articles nature publishing group

The Effect of Atazanavir and Atazanavir/

Ritonavir on UDP-Glucuronosyltransferase Using
Lamotrigine as a Phenotypic Probe
DM Burger1, A Huisman1, N Van Ewijk1, H Neisingh2, P Van Uden2, GA Rongen2,3, P Koopmans4,5
and RJ Bertz6

Atazanavir (ATV) is known to inhibit UGT1A1-mediated glucuronidation. Here we report the effect of ATV and
ATV/ritonavir (RTV) on another UGT1A isoenzyme, UGT1A4. Twenty-one healthy volunteers received a single
dose of 100 mg of oral lamotrigine on days 1, 13, and 27; on each occasion blood was sampled before the dose was
administered and through 120 h after ingestion of the drug. On days 8–17 the subjects received oral ATV 400 mg q.d.
On days 18–30 the subjects received oral ATV 300 mg plus oral RTV 100 mg q.d. Seventeen subjects were evaluable for
pharmacokinetic analysis. Geometric mean ratios (+90% confidence intervals (CIs)) of lamotrigine area under the plasma
concentration-time curve (AUC)0–inf and peak plasma concentration (Cmax) for ATV + lamotrigine and for lamotrigine
alone were 0.88 (0.86–0.91) and 0.99 (0.95–1.02), respectively; the corresponding ratios for ATV/RTV and for lamotrigine
were 0.68 (0.65–0.70) and 0.94 (0.90–0.97), respectively. The mean ratio of lamotrigine-2N-glucuronide to lamotrigine
AUC0–inf increased from 0.45 for lamotrigine to 0.71 for ATV/RTV + lamotrigine. ATV alone does not significantly
influence glucuronidation of lamotrigine. In contrast, ATV/RTV results in moderately decreased exposure to lamotrigine.

Antiretroviral agents are well known to cause a wide variety of
clinically relevant drug–drug interactions.1 Most of these inter-
actions are based on inhibition or induction of the cytochrome
P450 system. For instance, all human immunodeficiency virus
(HIV) protease inhibitors are able to inhibit the CYP3A isoen-
zyme,2 whereas a clinical dose of ritonavir (RTV) also increases
plasma concentrations of desipramine, indicating CYP2D6-
mediated inhibition.3 Also, in a phenotypic drug cocktail study
it was recently observed that lopinavir/RTV induces CYP1A2,
CYP2C9, and CYP2C19.4
Less attention has been paid in the past to the potential
influence of antiretroviral agents on other drug-metabolizing
enzyme systems such as UDP-glucuronosyl-transferases (UGTs).
Members of the UGT superfamily are responsible for the metab-
olism of many drugs, environmental chemicals, and endogenous
compounds.5 To our knowledge, no extensive overview is avail-
able on the potential influence of antiretroviral agents on UGT,
nor have all antiretroviral agents been screened for any effect
on UGT-mediated metabolism of drugs. To date, although 17
human UGT proteins have been identified, only UGT1A1, 1A3,
1A4, 1A6, 1A9, 2B7, and 2B15 are considered isoforms of impor-
tance in hepatic drug elimination.6
When reviewing potential drug–drug interactions including
those involving antiretroviral agents and members of the UGT
superfamily, the first candidate drug is one that was licensed as
the first antiretroviral agent more than 20 years ago, namely,
zidovudine. Zidovudine is glucuronidated through UGT2B7,7
and an increase in plasma concentrations of the drug (and there-
fore in toxicity) may occur after the use of zidovudine combined
with atovaquone, fluconazole, methadone, or valproic acid.8
Another good example of drug–drug interaction is the inhibi-
tion of UGT1A1 by both atazanavir (ATV) and indinavir, mak-
ing combined use of either of these HIV protease inhibitors with
the chemotherapeutic agent and UGT1A1 substrate irinotecan a
contraindication.9 Also, the recently introduced HIV integrase
inhibitor raltegravir (MK-0518) is primarily glucuronidated by

1Department of Clinical Pharmacy, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands; 2Clinical Research Centre, Radboud University

Nijmegen Medical Centre, Nijmegen, The Netherlands; 3Department of Pharmacology and Toxicology, Radboud University Nijmegen Medical Centre, Nijmegen,
The Netherlands; 4Department of General Internal Medicine, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands; 5Nijmegen University
Center for Infectious Diseases, Nijmegen, The Netherlands; 6Department of Research and Development, Bristol-Myers Squibb Co., Princeton, New Jersey, USA.
Correspondence: DM Burger (D.Burger@akf.umcn.nl)
Received 2 February 2008; accepted 2 April 2008; advance online publication 4 June 2008. doi:10.1038/clpt.2008.106

698 VOLUME 84 NUMBER 6 | DECEMBER 2008 | www.nature.com/cpt

 articles

UGT1A1, and therefore its pharmacokinetics can be influenced
by both inhibitors (e.g., ATV) and inducers (e.g., rifampin, efa-
virenz, and RTV) of this phase II metabolic enzyme.
On the basis of these observations, we postulate that antiret-
roviral agents should be screened for potential influences on
the most relevant UGT isoenzymes. This has already been car-
ried out to some extent with in vitro inhibition experiments,9
but almost no information has been collected on the poten-
tial induction of UGT isoenzymes by antiretroviral agents in
this manner. There is both in vitro10 and clinical evidence that
RTV is able to induce glucuronidation of other drugs (i.e., ethi-
nylestradiol,11 pravastatin,12 and levothyroxine13). These clinical
data, however, were collected with RTV being used as a single
protease inhibitor in therapeutic doses of 400–600 mg b.i.d.,
whereas currently RTV is almost exclusively used in a low dose
of 100–200 mg q.d. or b.d. to boost plasma concentrations of a
second HIV protease inhibitor. We have recently demonstrated
that the HIV protease inhibitor combination lopinavir/RTV was
able to reduce the plasma concentrations of lamotrigine, which
is most probably caused by lopinavir/RTV-mediated induction
of UGT1A4.14 Because lopinavir and RTV are coformulated (as
Kaletra), we were not able to determine the individual effects of
lopinavir and RTV.
ATV/RTV is another preferred protease inhibitor combi-
nation for treatment-naive patients, according to the most
recent guidelines by the US Department of Health and Human
Services panel.15 Predicting the effect of ATV/RTV on UGT1A4
is difficult because the potential inhibitory effects of ATV on
UGT1A49 can be counteracted by the potential inducing effects
of RTV on this enzyme (see earlier text). Moreover, ATV is also
licensed for use in treatment-naive patients as a single protease
inhibitor (that is, without RTV), and its effects on UGT1A4
in vivo have not been studied. We therefore conducted a study
was suspected during the last phase of the study, based on
undetectable plasma ATV and RTV concentrations on day
21 and day 24 and an incorrect number of returned medica-
tion units.
Pharmacokinetics
Eighteen subjects completed day 32 of the trial. One subject
showed a deviating effect of ATV/RTV on lamotrigine pharma-
cokinetic parameters when compared with the other subjects.
This was the same individual who had been suspected of non-
adherence during the last phase of the trial (see earlier text); the
data from this individual were excluded from all pharmacoki-
netic analyses.
Lamotrigine plasma concentration vs. time curves for the
three study phases in the remaining 17 subjects are presented
in Figure 1. Because lamotrigine plasma concentrations became
undetectable in most subjects by 72 h after intake, only the
pharmacokinetic curves up to 48 h are presented. The geomet-
ric means (95% confidence intervals (CIs)) of lamotrigine area
under the plasma concentration-time curve (AUC)0–inf, peak
plasma concentration (Cmax), and t1/2 for the 17 subjects on day
1 (reference) were 46.8 (41.3–53.0) h·mg/l, 1.09 (1.05–1.13) mg/l,
and 29.3 (26.2–32.8) h, respectively. Table 1 shows the geomet-
ric mean ratios of the AUC0–inf, Cmax, and t1/2, comparing days
13–18 (lamotrigine + ATV) vs. days 1–5 (reference: lamotrigine
alone), and days 27–32 (lamotrigine + ATV/RTV) vs. days 1–5
(reference: lamotrigine alone).
On the basis of bioequivalence criteria between ATV + lam-
otrigine and lamotrigine alone, no effect could be demonstrated
for any of the three pharmacokinetic parameters (AUC, Cmax,
t1/2); the 90% CIs of the geometric mean ratio fall completely
1.2
LTG

in healthy volunteers to evaluate the effect of ATV and ATV/ 1.0

LTG+ATV 400 mg
LTG plasma concentration (mg/l)

RTV on UGT1A4, using lamotrigine as a phenotypic probe for LTG+ATV/r 300/100 mg

this enzyme. 0.8

Results 0.6

Baseline characteristics
0.4
Twenty-one healthy male subjects were included in this trial.
The mean (+ range) age, body weight, and body mass index 0.2
were 35 (19–54) years, 77 (65–92) kg, and 24 (21–28) kg/m2,
respectively. There was one black subject and one of mixed racial 0.0
0 4 8 12 16 20 24 28 32 36 40 44 48 52
background; the others were Caucasians. Eighteen subjects com- Time after intake (h)
pleted the trial and were evaluable for statistical analyses. The
dropouts were attributed to personal reasons (n = 1), nonad- Figure 1  Pharmacokinetic curves of lamotrigine (LTG) on days 1 (LTG alone),
herence to instructions given by study personnel (n = 1), and 13 (LTG + atazanavir (ATV)), and 27 (LTG + atazanavir/ritonavir (ATV/r)); n = 17.
The data are geometric mean values + SD.
development of skin rashes (n = 1).
Table 1  Geometric mean ratios and 90% confidence intervals
Compliance
The compliance of 17 of the 18 evaluable subjects was good, Parameter LTG + ATV vs. LTG alone LTG + ATV/r vs. LTG alone
as indicated by the subjects’ statements about the intake of AUC 0.88 (0.86–0.91) 0.68 (0.65–0.70)
the drug doses, the number of capsules in the returned blis- Cmax 0.99 (0.95–1.02) 0.94 (0.90–0.97)
ters and boxes, ATV and RTV trough concentrations, and the t1/2 0.91 (0.88–0.94) 0.73 (0.70–0.76)
information recorded by the subjects in the booklets (data ATV, atazanavir; ATV/r, atazanavir/ritonavir; AUC, area under the plasma
not shown). There was one subject in whom nonadherence concentration-time curve; Cmax, peak plasma concentration; LTG, lamotrigine.

Clinical pharmacology & Therapeutics | VOLUME 84 NUMBER 6 | DECEMBER 2008 699

articles

within the predefined interval of 0.80–1.25. In contrast, the
criteria for bioequivalence were not met between ATV/RTV +
lamotrigine and lamotrigine alone, as indicated by a 32% (90%
CI: 30–35%) decrease in lamotrigine AUC and a 27% (90% CI:
24–30%) decrease in lamotrigine elimination half-life when
ATV/RTV was coadministered with single-dose lamotrigine
(Figure 1, Table 1).
The geometric mean (95% CI) AUC0–inf of lamotrigine-2N-
glucuronide was 20.5 (19.0–22.2) h·mg/l after intake of 100 mg of
lamotrigine alone, leading to a mean (SD) AUC ratio of metabo-
lite vs. parent compound of 0.45 (0.09). This mean AUC ratio
increased to 0.52 (0.13), when lamotrigine was taken with ATV
400 mg q.d., and further to 0.71 (0.11), when lamotrigine was
taken with ATV/RTV 300/100 mg q.d., indicating an increased
formation of the 2N-glucuronide metabolite (P < 0.001 for both
comparisons; paired samples t-test).
ATV (days 13 and 27) and RTV (day 27) plasma concentration
vs. time curves are presented in Figure 2; the pharmacokinetic
parameters are listed in Table 2. The pharmacokinetic param-
eters of ATV and RTV were comparable to those of historical
controls in healthy subjects for both the ATV 400 mg and ATV/
RTV 300/100 mg dosing regimens.16
Adverse events and safety assessments
One serious adverse event was reported: a subject developed a
skin rash and neurological symptoms and discontinued study
medication on day 19. The neurological symptoms were not
7.0
6.0
Plasma concentration (mg/l)
considered to be related to the study medication but, rather, to
cetirizine, which was administered to treat the rash.
Another subject stopped taking study medication at day 10
because of adverse events. This subject complained of irrita-
tion of the eyes, having a cold, and generally feeling unwell.
The events were judged by the investigator as not serious and
possibly/probably related to the study medication. The subject
discontinued the study medication on his own initiative.
There were 179 adverse events reported in the 21 subjects who
participated in this study. Fifty-nine (33.0%) of these events were
classified as possibly or probably related to the study medication.
A large majority (43 of 59 = 72.9%) were classified as grade I; the
remaining 16 events were classified as grade II (6, 10.2%), grade
III (2, 3.4%), or grade IV (8, 13.6%). All grade III/IV adverse
events were elevations of serum bilirubin levels, which is a well-
known side effect of ATV. In the subjects with grade III/IV eleva-
tions of serum unconjugated bilirubin, serum transaminases
remained normal, excluding toxic hepatitis as a cause for the
increase in bilirubin.
DISCUSSION
The primary objective of this study was to evaluate the effect of
ATV and ATV/RTV on the pharmacokinetics of the UGT1A4
probe lamotrigine. On the basis of in vitro studies,9 it was antici-
pated that ATV could possibly inhibit UGT1A4, although to a
lesser extent than UGT1A1, and this could theoretically have
led to higher plasma concentrations of lamotrigine after a single
dose. The data on lamotrigine pharmacokinetics in this study
(although the 90% CI for the ratio of geometric means did not
ATV 400 mg include 1.0) met the predefined bioequivalence criteria for no
ATV/r 300/100 mg
RTV 100 mg
effect between lamotrigine + ATV and lamotrigine alone, indi-

5.0 cating that ATV alone does not significantly influence glucuroni-
dation of the UGT1A4 probe lamotrigine.
4.0
In contrast to this, the combination of ATV/RTV caused a
3.0 significant reduction in lamotrigine exposure and elimination
half-life. In line with these observations, the ratio of AUCs of
2.0
metabolite (lamotrigine-2N-glucuronide) vs. parent compound
1.0 (lamotrigine) increased significantly, demonstrating that induc-
tion of UGT1A4 is the most likely mechanism of this drug inter-
0.0
0 2 4 6 8 10 12 14 16 18 20 22 24 26 action. On the basis of these observations, one can extrapolate
Time after intake (h) that ATV/RTV may modestly decrease plasma concentrations
(and hence the pharmacological effects) of other UGT1A4 sub-
Figure 2  Pharmacokinetic curves of atazanavir (ATV) and ritonavir (RTV) on strates such as amitriptyline, clozapine, doxepine, ketotifen,
days 13 (ATV alone), and 27 (ATV/RTV); n = 17. The data are geometric mean
values + SD.
olanzapine, tamoxifen, and others.17

Table 2  Pharmacokinetic parameters of atazanavir and ritonavir when compared to historical controls16
Drug (dose (mg)) Atazanavir (400 mg q.d.) Atazanavir (300 mg q.d. (+ritonavir 100 mg)) Ritonavir (100 mg q.d. (+atazanavir 300 mg))
Pharmacokinetic This study Historical controls This study Historical controls This study Historical controls
parameter (n = 17) (n = 52) (n = 17) (n = 48) (n = 17) (n = 48)
AUC0–24 h (h·mg/l) 30.5 (22) 28.2 (38) 49.9 (25) 44.5 (33) 6.0 (49) 10.3 (35)
Cmax (mg/l) 4.9 (22) 5.3 (24) 5.3 (19) 5.1 (31) 1.2 (39) 1.9 (32)
Cmin (mg/l) 0.19 (53) 0.19 (104) 0.67 (41) 0.71 (58) 0.03 (NC) 0.04 (64)
tmax (h) 3 (2–5) 2 (1–4) 3 (2–5) 2.5 (2–4) 4 (1–5) 4 (1–4)
AUC, area under the plasma concentration-time curve; Cmax, peak plasma concentration; Cmin, trough plasma concentration; tmax, time to reach Cmax; NL, not calculated.
Data are geometric means + coefficient of variation (%), except for tmax (median + range).

700 VOLUME 84 NUMBER 6 | DECEMBER 2008 | www.nature.com/cpt


Given that ATV alone does not significantly change lamotrig-
ine pharmacokinetics, it is most reasonable to conclude that
the effect of ATV/RTV is predominantly, if not fully, caused
by low-dose RTV. On the other hand, if ATV alone is able to
induce UGT1A4, the increased exposure to ATV when boosted
by RTV may contribute to the observed induction of lamotrig-
ine glucuronidation. The final proof for RTV’s role in inducing
UGT1A4 would come from a trial in which only low-dose RTV
(i.e., without a second protease inhibitor) is tested with a single-
dose of lamotrigine. Although this type of drug interaction study
has been conducted in the past,3 the assessment of the clinical
relevance of such an observation remains problematic because
low-dose RTV without a second protease inhibitor is currently
not applied in clinical practice.
Although our data suggest that ATV does not contribute to
the effect of RTV on UGT1A4, such a conclusion may not be
extrapolatable to other combinations of RTV + a second protease
inhibitor. As noted earlier, we have demonstrated that the combi-
nation lopinavir/RTV caused a 50% (90% CI: 47–54%) decrease in
lamotrigine AUC, which is substantially larger than the 32% (90%
CI: 30–35%) reduction observed in this study with ATV/RTV.
This could possibly be explained by the higher daily RTV dose in
the lopinavir combination than in the ATV combination (100 mg
b.d. vs. 100 mg q.d.) or through induction by lopinavir. This differ-
ence in the effects of ATV/RTV vs. lopinavir/RTV on lamotrigine
glucuronidation was also observed for another potential UGT
substrate, abacavir. Waters et al. found an average 17% (95% CI:
12–22%) reduction in abacavir plasma exposure when combined
with ATV/RTV in HIV-infected patients, whereas this was an
average 32% (95% CI: 22–41%) with lopinavir/RTV.18 It is worth
noting that previous reports have shown that RTV ­exposures are
similar in the ATV/RTV and lopinavir/RTV regimens despite the
higher daily RTV dose in the lopinavir arm, suggesting that RTV
does not account for these differences.14,16
This study is the second time we have used lamotrigine as a
probe for UGT1A4 activity. Because we had encountered a rela-
tively high incidence of rashes in the previous trial,14 we modi-
fied the design of the current study in two important ways. First,
we excluded female subjects from participation because we had
observed a higher incidence of rashes in female subjects than
in male subjects (41.6% vs. 8.3%). Another reason for excluding
female subjects from a study with lamotrigine is to avoid possi-
ble hormonal influences on lamotrigine pharmacokinetics.19–21
Second, when using only single doses instead of chronic dosing,
a titration phase is not needed, and the occurrence of rashes due
to too rapid a dose titration is no longer an issue. Indeed, rash
occurred in only one subject (4.8%).
Study Day 1 2 3 4 5 6 7 8 9 10 11 12 13 14
LTG dose (mg) 100 100
ATV dose (mg) 400 400 400 400 400 400 400 400 400 400 300 300 300 300 300 300 300 300 300 300 300 300 300 300 300
RTV dose (mg)
PK sampling X X X X X X X X
Figure 3  Study design. LTG, lamotrigine; ATV, atazanavir; RTV, ritonavir; PK, pharmacokinetic.
Clinical pharmacology & Therapeutics | VOLUME 84 NUMBER 6 | DECEMBER 2008
articles
Besides the improved tolerance profile of single-dose vs.
chronic dosing of lamotrigine, there is one additional factor that
differentiates the applicability of lamotrigine for UGT1A4 pheno-
typing, and that is the occurrence of autoinduction of lamotrig-
ine metabolism. For instance, the half-life of lamotrigine after
chronic dosing was 20.1 h in our earlier trial,14 whereas in this
study it was 29.1 h after a single dose. Before conducting this trial
we decided to have a washout of at least 13 days between study
assessments, assuming that autoinduction after repeated single
doses would not be present (see Figure 3). Our observation that
administration of ATV alone was also associated with a statisti-
cally significant increase in the lamotrigine-2N-glucuronide vs.
lamotrigine AUC ratio (0.52 vs. 0.45; paired-samples t-test: P <
0.001) makes it difficult to distinguish between a minor, clinically
irrelevant inductive effect of ATV on UGT1A4 and the possibility
of autoinduction after the first single dose of lamotrigine. Even so,
the 57.7% increase in lamotrigine-2N-glucuronide vs. lamotrig-
ine AUC ratio that was observed with ATV/RTV administered in
combination with the third single dose of lamotrigine could have
been caused, at least in part, by autoinduction. However, we think
that autoinduction did not play an important role in possibly
confounding the effect of ATV and ATV/RTV on lamotrigine
pharmacokinetics because a nonspecific marker of induced liver
enzymes, γ-glutamyltransferase, remained unchanged during the
study period: geometric means on days 1, 13, and 27 were 19, 19,
and 18 IU/L, respectively. When using single-dose lamotrigine
in future studies as a UGT1A4 probe, however, one should con-
sider a randomized rather than a sequential crossover design
so as to correct for potential autoinduction effects or build in a
longer washout period between doses than the 13 days we used
in this study.
Finally, there are two other aspects that need to be addressed
when discussing lamotrigine as a phenotypic probe. First, the
sampling period chosen was 5 days, but lamotrigine plasma con-
centrations were detectable only up to 48 h after dosing. A dose of
lamotrigine higher than 100 mg or a more sensitive assay would
be needed to follow the formation of the 2N-glucuronide over a
longer period than 48 h. However, when comparing AUC0–48 h vs.
AUC0–inf, the magnitude of the changes produced by ATV alone
and by ATV/RTV were similar, suggesting that the current design
and analytical techniques are adequate. Second, the effects of ATV
and ATV/RTV on lamotrigine Cmax were different than on lam-
otrigine AUC or t1/2 (Table 1); this indicates the inappropriateness
of using lamotrigine Cmax as a UGT1A4 probe, which is in line
with minimal first-pass metabolism of the drug.
It was recently demonstrated in in vitro studies that lamotrig-
ine 2N-glucuronidation takes place not only through UGT1A4
15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
100
100 100 100 100 100 100 100 100 100 100 100 100 100 100 100
X X X X X X X X X X
701
articles
but perhaps, at lower concentrations, through UGT2B7 as
well.22 This observation challenges the appropriateness of using
lamotrigine as a selective probe for UGT1A4 phenotyping. The
authors, however, acknowledge that the relative contributions of
UGT1A4 and UGT2B7 are dependent on the presence of bovine
serum albumin in the system, and this puts into question the
validity of extrapolating in vitro data to an in vivo context. As
Kiang et al. mention in their review of UGT drug interactions,
there are difficulties in predicting in vivo effects of UGT enzymes
on the basis of data from in vitro experiments; e.g., UGT
enzymes are located in the lumen of the endoplasmic reticulum,
and assay conditions may preclude the movement of substrates
and glucuronidated products to and from the active site.17 Until
more research is done, single-dose lamotrigine will continue to
be considered an attractive in vivo UGT1A4 substrate or (with
less certainty) a mixed UGT1A4/2B7 substrate.
In summary, our data showed a significant and clinically rel-
evant decrease in lamotrigine exposure when ATV/RTV was
administered to healthy volunteers; in contrast, ATV without
RTV did not influence lamotrigine pharmacokinetics in a clini-
cally relevant manner. Physicians should be aware of potential
suboptimal therapy when boosted protease inhibitors are used
together with UGT1A4 substrates.
METHODS
Study design. This open-label, sequential, three-period, single-center,
phase IV, multiple-dose trial was conducted in April and May 2006 at
the Radboud University Nijmegen Medical Centre (Nijmegen, The
Netherlands). The study was designed to examine the effect of ATV
and low-dose RTV + ATV on the pharmacokinetics of lamotrigine and
lamotrigine-2N-glucuronide as determined by intrasubject comparison.
The secondary objectives were to assess the effect of lamotrigine on ATV,
and on ATV/RTV in comparison with historical controls, and to deter-
mine the safety of the use of lamotrigine in combination with either ATV
or ATV/RTV. A single dose of 100 mg of lamotrigine was given alone on
day 1 of the trial. After 1 week, the subjects started receiving oral ATV
400 mg q.d. (taken with breakfast around 9 am) from day 8 to day 17.
On day 13 (together with the sixth dose of ATV) a second single dose
of 100 mg of lamotrigine was administered. On day 18, the ATV regi-
men was replaced with a combination of low-dose RTV (100 mg) + ATV
300 mg, and this new regimen continued up to day 32. After 10 days on
the new regimen (day 27), the subjects took their third and final dose of
100 mg of lamotrigine. The single doses of lamotrigine on days 1, 13, and
27 were taken together with a standardized breakfast that contained a
minimum of two and a maximum of three slices of wheat bread with but-
ter, ham, cheese, or jam, together with a free choice of water, milk, cof-
fee (with free choice of sugar and/or milk), tea (with free choice of sugar
and/or milk), or apple juice (~300 kcal; 10.7% fat on a g/g basis).
The trial was approved by the Review Board of the Radboud University
Nijmegen Medical Centre.
Study population. This trial was conducted in healthy men between
the ages of 18 and 55 years. For inclusion in the study, subjects were
required to be in a good, age-appropriate health condition as established
by medical history and physical examination and by electrocardiography,
biochemistry, hematology, and urinalysis testing within 3 weeks before
the first dose. Subjects had to be able and willing to sign the Informed
Consent Form before screening evaluations. The main exclusion criteria
were a history of sensitivity/idiosyncratic response to ATV, RTV, lam-
otrigine, or chemically related compounds or excipients, a positive HIV
test, a positive hepatitis B or hepatitis C test, or therapy with any drug (for
2 weeks preceding dosing), except for acetaminophen and loperamide.
702
Study drug and dosing. The subjects received a single dose of 100 mg
of lamotrigine on study days 1 (reference), 13 (on steady-state ATV
400 mg q.d.), and 27 (on steady-state ATV/RTV 300/100 q.d.). An ear-
lier study with lamotrigine as metabolic probe for UGT1A4 had used
chronic dosing with lamotrigine to attain steady-state conditions, but
that trial was associated with a relatively high incidence of rashes (25%).14
To reduce the incidence of rashes (and consequent withdrawal from the
study), we administered only single doses of lamotrigine. The dose was
selected on the basis of a previously published study that showed that
plasma concentrations of lamotrigine remained detectable for 120 h after
intake of a single dose of 100 mg of lamotrigine.23
ATV (400 mg q.d.) and ATV/RTV (300/100 mg q.d.) doses were
the same as those recommended in the US package insert for Reyataz
(­Bristor-Myers Squibb, Woerden, The Netherlands). Earlier studies dem-
onstrated that steady-state conditions were present after at least 6 d (ATV
alone) or 10 d (ATV/RTV),24 and therefore single doses of lamotrigine
were administered after the subjects had taken these medications for the
relevant duration. Figure 3 is a graphical depiction of the trial design.
Safety assessments and pharmacokinetic sampling. Blood samples for
pharmacokinetics were collected several times during a 120-h period
(at 0, 1, 2, 3, 4, 5, 6, 8, 24, 48, 72, 96, and 120 h) after dosing on days 1,
13, and 27 for lamotrigine and lamotrigine-2N-glucuronide, on day 13
for ATV, and on day 27 for ATV + RTV, to characterize drug absorption
and elimination. Blood samples were collected to determine trough levels
of ATV just before administering the drug on days 8, 11, 13–18, 21, 24,
and 27–32.
Serum biochemistry, hematology, and urinalysis were carried out
on days 1, 8, 11, 13, 16, 18, 21, 24, 27, 29, and 32. Adverse events were
assessed during the same visits. Screening for drugs of abuse was con-
ducted on days 1, 13, and 27.
Compliance. Study personnel supervised all intake of medication at the
clinical trial unit. The exact times of dosing were recorded. Drug intake
by the subjects at home was monitored by pill counts and trough-level
measurements at days 11, 13, 21, 24, and 27. The subjects were also asked
to record in a booklet the exact times of medication intake.
Bioanalysis of ATV, RTV, lamotrigine, and lamotrigine-2N-glucuronide
concentrations in plasma. Plasma levels of ATV and RTV were analyzed
using a validated high-performance liquid chromatography assay.16,25
The accuracy values for ATV were 101% at 0.15 mg/l, 102% at 1.5 mg/l,
and 103% at 7.5 mg/l. For RTV, the accuracy values were 101, 104, and
103%, respectively. The precision values (within-day coefficient of varia-
tion) for ATV were 2.06, 2.68, and 5.07%, respectively, for concentrations
of 0.15, 1.5 and 7.5 mg/l. For RTV, the corresponding precision values
were 3.22, 1.70, and 0.89%, respectively.
Plasma levels of lamotrigine and lamotrigine-2N-glucuronide were
analyzed using a validated reversed-phase high-performance liquid
chromatography method as described earlier.14 The accuracy values for
lamotrigine using this method were 103, 103, and 104%, respectively,
for concentrations of 0.358, 1.79, and 11.94 mg/l. The precision values
(within-day coefficient of variation) were 2.34, 1.82, and 1.87%, respec-
tively, for the same concentrations. For lamotrigine-2N-glucururonide,
the accuracy values were 99, 100, and 99%, respectively, for concentra-
tions of 0.218, 1.09, and 7.25 mg/l. At these same concentrations, the
precision values (within-day coefficient of variation) were 4.08, 2.43, and
1.06, respectively.
Pharmacokinetic analysis. Pharmacokinetic parameters for lamotrigine,
lamotrigine-2N-glucuronide, ATV, and RTV were calculated by noncom-
partmental methods using the WinNonlin software package (version 4.1;
Pharsight, Mountain View, CA) and the log/linear trapezoidal rule. On
the basis of the individual plasma concentration-time data, the following
pharmacokinetic parameters of lamotrigine were determined: the AUC
from 0 to infinity (AUC0–inf; in milligram/hour/liter), the maximum con-
centration of the drug in plasma (Cmax; in milligrams/liter), the time to
reach Cmax (tmax; in hours), and the apparent elimination half-life (t1/2; in
VOLUME 84 NUMBER 6 | DECEMBER 2008 | www.nature.com/cpt

hours). For ATV and RTV, the same pharmacokinetic parameters were
calculated, except that here AUC was calculated at steady state within an
interval between doses from 0 to 24 h after intake (AUC0–24); in addition,
the trough concentration was determined as being the predose 24-h con-
centration after an observed dose (Cmin; in milligrams/liter).
Sample size and statistical analysis. The study was powered to detect a
15% difference in lamotrigine AUC0–inf. The required number of partici-
pants was calculated as 16. With an estimated maximum dropout rate
of 20%, 20 subjects had to be included in this trial to ensure completed
data from 16 subjects.
To identify interactions that were not clinically relevant, we used the
bioequivalence approach.26 Geometric means were calculated for the
AUC0–inf, Cmax, and t1/2. Geometric mean ratios with 90% CIs were cal-
culated after log-transformation of within-patient ratios. A geometric
mean ratio with 90% CI falling entirely within 0.80–1.25 was considered
to indicate no significant effect.
We calculated AUC ratios of lamotrigine-N-glucuronide and lamo-
trigine on days 1 (reference), 13 (ATV alone), and 27 (low-dose RTV
+ ATV) to determine the effect of ATV with and without RTV on the
metabolism (glucuronidation) of lamotrigine.
Statistical evaluations were carried out using SPSS for Windows,
­version 12.0.1 (SPSS, Chicago, IL; 1989–2003).
Acknowledgments
We thank the healthy volunteers for participating in this trial. The technicians
of the Department of Clinical Pharmacy, Radboud University Nijmegen
Medical Centre, are gratefully acknowledged for processing and analyzing
the plasma samples for levels of lamotrigine, lamotrigine-2N-glucuronide,
atazanavir, and ritonavir. This study was funded by an educational grant
from Bristol-Myers Squibb.
Conflict of interest
D.M.B. has received honoraria for serving on advisory boards, speaker’s fees,
and educational grants for clinical research from Bristol-Myers Squibb, the
manufacturer of atazanavir. R.J.B. is an employee of Bristol-Myers Squibb. All
other authors declared no conflict of interest.
© 2008 American Society for Clinical Pharmacology and Therapeutics
1. Boffito, M. et al. Therapeutic drug monitoring and drug-drug interactions
involving antiretroviral drugs. Antivir. Ther. 10, 469–477 (2005).
2. Granfors, M.T., Wang, J.S., Kajosaari, L.I., Laitila, J., Neuvonen, P.J. &
Backman, J.T. Differential inhibition of cytochrome P450 3A4, 3A5 and 3A7 by
five human immunodeficiency virus (HIV) protease inhibitors in vitro. Basic
Clin. Pharmacol. Toxicol. 98, 79–85 (2006).
3. Aarnoutse, R.E. et al. Effect of low-dose ritonavir (100 mg twice daily) on the
activity of cytochrome P450 2D6 in healthy volunteers. Clin. Pharmacol. Ther.
78, 664–674 (2005).
4. Yeh, R.F. et al. Lopinavir/ritonavir induces the hepatic activity of cytochrome
P450 enzymes CYP2C9, CYP2C19, and CYP1A2 but inhibits the hepatic and
intestinal activity of CYP3A as measured by a phenotyping drug cocktail in
healthy volunteers. J. Acquir. Immune Defic. Syndr. 42, 52–60 (2006).
5. Miners, J.O., Knights, K.M., Houston, J.B. & Mackenzie, P.I. In vitro-in vivo
correlation for drugs and other compounds eliminated by glucuronidation in
humans: pitfalls and promises. Biochem. Pharmacol. 71, 1531–1539 (2006).
6. Miners, J.O., Smith, P.A., Sorich, M.J., McKinnon, R.A. & Mackenzie, P.I. Predicting
human drug glucuronidation parameters: application of in vitro and in silico
modeling approaches. Annu. Rev. Pharmacol. Toxicol. 44, 1–25 (2004).
7. Barbier, O. et al. 3’-azido-3’-deoxythimidine (AZT) is glucuronidated by human
UDP-glucuronosyltransferase 2B7 (UGT2B7). Drug Metab. Dispos. 28, 497–502
(2000).
Clinical pharmacology & Therapeutics | VOLUME 84 NUMBER 6 | DECEMBER 2008
articles
8. Trapnell, C.B., Klecker, R.W., Jamis-Dow, C. & Collins, J.M. Glucuronidation
of 3′-azido-3′-deoxythymidine (zidovudine) by human liver microsomes:
relevance to clinical pharmacokinetic interactions with atovaquone,
fluconazole, methadone, and valproic acid. Antimicrob. Agents Chemother.
42, 1592–1596 (1998).
9. Zhang, D., Chando, T.J., Everett, D.W., Patten, C.J., Dehal, S.S. & Humphreys, W.G.
In vitro inhibition of UDP glucuronosyltransferases by atazanavir and other
HIV protease inhibitors and the relationship of this property to in vivo bilirubin
glucuronidation. Drug Metab. Dispos. 33, 1729–1739 (2005).
10. Smith, C.M., Faucette, S.R., Wang, h. & LeCluyse, E.L. Modulation of UDP-
glucuronosyltransferase 1A1 in primary human hepatocytes by prototypical
inducers. J. Biochem. Mol. Toxicol. 19, 96–108 (2005).
11. Ouellet, D. et al. Effect of ritonavir on the pharmacokinetics of ethinyl
oestradiol in healthy female volunteers. Br. J. Clin. Pharmacol. 46, 111–116
(1998).
12. Fichtenbaum, C.J. et al. Pharmacokinetic interactions between protease
inhibitors and statins in HIV seronegative volunteers: ACTG Study A5047. AIDS
16, 569–577 (2002).
13. Tseng, A.L. & Fletcher, D. Interaction between ritonavir and levothyroxine.
AIDS 12, 2235–2236 (1998).
14. Van der Lee, M.J., et al. Lopinavir/ritonavir reduces lamotrigine plasma
concentrations in healthy subjects. Clin. Pharmacol. Ther. 80, 159–168
(2006).
15. Working Group of the office of AIDS Research Advisory Council (OARAC).
Guidelines for the Use of Antiretroviral Agents in HIV-1-Infected Adults and
Adolescents, version 1, December 2007. <http://www.aidsinfo.nih.gov/
guidelines/>. Accessed 26 January 2008.
16. Burger, D.M., Agarwala, S., Child, M., Been-Tiktak, A., Wang, Y., Bertz, & R. Effect
of rifampin on steady-state pharmacokinetics of atazanavir with ritonavir in
healthy volunteers. Antimicrob. Agents Chemother. 50, 3336–3342 (2006).
17. Kiang, T.K., Ensom, M.H. & Chang, T.K. UDP-glucuronosyltransferases and
clinical drug-drug interactions. Pharmacol. Ther. 106, 97–132 (2005).
18. Waters, L.J. et al. Abacavir plasma pharmacokinetics in the absence and
presence of atazanavir/ritonavir or lopinavir/ritonavir and vice versa in HIV-
infected patients. Antivir. Ther. 12, 825–830 (2007).
19. Sidhu, J., Job, S., Singh, S. & Philipson, R. The pharmacokinetic and
pharmacodynamic consequences of the co-administration of lamotrigine
and a combined oral contraceptive in healthy female subjects. Br. J. Clin.
Pharmacol. 61, 191–199 (2006).
20. Christensen, J. et al. Oral contraceptives induce lamotrigine metabolism:
evidence from a double-blind, placebo-controlled trial. Epilepsia 48, 484–489
(2007).
21. Contin, M. et al. Variation in lamotrigine plasma concentrations with
hormonal contraceptive monthly cycles in patients with epilepsy. Epilepsia
47, 1573–1575 (2006).
22. Rowland, A., Elliot, D.J., Williams, J.A., Mackenzie, P.I., Dickinson, R.G. &
Miners, J.O. In vitro characterization of lamotrigine N2-glucuronidation and
the lamotrigine-valproic acid interaction. Drug Metab. Dispos. 34, 1055–1062
(2006).
23. Odishaw, J. & Chen, C. Effects of steady-state bupropion on the
pharmacokinetics of lamotrigine in healthy subjects. Pharmacotherapy
20, 1448–1453 (2000).
24. L‘homme, R.F., Dijkema, T., van der Ven, A.J. & Burger, D.M. Brief report: enzyme
inducers reduce elimination half-life after a single dose of nevirapine in
healthy women. J. Acquir. Immune Defic. Syndr. 43, 193–196 (2006).
25. Droste, J.A., Verweij-Van Wissen, C.P. & Burger, D.M. Simultaneous
determination of the HIV drugs indinavir, amprenavir, saquinavir, ritonavir,
lopinavir, nelfinavir, the nelfinavir hydroxymetabolite M8 and nevirapine in
human plasma by reversed phase high performance liquid chromatography.
Ther. Drug Monit. 25, 393–399 (2003).
26. US Department of Health and Human Services, Center for Drug Evaluation
and Research, Food and Drug Administration. Guidance for Industry.
Statistical approaches to establishing bioequivalence. <http://www.fda.
gov/cder/guidance/index.htm>. Accessed 10 August 2007.
703