Bioequivalence Studies
In vivo Bioequivalence of Oral Antidiabetic
Agents: Pioglitazone Tablets
Hong Wong a, Yildiz Ozalp b, Audrey Lainessea, and Recep Serdar Alpan b
SFBC Anapharm a, Sainte-Foy, Quebec (Canada), and EIS Eczacıbaşı İlaç Sanayi ve Ticaret A.S.b, Istanbul (Turkey)
Key words
䊏 CAS 112529-15-4
䊏 Pioglitazone, bioequivalence,
clinical pharmacokinetics
Arzneim.-Forsch./Drug Res.
54, No. 9a, 618−624 (2004)
Zusammenfassung
In-vivo-Bioäquivalenz oraler Antidiabe-
tika: Pioglitazon-Tabletten
In der vorliegenden Studie sollte die
Bioverfügbarkeit von zwei Pioglitazon
(CAS 112529-15-4)-Zubereitungen unter-
sucht werden. Dazu wurde eine Bioäqui-
valenzprüfung an 26 gesunden männ-
lichen Probanden durchgeführt, denen
30 mg Pioglitazon-enthaltende Tabletten
618 Wong et al. − Pioglitazone
Summary
The study was designed to evaluate the
bioequivalence of two pioglitazone (CAS
112529-15-4) formulations. The trial was
performed in 26 healthy male volunteers
with the aim of comparing a new generic
product (tablets containing 30 mg pioglit-
azone hydrochloride, test) with the ori-
ginator product (reference). The trial was
performed according to an open, cross-
over design in one study centre. In each
of the two study periods (separated by a
wash-out of 14 days) a single oral dose of
30 mg (test or reference) formulation
was administered. Blood samples were
taken up to 120 h post dose, the plasma
was separated and the concentrations of
pioglitazone and its principal active me-
tabolite hydroxypioglitazone were deter-
mined by LC-MS-MS method. AUC0-inf,
AUC0-t, Cmax and Tmax were calculated for
both formulations.
The mean Cmax of pioglitazone ranged
between 1.01 µg/mL and 1.05 µg/mL,
while the mean AUC0-inf and AUC0-t
ranged between 10.89 µg · h/mL and
10.98 µg · h/mL as well as between 10.56
der Testzubereitung und der Referenzzu-
bereitung oral appliziert wurden. Die Prü-
fung war als offene, randomisierte, Cross-
over-Prüfung in einem Studienzentrum
angelegt. Blutentnahmen erfolgten bis zu
120 h nach der Applikation. Im abge-
trennten Plasma wurde die Bestimmung
von Pioglitazon und Hydroxypioglitazon
(aktiver Metabolit) mittels HPLC-MS-MS
vorgenommen. Die Parameter AUC0-inf,
Arzneim.-Forsch./Drug Res. 54, No. 9a, 618−624 (2004)
µg · h/mL and 10.62 µg · h/mL for the
Downloaded by: NYU. Copyrighted material.
test and reference formulations, respec-
tively. The median Tmax for the test tab-
lets was 1.50 h and for the reference was
1.75 h. The ratios test/reference formula-
tion for AUC0-inf, AUC0-t and Cmax were
99.70 %, 100.13 % and 99.17 %, respec-
tively. Furthermore, the 90 % geometric
confidence intervals of the mean ratio of
ln-transformed AUC0-inf were narrow and
symmetrical around 100 %, i.e. 90.59 %
to 109.72 %, for AUC0-t 90.69 % to
110.55 %, whereas for Cmax they were
87.52 % to 112.37 %. As in the case of pi-
oglitazone, mean values of the principal
bioequivalence parameters of hydroxypi-
oglitazone did not differ significantly
after administration of the test and refer-
ence formulations.
In the light of the present study it can
be concluded that the two evaluated piog-
litazone formulations, i.e. test formula-
tion of pioglitazone hydrochloride and
reference formulation, are bioequivalent
in terms of the rate and extent of absorp-
tion.
AUC0-t, Cmax und Tmax wurden für beide
Zubereitungen berechnet. Cmax für die Te-
st- bzw. Referenzzubereitung betrug im
Mittel 1,01 µg/mL bzw. 1,05 µg/mL, für
die AUC0-t 10,56 µg · h/mL bzw. 10,62
µg · h/mL. Der Mittelwert der AUC0-inf be-
trug 10,89 µg · h/mL (Test) bzw. 10,98
µg · h/mL (Referenz). Bis zum Erreichen
maximaler Plasmaspiegel (tmax, Median-
wert) vergingen 1,50 h (Test) bzw.
 Bioequivalence Studies

1,75 h (Referenz). Das mittlere Verhältnis
Test/Referenz lag bei 99.70 % (AUC0-inf ),
100.13 % (AUC0-t) bzw. 99.17 % (Cmax)
Die 90 %-Konfidenzintervalle für die
mittleren Verhältnisse der logarithmier-
ten Zielparameter waren eng und symme-
trisch um 100 % (90,59 %−109,72 % für
AUC0-inf; 90,69 %−110,55 % für AUC0-t;
87.52 %−112.37 % für Cmax). Wie für Pio- die Testzubereitung der Referenzuberei-
glitazon selbst waren die Mittelwerte der tung bioäquivalent ist.
Bioäquivalenzparameter für Hydroxypio-
glitazon nach Applikation der Test- und
der Referenzzubereitung nicht signifi-
kant voneinander verschieden. Auf der
Grundlage der vorliegenden Ergebnisse
kann der Schluß gezogen werden, daβ

Schlüsselwörter: CAS 112529-15-4 · Pioglitazon, Bioäquivalenz, klinische Pharmakokinetik

Özet

Pioglitazon hidroklorür aktif maddesi içe-
ren iki tablet formülasyonu ile yapılan in
vivo biyoeşdeğerlik çalışması
Bu klinik çalışma pioglitazon (CAS
112529-15-4) içeren iki tablet formülasyo-
nunun karşılaştırılması amacıyla 26
kadın ve erkek sağlıklı gönüllüde, her gö-
nüllüye aç karnına, tek doz 30 mg piogli-
trasyonları LC/MS/MS metodu ile öl-
çüldü. AUC0-inf, AUC0-t, Cmax ve Tmax her
iki formülasyon içinde hesaplandı. Piogli-
tazonun ortalama Cmax değeri 1.01 µg/
mL ile 1.05 µg/mL arasında iken, orta-
lama AUC0-inf ve AUC0-t test ve referans
ilaçlar için sırasıyla 10.89 ng · h/ml ile
10.98 ng · h/ml ve 10.56 ng . h/ml ile
10.62 ng · h/ml arasındaydı. Test ve refe-
güven aralıkları dardı ve % 100 civarında
simetrikti. Sözkonusu güvenlik aralıkları
AUC0-inf için % 90.59 ile % 109.72, AUC0-t
için % 90.69 ile % 110.55 arasında iken
Cmax için % 87.52 ile %112.37 arasın-
daydı. Pioglitazon ile birlikte değerlendi-
rilen aktif metaboliti hidroksipioglitazo-
nun biyoeşdeğerlilik göstergelerinin test
ve referans ürün karşılaştırmasında an-

Downloaded by: NYU. Copyrighted material.

tazon tablet verilerek; açık, tek merkezli,
randomize, çapraz tasarımlı ve arasında
14 günlük temizlenme dönemi olacak şe-
kilde iki periyodlu, olarak yapıldı. Kan ör-
nekleri oral dozu takip eden 120 saatlik
aralık içinde alındı ve plazmaları ayrıldık-
tan sonra pioglitazon ve onun aktif meta-
boliti hidroksipioglitazon plazma konsan-
rans tabletleri için medyan Tmax sırasıyla
1.50 saat ve 1.75 saatti. Test formülasyon-
larının ortalamalarının referans formü-
lasyon ortalamalarına olan oranları
AUC0-inf, AUC0-t ve Cmax için sırasıyla
% 99.70, % 100.13 ve % 99.17’ydi. Ayrıca,
AUC0-inf (ln-transformed) ve AUC0-t nin
ortalama oranlarının % 90’lik geometrik
lamlı değişiklikler göstermediği saptandı.
Mevcut çalışmanın ışığında, değerlendiri-
len iki piyoglitazon formülasyonunun
(test ve referans) emilim hızı ve kana ge-
çen miktar anlamında biyoeşdeğer oldu-
kları sonucuna varıldı.

Anahtar kelimeler: CAS 112529-15-4 · Pioglitazon, biyoeşdeğerlik, klinik farmakokinetik

1. Introduction pared with placebo, pioglitazone significantly reduced

Pioglitazone (CAS 112529-15-4) is an orally adminis-
tered insulin sensitising thiazolidinedione agent that
has been developed for the treatment of type 2 diabetes
mellitus. Pioglitazone activates the nuclear peroxisome
proliferator activated receptor-γ (PPAR-γ), which leads
to the increased transcription of various proteins regu-
lating glucose and lipid metabolism [1]. These proteins
amplify the post-receptor actions of insulin in the liver
and peripheral tissues, which leads to improved glyca-
emic control with no increase in the endogenous secre-
tion of insulin. Studies show that pioglitazone stimu-
lates the uptake of glucose and fatty acids into cells by
promoting the synthesis and expression of cellular glu-
cose and fatty acid transporters [2, 3]. In placebo-con-
trolled clinical studies, pioglitazone, whether used as
monotherapy or in combination with sulphonylurea,
metformin or insulin, effectively improved glycaemic
control in patients with type 2 diabetes as evidenced
by significant reductions in HbA1c and fasting plasma
glucose. Pioglitazone also had a beneficial effect on the
abnormal lipid profile seen in type 2 diabetes. Com-
serum triglycerides and increased high density lipo-
protein cholesterol with no change in low density lipo-
protein or total cholesterol [4].
In humans pioglitazone undergoes extensive metab-
olism, yielding primarily keto- (M-III) and hydroxy- (M-
IV) derivatives of pioglitazone. These two active metab-
olites mainly contribute to the extended glucose-lower-
ing effects. The hepatic metabolism of pioglitazone is
catalysed mainly by CYP2C8 and CYP3A [5]. However,
unlike troglitazone, studies have provided no evidence
to suggest that pioglitazone administration leads to in-
hibition or induction of any of the P450 isoenzymes in-
volved in drug metabolism. Therefore pioglitazone may
have a lower potential for drug interaction. In patients
with renal impairment a decrease in AUC of pioglita-
zone and its major metabolites MIII and MIV was ob-
served with increasing kidney failure. This suggests in-
creased hepatic clearance of the drug secondary to re-
duced protein binding in plasma, with no net change
in free plasma drug concentrations [6].

Arzneim.-Forsch./Drug Res. 54, No. 9a, 618−624 (2004)

Wong et al. − Pioglitazone 619
Bioequivalence Studies
The present paper describes the results of a bioequi-
valence trial of two pioglitazone formulations, i.e. a new
pioglitazone hydrochloride generic product as the test
formulation and an orginator product used as the refer-
ence formulation.
2. Methods
2.1. Subjects
Twenty-six healthy, adult, non-smoking male volunteers, aged
between 18 and 55 years, participated in this single-center,
single-dose, randomised, open-label, two-way cross-over bio-
equivalence study conducted at Anapharm Inc. (Sainte-Foy,
Quebec, Canada). All subjects met the inclusion and exclusion
criteria described in the protocol and were judged eligible for
the study, based on medical history, demographic data, med-
ication history, physical examination, vital signs and clinical
laboratory tests. The study was conducted in accordance with
internationally-accepted standards of Good Clinical Practices,
Good Laboratory Practices, local regulatory requirements, and
the principles of the Declaration of Helsinki (Edinburgh, Scot-
land, 2000).
2.2. Pharmacokinetic study
Each subject received either Treatment A [pioglitazone hydro-
chloride 30 mg tablet, manufactured by Eczacıbaşı Pharmaceu-
ticals, Istanbul, Turkey (test)] or Treatment B [30 mg tablet (ref-
erence)] after a supervised overnight fasting of at least 10 h, in
accordance with the randomisation schedule, with a wash-out
period of 14 days between doses.
During each study period, blood samples were taken from
each subject pre-dose and at 0.5, 1.0, 1.5, 1.75, 2.0, 2.5, 3.0, 3.5,
4.0, 4.5, 5.0, 6.0, 8.0, 10.0, 12.0, 16.0, 24.0, 36.0, 48.0, 72.0, 96.0
and 120 h post-dose (n = 23). Plasma was obtained from each
blood sample and stored in a −80 °C (−65 °C to −85 °C) freezer,
pending assay for pioglitazone and hydroxypioglitazone (M-IV)
plasma concentrations.
2.3. Analytical method
Plasma concentrations of pioglitazone and hyroxy-pioglitazone
were measured by a sensitive and specific LC/MS/MS method.
Analysis was perfomed on a PE SCIEX API 4000 (Perkin Elmer,
Toronto, Canada). Positive ions were measured using MRM
(Multiple Reaction Monitoring) mode with m/z 357.1 → 135.1
for pioglitazone and 373.1 → 150.1 for hydroxy-pioglitazone.
The lower limit of quantitation for pioglitazone and hydroxy-
pioglitazone were 10.12 ng/mL and 0.99 ng/mL, respectively.
The standard curve ranges of the plasma concentrations for
pioglitazone and hydroxy-pioglitazone were 10.12 ng/mL to
1517.40 ng/mL and 0.99 ng/mL to 742.05 ng/mL, respectively.
Pioglitazone and hydroxy-pioglitazone were extracted from
plasma by using liquid-liquid phase extraction. Chromato-
graphic conditions: The mobile phase used was a mixture of
methaol/ammonium acetate in a 1/1 ratio. The reverse phase
chromatography was performed on a Zorbax SB-C18 (2), 75 ×
4.6 mm, 3.5 µm (CS, Brockville, Canada), with a flow rate of 1
mL/min and an injection volume of 20 µL. The chromato-
graphic run time lasted 4.5 min and linearity over the calibra-
tion range was 0.9971 for both analytes. The between-run accu-
racy ranged from 90.20 % to 100.65 % for pioglitazone with pre-
cision ranged from 3.33 % to 4.34 %. For hydroxy-pioglitazone,
620 Wong et al. − Pioglitazone
the between-run accuracy ranged from 90.52 % to 96.53 % with
precision ranged from 2.61 % to 4.35 %. The within-run accu-
racy for pioglitazone ranged from 87.83 % to 96.57 % with pre-
cision ranging form 1.47 % to 6.91 %. For hydroxy-pioglitazone,
the within-run accuracy ranged from 87.80 to 96.33 % with pre-
cision ranging form 0.96 to 5.78 %. The method has been
shown to be accurate and repoducible and was successfully
applied for the analysis of clinical samples.
2.4. Pharmacokinetic analysis
Pharmacokinetic parameters, for both pioglitazone and
hydroxypioglitazone, were obtained using non-compartmental
methods. Cmax, the maximum observed concentration, and
Tmax, the time to reach that peak concentration, were deter-
mined for each subject and for each treatment. AUC0-t, the area
under the plasma concentration-time curve from time zero to
the time of the last non-zero concentration was calculated us-
ing the linear trapezoidal rule. The AUC0-inf, the area under the
concentration-time curve from time zero to infinity (extrapo-
lated) was calculated as: AUC0-t + Ct/Kel, where: Ct = the last
non-zero concentration. The residual area, the percentage of
extrapolated area under the curve, was calculated as: 100*(1-
AUC0-t/AUC0-inf ). To calculate the elimination rate constant
(Kel), regression analyses were performed on the natural log
Downloaded by: NYU. Copyrighted material.
(ln) of plasma concentration values (y) versus time (x). The Kel
was taken as the slope multiplied by (−1) and the apparent
half-life (T1/2 el) as (ln 2)/Kel.
2.5. Statistical analysis
Using General Linear Models procedure (GLM, SAS Institute,
Cary, NC, USA), ANOVA was performed on Kel and T1/2 el and
on ln-transformed AUC0-t, AUC0-inf and Cmax at the alpha level
of 0.05. The model included sequence, subject within se-
quence, period and treatment as factors. A non-parametric
test, the Wilcoxon’s test was carried out to compare the Tmax
between treatments. Ratio of means test/reference (Treatment
A/Treatment B) and 90 % geometric confidence interval for the
ratio of means, based on the least-squares means from the
ANOVA of the ln-transformed data, were calculated for AUC0-t,
AUC0-inf and Cmax.
3. Results
3.1. Pharmacokinetic study
All 26 subjects who were enrolled completed the study.
In accordance with the study protocol, plasma samples
of the first 24 subjects completing the study were ana-
lysed and used for pharmacokinetic and statistical anal-
yses. The plots of the mean plasma pioglitazone and
hydroxypioglitazone levels over 120-h sampling period
for both untransformed and ln-transformed data are
presented in Fig. 1 and in Fig. 2. At it is observed in
these figures, the plasma concentrations of pioglitazone
and its main metabolite hydroxypioglitazone did not
differ significantly after administration of both formula-
tions, i.e. pioglitazone test and reference tablets.
Table 1 summarises the pharmacokinetic parameters
of pioglitazone in 24 fasted healthy male subjects who
received a single dose of the two pioglitazone formula-
tions in two different occasions, separated by a washout
Arzneim.-Forsch./Drug Res. 54, No. 9a, 618−624 (2004)
 Bioequivalence Studies

C [ng/mL]
1200

1000

800

600

400

200

0
0.0 6.0 12.0 18.0 24.0 30.0 36.0 42.0 48.0 54.0 60.0 66.0 72.0
Time (h)
lnC (ng/mL)
8.00

7.00

6.00

5.00

Downloaded by: NYU. Copyrighted material.

4.00

3.00

2.00

1.00

0.00
0.0 6.0 12.0 18.0 24.0 30.0 36.0 42.0 48.0 54.0 60.0 66.0 72.0
Time (h)

Fig. 1: Top: Mean concentration-time curves of pioglitazone after administration of the test (pioglitazone hydrochloride, 30 mg) (䊐)
and reference formulations (originator product, 30 mg) (+). Bottom: Mean ln-transformed concentration-time curves of pioglitazone
after administration of the test (pioglitazone hydrochloride, 30mg) (䊐) and reference formulations (originator product, 30 mg) (+).

period of 14 days. The mean Cmax were 1015.46 ng/mL
and 1051.75 ng/mL, while the mean AUC0-inf were
10887.82 ng · h/mL and 10984.58 ng · h/mL for the test
and reference formulations, respectively. The mean
AUC0-t were 10560.15 ng · h/mL and 10619.09 ng · h/
mL for the test and reference formula, respectively. The
median Tmax was 1.50 h for the test tablet and 1.75 h
for the reference formula. The mean T1/2 el was 10.71 h
for the test formulation and 9.96 h for the reference
formulation. Mean Kel values for the test and reference

Table 2: Pharmacokinetic parameters (means ± SD) of hydroxy-

Table 1: Pharmacokinetic parameters (mean ± SD) of pioglitazone
pioglitazone after a single 30 mg oral dose of pioglitazone hy-
after a single 30 mg oral dose of pioglitazone hydrochloride to 24
drochloride to 24 healthy male volunteers under fasting condi-
healthy male volunteers under fasting conditions.
tions.
Parameter (unit) Test Reference Parameter (unit) Test Reference
AUC0-t (ng · h/mL) 10560.15 ± 3001.05 10619.09 ± 2971.27 AUC0-t (ng · h/mL) 21651.75 ± 5984.51 21961.09 ± 6246.84
AUC0-inf (ng · h/mL) 10887.82 ± 3042.26 10984.58 ± 3011.16 AUC0-inf (ng · h/mL) 22592.14 ± 6486.82 22912.12 ± 6819.22
Cmax (ng/mL) 1015.46 ± 238.15 1051.75 ± 351.89 Cmax (ng/mL) 404.75 ± 102.29 410.21 ± 115.38
Residual area (%) 3.11 ± 2.29 3.53 ± 2.04 Residual area (%) 3.93 ± 1.41 3.91 ± 1.71
Tmax (h)a) 1.50 ± 1.13 1.75 ± 0.75 Tmax (h)a) 16.0 ± 4.0 14.0 ± 6.0
Kel (h-1) 0.0764 ± 0.0272 0.0811 ± 0.0278 Kel (h-1) 0.0300 ± 0.0033 0.0303 ± 0.0039
T1/2 el (h) 10.71 ± 5.27 9.96 ± 4.78 T1/2 el (h) 23.40 ± 2.72 23.24 ± 3.28
AUC0-t = the area under the plasma concentration-time curve AUC0-t = the area under the plasma concentration-time curve
from time zero to the time of the last non-zero concentration, from time zero to the time of the last non-zero concentration,
AUC0-inf = the area under the concentration-time curve from time AUC0-inf = the area under the concentration-time curve from time
zero to infinity, Cmax = the maximum observed concentration, Resi- zero to infinity, Cmax = the maximum observed concentration, Resi-
dual area = the percentage of extrapolated area under the curve dual area = the percentage of extrapolated area under the curve
calculated as: 100 · (1 − AUC0-t/AUC0-inf ), Tmax = the time to reach calculated as: 100 · (1 − AUC0-t/AUC0-inf ), Tmax = the time to reach
that peak concentration, Kel = the elimination rate constant, T1/2 that peak concentration, Kel = the elimination rate constant, T1/2
el = the elimination rate constant. el = the elimination rate constant.
a) a)
For Tmax, medians and interquartile ranges are presented instead For Tmax, medians and interquartile ranges are presented instead
of means and SD. of means and SD.

Arzneim.-Forsch./Drug Res. 54, No. 9a, 618−624 (2004)

Wong et al. − Pioglitazone 621
Bioequivalence Studies

C (ng/mL)
450

400

350

300

250

200

150

100

50

0
0 10 20 30 40 50 60 70 80 90 100 110 120
Time (h)
lnC (ng/mL)
7.00

6.00

5.00

Downloaded by: NYU. Copyrighted material.

4.00

3.00

2.00

1.00

0.00
0 10 20 30 40 50 60 70 80 90 100 110 120
Time (h)

Fig. 2: Top: Mean concentration-time curves of hydroxypioglitazone after administration of the test (Pioglitazone hydrochloride, 30
mg) (䊐) and reference formulations (originator product, 30 mg) (+). Bottom: Mean ln-transformed concentration-time curves of
hydroxypioglitazone after administration of the test (Pioglitazone hydrochloride, 30 mg) (䊐) and reference formulations (originator
product, 30 mg) (+).

formulations were 0.0764 h-1 and 0.0811 h-1, respec-
tively.
The same pharmacokinetic parameters for hydroxy-
pioglitazone are presented in Table 2. The mean Cmax
were 404.75 ng/mL and 410.21 ng/mL, while the mean
AUC0-inf were 22592.14 ng · h/mL and 22912.12 ng · h/
mL for the test and reference formulations, respectively.
The mean AUC0-t were 21651.75 ng · h/mL and 21961.09
ng · h/mL for the test and reference formula, respec-
tively. The median Tmax were 16.0 h and 14.0 h for the
test and reference formulations, respectively. The mean
T1/2 el was 23.40 h for the test formulation and 23.24 h
for the reference formulation. Mean Kel values for the
test and reference tablet were 0.0300 h-1 and 0.0303 h-1, re-
spectively. For both pioglitazone and hydroxypioglita-
zone the mean residual areas were lower than 20 %.

Table 4: Hydroxypioglitazone ratios of least-squares means and

Table 3: Pioglitazone ratios of least-squares means and the 90 % the 90 % geometric confidence intervals (CI) based on the pair-
geometric confidence intervals (CI) based on the pairwise com- wise comparisons of the ln-transformed AUC0-t, AUC0-inf and
parisons of the ln-transformed AUC0-t, AUC0-inf and Cmax. Cmax.
AUC0-t AUC0-inf Cmax AUC0-t AUC0-inf Cmax
a)
Ratio 100.13 99.70 99.17 Ratioa)
98.87 98.89 99.61
90 % Geom. CI (%)b) 90.69 to 110.55 90.59 to 109.72 87.52 to 112.37 90 % Geom. CI (%) b) 91.35 to 107.01 91.28 to 107.13 91.25 to 108.72
Intra-subject CV (%) 20.18 19.51 25.61 Intra-subject CV (%) 16.05 16.29 17.81
a) a)
Calculated using least-squares means according to the formula: Calculated using least-squares means according to the formula:
e(test-reference) × 100. e(test-reference) × 100.
b) b)
90 % Geometric confidence interval using ln-transformed data. 90 % Geometric confidence interval using ln-transformed data.
CV = coefficient of variation. CV = coefficient of variation.

Arzneim.-Forsch./Drug Res. 54, No. 9a, 618−624 (2004)

622 Wong et al. − Pioglitazone

ANOVA did not detect any statistically significant dif-
ference between treatments for ln-transformed AUC0-t,
AUC0-inf and Cmax, and untransformed T 1/2el and Kel for
both pioglitazone and hydroxy-pioglitazone. Moreover,
Wilcoxon’s test did not detect any statistically signifi-
cant difference between treatments for untransformed
Tmax for both pioglitazone and hydroxypioglitazone.
The mean ratios as well as the 90 % geometric con-
fidence intervals of the mean ratio of ln-transformed
AUC0-t, AUC0-inf and Cmax of the test and reference for-
mulation for pioglitazone and for hydroxypioglitazone
are displayed in Table 3 and Table 4, respectively. In
accordance with the study protocol, the hypothesis of
bioequivalence of the formulations was accepted if the
90 % confidence intervals of the mean ratio of the test
to reference products were within the acceptance range
of 80 % to 125 % for ln-transformed AUC0-t and within
the acceptance range of 75 % to 133 % for ln-trans-
formed Cmax for both pioglitazone and hydroxypioglita-
zone. This was evidently true for pioglitazone [test
product vs. reference product: (90.59 % to 109.72 %)] for
AUC0-inf, [test product vs. reference product: (90.69 % to
110.55 %)] for AUC0-t and even met the more restrictive
requirement of 80 % to 125 % [test product vs. reference
product: (87.52 % to 112.37 %)] for Cmax. Bioequivalence
was also demonstrated with hydroxypioglitazone [test
product vs. reference product: (91.28 % to 107.13 %)] for
AUC0-inf, [test product vs. reference product: (91.35 % to
107.01 %)] for AUC0-t and even met the more restrictive
requirement of 80 % to 125 % [test product vs. reference
product: (91.25 % to 108.72 %)] for Cmax. The pioglita-
zone intra-subject CVs for AUC0-t, AUC0-inf and Cmax
were respectively 20.18 %, 19.51 % and 25.61 %. The
hydroxypioglitazone intra-subject CVs for AUC0-t, AUC0-
inf and Cmax were respectively 16.05 %, 16.26 % and
17.81 %.
3.2. Drug tolerance
During the study, adverse events were mild or moderate
in severity and no serious, severe or significant adverse
events were reported. The laboratory screening pre-
sented no signs of adverse events or adverse drug reac-
tions. Subjects were monitored for adverse events as
specified in the protocol. A total of 38 post-dose adverse
events were recorded: 12 following administration of
the test product, 24 following administration of the ref-
erence product, and 2 associated with clinically signifi-
cant post-study laboratory results (exact onset time and
date unknown). All adverse events were followed by
complete restitution within a short time period. Thus,
the products were well tolerated overall.
4. Discussion
Pioglitazone is a thiazolidinedione that increases insu-
lin sensitivity in target tissues. It is well-absorbed, with
a mean absolute bioavailability of 83 % and reaching
maximum concentrations in around 1.5 h. It is metab-
olised by the hepatic cytochrome P450 enzyme system
with the formation of two principal active metabolites,
i.e. keto- (M-III) and hydroxy- (M-IV) derivatives of pi-
oglitazone. The half-life of the drug is about 9 h, but
Arzneim.-Forsch./Drug Res. 54, No. 9a, 618−624 (2004)
Bioequivalence Studies
the two active metabolites mainly contribute to the ex-
tended glucose-lowering effects [7].
The aim of the present study was to evaluate the
bioavailability of the tested pioglitazone hydrochloride
30 mg tablet manufactured by Eczacıbaşı Pharmaceu-
ticals and a reference pioglitazone 30 mg tablets admin-
istered as a single oral dose.
The mean residual area for pioglitazone and
hydroxypioglitazone lower than 20 % for both evaluated
formulations indicated that both the duration of sam-
pling time and the sensitivity of the analytical method
were sufficient. The plots of the mean pioglitazone and
its main metabolite − hydroxypioglitazone did not differ
significantly, indicating comparative pharmacokinetic
profile of the two evaluated drug formulations.
For the parent compound, i.e. pioglitazone, the mean
Cmax were 1015.46 ng/mL and 1051.75 ng/mL, while the
mean AUC0-t were 10560.15 ng · h/mL for the test tab-
lets and 10619.09 ng · h/mL for the reference formula-
tion. The results of the present trial regarding pharma-
cokinetics parameters correspond to published data [4,
7, 9]. A median time to maximum serum concentration
of 1.5 h was reported (range 0.5 to 3.0 h), which was not
influenced by an administered dose of the drug from 2
Downloaded by: NYU. Copyrighted material.
to 60 mg. For both single and multiple doses, both the
maximum serum concentration and the area under the
curve of pioglitazone serum concentration against time
increased linearly with increasing dose.
Pioglitazone is an enantiomeric parent drug given as
a racemate. The ratio of the enantiomers in human
plasma is 1:1 [7]. This information and the fact that pi-
oglitazone exhibits linear kinetic behavior was taken
into account when planning the current trial. It was
therefore decided that no enantioselective analytical
method will be needed.
As for the secondary bioequivalence parameter Tmax,
the median values of the two analysed pioglitazone for-
mulas were comparable, i.e. 1.50 for the test tablet and
1.75 h for reference one. These observations testify that
the two pioglitazone formulations are very similar in
terms of the rate of absorption. Furthermore, the sim-
ilar values of T1/2 el and Kel of the evaluated tablets in-
dicate comparable rate of pioglitazone elimination from
the body.
Since hydroxypioglitazone is an active metabolite of
pioglitazone [4, 7], its kinetics was also evaluated. Ac-
cording to the currently valid CPMP Note for Guidance
for the Evaluation of Bioavailability and Bioequivalence,
the evaluation of hydroxypioglitazone would not have
been absolutely necessary due to the linear kinetics of
pioglitazone. The results of the present trial demon-
strate that the 90 % confidence intervals for the metab-
olite were narrower and the intra-individual CV’s were
lower compared to those of the parent compound. It
can be thus concluded that in the case of pioglitazone,
the parent compound is more appropriate for the
evaluation of bioequivalence and a simultaneous evalu-
ation of the metabolite is not absolutely necessary.
Based on the results of the present studies it can be
concluded that the two pioglitazone formulations, i.e.
the test formulation manufactured by Eczacıbaşı Phar-
maceuticals and the reference preparation, are bioequi-
valent.
Wong et al. − Pioglitazone 623
Bioequivalence Studies

5. References [7] Hanefeld, M. Pharmacokinetics and clinical efficacy of

pioglitazone. Int. J. Clin. Pract. Suppl. 121, 19 (2001)
[1] Lehmann, J., Moore, L., Smith-Oliver, T. et al., An antidia-
[8] CPMP Note for guidance: Investigation of bioavailability
betic thiazolidinedione is a high affinity ligand for peroxisome
and bioequivalence. CPMP/EWP/QWP/1401/98 (1998)
proliferator-activated receptor γ (PPAR γ). J. Biol. Chem. 270,
[9] Eckland, D., Danhoh, M., Clinical pharmacokinetics of
12953 (1995)
pioglitazone. Exp. Clin. Endocrinol. Diabetes. 108 (Suppl. 2),
[2] Sandouk, T., Reda, D., Hofman, C., The antidiabetic agent
S234 (2000)
pioglitazone increases expression of glucose transporters in
3T3-F442A cells by increasing messenger ribonucleic acid tran-
script stability. Endocrinology 133, 352 (1993)
[3] Motojima, K., Passilly, P., Peters, J., Expression of putative
fatty acid transporter genes are regulated by peroxisome pro-
liferator-activated receptor α and γ activators in a tissue- and
Correspondence:
inducer specific manner. J. Biol. Chem. 273, 16710 (1998)
Recep Serdar Alpan, MD, PhD, MSc,
[4] Gillies, P., Dunn, Ch,. Pioglitazone. Drugs 60, 333 (2000)
[5] Nowak, S., Edwards, D., Clarke, A., et. al., Pioglitazone: Eczacıbaşı Pharmaceuticals,
effect on CYP3A4 activity. J. Clin. Pharmacol. 42, 1299 (2002) Büyükdere Caddesi, Ali Kaya Sk. No: 7,
[6] Edwards, G., Eckland, D., Pharmacokinetics of pioglita- Levent, 34394, Istanbul (Turkey)
zone in patients with renal impairment. Diabetologia 48 Fax: +90 212 3508650
(Suppl. 1), A230 (1999) [abstract] E-mail: serdara@eczacibasi.com.tr

Downloaded by: NYU. Copyrighted material.

Bioequivalence Study of Rofecoxib
Tablets
Rossen Koytchev a, Yildiz Ozalp b, Aydin Erenmemisogluc, Mike John van der Meer d, and Recep Serdar Alpan b

Cooperative Clinical Drug Research and Development a, Neuenhagen (Germany), EIS Eczacıbaşı
İlaç Sanayi ve Ticaret A.S.b, Istanbul (Turkey), Erciyes University School of Medicinec, Kayseri (Turkey),
and Trident Bioanalytics Ltd.d, Cork (Ireland)

Summary

Key words
䊏 CAS 162011-90-7
䊏 Ecrox
䊏 Rofecoxib, bioequivalence,
clinical pharmacokinetics
Arzneim.-Forsch./Drug Res.
54, No. 9a, 624−628 (2004)
The study was designed to evaluate the
bioavailability of two rofecoxib (CAS
162011-90-7) tablet formulations.
Twenty-four healthy male volunteers
were administered a 25 mg tablet of the
test formulation (Ecrox) containing ro-
fecoxib or the originator product (refer-
ence). The trial was performed according
to an open, cross-over design with a
wash-out period of 7 days. Blood
samples were taken up to 72 h post dose,
the plasma was separated and the con-
centrations of rofecoxib were determined
by an HPLC method. The mean Cmax
were 192.07 ng/mL and 187.35 ng/mL,
while the mean AUC0-t were 3613.84 ng ·
h/mL and 3501.56 ng · h/mL for the test
and reference formulations, respectively.
The median tmax was 3.75 h for the test
tablet and 4.00 h for the reference formu-
lation. The mean t1/2 el was 10.66 h and
10.61 h for the test and reference formu-
lation, respectively. Mean MRT values for
the test and reference tablets were
15.34 h and 15.33 h, respectively. No sig-
nificant differences of pharmacokinetic

Arzneim.-Forsch./Drug Res. 54, No. 9a, 624−628 (2004)

624 Koytchev et al. − Rofecoxib