able (for example, in resource-limited settings), we con-
ducted a population pharmacokinetic study of efavirenz
in children. Our main objective was to adequately
describe the population pharmacokinetics of efavirenz
in this population and to identify important covariates
influencing the pharmacokinetics. Our secondary objec-
tive was to explore efavirenz exposure on the basis of
the current paediatric guidelines and alternative dosing
strategies using the developed model.
Methods
Patient recruitment
All patients were recruited from the Emma Children’s
Hospital (Amsterdam, the Netherlands). All par-
ticipated in an ongoing open-label study to evalu-
ate the efficacy and safety of single daily treatment
with efavirenz, abacavir, didanosine and lamivudine.
HIV-1-infected children were eligible when they had a
CD4
+
T-cell count of <1,750 cells/µl (<1 year of age),
<1,000 cells/µl (1–2 years of age), <750 cells/µl (3–6
years of age) or <500 cells/µl (>6 years of age). Pre-
vious exposure to antiretroviral therapy was allowed.
Exclusion criteria were presence of mutations associ-
ated with efavirenz resistance or with two or more of
the nucleoside reverse transcriptase inhibitors used. No
restrictions were made with respect to ethnicity, gender,
route of HIV acquisition or disease stage. The medical
ethics committee approved the protocol and parents or
caregivers provided written informed consent.
Dosing strategy
All children were dosed according to paediatric dosing
guidelines listed in Table 1 and there were no restrictions
considering food intake. For adherence, the children’s
guardians were counselled regarding the importance
of treatment adherence. Where appropriate, children
were counselled accordingly. Adherence was monitored
by telephone and at each follow-up clinic visit. TDM
was applied throughout. The dose was increased when
a child was exposed to subtherapeutic C
min
of efavirenz
(<1 mg/l). The dose was decreased when a child was
suspected to have efavirenz-related side effects in com-
bination with high efavirenz C
min
(>4 mg/l).
Sampling and bioanalysis
All children were admitted to the hospital at the start
of therapy. During this first day of treatment, a full
pharmacokinetic curve during a dosing interval was
assessed. Samples were drawn just before and after
approximately 0–1, 1–2, 2–4, 4–7, 7–12 and 24 h after
drug intake. Subsequently, after 2 and 6 weeks of treat-
ment a trough sample (20–24 h after drug intake), and
preferably a sample between 2–4 h after drug ingestion
were taken. The efavirenz concentrations were quanti-
fied using a validated HPLC assay with UV detection.
Validated concentration ranges were 0.05–15 mg/l.
Intra- and interassay precision were <5.9%, whereas
accuracies varied between -12.7 and 8.5% [13].
CYP2B6 genotyping
The presence of the c.516G>T polymorphism in the
CYP2B6 gene was analysed using PCR and sequenc-
ing. DNA cycle sequencing was carried out essentially
as described by the manufacturer (Applied Biosystems,
Foster City, CA, USA). Genomic DNA was extracted
from plasma using the QIAamp DNA mini kit (Qiagen,
Inc., Valencia, CA, USA), following the manufacturer’s
instructions. The methods used for the amplification of
the CYP2B6 genes have been previously described by
Table 1. Efavirenz paediatric dosing guidelines
Liquid formulation (30 mg/ml) Capsules/tablets
Body weight, kg Age 3<5 years dosage, ml Age >5 years dosage, ml dosage, mg
13<15 12 9 200
15<20 13 10 250
20<25 15 12 300
25<32.5 17 15 350
32.5<40 17 400
>40 24 600
ter Heine.indd 780 5/9/08 16:59:25
Pharmacokinetics of efavirenz in children
Antiviral Therapy 13.6 781
Lang et al. [14]. After amplification and purification,
sequences were analysed on an Applied Biosystems
3100-Avant DNA sequencer. For sequence alignment
Seqscape version 2.1 (Applied Biosystems) was used.
Pharmacokinetic analysis
The nonlinear mixed effect modelling program
(NONMEM) version VI was used to perform the
analysis [15]. The model was fitted using the first-or-
der conditional estimation procedure with interaction
between interindividual, intraindividual and residual
variability. The minimal value of the objective function
(OFV, equal to minus twice the log likelihood) pro-
vided by NONMEM was used as goodness of fit char-
acteristic to discriminate between hierarchical models
using the log likelihood ratio test [15]. A P-value of
0.05, representing a decrease in OFV of 3.84 points
was considered statistically significant. Furthermore,
XPose and Perl speaks NONMEM (PsN) were used
for graphical and statistical model diagnostics [16,17].
Conditional weighted residuals for model diagnosis
were determined as described by Hooker et al. [18].
We used Piraña (interface to NONMEM, PsN and our
cluster) for run deployment and analysis [19]. Preci-
sion of parameter estimates were estimated using the
covariance step in NONMEM.
Basic pharmacokinetic model
Data were described with a one-compartment model.
The oral clearance (Cl/F, where Cl represents the
clearance in l/h and F represents the oral bioavail-
ability) and volume of distribution (V/F) were allom-
etrically scaled for body weight using two equations:
Cl/ F=θ
1
×(weight/70)
0.75
and V/F=θ
2
×(weight/70)
1
, in
which Cl/F and V/F represent the clearance and volume
of distribution scaled to a person weighing 70 kg and
θ
1
and θ
2
are the respective typical values of clearance
and volume of distribution. Allometric scaling includes
weight as a covariate in the model a priori, allowing
exploration of other possible covariates independent of
size. Scaling to a weight of 70 kg allows comparison of
the pharmacokinetic estimates with adults [20–22].
The absorption phase was described using two
transition compartments between the depot com-
partment and the central compartment, as described
before by Kappelhoff et al. [23] for the determination
of the mean absorption time (MAT). Autoinduction
of clearance was accounted for by estimation of the
proportional increase of clearance of efavirenz during
the first 2 weeks of treatment.
Interindividual variability in the different pharmaco-
kinetic parameters and interoccasion variability on
relative bioavailability were estimated with an expo-
nential error model as proposed by Karlsson et al. [24].
The residual variability was modelled with a combined
additive and proportional error. The basic pharmaco-
kinetic model is depicted schematically in Figure 1.
Covariate model building
Gender, age, drug formulation (liquid or solid), the
presence of the c.516G>T polymorphism in the CYP2B6
gene and the liver enzymes alanine aminotransferase
(ALT) and aspartate aminotransferase (AST) at baseline
were collected. The effect of body size was accounted
for a priori by allometric scaling of Cl/F and V/F on
body weight. When weight data of a patient were not
Mean absorption time
V/F allometrically scaled
Cl/F allometrically scaled
Autoinduction
Cl/F
Central
Transition
compartment
Depot
Transition
compartment
Figure 1. The basic pharmacokinetic model
Cl/F, oral clearance where Cl represents clearance in l/h and F represents the oral bioavailability; V/F, volume of distribution.
ter Heine.indd 781 5/9/08 16:59:25
R ter Heine et al.
© 2008 International Medical Press
782
available, a linear extrapolation of the weight was made
on the basis of a previous and a later weight observa-
tion. All covariates were tested univariately (P<0.05)
for their inclusion in an intermediate model: gender and
age were tested on volume of distribution and clearance,
drug formulation was tested on relative bioavailability,
liver enzymes ALT and AST were tested on clearance
and the presence of the c.516G>T polymorphism in
the CYP2B6 gene was tested on reduction of clearance.
The clearance for the group of extensive metabolizers
(G/G genotype) was estimated. The decrease in Cl/F for
the group of intermediate metabolizers (G/T genotype)
and poor metabolizers (T/T genotype) was estimated
separately. After inclusion of all significant covariates
into the intermediate model, a stepwise backward elim-
ination procedure was carried out. A parameter was
retained in the model when the influence of this param-
eter was statistically significant with a P-value <0.005,
representing a decrease in OFV of 7.88.
Simulation study
The probability of developing a C
min
<1 mg/l or >4 mg/l
at 24 h after intake was estimated using the developed
model. For each dose level, 1,000 virtual patients from
the corresponding weight category were simulated and
the probability of developing a C
min
<1 mg/l or >4 mg/l
for each weight category was expressed as a percentage
of the number of simulated patients. Using this strat-
egy, efavirenz exposure was evaluated using the current
paediatric dosing guidelines. Alternative dosing options
were explored as well. An alternative dose was consid-
ered acceptable if >75% of the patients had a efavirenz
C
min
of at least 1 mg/l at steady state.
Results
Pharmacokinetic analyses
All patient characteristics have been previously described
by Scherpbier et al. [1]. In the study by Scherpbier et al.
a total of 36 children were included. In our study, 33
children were evaluable for pharmacokinetic analy-
ses. The median age of these 33 children was 6.5 years
(range 0.9–19 years), the median weight was 20.5 kg
(range 8.7–83 kg), 16 of 33 patients were female and
11, 15 and 7 patients were wild type, heterozygous or
homozygous mutant for the c.516G>T polymorphism,
respectively. None of the children was previously
exposed to efavirenz. A total of 30 full pharmacokinetic
curves and 302 plasma concentrations at single time
points were available. Samples without a known time
after drug intake were excluded. Efavirenz was given as
a liquid formulation in 95 out of a total of 329 dosing
events. The median dose of all dosing events scaled to a
person weighing 70 kg was 754 mg (interquartile range
636–901 mg).
Plasma concentrations of efavirenz were highly
variable. On the first day of treatment the concentra-
tions ranged from undetectable to approximately 5 mg/l.
Maximum plasma concentrations were achieved 2–8 h
after drug intake and declined gradually afterwards.
As shown in Figure 2, most observed efavirenz concen-
trations during follow-up were within the therapeutic
window of 1–4 mg/l (11% <1 mg/l and 23% >4 mg/l),
probably because of TDM-guided dosing interventions.
Two children with efavirenz levels >4 mg/l had sleep-
ing abnormalities. In one child, the dose was therefore
reduced from 600 to 400, then to 200 mg once daily.
One child developed grade 3 increases in liver enzymes,
which led to the discontinuation of efavirenz.
For the basic model, Cl/F at baseline scaled to a
person weighing 70 kg was estimated to be 14.5 l/h.
In the basic model, autoinduction could not be esti-
mated. V/F scaled to a person weighing 70 kg was
523 l and MAT was 2.21 h. The results of the basic
pharmacokinetic model are presented in Table 2.
The covariates of interest were introduced into the
model separately and drug formulation (liquid formu-
lation versus tablets or capsules) as well as CYP2B6
genotype significantly influenced efavirenz pharmaco-
kinetics in the univariate analysis. The reduction
in clearance of a heterozygous and a homozygous
CYP2B6 c.516G>T mutation was 25.4% and 58.7%,
respectively, indicating that the effect of a homozygous
mutation was approximately twice the effect of a heter-
ozygous mutation. Separate estimation of these effects
did not improve the model (as judged by the graphical
and statistical model diagnostics). Because more degrees
of freedom resulted in model instability, the influence of
a homozygous mutation was fixed to have twice the
effect of a heterozygous mutation.
Figure 2. Observed efavirenz concentrations during follow-up
0612 18 24
0
2
4
6
8
10
12
14
16
18
20
Time after ingestion, h
Efavirenz concentration, mg/l
ter Heine.indd 782 5/9/08 16:59:26
Pharmacokinetics of efavirenz in children
Antiviral Therapy 13.6 783
In the multivariate analysis both covariates were
retained in the model. For the final model, Cl/F at base-
line scaled to a person weighing 70 kg was estimated to
be 13 l/h and increased by 21% during the first 2 weeks
of treatment. V/F scaled to a person weighing 70 kg was
424 l and MAT was 1.96 h. The relative bioavailability
for the efavirenz liquid formulation was only 46.6%
and Cl/F was reduced by 29.7% due to a heterozygous
G516T mutation in the CYP2B6 gene. A homozygous
mutation was therefore estimated to reduce the Cl/F by
59.4% (twice the effect of a heterozygous mutation).
No additional age effect was found on clearance. The
results of the final model are presented in Table 2. Fig-
ure 3 shows the goodness of fit plots of the final model.
Figure 3A shows the observed concentrations versus the
population predicted concentrations. Figure 3B shows
the observed concentrations versus individual predicted
concentrations, indicating that individual efavirenz
pharmacokinetics were well-described by the model.
Lastly, Figure 3C shows the conditional weighted resid-
uals versus time. All conditional weighted residuals
were within ±3, indicating that the variability in efa-
virenz exposure was adequately captured by the model.
Moreover, equal distribution of positive and negative
conditional weighted residuals during time indicated
the absence of bias in the model.
Results simulation study
Model-derived probabilities of developing a C
min
<1
or >4 mg/l for each genotype and dosing schedule
(tablets/capsules, oral liquid 3–<5 years old and oral
liquid >5 years old) according to the manufacturer’s
dosing guidelines are depicted in Table 3. Probabili-
ties for each weight category were pooled due to simi-
lar results at each dose level. It can be noted that the
group of extensive metabolizers and the group receiv-
ing the liquid formulation showed a high risk for sub-
therapeutic C
min
. The probability of developing a C
min
>4 mg/l was estimated to be relatively small, except in
the group of poor metabolizers taking efavirenz as a
capsule or tablet.
Alternative dosing options were explored on the
basis of the group wild type for the c.516G>T CYP2B6
genotype, due to the high probability in that group of
developing subtherapeutic concentrations. No addi-
tional age effect on pharmacokinetics was found in our
model, therefore alternative dosing regimes were on the
basis of weight only. We tested the regimen where one
should give an adult dose of 600 mg to children weigh-
ing 25 kg. Doses for children weighing <25 kg should
be allometrically scaled. Furthermore, a dose increase
of 215% when switching from a solid efavirenz for-
mulation to the oral solution because the diminished
bioavailability of the latter was proposed. This regimen
is shown in Table 4. Simulation derived probabilities of
developing a C
min
<1 or >4 mg/l for each genotype using
the proposed alternative dosing regimen are shown in
Table 5. As can be seen in this table, the risk for devel-
oping a subtherapeutic C
min
has been highly reduced.
However, children carrying the c.516G>T mutation in
the CYP2B6 gene show an increased risk for developing
a C
min
>4 mg/l.
Discussion
We have successfully developed a population pharmaco-
kinetic model for efavirenz in children. Weight was the
main determinant for the development of clearance and
volume of distribution and no additional age effect was
found. In addition, drug formulation and the c.516G>T
Table 2. Final parameter estimates of the basic and final pharmacokinetic model for efavirenz
Cl/F, oral clearance; F, oral bioavailability; MAT, mean absorption time; RSE, relative standard error; V/F, volume of distribution.
Basic model Final model
Parameter Estimate RSE, % Estimate RSE, %
MAT, h 2.21 15.1 1.96 11.9
Cl/F at baseline, l/h 14.5 13.6 13.0 11.5
Autoinduction, % 21.0 4.54
V/F, l 523 14.5 424 8.87
Decrease in clearance due to c.516G>T 29.7 19.4
heterozygous mutation, %
Relative bioavailability liquid formulation, % 46.6 11.8
Interindividual variability MAT, % 57.5 44.7 54.4 47.0
Interindividual variability Cl/F, % 73.5 30.7 52.4 31.2
Interindividual variability V/F, % 46.6 61.3 40.0 26.4
Interoccasion variability F, % 37.4 32.5 46.9 51.4
Additive error, mg/l 0.298 15.4 0.200 21.0
Proportional error, % 20.3 25.0 31.9 6.36
ter Heine.indd 783 5/9/08 16:59:26
R ter Heine et al.
© 2008 International Medical Press
784
polymorphism were found to have a large effect on the
C
min
of efavirenz in our cohort.
Our data suggests that an unexpected Cl/F is the
main determinant of decreased efavirenz exposure
in children. The Cl/F in a child wild type for the
c.516G>T CYP2B6 genotype scaled to a person of
70 kg was 13 l/h, and increased by 21% to 15.7 l/h
during the first 2 weeks of treatment due to autoinduc-
tion. The MAT was 1.96 h. In the final model the V/F
scaled to a person weighing 70 kg was 424 l. The V/F
is comparable with previously found V/F in popula-
tion pharmacokinetic analyses in adults ranging from
150 to 418 l [23,26–28]. However, the Cl/F of 15.7 l/h
in children wild type for the c.516G>T CYP2B6 poly-
morphism is higher than the previously reported 9.4 l/h
in adults wild type for this genotype [28].
Drug formulation had a large effect on the bioavail-
ability of efavirenz. The estimated relative bioavail-
ability of the oral liquid compared with solid liquid
formulations was 46.6%. This finding conflicts with
the 20% dose increase suggested by the manufacturer
when switching from a solid formulation to the liquid.
This dose increase was based on a single dose study in
healthy adults [12]. A previous pharmacokinetic study
of the efavirenz liquid formulation in HIV- infected
children showed that its relative bioavailability was
62%, which supports our conclusion that a 20% dose
increase is not sufficient [3]. This could indicate that
there is a physiological difference between healthy
adults and HIV-infected children that could influence
drug absorption. Moreover, the absence of food restric-
tions might also have influenced the relative bioavail-
ability because simultaneous food intake is known to
increase efavirenz exposure [12].
Efavirenz Cl/F was reduced markedly in children by
the c.516G>T polymorphism in the CYP2B6 gene. A
heterozygous mutation resulted in a 29.7% decrease in
clearance. CYP2B6 genotype has been previously shown
to significantly decrease efavirenz clearance in adults as
well as children. Haas et al. [9] showed an increase of
efavirenz exposure of approximately 36% in hetero-
zygous patients and even a threefold increase in patients
who were homozygous for the CYP2B6 c.516G>T
mutation. A study by Saitoh et al. [10] investigating the
influence of CYP2B6 on efavirenz pharmacokinetics in
children showed similar results. Our study confirms the
importance of the CYP2B6 genotype in the clearance of
efavirenz in children.
The mean administered dose scaled to a person
weighing 70 kg of 754 mg already indicated that, in
our population, normal efavirenz dosing (600 mg once
daily allometrically scaled to a person weighing 70 kg)
would result in inadequate exposure. Using a simula-
tion study, it was confirmed that current paediatric
dosing guidelines leaves certain populations (children
Figure 3. Goodness of fit in the final pharmacokinetic model
(A) Observed concentrations versus population predicted concentrations.
(B) Observed concentrations versus individual predicted concentrations.
(C) Conditional weighted residuals versus time.
0
0
2
4
6
8
10
12
14
16
18
20
246810 12 14 16 18 20
Observed concentrations, mg/l
Individual predicted concentrations, mg/l
30
4
3
2
1
0
0
5
10 15 20 25
-1
-2
-3
-4
Time, h
Conditional weighted residual
0
0
2
4
6
8
10
12
14
16
18
246810 12
Population predicted concentrations, mg/l
Observed concentrations, mg/l
A
B
C
ter Heine.indd 784 5/9/08 16:59:27
Pharmacokinetics of efavirenz in children
Antiviral Therapy 13.6 785
wild type for the c.516G>T mutation as well as children
dosed with the liquid formulation) at risk for obtaining
subtherapeutic efavirenz C
min
. This is in concordance
with results of earlier observational studies, reporting
a high incidence of subtherapeutic efavirenz concentra-
tions or a high incidence of TDM-guided dose increases
in children [6–8].
Currently, CYP2B6 genotyping is not part of the
standard of care. The reported allele frequency of the
c.516G>T mutation in the CYP2B6 gene is reported
to be 20–50%, with the higher frequencies mainly
reported in patients of African origin [9,14,28,29]. We
have shown that the group wild type for the CYP2B6
c.516G>T mutation is at high risk of developing sub-
therapeutic concentrations. This is approximately
50–80% of all children. TDM should be standard of
care but it might not always be available, for instance in
resource-limited settings. Adequate drug exposure from
the start of therapy is desirable to minimize the chance
of developing resistance against efavirenz. Side effects
due to high efavirenz exposure are reversible, however,
drug resistance is not. Although it increases the probabil-
ity of developing C
min
>4 mg/l in patients heterozygous
or homozygous for the CYP2B6 c.516G>T mutation,
we propose to increase the dose of efavirenz a priori
when the presence of the c.516G>T polymorphism is
unknown or known to be absent. Furthermore, when
administered as the liquid formulation, the efavirenz
dose should be increased at least twofold, instead of
20% as proposed by the manufacturer.
Two previous pharmacogenetic studies of efavirenz
in adults proposed an a priori efavirenz dose decrease
in patients with the CYP2B6 c.516G>T genotype, but
did not report subtherapeutic concentrations in patients
Table 3. Simulation-derived probabilities of developing a trough level <1 or >4 mg/l for each genotype using current paediatric
dosing guidelines
C
min
, trough level.
Tablets/capsules Oral liquid 3–5 years Oral liquid >5 years
Probability of Probability of Probability of Probability of Probability of Probability of
Genotype C
min
<1 mg/l, % C
min
>4 mg/l, % C
min
<1 mg/l, % C
min
>4 mg/l, % C
min
<1 mg/l, % C
min
>4 mg/l, %
Extensive metabolizer 49 3 64 2 71 1
(c.516G>T wild type)
Intermediate metabolizer 24 11 43 7 51 4
(c.516G>T heterozygote)
Poor metabolizer 5 43 14 27 20 18
(c.516G>T homozygote)
Table 4. Alternative efavirenz dosing regimen
Body weight, kg Capsules/tablets dose, mg Oral liquid dose of 30 mg/ml, ml
13<15 400 29
15<20 450 32
20<25 500 36
>25 600 43
Table 5. Simulation derived probabilities of developing a trough level <1 or >4 mg/l for each genotype using the proposed
alternative dosing regimen
C
min
, trough level.
Tablets/capsules Oral liquid
Probability of Probability of Probability of Probability of
Genotype C
min
<1 mg/l, % C
min
>4 mg/l, % C
min
<1 mg/l, % C
min
>4 mg/l, %
Extensive metabolizer 22 13 24 16
(G516T wild type)
Intermediate metabolizer 9 30 11 31
(G516T heterozygote)
Poor metabolizer 1 59 2 58
(G516T homozygote)
ter Heine.indd 785 5/9/08 16:59:27
R ter Heine et al.
© 2008 International Medical Press
786
wild type for this genotype [28,30]. This underlines our
finding that the paediatric population is a different pop-
ulation than the adult population taking efavirenz. Our
model adequately described the population pharmaco-
kinetics of efavirenz in children, but additional data are
needed to fully describe the development of efavirenz
pharmacokinetics from infancy to adulthood.
Our study is limited by the number of children
included. However, this is the largest study of its kind
performed so far. Although there was thorough moni-
toring of compliance, non-compliance could not be
ruled out and could have influenced the results.
In conclusion, we have successfully developed a
population pharmacokinetic model of efavirenz in
children. Weight, CYP2B6 genotype and drug formu-
lation were important covariates influencing efavirenz
pharmacokinetics and using current paediatric dos-
ing guidelines, a large group is at risk for developing
subtherapeutic efavirenz concentrations. An a priori
dose increase in children seems therefore feasible and
needs prospective validation.
Acknowledgements
Ron Keizer is greatly acknowledged for technical
assistance, Atie van der Plas and Marleen Kemper are
acknowledged for their help in gathering clinical infor-
mation and Valerie Doodeman is acknowledged for her
help genotyping the samples.
Disclosure statement
The authors declare no competing interests.
References
1. Scherpbier HJ, Bekker V, Pajkrt D, Jurriaans S, Lange JM,
Kuijpers TW. Once-daily highly active antiretroviral therapy
for HIV-infected children: safety and efficacy of an efavirenz-
containing regimen. Pediatrics 2007; 119:e705–e715.
2. Back D, Gibbons S, Khoo S. An update on therapeutic
drug monitoring for antiretroviral drugs. Ther Drug Monit
2006; 28:468–473.
3. Brundage R, Fletcher CV, Fiske WD, et al. Pharmacokinetics
of an efavirenz suspension in children. 6th Conference on
Retroviruses and Opportunistic Infections. 31 January–4
February 1999, Chicago, IL, USA. Abstract 424.
4. Joshi AS, Barrett JS, Fiske WD, Pieniaszek HJLTM,
Bacheler LT, Saitoh A. Population pharmacokinetics of
efavirenz in Phase II studies and relationship with efficacy.
39th Interscience Conference Antimicrobial Agents and
Chemothrapy. 26–29 September 1999, San Francisco, CA,
USA. Abstract 1201.
5. Marzolini C, Telenti A, Decosterd LA, Greub G, Biollaz J,
Buclin T. Efavirenz plasma levels can predict treatment
failure and central nervous system side effects in HIV-1-
infected patients. AIDS 2001; 15:71–75.
6. Fletcher CV, Brundage RC, Fenton T, et al. Pharmacokinetics
and Pharmacodynamics of efavirenz and nelfinavir in HIV-
infected children participating in an area-under-the-curve
controlled trial. Clin Pharmacol Ther 2007; 83:300–306.
7. Ren Y, Nuttall JJ, Egbers C, et al. High prevalence of
subtherapeutic plasma concentrations of efavirenz in
children. J Acquir Immune Defic Syndr 2007; 45:133–136.
8. von Hentig N, Koenigs C, Elanjikal S, et al. Need for
therapeutic drug monitoring in HIV-1 infected children
receiving efavirenz doses according to international
guidelines. Eur J Med Res 2006; 11:377–380.
9. Haas DW, Ribaudo HJ, Kim RB, et al. Pharmacogenetics
of efavirenz and central nervous system side effects: an
Adult AIDS Clinical Trials Group study. AIDS 2004;
18:2391–2400.
10. Saitoh A, Fletcher CV, Brundage R, et al. Efavirenz
pharmacokinetics in HIV-1-infected children are associated
with CYP2B6-G516T polymorphism. J Acquir Immune
Defic Syndr 2007; 45:280–285.
11. Hofmann MH, Blievernicht JK, Klein K, et al. Aberrant
splicing caused by single nucleotide polymorphism
c.516G>T [Q172H], a marker of CYP2B6*6, is responsible
for decreased expression and activity of CYP2B6 in liver.
J Pharmacol Exp Ther 2008; 325:284–292.
12. European Medicines Agency website 2008. Scientific
discussion for the approval of Stocrin. (Updated 3
January 2004. Accessed8 August 2008.) Available from
http://www.emea.europa.eu/humandocs/PDFs/EPAR/
Stocrin/127599en6.pdf
13. Kappelhoff BS, Rosing H, Huitema AD, Beijnen JH. Simple
and rapid method for the simultaneous determination of the
non-nucleoside reverse transcriptase inhibitors efavirenz and
nevirapine in human plasma using liquid chromatography.
J Chromatogr B Analyt Technol Biomed Life Sci 2003;
792:353–362.
14. Lang T, Klein K, Fischer J, et al. Extensive genetic
polymorphism in the human CYP2B6 gene with impact on
expression and function in human liver. Pharmacogenetics
2001; 11:399–415.
15. Beal SL, Sheiner LB. NONMEM User’s Guides 1998.
NONMEM project group, University of California, San
Francisco, CA, USA.
16. Jonsson EN, Karlsson MO. Xpose-an S-PLUS based
population pharmacokinetic/pharmacodynamic model
building aid for NONMEM. Comput Methods Programs
Biomed 1999; 58:51–64.
17. Lindbom L, Ribbing J, Jonsson EN. Perl-speaks-NONMEM
(PsN)-a Perl module for NONMEM related programming.
Comput Methods Programs Biomed 2004; 75:85–94.
18. Hooker AC, Staatz CE, Karlsson MO. Conditional
weighted residuals (CWRES): a model diagnostic for the
FOCE method. Pharm Res 2007; 24:2187–2197.
19. Keizer RJ. Piraña 0.99beta 2008. (Updated 29 July 2008.
Accessed 8 August 2008.) Available from http://sourceforge.
net/projects/pirana/
20. Anderson BJ, Allegaert K, Holford NH. Population clinical
pharmacology of children: modelling covariate effects. Eur
J Pediatr 2006; 165:819–829.
21. Anderson BJ, Allegaert K, Holford NH. Population clinical
pharmacology of children: general principles. Eur J Pediatr
2006; 165:741–746.
22. Anderson BJ, Holford NH. Mechanism-based concepts
of size and maturity in pharmacokinetics. Annu Rev
Pharmacol Toxicol 2007; 48:303–332.
23. Kappelhoff BS, Huitema AD, Yalvac Z, et al. Population
pharmacokinetics of efavirenz in an unselected cohort of HIV-
1-infected individuals. Clin Pharmacokinet 2005; 44:849–861.
24. Karlsson MO, Sheiner LB. The importance of modeling
interoccasion variability in population pharmacokinetic
analyses. J Pharmacokinet Biopharm 1993; 21:735–750.
25. Zandvliet AS, Huitema AD, Copalu W, et al. CYP2C9
and CYP2C19 polymorphic forms are related to increased
indisulam exposure and higher risk of severe hematologic
toxicity. Clin Cancer Res 2007; 13:2970–2976.
26. Csajka C, Marzolini C, Fattinger K, et al. Population
pharmacokinetics and effects of efavirenz in patients with
human immunodeficiency virus infection. Clin Pharmacol
Ther 2003; 73:20–30.
ter Heine.indd 786 5/9/08 16:59:27
Pharmacokinetics of efavirenz in children
Antiviral Therapy 13.6 787
27. Kappelhoff BS, van LF, MacGregor TR, Lange J,
Beijnen JH, Huitema AD. Nevirapine and efavirenz
pharmacokinetics and covariate analysis in the 2NN study.
Antivir Ther 2005; 10:145–155.
28. Nyakutira C, Roshammar D, Chigutsa E, et al. High
prevalence of the CYP2B6 516G-–>T(*6) variant and effect
on the population pharmacokinetics of efavirenz in HIV/
AIDS outpatients in Zimbabwe. Eur J Clin Pharmacol
2008; 64:357–365.
29. Klein K, Lang T, Saussele T, et al. Genetic variability of
CYP2B6 in populations of African and Asian origin:
allele frequencies, novel functional variants, and possible
implications for anti-HIV therapy with efavirenz.
Pharmacogenet Genomics 2005; 15:861–873.
30. Gatanaga H, Hayashida T, Tsuchiya K, et al. Successful
efavirenz dose reduction in HIV type 1-infected individuals
with cytochrome P450 2B6 *6 and *26. Clin Infect Dis
2007; 45:1230–1237.
Accepted for publication 9 June 2008
ter Heine.indd 787 5/9/08 16:59:27
    • "where k tr is the transit rate constant and n equals the number of transition compartments, as described previously [19, 20] . The relationship between presystemic and systemic hepatic metabolism was accounted for by implementing a physiological well-stirred liver model, as previously proposed by Gordi et al. [21]. "
    [Show abstract] [Hide abstract] ABSTRACT: Introduction and objective: Everolimus (a drug from the class of mammalian target of rapamycin [mTOR] inhibitors) is associated with frequent toxicity-related dose reductions. Everolimus accumulates in erythrocytes, but the extent to which hematocrit affects everolimus plasma pharmacokinetics and pharmacodynamics is unknown. We aimed to investigate the everolimus pharmacokinetics/pharmacodynamics and the influence of hematocrit in cancer patients. Methods: A semi-physiological pharmacokinetic model for everolimus was developed from pharmacokinetic data from 73 patients by non-linear mixed-effects modeling. Using a simulation study with a known pharmacodynamic model describing S6K1 (a downstream mTOR effector) inhibition, we investigated the impact of hematocrit. Results: The apparent volume of distribution of the central and peripheral compartment were estimated to be 207 L with a relative standard error (RSE) of 5.0 % and 485 L (RSE 4.2 %), respectively, with an inter-compartmental clearance of 72.1 L/h (RSE 3.2 %). The apparent intrinsic clearance was 198 L/h (RSE 4.3 %). A decrease in hematocrit from 45 % to 20 % resulted in a predicted reduction in whole-blood exposure of ~50 %, but everolimus plasma pharmacokinetics and pharmacodynamics were not affected. The predicted S6K1 inhibition was at a plateau level in the approved dose of 10 mg once daily. Conclusions: A population pharmacokinetic model was developed for everolimus in cancer patients. Hematocrit influenced whole-blood pharmacokinetics, but not plasma pharmacokinetics or pharmacodynamics. Everolimus whole-blood concentrations should always be corrected for hematocrit. Since predicted mTOR inhibition was at a plateau level in the approved dose, dose reductions may have only a limited impact on mTOR inhibition.
    Full-text · Article · Jun 2016
    • "This is similar to observations by Marzolini et al. (2001), Ståhle et al. (2004), Lubomirov et al. (2011), Mukonzo et al. (2013, and Sanchez Martin et al. (2013). Poor virologic suppression and ADRs lead to patient's health deteriorating which could lead to increased non-adherence to treatment (ter Heine et al., 2008). Drug response is a complex phenotype and multiple factors including ethnicity, sex, age, body weight, drug-drug, and drugfood interactions, hepatic impairment, disease state, pregnancy, TABLE 2 | Multivariate logistic regression analysis of efavirenz plasma concentrations above or below 4 µg/mL. "
    [Show abstract] [Hide abstract] ABSTRACT: Introduction: Efavirenz (EFV) is a non-nucleoside reverse transcriptase inhibitor prescribed as part of first-line highly active antiretroviral therapy (HAART) in South Africa. Despite administration of fixed doses of EFV, inter-individual variability in plasma concentrations has been reported. Poor treatment outcomes such as development of adverse drug reactions or treatment failure have been linked to EFV plasma concentrations outside the therapeutic range (1-4 μg/mL) in some studies. The drug metabolizing enzyme (DME), CYP2B6, is primarily responsible for EFV metabolism with minor contributions by CYP1A2, CYP2A6, CYP3A4, CYP3A5, and UGT2B7. DME coding genes are also regulated by microRNAs through targeting the 3'-untranslated region. Expanded analysis of 30 single nucleotide polymorphisms (SNPs), including those in the 3'-UTR, was performed to identify pharmacogenetics determinants of EFV plasma concentrations in addition to CYP2B6 c.516G>T and c.983T>C SNPs. Methods: SNPs in CYP1A2, CYP2B6, UGT2B7, and NR1I2 (PXR) were selected for genotyping among 222 Bantu-speaking South African HIV-infected patients receiving EFV-containing HAART. This study is a continuation of earlier pharmacogenetics studies emphasizing the role of genetic variation in the 3'-UTR of genes which products are either pharmacokinetic or pharmacodynamic targets of EFV. Results: Despite evaluating thirty SNPs, CYP2B6 c.516G>T and c.983T>C SNPs remain the most prominent predictors of EFV plasma concentration. Conclusion: We have shown that CYP2B6 c.516G>T and c.983T>C SNPs are the most important predictors of EFV plasma concentration after taking into account all other SNPs, including genetic variation in the 3'-UTR, and variables affecting EFV metabolism.
    Full-text · Article · Jan 2016
    • "Similarly, an EFV pharmacokinetics study in Ugandan adults suggested that a daily EFV dose of 300 mg may be adequate for individuals carrying the CYP2B6 c.516 T/T genotype [47]. Ter Heine et al. [28], showed that children carrying the CYP2B6 c516G/G genotype had a 50–70 % probability of developing sub-therapeutic EFV concentrations, pointing towards the need for dose optimisation in both adults and children. Cases 1, 2 and 3, as compared to case 4, carried not only the CYP2B6 c. 516G > T SNP but additional heterozygous SNP's conferring impaired CYP2B6 activity resulting in higher plasma EFV concentrations. "
    [Show abstract] [Hide abstract] ABSTRACT: Background Efavirenz, widely used as part of antiretroviral drug regimens in the treatment of paediatric human immunodeficiency virus infection, has central nervous system side effects. We describe four children presenting with serious, persistent central nervous system adverse events who were found to have elevated plasma efavirenz concentrations as a result of carrying CYP2B6 single nucleotide polymorphisms, known to play a role in the metabolism of EFV. None of the children had a CYP2B6 wildtype haplotype. We believe this is the first case of cerebellar dysfunction associated with efavirenz use to be described in children. Case presentation Four black African children, between the ages of 4 and 8 years presenting between 1 and 20 months post-efavirenz initiation, are described. Cerebellar dysfunction, generalised seizures and absence seizures were the range of presenting abnormalities. Plasma efavirenz levels ranged from 20-60 mg/L, 5–15 times the upper limit of the suggested reference range. All abnormal central nervous system manifestations abated after efavirenz discontinuation. Conclusion Efavirenz toxicity should always be considered in human immunodeficiency virus-infected children with unexplained central nervous system abnormalities. Our findings further our understanding of the impact of genetic variants on antiretroviral pharmacokinetics in children across various ethnic groups. Screening for potential EFV-toxicity based on the CYP2B6 c.516 SNP alone, may not be adequate.
    Full-text · Article · Dec 2015
Show more