Candesartan Has No Clinically Meaningful Effect on the Plasma
Concentrations of Cytochrome P450 2C8 Substrate Repaglinide
in HumansS
Mikael O.W. Piha, Kristiina Cajanus, Marica T. Engstr€om, Mikko Neuvonen,
Troels K. Bergmann, Mikko Niemi, Janne T. Backman, Anne M. Filppula, and
Aleksi Tornio
Integrative Physiology and Pharmacology, Institute of Biomedicine, University of Turku, Turku, Finland (M.O.W.P., K.C., A.T.);
Bioanalytical Laboratory, Institute of Biomedicine, University of Turku, Turku, Finland (M.T.E.); Unit of Clinical Pharmacology, Turku
University Hospital, Turku, Finland (M.O.W.P., K.C., A.T.); Department of Clinical Pharmacology, University of Helsinki, Helsinki,
Finland and Individualized Drug Therapy Research Program, Faculty of Medicine, University of Helsinki, Helsinki, Finland (M.Ne.,
M.Ni., J.T.B.); Department of Regional Health Research, University of Southern Denmark, Esbjerg, Denmark and Department of
Clinical Pharmacology, Odense University Hospital, Odense, Denmark (T.K.B.); Department of Clinical Pharmacology, HUS
Diagnostic Center, Helsinki University Hospital, Helsinki, Finland (M.Ni., J.T.B.); and Pharmaceutical Sciences Laboratory, Faculty
of Science and Engineering, Åbo Akademi University, Turku, Finland (A.M.F.)
Received May 15, 2024; accepted October 3, 2024
ABSTRACT
In vitro evidence shows that the acyl-b-D-glucuronide metabolite
of candesartan inhibits cytochrome P450 (CYP) 2C8 with an inhibi-
tion constant of 7.12 lM. We investigated the effect of candesartan
on the plasma concentrations and glucose-lowering effect of repa-
glinide, a sensitive clinical CYP2C8 index substrate. In a random-
ized crossover study, ten healthy volunteers ingested 8 mg of
candesartan or placebo daily for three days, and on day 3, they
also ingested 0.25 mg of repaglinide one hour after candesartan or
placebo. We measured the plasma concentrations of repaglinide,
candesartan, and candesartan acyl-b-D-glucuronide, and blood
glucose concentrations for up to nine hours after repaglinide in-
take. Candesartan had no effect on the area under the plasma
concentration-time curve and peak plasma concentration of repaglinide
compared with placebo, with ratios of geometric means of 1.02 [P 5
0.809; 90% confidence interval (CI) 0.90–1.15] and 1.13 (P5 0.346; 90%
CI 0.90–1.43), respectively. Other pharmacokinetic variables and blood
glucose concentrations were neither affected. Candesartan acyl-b-D-
glucuronide was detectable in seven subjects, in whom the peak con-
centration of repaglinide was 1.32-fold higher in the candesartan
phase than in the placebo phase (P 5 0.041; 90% CI 1.07–1.62). Sys-
temic concentrations of candesartan acyl-b-D-glucuronide were very
low compared with its CYP2C8 inhibition constant (ratio 
 0.1). Fur-
thermore, in a cohort of 93 cancer patients, no indication of decreased
paclitaxel clearance was found in four patients using candesartan
concomitantly. In conclusion, candesartan therapy is unlikely to in-
hibit CYP2C8-mediated metabolism of other drugs to any clinically
significant extent.
SIGNIFICANCE STATEMENT
The findings of this study suggest that candesartan is unlikely to
cause drug-drug interactions via inhibition of cytochrome P450
(CYP) 2C8. Although candesartan acyl-b-D-glucuronide has been
shown to inhibit CYP2C8 in vitro, it shows no clinically relevant
CYP2C8 inhibition in humans due to low systemic concentrations.
Introduction
Candesartan, an angiotensin II receptor inhibitor, is widely used to
treat elevated blood pressure and systolic heart failure (Cernes et al.,
2011). Administered as the prodrug candesartan cilexetil, active cande-
sartan is formed through hydrolysis during absorption, yielding an abso-
lute bioavailability of 14% for candesartan from the tablet formulation.
Peak plasma concentration (Cmax) is reached approximately four hours
after ingestion, and more than 99% of candesartan is protein-bound.
Candesartan is excreted into both feces and urine, primarily as candesar-
tan (70–80% of radioactivity), with an elimination half-life of approxi-
mately nine hours. In addition, a small proportion is metabolized through
O-deethylation and glucuronidation (Kondo et al., 1996a,b; United States
Food and Drug Administration, 1998; Alonen et al., 2008a,b; Katsube
This work was supported by State funding for university-level health research
(the Hospital District of Southwest Finland, Finland) and the TYKS Foundation
(Turku, Finland).
No author has an actual or perceived conflict of interest with the contents of
this article.
Part of the results of this work were presented at the Second Nordic Confer-
ence on Personalized Medicine in Turku, Finland, from 14 to June 16, 2023,
and at the 19th World Congress of Basic & Clinical Pharmacology in Glasgow,
United Kingdom, on July 6, 2023.
dx.doi.org/10.1124/dmd.124.001798.
S This article has supplemental material available at dmd.aspetjournals.org.
ABBREVIATIONS: AUC, area under the plasma concentration-time curve; AUCR, area under the plasma concentration-time curve ratio; CI,
confidence interval; Cmax, peak plasma concentration; CYP, cytochrome P450; DDI, drug-drug interaction; Ki, inhibition constant; OATP, organic
anion-transporting polypeptide; t1=2, terminal half-life; tmax, time to Cmax; UGT, uridine 5’-diphospho-glucuronosyltransferase.
1388
1521-009X/52/12/1388–1395$35.00 dx.doi.org/10.1124/dmd.124.001798
DRUG METABOLISM AND DISPOSITION Drug Metab Dispos 52:1388–1395, December 2024
Copyright © 2024 by The Author(s)
This is an open access article distributed under the CC BY-NC Attribution 4.0 International license.
http://dmd.aspetjournals.org/content/suppl/2024/11/01/dmd.124.001798.DC1
Supplemental material to this article can be found at: 
 at A
SPET Journals on N
ovem
ber 21, 2024
dm
d.aspetjournals.org
D
ow
nloaded from
 
et al., 2021). Candesartan has been thought to possess low drug–drug in-
teraction (DDI) potential based on in vitro and in vivo studies (Jonkman
et al., 1997; Taavitsainen et al., 2000; Pietruck et al., 2005; Miura et al.,
2009; Aberg et al., 2011; Brendel et al., 2013; Senda et al., 2017; Kim
et al., 2018; Gundlach et al., 2021).
Previously, the cytochrome P450 (CYP) 2C8 inhibitory potency of can-
desartan was shown to be relatively low in vitro with a half-maximal in-
hibitory concentration of 36.2 lM (Walsky et al., 2005). Recently,
however, Katsube et al. (2018) found an increased incidence of neutrope-
nia in patients receiving candesartan at a median dose of 8 mg, and the
anticancer agent paclitaxel, a CYP2C8 substrate, concomitantly. In a
subsequent study, the authors demonstrated that candesartan and its glucu-
ronide metabolites inhibited the CYP2C8 and CYP3A4-mediated hydrox-
ylation of paclitaxel, as well as organic anion-transporting polypeptide
(OATP) 1B1 and OATP1B3 in vitro. Of note, the acyl-b-D-glucuronide
of candesartan inhibited the CYP2C8-mediated hydroxylation of paclitaxel
with a particularly low inhibition constant (Ki) of 7.12 lM. This suggests
that candesartan could increase the concentrations of CYP2C8 substrate
drugs in humans, which could also explain the observed cases of neu-
tropenia during concomitant treatment with candesartan and paclitaxel
(Katsube et al., 2021). In rats and dogs, candesartan acyl-b-D-glucuronide
was present in the urine and feces, but it was undetectable in the plasma of
rats (Kondo et al., 1996a,b). Recombinant human uridine 50-diphospho-
glucuronosyltransferase (UGT) enzymes as well as human liver microsomes
also metabolize candesartan into its acyl-b-D-glucuronide (Alonen et al.,
2008a,b), but the presence of this metabolite in humans is not known.
Glucuronide metabolites of drugs have previously been shown to me-
diate clinically significant DDIs, especially those involving CYP2C8
(Niemi et al., 2003; Shitara et al., 2004; Ogilvie et al., 2006; Tornio et al.,
2014, 2022; Backman et al., 2016). For example, clopidogrel increased
the plasma concentrations of repaglinide, a clinical index substrate of
CYP2C8, up to 5-fold, with clopidogrel acyl-b-D-glucuronide identified
as the actual inhibitor according to in vitro data and pharmacokinetic
modeling (Tornio et al., 2014). In the current trial, we aimed to investigate
the effect of the concomitant administration of candesartan on repaglinide
in healthy volunteers. We also measured the plasma concentrations of can-
desartan acyl-b-D-glucuronide to assess its role in the possible DDI.
Materials and Methods
Subjects and Study Design. Ten healthy nonsmoking volunteers (six men
and four women; age range, 19–28 years; body mass index range, 18.9–28.1 kg/m2)
not using any continuous medication, including hormonal contraception, were re-
cruited to the trial. Before inclusion, they gave written informed consent, and
their health was confirmed by medical history interview, physical examination,
and routine laboratory tests (basic blood count and blood platelets, plasma ala-
nine aminotransferase, plasma alkaline phosphatase, plasma glutamyl transferase,
plasma creatinine, estimated glomerular filtration rate, plasma sodium, plasma
potassium, and plasma glucose, as well as serum human chorionic gonadotropin
for females). All subjects had their plasma creatinine, potassium, and sodium
concentrations within reference ranges, as well as their systolic blood pressure at
or above 110 mmHg. The study protocol was approved by the Ethics Committee
of the Hospital District of Southwest Finland (ETMK 79/2021) and the Finnish
Medicines Agency Fimea (EudraCT number 2021-003178-29). The subjects
were randomized to ingest either placebo (Placebo 9 mm tablet, University Phar-
macy, Helsinki, Finland) or 8 mg of candesartan cilexetil (Atacand 8 mg tablet,
Cheplapharm Arzneimittel GmbH, Greifswald, Germany) once daily at 8AM for
3 days per an open-label, two-phase, crossover design. A wash-out period of at
least two weeks separated the trial phases. In the morning of day 3, after an over-
night fast, the subjects ingested 0.25 mg of repaglinide (prepared as capsules by
Tyks Hospital Pharmacy from Repaglinide Krka 0.5 mg tablets, KRKA, d.d.,
Novo mesto, Slovenia) at 9AM, precisely 1 hour after the final dose of placebo
or candesartan. The subjects received a standardized light breakfast 15 minutes
after repaglinide ingestion, snacks after 1 and 2 hours, a warm meal after
3 hours, and snacks after 7 and 9 hours. Oral carbohydrates, intramuscular
glucagon and intravenous glucose solution were also available but were not
needed. Consumption of grapefruit and its juice was prohibited for one week
prior to and throughout the trial, and consumption of other drugs was prohib-
ited for one week prior to and during the days of repaglinide administration.
Blood Sampling. On the study days (day 3), blood was sampled from an an-
tebrachial vein through a cannula or needle at our laboratory: before administer-
ing placebo or candesartan, as well as 5 minutes before and 15, 30, 45, 60, 80,
and 100 minutes after and 2, 2 1/2, 3, 4, 5, 7, and 9 hours after administering repa-
glinide. Blood samples were collected into EDTA-containing tubes. As a safety
measure, glucose concentrations were immediately measured in the whole blood
samples using an instant glucose meter (CareSense Dual; I-Sense Inc, Seoul, Korea).
Plasma was then separated by centrifugation and stored at -80C until analysis.
Drug Concentration Measurements. The plasma concentrations of repagli-
nide, candesartan, and candesartan acyl-b-D-glucuronide were measured in the
collected samples using liquid chromatography-tandem mass spectrometry. Ref-
erence repaglinide, candesartan, and the corresponding stable isotope-labeled in-
ternal standards repaglinide-d5 and candesartan-d5 were purchased from Toronto
Research Chemicals (North York, ON, Canada). Candesartan acyl-b-D-glucuro-
nide was purchased from SynInnova (Edmonton, AB, Canada). Other reagents
and organic solvents were of commercially available analytical grade.
For repaglinide, reference calibration standards and quality control samples were
prepared in blank human plasma and processed along with the study samples. The
samples were purified using a 96-well Oasis MAX lElution plate (Waters Corpora-
tion, Milford, MA, USA) according to the manufacturer’s instructions. An aliquot
of 100 ll of sample was mixed with 30 ll of 5% phosphoric acid containing the in-
ternal standard (5 ng/ml), and the sample mixture was drawn through the precondi-
tioned extraction plate. The plate was then washed with 200 ll of 5% ammonium
hydroxide in water, and the compounds were eluted three times with 30 ll of 2.5%
formic acid in methanol. The chromatographic separation was performed on a Sym-
metry C8 column (150 × 2.1 mm internal diameter, particle size 3.5 lm; Waters
Corporation, Milford, MA, USA) using a mobile phase of 5 mM ammonium for-
mate, pH 3.6 adjusted with glacial formic acid (channel A) and acetonitrile
(channel B). The mobile phase gradient was set as follows: 1 minute at 40% B, a
linear ramp to 65% B over 2 minutes, 2 minutes on hold at 65% B, a second
linear ramp to 95% over 2 minutes, and 2.5 minutes at 95% B followed by equil-
ibration at 40% B. The flow rate was set at 300 ll/min and the column tempera-
ture was maintained at 35C. Quantification of drug concentrations was carried
out using a 5500 QTRAP liquid chromatography-tandem mass spectrometry sys-
tem with an electrospray ion source (AB Sciex, Toronto, ON, Canada). The
mass spectrometer was operated in positive multiple reaction monitoring mode,
and the quantification was based on the mass-to-charge (m/z) ion transition m/z
453! m/z 230 for repaglinide. The lower limit of quantification was 0.01 ng/ml
for repaglinide. The between-day precisions (expressed as coefficients of varia-
tion) for the quality control samples (0.1 ng/ml and 2 ng/ml) were below 10%
and the between-day accuracies were within ± 15% for repaglinide.
For candesartan and candesartan acyl-b-D-glucuronide, reference calibration
standards and quality control samples were prepared in blank human plasma and
processed along the study samples. The samples were purified using protein pre-
cipitation and phospholipid removal. An aliquot of 50 ll of the plasma sample
was mixed with 150 ll of the protein precipitation solution containing the internal
standard (12 ng/ml) and vortexed for 1 minute. After centrifugation, supernatant
aliquots of 100 ll were treated with Ostro Protein Precipitation & Phospholipid
Removal Plate (Waters Corporation, Milford, MA, USA). The wells were then
washed with 100 ll methanol/water (1:1, v/v) and the solvent was collected into
the same well with the previous aliquot. The chromatographic separation was
performed on a Kinetex C18 column (100 × 2.1 mm, 2.6 lm; Phenomenex, Inc.,
Torrance, CA, USA) coupled with a SecurityGuard ULTRA C18 guard column
(2.1 × 4 mm; Phenomenex, Inc., Torrance, CA, USA). The mobile phase consisted
of 0.1% formic acid in water (channel A) and 0.1% formic acid in acetonitrile
(channel B). The mobile phase gradient was as follows: 0.0–0.5 minutes: 30%
B in A; 0.5–4.0 minutes: 30–70% B in A (linear gradient); 4.0–4.5 minutes:
70–90% B in A (linear gradient); 4.5–6.0 minutes: 90% B in A; 6.0–6.5 minutes:
90–30% B in A (linear gradient) and 6.5–8.0 minutes: 30% B in A. The flow rate
was set at 300 ll/min and column temperature was set at 40C. Injection volume
was 10 ll. Quantification of drug concentrations was carried out using a QTRAP
65001 liquid chromatography-tandem mass spectrometry system (AB Sciex,
Toronto, ON, Canada), using positive Turbo Ion Spray ionization and single
reaction monitoring mode. The single reaction monitoring transitions were
Piha et al. 1389
 at A
SPET Journals on N
ovem
ber 21, 2024
dm
d.aspetjournals.org
D
ow
nloaded from
 
m/z 441.2! m/z 263.0 for candesartan, m/z 617.2! m/z 441.3 for candesartan
acyl-b-D-glucuronide, and m/z 446.2 ! m/z 268.3 for candesartan-d5 which
was the internal standard for both analytes. The lower limit of quantification
was 2 ng/ml for candesartan and 1 ng/ml for candesartan acyl-b-D-glucuronide.
The interassay precision (as coefficients of variation) for the quality control
samples (6 ng/ml, 40 ng/ml and 150 ng/ml for candesartan; 1.5 ng/ml, 7.5 ng/ml
and 40 ng/ml for candesartan acyl-b-D-glucuronide) were below 15% for both
analytes and interassay accuracies were within ± 15% for both analytes.
Measurement of Unbound Plasma Fraction of Candesartan-Acyl-b-D-
Glucuronide. Samples at concentration levels 7.5 ng/ml and 40 ng/ml were pre-
pared by spiking adequate amounts of candesartan acyl-b-D-glucuronide working
solutions into analyte-free human K2EDTA plasma and ultrafiltered analyte-free
human K2EDTA plasma (prepared with Amicon Ultra-15 Centrifugal Filters
30 kDa MWCO, Merck KGaA, Darmstadt, Germany). For sample analysis, 950 ll
aliquots of human K2EDTA plasma or ultrafiltered analyte-free human K2
EDTA plasma spiked with candesartan acyl-b-D-glucuronide were pipetted into
Centrifree Ultrafiltration Centrifugal Filters (30 kDa MWCO, Merck KGaA,
Darmstadt, Germany) and centrifuged for 15 minutes at 1900 g until approxi-
mately 100 ll of ultrafiltrate was recovered. Thereafter, sample analysis was
continued as in the determination of total candesartan and total candesartan
acyl-b-D-glucuronide concentrations. The amount of protein-free (unbound)
candesartan acyl-b-D-glucuronide was determined as % from the total amount
of candesartan acyl-b-D-glucuronide.
Pharmacokinetic Analysis. The areas under the plasma concentration-time
curves from zero to nine (AUC0-9 hours) or ten hours (AUC0-10 hours) and to infin-
ity (AUC0-1), Cmax, times to peak concentrations (tmax), as well as the elimina-
tion half-lives (t1/2) of repaglinide and candesartan, were calculated using
standard noncompartmental analysis with Phoenix WinNonlin version 8.3
(Certara, Princeton, NJ, USA).
Pharmacodynamic Analysis. The safety blood glucose measurements were
also treated as a secondary pharmacodynamic outcome measure. The AUCĆ0-3
hours and AUC0-9 hours of blood glucose were calculated using the trapezoidal
method. Baseline, minimum, and maximum concentrations of blood glucose
were derived directly from the data, and the mean concentrations from zero to
three hours and nine hours were also calculated by dividing AUC0-3 hours and
AUC0-9 hours by 3 hours and 9 hours, respectively.
Static Drug-Drug Interaction Predictions. To predict the overall effect of
candesartan and candesartan acyl-b-D-glucuronide on the plasma concentrations
of repaglinide via different inhibition mechanisms, the following eq. (1) was used.
AUCR5
1
A
 1
B
 1
C
(1)
where
A5
1 Fg, REP
11
½Ig, CAN
Ki, CYP3A4, CAN
1Fg, REP
B5
fm,CYP3A4, REP
11
½Ih, CAN
Ki, CYP3A4, CAN
1
½Ih, GLU
Ki, CYP3A4,GLU
1
fm,CYP2C8, REP
11
½Ih, CAN
Ki, CYP2C8, CAN
1
½Ih, GLU
Ki, CYP2C8,GLU
1 1 fm,CYP3A4, REP1 fm,CYP2C8, REPð Þ½ 
C5
ft, OATP1B1, REP
11
½Ipv, CAN
Ki, OATP1B1, CAN
1
½Ipv, GLU
Ki, OATP1B1,GLU
1 1 ft, OATP1B1, REPð Þ
AUCR refers to the ratio of the areas under the concentration-time curves of
the victim drug in the presence and absence of the perpetrator compounds. Factor
A predicts the effect of candesartan (CAN) on the CYP3A4-mediated metabo-
lism of repaglinide (REP) in the gut; factor B predicts the effects of candesartan
and candesartan acyl-b-D-glucuronide (GLU) on the CYP2C8- and CYP3A4-
mediated metabolism of repaglinide in the liver; and factor C predicts the effects
of candesartan and candesartan acyl-b-D-glucuronide on the OATP1B1-mediated
transport of repaglinide into the liver. The prediction equation was customized
from previously proposed mathematical models (Wang et al., 2004; Ito et al.,
2005; Fahmi et al., 2008; Templeton et al., 2008; Zamek-Gliszczynski et al.,
2009). Since there is no evidence of time-dependent inhibition or induction of
CYP enzymes or transporters by candesartan or candesartan acyl-b-D-glucuro-
nide, only reversible inhibition was included in the equation. Fg expresses the
fraction of intact victim drug entering the portal vein; [I]g depicts the concentra-
tion of the perpetrator drug in enterocytes; Ki denotes the inhibition constant of
the interaction between the perpetrator and the enzyme or transporter; [I]pv and
[I]h are the unbound concentrations of the perpetrator in the portal vein and liver,
respectively; and fm and ft express the fraction of the victim drug metabolized
and transported by the indicated enzyme or transporter, respectively. The highest
TABLE 1
Pharmacokinetic and biochemical parameters of candesartan, candesartan acyl-
b-D-glucuronide, and repaglinide, and physiological parameters used in static
drug-drug predictions
Parameter Value Reference/comment
Candesartan
Dose 8 mg (18.16 lmol) or
32 mg (72.64 lmol)a
Current trial or highest
clinically used dose
Fa 1.00 Worst-case scenario
Fg 1.00 Worst-case scenario
ka 6.00 1/h
b Ito et al., 1998
RB 0.53
c United States Food and Drug
Administration, 1998
fu,p 0.01
d United States Food and Drug
Administration, 1998
Cmax (total) 0.33 or 1.32 lM
e Current trial
RL 1 or 3.8793
f Supplemental Methods,
Supplemental Table 1
[I]g 6.05 or 24.22 lM
g Rostami-Hodjegan and
Tucker, 2004
[I]pv 0.024 or 0.098 lM
h Ito et al., 1998
[I]h 0.024, 0.095, 0.098 or
0.38 lMi
Ki,CYP3A4 125 lM
j Katsube et al., 2021
Ki,CYP2C8 18.1 lM
k Walsky et al., 2005
Ki,OATP1B1 10.0 lM
l Karlgren et al., 2012
Candesartan acyl-
b-D-glucuronide
fu,p 0.01
m Current trial
Cmax (total) 0.01 or 0.04 lM
n Current trial
RL 1
f Supplemental Methods,
Supplemental Table 1
[I]pv 0.00009 or 0.0004 lM
o
[I]h 0.00009 or 0.0004 lM
i
Ki,CYP3A4 79.50 lM
p Katsube et al., 2021
Ki,CYP2C8 7.12 lM Katsube et al., 2021
Ki,OATP1B1 5.00 lM
q Katsube et al., 2021
Repaglinide
Fg 0.89 Gertz et al., 2010
fm,CYP3A4 0.16
r
fm,CYP2C8 0.84 Honkalammi et al., 2011
ft,OATP1B1 0.65
s Niemi et al., 2005
Physiology
Qen 18.0 l/h/70 kg Yang et al., 2007a
Qh 97.0 l/h/70 kg Yang et al., 2007b
a 8 mg or 32 mg of candesartan; molar mass 440.5 g/mol (National Center for Biotechnology In-
formation, 2023a).
b FDA-recommended surrogate value (0.1 1/min).
c Ratio of radioactivity in whole blood vs. plasma.
d Actually 0.0016.
e 145.0 ng/ml at 8 mg dose (highest measured Cmax in the current trial) and 580.0 ng/ml
(linearly scaled) at 32 mg dose; molar mass 440.5 g/mol (National Center for Biotechnol-
ogy Information, 2023a).
f Ratio of concentration in the liver vs. plasma.
g [I]g 5 Fa × ka × Dose/Qen.
h [I]pv 5 fu,p × [Cmax 1 (Fa × Fg × ka × Dose)/Qh/RB].
i [I]h 5 RL × [I]pv.
j IC50 > 250 lM; approximated worst-case scenario Ki  250 lM/2  125 lM.
k IC50 5 36.2 lM at Km,substrate, hence Ki  36.2 lM/2  18.1 lM.
l 52.1% inhibition at Ccandesartan 5 20 lM, hence IC50  20 lM and Ki  20 lM/2  10 lM.
mActually 0.0099.
n 5.4 ng/ml at 8 mg dose (highest measured Cmax in the current trial) and 21.6 ng/ml (line-
arly scaled) at 32 mg; molar mass 616.6 g/mol (National Center for Biotechnology Infor-
mation, 2023b).
o [I]
pv
5 f
u,p
× C
max
.
p IC50 5 159 lM, hence Ki  159 lM/2  79.5 lM.
q 40% inhibition at Cglucuronide 5 20 lM, hence IC50  10 lM and Ki  10 lM/2  5 lM.
rAssuming CYP3A4 metabolizes the fraction that CYP2C8 does not.
s Repaglinide AUC 2.88-fold higher in subjects with SLCO1B1 521CC vs.521TT genotype (Niemi
et al., 2005), and AUCR 5 1/(1  ft)$ ft 5 1 – (1/AUCR) (Zamek-Gliszczynski et al., 2009).
1390 No Interaction Between Candesartan and Repaglinide in Humans
 at A
SPET Journals on N
ovem
ber 21, 2024
dm
d.aspetjournals.org
D
ow
nloaded from
 
individual plasma concentrations of candesartan and candesartan acyl-b-D-glucu-
ronide measured in the present trial were used, while the other parameters
were derived from literature (Table 1). Four scenarios were predicted: with the
currently measured concentrations, and with these concentrations linearly scaled
up according to the largest 32 mg clinical dose of candesartan, and assuming
liver-to-plasma ratios of either 1 or worst-case scenario (Supplemental Table 1),
for candesartan and candesartan acyl-b-D-glucuronide.
Paclitaxel Clearance in a Cancer Patient Cohort. To evaluate the effect
of candesartan on paclitaxel pharmacokinetics, we identified candesartan users in
a previously described population pharmacokinetic study (Bergmann et al.,
2011). The data contained individual clearance values of unbound paclitaxel esti-
mated in 93 women (median age, 60 years) with ovarian cancer treated with
175 mg/m2 of paclitaxel as a three-hour infusion. Candesartan users were iden-
tified by reviewing the recorded concomitant medications of the patients. The
previously reported single clopidogrel (a known CYP2C8 inhibitor) user was
also identified (Bergmann et al., 2016). Patients not using candesartan, clopi-
dogrel, or any known CYP2C8 inhibitors were designated as the control group.
Statistical Analysis. It was estimated that ten subjects would be sufficient to
demonstrate a change of more than 30% in the AUC0-1 of repaglinide between
the placebo and candesartan phases, with a statistical power of more than 80%.
Pharmacokinetic results are expressed as geometric means with geometric coeffi-
cients of variation, and as ratios of geometric means with 90% confidence intervals
(CIs), except for tmax for which median with range is given. Except for tmax, phar-
macokinetic data were transformed into natural logarithms for statistical testing.
Blood glucose results are presented as arithmetic means with standard deviations.
Estimated paclitaxel clearance values from the patient cohort are presented as scat-
ter plots, with median and interquartile range reported for the patients not using
candesartan or clopidogrel. Paired t test or Wilcoxon signed-rank test was used to
test the statistical significances of differences between the pharmacokinetic trial
phases. A P-value below 0.05 was considered statistically significant. All statistical
tests were performed with JMP Pro version 17 (SAS Institute, Cary, NC, USA).
Results
Effect of Candesartan on Repaglinide. Compared with placebo,
candesartan had no relevant effect on any of the pharmacokinetic or
pharmacodynamic variables of repaglinide (Figs. 1 and 2, Table 2). The
geometric mean ratios of the AUC0-1 and Cmax of repaglinide in the
candesartan phase compared with the placebo phase were 1.02 (P 5
0.809; 90% CI 0.90–1.15) and 1.13 (P 5 0.346; 90% CI 0.90–1.43), re-
spectively. There were no differences in the maximum blood glucose
concentrations nor in the mean blood glucose concentrations between
the candesartan and placebo phases. However, the minimum blood glu-
cose concentration was slightly higher in the candesartan phase than in
the placebo phase (P 5 0.020; 4.9 mM, S.D. ± 0.6 mM, vs. 4.5 mM,
S.D. ± 0.5 mM).
Plasma Concentrations of Candesartan and Candesartan Acyl-
b-D-Glucuronide. On the day of repaglinide administration, the geomet-
ric mean of candesartan Cmax was 59.8 ng/ml (range 34.2–145 ng/ml;
Fig. 3A, Table 3). There was also considerable interindividual variation
Fig. 1. The effect of candesartan on the plasma concentrations of repaglinide (A) and blood glucose concentrations (B). Ten healthy volunteers ingested 8 mg of candesartan or
placebo once daily for three days. On day 3, the subjects ingested 0.25 mg of repaglinide one hour after the day’s candesartan or placebo dose. The ingestion of repaglinide
marks time zero. (A) Data are presented as geometric means with 90% confidence intervals. The inset is a presentation of the same data on a semilogarithmic scale. Some error
bars have been omitted for clarity. (B) Data are presented as arithmetic means with standard deviations. Some error bars have been omitted for clarity.
Fig. 2. The effect of candesartan on the individual areas under the plasma con-
centration-time curves (AUC0-1) (A) and peak concentrations (Cmax) (B) of repa-
glinide. Ten healthy volunteers ingested 8 mg of candesartan or placebo once
daily for three days. On day 3, the subjects ingested 0.25 mg of repaglinide one
hour after the day’s candesartan or placebo dose. The connected dots represent
individual values in the placebo and candesartan phases, respectively. The solid
lines indicate subjects with at least one quantifiable ($1.00 ng/ml) candesartan
acyl-b-D-glucuronide plasma concentration measurement, whereas the dashed
lines indicate subjects below that limit.
Piha et al. 1391
 at A
SPET Journals on N
ovem
ber 21, 2024
dm
d.aspetjournals.org
D
ow
nloaded from
 
in the plasma concentrations of candesartan acyl-b-D-glucuronide. With
a lower limit of quantification of 1.00 ng/ml, candesartan acyl-b-D-
glucuronide was quantifiable in the plasma of 7 out of 10 subjects, with
the highest measurement in any subject reaching 5.4 ng/ml (Fig. 3B).
The unbound fraction of candesartan acyl-b-D-glucuronide in plasma
was 0.99%. Due to low plasma concentrations, we were unable to
calculate the pharmacokinetic parameters of candesartan acyl-b-D-
glucuronide.
Predicted Effects of Candesartan on Repaglinide Plasma Con-
centrations Based on In Vitro Data. The prediction eq. (1) indicated
no clinically significant inhibition of CYP2C8, CYP3A4, or OATP1B1
by candesartan or candesartan acyl-b-D-glucuronide, even in the worst-
case scenario at a 32 mg candesartan dose (Table 4). Thus, the static
predictions were in agreement with the observation of no significant ef-
fect of candesartan on the plasma concentrations of repaglinide.
Pharmacokinetics of Repaglinide in Subjects with Quantifi-
able Candesartan Acyl-b-D-Glucuronide Plasma Concentrations.
A post hoc analysis revealed that the Cmax of repaglinide was 1.32-fold
higher (P5 0.041; 90% CI 1.07–1.62) in the candesartan phase compared
with placebo phase in the subset of the seven subjects who had quantifiable
candesartan acyl-b-D-glucuronide concentrations in plasma. With the ex-
ception of one individual, the Cmax of repaglinide was greater in the cande-
sartan phase than in the placebo phase in these subjects (Fig. 2). However,
we found no statistically significant differences in other pharmacokinetic or
pharmacodynamic variables of repaglinide (data not shown).
Estimated Clearance of Unbound Paclitaxel in Ovarian Can-
cer Patients with Concomitant Perpetrators. In patients not using
candesartan or clopidogrel, the median of estimated unbound paclitaxel
clearance values was 383.12 l/h (interquartile range 329.65–450.50 l/h)
(Fig. 4). Four patients were using candesartan, specifically Atacand
16 mg (candesartan cilexetil), Atacand at an unrecorded dose, Atacand
Zid (candesartan cilexetil and hydrochlorothiazide) at unrecorded doses,
and Atazid 8 mg/12.5 mg (8 mg of candesartan cilexetil and 12.5 mg of
hydrochlorothiazide). The estimated unbound paclitaxel clearance val-
ues in these candesartan users were 469.98 l/h, 641.41 l/h, 317.22 l/h,
and 331.72 l/h, respectively. As previously described, one patient was
using clopidogrel and presented with the second lowest paclitaxel clear-
ance in the cohort (Bergmann et al., 2016).
Discussion
In the present study, typical therapeutic doses of candesartan had no
clinically meaningful effect on the plasma concentrations of repaglinide
in healthy volunteers. The AUC0-1, Cmax, and t1/2 of repaglinide re-
mained practically unchanged in the candesartan phase compared with
the placebo phase. Our static prediction results are in line with these
TABLE 2
Pharmacokinetic variables of repaglinide and blood glucose variables in ten healthy volunteers following three days of daily doses of placebo or 8 mg of candesartan,
and a single 0.25 mg dose of repaglinide one hour after placebo or candesartan on day 3. Pharmacokinetic values are expressed as geometric means with geometric
coefficients of variation, except for tmax, for which median with range is given. Blood glucose values are expressed as arithmetic means with standard deviations.
Pharmacokinetic data, except for tmax, were logarithmically transformed for statistical testing. Paired t test was applied on pharmacokinetic and blood glucose data, ex-
cept for Wilcoxon signed rank test for tmax
Variable Placebo Candesartan 8 mg Ratio (90% CI) P value
Repaglinide
Cmax (ng/ml) 3.30 (60.3) 3.75 (57.5) 1.13 (0.90; 1.43) 0.346
tmax (h) 0.50 (0.50–0.75) 0.50 (0.50–1.33) N/A >0.999
t1/2 (h) 1.24 (15.1) 1.27 (12.4) 1.02 (0.93; 1.13) 0.691
AUC0-9 h (ng*h/ml) 4.15 (42.8) 4.21 (36.9) 1.02 (0.90; 1.15) 0.821
AUC0-1 (ng*h/ml) 4.18 (43.0) 4.24 (37.3) 1.02 (0.90; 1.15) 0.809
Blood glucose
Baseline concentration (mM) 5.9 (0.7) 5.8 (0.7) N/A 0.748
Minimum concentration (mM) 4.5 (0.5) 4.9 (0.6) N/A 0.020*
Maximum concentration (mM) 8.1 (1.0) 8.1 (1.1) N/A 0.959
Mean concentration from 0 to 3 h (mM) 6.1 (0.5) 6.1 (0.6) N/A 0.704
Mean concentration from 0 to 9 h (mM) 6.2 (0.4) 6.2 (0.4) N/A 0.923
N/A, not applicable.
*P < 0.05 vs. placebo phase.
Fig. 3. Individual plasma concentrations of candesartan (A) and candesartan acyl-b-D-glucuronide (B) in ten healthy volunteers on the third day of 8 mg of candesar-
tan once daily. (A) The bold curve indicates the geometric means of individual values. (B) The dashed horizontal line indicates the lower limit of quantification for
candesartan acyl-b-D-glucuronide.
1392 No Interaction Between Candesartan and Repaglinide in Humans
 at A
SPET Journals on N
ovem
ber 21, 2024
dm
d.aspetjournals.org
D
ow
nloaded from
 
findings, indicating minimal effect of candesartan on the plasma
concentrations of repaglinide even at the highest clinically approved
candesartan doses.
We used the antidiabetic agent repaglinide as a clinical index sub-
strate of CYP2C8. Repaglinide is extensively and rapidly metabolized
to a number of phase I metabolites, including the aromatic amine M1
and dicarboxylic acid M2 mainly formed by CYP3A4, and the piperi-
dine-hydroxylated M4 and isopropyl-hydroxylated M0-OH primarily
formed by CYP2C8 (Guay, 1998; Bidstrup et al., 2003). It has been es-
timated that up to 84% of repaglinide is metabolized by CYP2C8, and
its sensitivity to CYP2C8 inhibitors makes it a recommended clinical
index substrate of CYP2C8 (Honkalammi et al., 2011; Tornio et al.,
2014, 2019; United States Food and Drug Administration, 2023). How-
ever, repaglinide is also transported by OATP1B1, which may have
clinical implications (Niemi et al., 2005; Honkalammi et al., 2011).
Therefore, along with CYP2C8, OATP1B1 and CYP3A4 can also play
a role in DDIs of repaglinide.
In their in vitro assays with recombinant CYP enzymes, Katsube et al.
(2021) showed that candesartan, candesartan N2-glucuronide, and
candesartan acyl-b-D-glucuronide all inhibited the CYP2C8- and CYP3A4-
mediated hydroxylation of paclitaxel, and the OATP1B1-mediated trans-
port of 2’,7’-dichlorofluorescein. The strongest observed effect was the
reversible inhibition of CYP2C8 by candesartan acyl-b-D-glucuronide (Ki
7,120 nM). The authors suggested that this mechanism could be the un-
derlying cause of the high incidence of severe neutropenia previously ob-
served in patients receiving candesartan and paclitaxel (Katsube et al.,
2018, 2021). In the current trial, however, we did not observe any clini-
cally significant DDI between candesartan and repaglinide. Compliance
of the subjects to candesartan was confirmed with trough concentration
measurements and controlled administration of the last candesartan dose
on the days of repaglinide ingestion. Moreover, despite a small number
of candesartan users, real-world paclitaxel clearance data from a patient
cohort offers no support for a pharmacokinetic DDI between candesartan
and paclitaxel. Thus, inhibition of CYP2C8 is unlikely to explain the
previously suggested DDI between candesartan and paclitaxel, either.
Further work is required to unveil the mechanisms predisposing some
patients to severe paclitaxel-induced neutropenia.
There is an increasing awareness of the involvement of drug metabo-
lites in pharmacokinetic DDIs (Isoherranen et al., 2009; VandenBrink
and Isoherranen, 2010; Yeung et al., 2011; European Medicines
Agency, 2012; United States Food and Drug Administration, 2020).
Several glucuronide metabolites inhibit CYP2C8, which is well illus-
trated by the drug interactions of gemfibrozil and clopidogrel (Backman
et al., 2016). In early studies, the concomitant administration of gemfi-
brozil increased the AUC0-1 of cerivastatin almost 6-fold and that of re-
paglinide approximately 8-fold (Backman et al., 2002; Niemi et al.,
2003). Subsequently, it was found that the 1-O-b-glucuronide metabo-
lite of gemfibrozil is a strong, metabolism-based inhibitor of CYP2C8
in vitro (Shitara et al., 2004; Ogilvie et al., 2006). Later, it was found
that clopidogrel increased the AUC0-1 of repaglinide up to 5-fold in
healthy volunteers. In vitro experiments revealed that clopidogrel acyl-
b-D-glucuronide, like gemfibrozil 1-O-b-glucuronide, is an irreversible
mechanism-based inhibitor of CYP2C8. This was the most important
mechanism of the clinical interaction according to physiologically based
pharmacokinetic modeling (Tornio et al., 2014). This contrasts with the
reversible CYP2C8 inhibition of candesartan acyl-b-D-glucuronide
which was not associated with increased repaglinide plasma concentra-
tions in the present study. Of note, recent studies indicate that, apart
from CYP2C8, glucuronide metabolites can cause time-dependent inhi-
bition of other CYP enzymes as well (Kahma et al., 2024).
The circulating plasma concentrations [I] of candesartan acyl-b-D-
glucuronide were found to be several orders of magnitude lower than
those of gemfibrozil 1-O-b-glucuronide and clopidogrel acyl-b-D-
glucuronide, which might also contribute to the different DDI potential.
To compare, the mean steady state Cmax of gemfibrozil 1-O-b-glucuronide
is approximately 14,500 ng/ml (34,000 nM) (Itkonen et al., 2019), and
that of clopidogrel acyl-b-D-glucuronide approximately 730 ng/ml
(1,500 nM) (Tornio et al., 2014). In contrast, the highest measured
concentration of candesartan acyl-b-D-glucuronide was only 5.4 ng/ml
(8.8 nM) in the present trial. Thus, the low [I]/Ki ratio 
 0.1 of cande-
sartan acyl-b-D-glucuronide and its competitive inhibition mechanism
are likely to result in only weak and transient CYP2C8-inhibiting effect.
Our clinical data are limited to one candesartan dose level only, but
TABLE 3
Pharmacokinetic variables of candesartan in ten healthy volunteers following
three days of ingesting daily 8 mg doses of candesartan. Values are expressed as
geometric means with geometric coefficients of variation, except for tmax, for
which median with range is given
Cmax (ng/ml) 59.78 (49.84)
tmax (h) 2.67 (1.50–4.00)
t1/2 (h) 3.69 (18.67)
AUC0-10 h (ng*h/ml) 329.64 (42.49)
AUC0-1 (ng*h/ml) 418.64 (42.75)
TABLE 4
Predicted ratios of the areas under the plasma concentration-time curves of repa-
glinide in the presence and absence of candesartan and candesartan acyl-b-D-
glucuronide at two different candesartan doses, with liver-to-plasma ratios (RL)
of either 1 or worst-case scenario for candesartan and candesartan acyl-b-D-
glucuronide (Supplemental Data)
Candesartan dose
Candesartan RL 5 1;
Candesartan
acyl-b-D-glucuronide
RL 5 1
Candesartan RL 5 3.8793;
Candesartan
acyl-b-D-glucuronide
RL 5 1
8 mg 1.008 1.011
32 mg 1.030 1.043
Fig. 4. Estimated individual clearance values of unbound paclitaxel in a cohort of
93 ovarian cancer patients using and not using candesartan or clopidogrel concomi-
tantly. The patients were treated with 175 mg/m2 of paclitaxel as a three-hour infu-
sion. Four patients were using candesartan, and a single patient was using clopidogrel
concomitantly; control patients were not using candesartan, clopidogrel, or any known
CYP2C8 inhibitors. The horizontal line indicates median, and the whiskers indicate
interquartile range in control patients.
Piha et al. 1393
 at A
SPET Journals on N
ovem
ber 21, 2024
dm
d.aspetjournals.org
D
ow
nloaded from
 
since the pharmacokinetics of candesartan are linearly dose-proportional
in the range of 2–64 mg (United States Food and Drug Administra-
tion, 1998), we scaled up our plasma concentration data to predict
the effect of the maximum 32 mg clinical dose to simulate the
worst-case scenario. Our static DDI predictions offer little support
for clinically meaningful CYP2C8 inhibition by candesartan or
candesartan acyl-b-D-glucuronide at any clinical dose. Even at the
32 mg dose and corresponding plasma and liver concentrations of
candesartan and candesartan acyl-b-D-glucuronide, the predicted
effect on repaglinide plasma concentrations was negligible (AUCR5 1.04).
It should be noted, as Katsube et al. (2021) also speculated, that the circulat-
ing systemic concentrations of candesartan acyl-b-D-glucuronide do not
necessarily accurately reflect those in the portal vein and liver, which causes
uncertainty in in vitro-to-in vivo extrapolations. However, we predicted the
liver-to-plasma ratios of candesartan and candesartan-acyl-b-D-glucuronide
and used the highest predicted hepatic concentrations in our prediction
equation.
Interestingly, we found that candesartan increased the Cmax of repa-
glinide by 32% in those seven subjects with quantifiable candesartan
acyl-b-D-glucuronide plasma concentrations. This suggests that in some
individuals, candesartan acyl-b-D-glucuronide might indeed inhibit
CYP2C8 weakly. According to in vitro studies, candesartan is conju-
gated to its acyl-b-D-glucuronide by UGT1A7, UGT1A8, UGT1A9,
UGT1A10, and UGT2B7 (Alonen et al., 2008b; Katsube et al., 2021).
The activities of these UGT enzymes are subject to genetic variation with
potentially clinically significant implications (Stingl et al., 2014). While
several UGT enzymes can contribute to the glucuronidation of candesar-
tan, genetic variability in UGT activity might produce variability in the
systemic concentrations of candesartan acyl-b-D-glucuronide and conse-
quently in the susceptibility to CYP2C8-mediated DDIs. Moreover, in-
flammatory disease states, including cancer, have been linked with altered
pharmacokinetics of various drugs consistent with reduced CYP enzyme
and transporter activity (Gatti and Pea, 2022), further predisposing certain
individuals to DDIs. However, since the AUC0-1 of repaglinide remained
unaffected even in the subjects with quantifiable candesartan acyl-b-D-
glucuronide plasma concentrations, and no decrease in paclitaxel clear-
ance was evident in any of the ovarian cancer patients using candesartan,
the potential for clinically meaningful DDIs still appears to be low.
In conclusion, the coadministration of candesartan has no clinically
relevant effect on the plasma concentrations or hypoglycemic effect of
the CYP2C8 substrate repaglinide. We confirmed the presence of can-
desartan acyl-b-D-glucuronide in human plasma but, considering its
low systemic concentrations and reversible mechanism of CYP2C8 in-
hibition, its potential to inhibit CYP2C8 to a clinically relevant degree
appears low. Therefore, candesartan therapy is unlikely to cause phar-
macokinetic DDIs via inhibition of CYP2C8.
Acknowledgments
The authors thank Marika Iljamo, Tiina Ceder, and Karla Saukkonen
for their technical assistance.
Data Availability
The individual data supporting the findings of this study cannot be
made publicly available to protect the privacy of the study subjects in accor-
dance with legal and ethical requirements.
Authorship Contributions
Participated in research design: Piha, Cajanus, Niemi, Backman, Filp-
pula, Tornio.
Conducted experiments: Piha, Cajanus, Engstr€om, Neuvonen, Tornio.
Performed data analysis: Piha, Filppula, Tornio.
Wrote or contributed to the writing of the manuscript: Piha, Cajanus,
Engstr€om, Neuvonen, Bergmann, Niemi, Backman, Filppula, Tornio.
References
Aberg JG, Olofsson B, and Karlson BW (2011) Pharmacokinetic interaction study with fixed high
dose combinations of candesartan cilexetil and hydrochlorothiazide. Int J Clin Pharmacol Ther
49:750–755.
Alonen A, Jansson J, Kallonen S, Kiriazis A, Aitio O, Finel M, and Kostiainen R (2008a) En-
zyme-assisted synthesis and structure characterization of glucuronic acid conjugates of losartan,
candesartan, and zolarsartan. Bioorg Chem 36:148–155.
Alonen A, Finel M, and Kostiainen R (2008b) The human UDP-glucuronosyltransferase UGT1A3
is highly selective towards N2 in the tetrazole ring of losartan, candesartan, and zolarsartan. Bio-
chem Pharmacol 76:763–772.
Backman JT, Filppula AM, Niemi M, and Neuvonen PJ (2016) Role of Cytochrome P450 2C8 in
Drug Metabolism and Interactions. Pharmacol Rev 68:168–241.
Backman JT, Kyrklund C, Neuvonen M, and Neuvonen PJ (2002) Gemfibrozil greatly increases
plasma concentrations of cerivastatin. Clin Pharmacol Ther 72:685–691.
Bergmann TK, Brasch-Andersen C, Green H, Mirza M, Pedersen RS, Nielsen F, Skougaard K,
Wihl J, Keldsen N, Damkier P, et al. (2011) Impact of CYP2C8*3 on paclitaxel clearance: a
population pharmacokinetic and pharmacogenomic study in 93 patients with ovarian cancer.
Pharmacogenomics J 11:113–120.
Bergmann TK, Filppula AM, Launiainen T, Nielsen F, Backman JT, and Brosen K (2016) Neuro-
toxicity and low paclitaxel clearance associated with concomitant clopidogrel therapy in a
60-year-old Caucasian woman with ovarian carcinoma. Br J Clin Pharmacol 81:313–315.
Bidstrup TB, Bjørnsdottir I, Sidelmann UG, Thomsen MS, and Hansen KT (2003) CYP2C8 and
CYP3A4 are the principal enzymes involved in the human in vitro biotransformation of the in-
sulin secretagogue repaglinide. Brit J Clinical Pharma 56:305–314.
Brendel E, Weimann B, Dietrich H, Froede C, and Thomas D (2013) Investigation of bioequiva-
lence of a new fixed-dose combination of nifedipine and candesartan with the corresponding
loose combination as well as the drug-drug interaction potential between both drugs under fast-
ing conditions. CP 51:753–762.
Cernes R, Mashavi M, and Zimlichman R (2011) Differential clinical profile of candesartan com-
pared to other angiotensin blockers. Vasc Health Risk Manag 7:749–759.
European Medicines Agency, Committee for Human Medicinal Products (2012) Guideline on the
investigation of drug interactions (CPMP/EWP/560/95/Rev. 1 Corr. 2**).
Fahmi OA, Maurer TS, Kish M, Cardenas E, Boldt S, and Nettleton D (2008) A combined model
for predicting CYP3A4 clinical net drug-drug interaction based on CYP3A4 inhibition, inactiva-
tion, and induction determined in vitro. Drug Metab Dispos 36:1698–1708.
Gatti M and Pea F (2022) The Cytokine Release Syndrome and/or the Proinflammatory Cytokines
as Underlying Mechanisms of Downregulation of Drug Metabolism and Drug Transport: A Sys-
tematic Review of the Clinical Pharmacokinetics of Victim Drugs of this Drug-Disease Interac-
tion under Different Clinical Conditions. Clin Pharmacokinet 61:1519–1544.
Gertz M, Harrison A, Houston JB, and Galetin A (2010) Prediction of human intestinal first-pass
metabolism of 25 CYP3A substrates from in vitro clearance and permeability data. Drug Metab
Dispos 38:1147–1158.
Guay DRP (1998) Repaglinide, A Novel, Short-Acting Hypoglycemic Agent for Type 2 Diabetes
Mellitus. Pharmacotherapy 18:1195–1204.
Gundlach K, Wolf K, Salem I, Randerath O, and Seiler D (2021) Safety of Candesartan, Amlodi-
pine, and Atorvastatin in Combination: Interaction Study in Healthy Subjects. Clin Pharmacol
Drug Dev 10:190–197.
Honkalammi J, Niemi M, Neuvonen PJ, and Backman JT (2011) Dose-Dependent Interaction be-
tween Gemfibrozil and Repaglinide in Humans: Strong Inhibition of CYP2C8 with Subthera-
peutic Gemfibrozil Doses. Drug Metab Dispos 39:1977–1986.
Isoherranen N, Hachad H, Yeung CK, and Levy RH (2009) Qualitative analysis of the role of me-
tabolites in inhibitory drug-drug interactions: literature evaluation based on the metabolism and
transport drug interaction database. Chem Res Toxicol 22:294–298.
Itkonen MK, Tornio A, Neuvonen M, Neuvonen PJ, Niemi M, and Backman JT (2019) Clopidog-
rel and Gemfibrozil Strongly Inhibit the CYP2C8-Dependent Formation of 3-Hydroxydeslorata-
dine and Increase Desloratadine Exposure In Humans. Drug Metab Dispos 47:377–385.
Ito K, Hallifax D, Obach RS, and Houston JB (2005) Impact of Parallel Pathways of Drug Elimi-
nation and Multiple Cytochrome P450 Involvement on Drug-Drug Interactions: CYP2D6 Para-
digm. Drug Metab Dispos 33:837–844.
Ito K, Iwatsubo T, Kanamitsu S, Ueda K, Suzuki H, and Sugiyama Y (1998) Prediction of phar-
macokinetic alterations caused by drug-drug interactions: metabolic interaction in the liver.
Pharmacol Rev 50:387–412.
Jonkman JH, van Lier JJ, van Heiningen PN, Lins R, Sennewald R, and H€ogemann A (1997)
Pharmacokinetic drug interaction studies with candesartan cilexetil. J Hum Hypertens 11 Suppl
2:SupplS31–35.
Kahma H, Paludetto M-N, Neuvonen M, Kurkela M, Filppula AM, Niemi M, and Backman JT
(2024) Screening of 16 major drug glucuronides for time-dependent inhibition of nine drug-me-
tabolizing CYP enzymes - detailed studies on CYP3A inhibitors. Eur J Pharm Sci 198:106735.
Karlgren M, Ahlin G, Bergstr€om CAS, Svensson R, Palm J, and Artursson P (2012) In vitro and
in silico strategies to identify OATP1B1 inhibitors and predict clinical drug-drug interactions.
Pharm Res 29:411–426.
Katsube Y, Hira D, Tsujimoto M, Koide H, Minegaki T, Ikeda Y, Morita S-Y, Nishiguchi K, and
Terada T (2018) Concomitant administration of candesartan cilexetil in patients on paclitaxel
and carboplatin combination therapy increases risk of severe neutropenia. Int J Clin Pharmacol
Ther 56:328–336.
Katsube Y, Tsujimoto M, Koide H, Hira D, Ikeda Y, Minegaki T, Morita S-Y, Terada T, and
Nishiguchi K (2021) In Vitro Evidence of Potential Interactions between CYP2C8 and Cande-
sartan Acyl-b-D-glucuronide in the Liver. Drug Metab Dispos 49:289–297.
Kim J-R, Kim S, Huh W, and Ko J-W (2018) No pharmacokinetic interactions between candesar-
tan and amlodipine following multiple oral administrations in healthy subjects. Drug Des Devel
Ther 12:2475–2483.
Kondo T, Yoshida K, Yoshimura Y, Motohashi M, and Tanayama S (1996a) Disposition of the
new angiotensin II receptor antagonist candesartan cilexetil in rats and dogs. Arzneimittelfor-
schung 46:594–600.
1394 No Interaction Between Candesartan and Repaglinide in Humans
 at A
SPET Journals on N
ovem
ber 21, 2024
dm
d.aspetjournals.org
D
ow
nloaded from
 
Kondo T, Yoshida K, Yoshimura Y, Motohashi M, and Tanayama S (1996b) Characterization of
Conjugated Metabolites of a New Angiotensin II Receptor Antagonist, Candesartan Cilexetil, in
Rats by Liquid Chromatography/Electrospray Tandem Mass Spectrometry Following Chemical
Derivatization. J Mass Spectrom 31:873–878.
Miura M, Satoh S, Kagaya H, Saito M, Inoue T, Ohkubo T, Habuchi T, and Suzuki T (2009) Ef-
fect of telmisartan, valsartan and candesartan on mycophenolate mofetil pharmacokinetics in
Japanese renal transplant recipients. J Clin Pharm Ther 34:683–692.
National Center for Biotechnology Information (2023a) PubChem Compound Summary for CID
2541, Candesartan.
National Center for Biotechnology Information (2023b) PubChem Compound Summary for CID
131769984, Candesartan O-glucuronide.
Niemi M, Backman JT, Kajosaari LI, Leathart JB, Neuvonen M, Daly AK, Eichelbaum M, Kivist€o
KT, and Neuvonen PJ (2005) Polymorphic organic anion transporting polypeptide 1B1 is a ma-
jor determinant of repaglinide pharmacokinetics. Clin Pharmacol Ther 77:468–478.
Niemi M, Backman JT, Neuvonen M, and Neuvonen PJ (2003) Effects of gemfibrozil, itracon-
azole, and their combination on the pharmacokinetics and pharmacodynamics of repaglinide:
potentially hazardous interaction between gemfibrozil and repaglinide. Diabetologia 46:
347–351.
Ogilvie BW, Zhang D, Li W, Rodrigues AD, Gipson AE, Holsapple J, Toren P, and Parkinson A
(2006) Glucuronidation converts gemfibrozil to a potent, metabolism-dependent inhibitor of
CYP2C8: implications for drug-drug interactions. Drug Metab Dispos 34:191–197.
Pietruck F, Kiel G, Birkel M, Stahlheber-Dilg B, and Philipp T (2005) Evaluation of the effect of
candesartan cilexetil on the steady-state pharmacokinetics of tacrolimus in renal transplant pa-
tients. Biopharm Drug Dispos 26:135–141.
Rostami-Hodjegan A and Tucker G (2004) ‘In silico’ simulations to assess the ‘in vivo’ conse-
quences of ‘in vitro’ metabolic drug-drug interactions. Drug Discov Today Technol 1:441–448.
Senda A, Mukai Y, Hayakawa T, Kato Y, Eliasson E, Rane A, Toda T, and Inotsume N (2017)
Angiotensin II Receptor Blockers Inhibit the Generation of Epoxyeicosatrienoic Acid from Ara-
chidonic Acid in Recombinant CYP2C9, CYP2J2 and Human Liver Microsomes. Basic Clin
Pharma Tox 121:239–245.
Shitara Y, Hirano M, Sato H, and Sugiyama Y (2004) Gemfibrozil and its glucuronide inhibit the or-
ganic anion transporting polypeptide 2 (OATP2/OATP1B1:SLC21A6)-mediated hepatic uptake and
CYP2C8-mediated metabolism of cerivastatin: analysis of the mechanism of the clinically relevant
drug-drug interaction between cerivastatin and gemfibrozil. J Pharmacol Exp Ther 311:228–236.
Stingl JC, Bartels H, Viviani R, Lehmann ML, and Brockm€oller J (2014) Relevance of
UDP-glucuronosyltransferase polymorphisms for drug dosing: A quantitative systematic
review. Pharmacol Ther 141:92–116.
Taavitsainen P, Kiukaanniemi K, and Pelkonen O (2000) In vitro inhibition screening of human
hepatic P450 enzymes by five angiotensin-II receptor antagonists. Eur J Clin Pharmacol
56:135–140.
Templeton IE, Thummel KE, Kharasch ED, Kunze KL, Hoffer C, Nelson WL, and Isoherranen N
(2008) Contribution of itraconazole metabolites to inhibition of CYP3A4 in vivo. Clin Pharma-
col Ther 83:77–85.
Tornio A, Filppula AM, and Backman JT (2022) Translational aspects of cytochrome P450-medi-
ated drug-drug interactions: A case study with clopidogrel. Basic Clin Pharma Tox 130:48–59.
Tornio A, Filppula AM, Kailari O, Neuvonen M, Nyr€onen TH, Tapaninen T, Neuvonen PJ, Niemi
M, and Backman JT (2014) Glucuronidation Converts Clopidogrel to a Strong Time-Dependent
Inhibitor of CYP2C8: A Phase II Metabolite as a Perpetrator of Drug-Drug Interactions. Clin
Pharmacol Ther 96:498–507.
Tornio A, Filppula AM, Niemi M, and Backman JT (2019) Clinical Studies on Drug-Drug Interac-
tions Involving Metabolism and Transport: Methodology, Pitfalls, and Interpretation. Clin Phar-
macol Ther 105:1345–1361.
United States Food and Drug Administration. (2023) Drug Development and Drug Interactions:
Table of Substrates, Inhibitors and Inducers.
United States Food and Drug Administration, Center for Drug Evaluation and Research (1998)
Atacand clinical pharmacology and biopharmaceutics review. NDA 20-838:1-30.
United States Food and Drug Administration, Center for Drug Evaluation and Research. (2020) In
vitro drug interaction studies — cytochrome P450 enzyme- and transporter-mediated drug inter-
actions: guidance for industry.
VandenBrink BM and Isoherranen N (2010) The role of metabolites in predicting drug-drug inter-
actions: focus on irreversible P450 inhibition. Curr Opin Drug Discov Devel 13:66–77.
Walsky RL, Gaman EA, and Obach RS (2005) Examination of 209 drugs for inhibition of cyto-
chrome P450 2C8. J Clin Pharmacol 45:68–78.
Wang Y-H, Jones DR, and Hall SD (2004) Prediction of cytochrome P450 3A inhibition by verap-
amil enantiomers and their metabolites. Drug Metab Dispos 32:259–266.
Yang J, Jamei M, Yeo KR, Rostami-Hodjegan A, and Tucker GT (2007a) Misuse of the well-
stirred model of hepatic drug clearance. Drug Metab Dispos 35:501–502.
Yang J, Jamei M, Yeo KR, Tucker GT, and Rostami-Hodjegan A (2007b) Prediction of Intestinal
First-Pass Drug Metabolism. Curr Drug Metab 8:676–684.
Yeung CK, Fujioka Y, Hachad H, Levy RH, and Isoherranen N (2011) Are Circulating Metabo-
lites Important in Drug-Drug Interactions?: Quantitative Analysis of Risk Prediction and Inhibi-
tory Potency. Clin Pharmacol Ther 89:105–113.
Zamek-Gliszczynski MJ, Kalvass JC, Pollack GM, and Brouwer KLR (2009) Relationship be-
tween Drug/Metabolite Exposure and Impairment of Excretory Transport Function. Drug Metab
Dispos 37:386–390.
Address correspondence to: Dr. Aleksi Tornio, Institute of Biomedicine, Uni-
versity of Turku, and Unit of Clinical Pharmacology, Turku University Hospital,
Kiinamyllynkatu 10, FI-20520 Turku, Finland. E-mail: aleksi.tornio@utu.fi
Piha et al. 1395
 at A
SPET Journals on N
ovem
ber 21, 2024
dm
d.aspetjournals.org
D
ow
nloaded from