Bioequivalence study of two fluoxetinecapsule formulations in healthy MiddleEastern volunteers

A.-N. Zaid1, A. Bowirrat1,2, J.J. Kort3 and M. Oscar-Berman4

1Faculty of Pharmacy, An-Najah National University, Nablus, PA, 2The Regional Research and Development Center, The Galilee Society, Shefa-Amr,

PA, 3R and D Manager, Pharmacare PLC Company, Ramallah, Palestine,4Departments of Psychiatry, Neurology and Anatomy and Neurobiology, Boston

University School of Medicine and Boston VA Healthcare System, Boston, MA, USA

Bioequivalence of fluoxetine

Ab stract. Ob jec tive: To as sess the bio -equi valence of two fluoxetine hy dro chlo ridecap sule (20 mg) for mu la tions (Fluoxicarecap sule from Pharmacare Ltd., Chem i calsand Cos me tics, Ramallah, Pal es tine, as testfor mu la tion, and Prozac from Eli Lilly Ltd.,Basingstoke, UK, as ref er ence for mu la tion).De sign and meth ods: The study was con -ducted open with a ran dom ized 2-pe riodcross over de sign and a 6-week wash out pe -riod. Par tic i pants were 24 healthy male vol -un teers aged 18 – 28 years, di vided into2 groups of 12 sub jects. One group was giventhe orig i na tor drug (ref er ence for mu la tion),and the other was given the test for mula (testfor mu la tion). Blood sam ples were ob tained at base line and at 14 time points dur ing the in -ter val 0 – 96 hours af ter drug ad min is tra tion.The con cen tra tions of the sam ples were as -sayed spec tro pho to met ri cally at 220 nm us -ing a Shimadzu 160 A UV-vis i ble spec trom e -ter. We cal cu lated the plasma con cen tra -tion-time curve (AUC), max i mum plasmacon cen tra tion (Cmax), and time of max i mumplasma con cen tra tion (tmax) for each sub ject.Log a rith mic trans for ma tion of the AUC andCmax was used for the sta tis ti cal anal y ses andto as sess the bioavailability of the 2 for mu la -tions, us ing anal y ses of vari ance (ANOVA)and Sather wait t-tests for un equal vari ances.The ANOVA per formed of tmax in Cmax, and in AUC pro vided the ap pro pri ate intra-sub jectvari ance es ti mates to eval u ate the 90% con fi -dence in ter vals for the dif fer ences be tweenstudy vari ables af ter ad min is tra tion of the test and ref er ence for mu la tions. Sta tis ti cal anal y -ses were con ducted on AUC0-4 as the ex trap o -lated part of the AUC, a trun cated area ap -proach was adapted. Re sults: The mean phar -ma cokinetic pa ram e ters for both of the drugsun der study were as fol lows: Cmax = 61.24(± 12.96) ng/ml for the test for mu la tion, and

for the ref er ence for mu la tion Cmax = 61.39 (±14.1) ng/ml, the ef fects were sta tis ti callyequiv a lent. The tmax for the test for mu la tionwas 8.25 (± 1.7) and 7.33 (± 0.96) for the ref -er ence for mu la tion. The ar eas un der the curve to in fin ity (AUC0-¥ (ng, day/ml)) for the testfor mu la tion and for the ref er ence for mu la tion were 293.02 (± 52.69) and 296.15 (± 61.69),re spec tively. Con clu sions: The 2 for mu la -tions had equiv a lent pharmacokine tic pa ram -e ters, were well- tolerated, and their rel a tivebioavailability was 98.94%.

Introduction

Se lec tive se ro to nin reuptake in hib i tors

(SSRIs) have emerged as a ma jor ther a peu tic

ad vance in treat ing psy chi at ric dis eases, es -

pe cially af fec tive dis or ders [Gourion et al.

2004]. An im por tant fea ture of all SSRIs is

that they act as an ti de pres sant drugs be cause

of their abil ity to re vers ibly block the reup -

take of se ro to nin (5-hydroxytryptamine, 5- HT)

in the syn ap tic cleft of neu rons in the cen tral

ner vous sys tem. From a chem i cal per spec -

tive, how ever, dif fer ent SSRI agents have dis -

tinctly dif fer ent pharmacokinetic pro files

which, in turn, can have a ma jor in flu ence on

their clin i cal modes of ac tions. All SSRIs

have a great af fin ity for the 5-HT reup take

car rier in the syn ap tic cleft in the cen tral ner -

vous sys tem [van Harten 1993], a prop erty

shared with the tricyclic antidepressants

(TCAs), with out af fect ing var i ous other cen -

tral neuroreceptors (e.g. his ta mine, ace tyl cho -

line and adrenergic re cep tors) that are re spon -

si ble for many of the safety and toler ability

CPH-5591 / 24.08.2006

Key words

bioequivalence – fluoxe -

tine – bioavailability –

spectrophotometer –

pharmacokinetics

Re ceived

No vem ber 15, 2005;

ac cepted

June 28, 2006

Cor re spon dence to

M. Os car-Berman, PhD

Boston Uni ver sity

School of Med i cine,

L815 715 Al bany Street,

Boston, MA 02118, USA

os car@bu.edu

Orig i nal©2006 Dustri-Verlag Dr. K. Feistle

ISSN 0301-0430

In ter na tional Jour nal of Clin i cal Phar ma col ogy and Ther a peu tics, Vol. 44 – No. n/2006 (nnn-nnn)

prob lems as so ci ated with TCAs [Gourion et

al. 2004].

Most SSRIs have a half-life (t1/2) of ap -

prox i mately 1 day [Preskorn 1997]. One of

the im por tant SSRIs is fluoxetine, which has

a lon ger t1/2 of up to 6 days. Fluoxetine in hib -

its its own me tab o lism [DeVane 1994], and its

ac tive me tab o lite, norfluoxetine, has an ex -

tended t1/2 of 7 – 15 days. This re sults in an ex -

tended time to steady state and a pro longed

wash out pe riod when dos ing is dis con tin ued.

In deed, fluoxetine is me tab o lized ex ten sively

in the liver, pri mar ily via N-demethylation to

norfluoxetine [Aronoff et al. 1984, Fuller and

Wong 1987, Lemberger et al. 1985]. The ac -

tive me tab o lite of fluoxetine has an in sig nif i -

cant ef fect on fluoxetine plasma lev els by in -

hib it ing fluoxetine me tab o lism. Glucuronide

con ju gates are also found but in small quan ti -

ties. In healthy in di vid u als ap prox i mately

60% of an oral dose is ex creted in the urine

within 35 days, with only 2.5% as un changed

drug and 5.2% as its glucuronide con ju gate,

10% as norfluoxetine, and 9.5% as norfluoxe -

tine glucuronide. Most of the ex creted drug

(72.8%) consists of unidentified metabolites

[Altamura et al. 1994].

Fluoxetine is a bicyclic de riv a tive of phenyl -

propylamine, and it is a widely used se lec tive

5HT reuptake in hib i tor with an ti de pres sant

prop er ties. Fluoxetine is well-ab sorbed af ter

oral in take, is highly protein- bound and has a

large vol ume of dis tri bu tion. It is pre scribed

for a va ri ety of psychopathological con di tions,

in clud ing mood dis or ders, eat ing dis or ders,

ob ses sive-com pul sive dis or ders, de press ion in

the el derly, and dysthymia [Alta mura and

Mauri 1991, Altamura et al. 1994, Benfield et

al. 1986]. In con trast with its ef fect on the

pharmacokinetics of other an ti de pres sants,

age does not af fect fluoxetine pharmacoki -

netics. This find ing to gether with the better

tolerability pro files of fluoxetine (com pared

with tricyclic an ti de pres sants) makes this

drug par tic u larly suit able for use in el derly

pa tients with de pres sion.

Fluoxetine un der goes ex ten sive first-pass

me tab o lism in the liver. Be cause of this,

marked interindividual dif fer ences in av er age

peak plasma con cen tra tion (Cmax) val ues are

ap par ent af ter a stan dard daily dos age. For

ex am ple, a 3- to 4-fold interindividual vari a -

tion in Cmax is seen af ter ad min is tra tion of sin -

gle doses to healthy vol un teers [Aronoff et al.

1984]. Fluoxetine has a large vol ume of dis -

tri bu tion (20 – 40 l/kg) prob a bly as a re sult of

ex ten sive tis sue dis tri bu tion and bind ing

[Benfield et al. 1986]. Fluoxetine is 94.5%

bound to plasma pro teins, mainly to al bu min

and a1-glycoprotein [Aronoff et al. 1984,

Lemberger et al. 1985]. The ex tent of bind ing

ap pears to be in de pend ent of plasma con cen -

tra tion. Fluoxetine stor age in body fat ap pears

to be lim ited since dis tri bu tion in obese and

non-obese (lean) in di vid u als is sim i lar, and

the pharmacokinetics of fluoxetine are not af -

fected by ei ther obe sity or re nal im pair ment

[Altamura et al. 1994]. How ever, fluoxetine

has a non-lin ear pharmacokinetic pro file and,

there fore, the drug is used with cau tion in

patients with a reduced metabolic capability

(i.e. hepatic dysfunction).

Fluoxetine is elim i nated slowly from the

body. Fol low ing a sin gle oral dose in healthy

adults, the t1/2 elim i na tion of fluoxetine re -

port edly av er ages 4 – 6 days. How ever, the

plasma half-life ex hib its con sid er able interin -

divi dual vari a tion with a re ported range be -

tween 1 and 13 days [Aronoff et al. 1984].

Fluoxetine is me tab o lized pri mar ily via N-de -

methylation to norfluoxetine [Aronoff et al.

1984, Fuller and Wong 1987, Lemberger et

al. 1985]. In ad di tion, glucuronide con ju gates

are also found in small quan ti ties. In healthy

in di vid u als ap prox i mately 60% of an oral

dose is ex creted in the urine within 35 days,

with only 2.5% as un changed drug and 5.2%

as its glucuronide con ju gate, 10% as nor flu -

oxetine, and 9.5% as norfluoxetine glucu -

ronide. Most of the ex creted drug (72.8%)

con sists of un iden ti fied me tab o lites [Alta -

mura et al. 1994]. The con sid er able interin -

dividual vari abil ity in fluoxetine elim i na tion

may be re lated to ge netic dif fer ences in the

rate of N-demethylation of the drug in the

liver.

Down-reg u la tion of 5-HT1 re cep tors is

the most com monly re ported CNS ef fect of

subchronic ex po sure (i.e. for at least 10 days),

but with out a change in re cep tor af fin ity for

its ligand [Beasley et al. 1992]. How ever, its

ef fect on the 5-HT2 re cep tors is con tro ver sial

[Altamura et al. 1994]. Fluoxetine seems to

facilitate serotonergic trans mis sion by down-

regulation of presynaptic in hib i tory autore -

ceptors [Beasley et al. 1992]. Ini tially, it was

sug gested that the ef fec tive dose of fluoxetine

was 80 mg/day, but a dos age of 20 mg/day has

Zaid, Bowirrat, Kort and Oscar-Berman 2

CPH-5591 / 24.08.2006

also been shown to be ef fec tive with a good

risk-ben e fit ra tio [Altamura et al. 1988].

In the pres ent study, we sought to com pare

the bioequivalence of two fluoxetine hy dro -

chlo ride cap sule for mu la tions: Fluoxicare as

a test for mu la tion, and Prozac as the ref er ence

for mulation.

Materials and methods

Subjects

The par tic i pa tion of vol un teers was so lic -

ited through news pa per ad ver tise ments. Be -

cause no women re sponded (per haps re lated

to cul tural tra di tions), the re search par tic i -

pants were 24 healthy young male adults (age

range 18 – 28 years). The na ture and the aims

of the study were ex plained to all the par tic i -

pants, and writ ten in formed con sents were

ob tained. All vol un teers were ex am ined clin i -

cally by a qual i fied at tend ing phy si cian who

judged el i gi bil ity for par tic i pa tion. The clin i -

cal as sess ment was com ple mented by lab o ra -

tory ex am i na tion, which con firmed that all of

the par tic i pants had nor mal he ma to log i cal

val ues, and re nal and hepatic func tions were

within nor mal lim its. Ex clu sion cri te ria in -

cluded ex treme weight ranges (over weight or

un der weight), ane mia, liver or re nal in fec tion

and par a sitic or other dis eases or con di tions

that might af fect ab sorp tion, dis tri bu tion

and/or elim i na tion of fluoxetine. Ad di tion -

ally, in di vid u als with psy chi at ric dis or ders or

those tak ing pre scrip tion med i ca tions also

were ex cluded. The sub jects were re quired to

ab stain from tak ing drugs or al co hol for at

least 3 days prior to the ex per i ment and

throughout the study period. The volunteers

also were instructed to fast for at least 8 hours

on the night before drug administration.

Drug administration

The par tic i pants were ran domly as signed

to 1 of 2 groups of 12 sub jects. One group was

given the ref er ence stan dard prep a ra tion and

the other was given the test prep a ra tion with a

cross over af ter the drug wash out pe riod of 6

weeks. On the morn ing of the ex per i ment, a

blood sam ple was taken from each vol un teer

to serve as a zero-time blank for the drug as -

say. Each of the 24 vol un teers then took a

stan dard dose of 1 cap sule (equiv a lent to 20

mg of fluoxetine), ei ther as the test Fluoxicare

(Pharmacare Ltd., Ramallah, Palestine) or the

ref er ence Prozac (Eli Lilly Com pany, Ltd.,

Basingstoke, UK), fol lowed by 200 ml of wa -

ter. 4 hours af ter drug ad min is tra tion, the sub -

jects were al lowed a stan dard break fast of

bread, jam, low-fat white cheese, fruit juice

and tea. From then on, they were al lowed free

ac cess to fruit juice and other non-al co holic

bev er ages. They had their sec ond meal 4

hours later.

It should be noted that fluoxetine is well-

absorbed from the gas tro in tes ti nal tract af ter

oral ad min is tra tion, and its bioavail ability is

not af fected by the pres ence of food (12, 13).

Af ter a sin gle oral dose of fluoxetine the av er -

age peak plasma con cen tra tion (Cmax) oc curs

be tween 6 and 8 hours post dose (range 1.5 –

12 hours) [Lemberger et al. 1985]. The mean

time taken to achieve Cmax val ues (tmax) is de -

layed by 3 – 5 hours when fluoxetine is ad -

min is tered with food. How ever, the ex tent of

ab sorp tion (bioavailability) and the Cmax val -

Bioequivalence of fluoxetine 3

CPH-5591 / 24.08.2006

Fig ure 1. Rep re sen ta tive chromatogram of a pa -

tient sam ple.

ues are vir tu ally un changed with food. Be -

sides, over an oral dose range of 20 – 80 mg,

Cmax val ues are pro por tional [Lemberger et

al. 1985].

Product description

Test drug

Fluoxicare cap sule, 20 mg fluoxetine HCl,

batch num ber 36D8, man u fac tur ing date

04/1998, and ex pi ra tion date 04/2000, pro -

duced by Pharmacare Ltd., Pharmaceuticals,

Chem i cals and Cos me tics, Ramallah, Pal es -

tine.

Reference drug

Prozac cap sule, 20 mg fluoxetine HCl,

batch num ber B4125GE, man u fac tur ing date

05/1996, and ex pi ra tion date 06/1999, pro -

duced by Eli Lilly and Com pany Ltd.,

Basingstoke, UK.

Sample collection and drug

assay

Blood sam ples were col lected from an in -

dwell ing cath e ter in the antecubital vein of

the arm. Sam pling times were pre-spec i fied at

0.0, 0.5, 1.0, 1.5, 2.0, 3.0, 4.0, 6.0, 8.0, 10.0,

12.0, 24.0, 72.0 and 96.0 hours af ter drug ad -

min is tra tion. The sam ples were col lected in

heparinized tubes, and the plasma was sep -

arated by centrifugation at 3,000 rpm for

10 min, and was stored at –20 °C until as -

sayed.

Instrumentation and accessories

The chro mato graphic sys tem con sisted of

wa ters high-pres sure liq uid chro mato graphic

analysis (HPLC) sys tem equipped with a 600

pump, 486 UV de tec tor (226 nm), and

u-Bondapak C18 col umn (30 cm) length, and

a Mil len nium 2.1 com puter soft ware sys tem.

The flow rate was ad justed at 2 ml/min.

Mobile phase

The mo bile phase was a mix ture of wa ter :

acetonitrile : triethylamine (40 : 50 : 10 v/v).

The pH of the wa ter and triethylamine mix -

ture was ad justed to 6.5 by phos pho ric acid.

The mo bile phase com po nents were fil trated

through a 0.45 mm Millipore fil ter sep a rately.

Af ter mix ing of com po nents, the mix ture was

de gassed us ing a Sonicator for 10 min.

Preparation of plasma samples

The frozen plasma sam ples were thawed

and then mixed by vor tex: an aliquot of 1,200

ml of each plasma sam ple, vortexed for 20 sec

and cen tri fuged for 15 min at 5,000 rpm in a

cool ing cen tri fuge. The supernatant was sep -

a rated and then fil trated through a 0.45 mm

Millipore fil ter. 20 ml of the fil trate were in -

jected into the HPLC sys tem for anal y sis.

Quantification

A stan dard curve was pre pared by spik ing

plasma with stan dard drug so lu tions pre pared

Zaid, Bowirrat, Kort and Oscar-Berman 4

CPH-5591 / 24.08.2006

Fig ure 2. Mean plasma fluoxetine con cen tra tion-

time pro file fol low ing oral ad min is tra tion of ei ther

one 20 mg test for mu la tion or ref er ence for mu la tion

cap sule (mean ± SD).

in acetonitrile, and the best fit ting straight line

was ob tained by the method of least squares

[Mandel 1984]. The un known sam ple con -

cen tra tions were cal cu lated from the stan dard

curve.

Pharmacokinetic and statistical

analysis

The con cen tra tions of fluoxetine in the

plasma at the spec i fied times of sam ple col -

lec tion fol low ing ad min is tra tion of ei ther a

test or a refernece cap sule were plot ted for

each in di vid ual vol un teer. The val ues for

peak con cen tra tion (Cmax) and the time re -

quired for reach ing it (tmax) were re corded.

Our study de sign fol lowed a “trun cated area

ap proach” due to the long elim i na tion half

time of the drug (4 – 6 days). The area un der

the plasma drug con cen tra tion “C” ver sus

time (t) curves from 0 – 4 days (AUC0-4) were

cal cu lated di rectly from the plasma drug val -

ues by the “trap e zoidal rule,” us ing the re -

lationship:

( )AUC

C C 1

2 t t0 4

n n

n 1 n® +=+ +

-å (Eq. 1)

The area un der the curve from 4 days to

in fin ity was ob tained from the re la tion ship:

AUC C /0 4® ¥ = b (Eq. 2)

where C4 is the con cen tra tion of fluoxe -

tine at the 4-day time point (last de ter mined

con cen tra tion), and b is the elim i na tion-rate

con stant of fluoxetine in a par tic u lar indi -

vidual.

The elim i na tion rate con stant (b) was cal -

cu lated from the slope of the straight part of

the tail-end of the curve of log C vs. t, where

the slope of the straight line equals (b/2.303).

To ob tain this value, the data for each in di vid -

Bioequivalence of fluoxetine 5

CPH-5591 / 24.08.2006

Ta ble 1. Pharmacokinetic pa ram e ters de rived af ter a sin gle dose ad min is tra tion of the test for mu la tion (20 mg fluoxetine) cap sule to 24

healthy male vol un teers. Min i mum and max i mum val ues are sig ni fied by min and maxi, re spec tively. The range for to tal AUC (min i mum –

max i mum) is 202.16 – 399.88, and the me dian is 296.14.

Sub ject Cmax tmax b t1/2 AUC AUC To tal AUC

(ng/ml) (h) (h–1) (day) 0 – 4 4 – ¥ 0 – ¥

(ng/day/ml) (ng/day/ml) (ng/day/ml)

1 42.91 8.0 5.6–03 5.17 108.90 149.34 258.24

2 40.44min 10maxi 7.3–03 3.97 123.07 125.49 248.56

3 61.93 10.0 3.8–03 7.53 125.15 271.68 396.83

4 60.61 6.0min 4.4–03 6.53 112.48 207.15 319.63

5 44.86 8.0 2.9–03min 9.85maxi 101.42min 298.46maxi 399.88maxi

6 53.84 10.0 8.3–03 3.46 123.01 102.49 225.50

7 59.40 10.0 7.8–03 3.67 157.22maxi 143.15 300.37

8 60.16 10.0 9.7–03 2.99 120.19 81.97 min 202.16 min

9 69.13 10.0 6.5–03 4.42 152.62 169.48 322.10

10 71.10 10.0 8.9–03 3.21 165.98 120.33 286.31

11 48.79 10.0 5.6–03 5.16 120.56 171.35 291.91

12 43.39 10.0 6.5–03 4.41 101.69 114.59 216.28

13 66.26 6.0 5.1–03 5.75 127.49 189.51 317.00

14 57.4 8.0 7.3–03 3.97 154.22 154.77 308.99

15 63.71 6.0 6.4–03 4.52 140.91 188.69 329.60

16 72.19 10.0 6.9–03 4.18 126.95 125.05 252.00

17 54.15 8.0 5.7–03 5.19 144.73 224.77 369.50

18 59.37 6.0 6.2–03 4.64 150.15 127.01 327.16

19 56.89 6.0 6.8–03 4.23 143.57 131.66 275.23

20 63.71 8.0 6.4–03 4.46 149.50 171.32 320.82

21 75.83 8.0 9.9–03maxi 2.89min 145.38 89.76 235.14

22 69.16 8.0 9.2–03 3.12 147.96 103.73 251.69

23 94.35maxi 6.0 6.2–03 4.65 143.99 167.96 311.95

24 80.09 6.0 7.1–03 4.09 132.44 143.20 265.64

Me dian 60.385 8 6.5–03 4.415 136.675 146.27 296.14

Mean 61.24 8.25 6.69–03 4.67 134.15 158.87 293.02

SD 12.69 1.70 1.74–03 1.55 18.03 53.53 52.69

ual vol un teer were plot ted on semilog graph

pa per (to en sure lin ear ity of the tail-end), and

the best fit ting line was de ter mined by the

method of least squares [Mandel 1984].

The to tal area un der the curve from 0 to in -

fin ity (AUC0-¥) was cal cu lated as the sum of

the ar eas ob tained:

AUC0-¥ = AUC0-4 + AUC4-¥ (Eq.3)

The rel a tive bioavailability was cal cu -

lated us ing the for mula: Rel a tive Bioavail -

ability = [AUC0-¥ of test for mu la tion] ÷

[AUC0-¥ of ref er ence for mu la tion × 100.

The soft ware pro gram used for pharma -

cokinetic and sta tis ti cal anal y sis eval u a tions

was a Minitab Sta tis ti cal Pack age 13, per -

formed on an IBM PC.

Assay validation report

Stan dard cal i bra tion curve for plasma

sam ples spiked by stan dard fluoxetine is lin -

ear in the range of 15 – 1,000 ng/ml, with a re -

gres sion equa tion of y = 2.091 x –43.2339

and a cor re la tion co ef fi cient of 0.9855.

The val i da tion of the as say method was

de ter mined in terms of y = 2.091 x –43.2339

and a cor re la tion co ef fi cient of 0.9855.

The val i da tion of the as say method is

shown in the ta ble be low.

The per for mance of the method was mon -

i tored us ing 3 con cen tra tions of qual ity con -

Zaid, Bowirrat, Kort and Oscar-Berman 6

CPH-5591 / 24.08.2006

Ta ble 2. Pharmacokinetic pa ram e ters de rived af ter a sin gle dose ad min is tra tion of the ref er ence for mu la tion (20 mg fluoxetine) cap -

sule to 24 healthy male vol un teers. Min i mum and max i mum val ues are sig ni fied by min and maxi, re spec tively. The range for to tal AUC

(min i mum to max i mum) is 176.98 – 480.07, and the me dian is 292.335.

Sub ject Cmax tmax b t1/2 AUC AUC To tal AUC

(ng/ml) (h) (h–1) (day) 0 – 4 4 – ¥ 0 – ¥

(ng/day/ml) (ng/day/ml) (ng/day/ml)

1 54.64 8.0maxi 5.3–03 5.47 116.35 169.67 286.022 33.89min 8.0 5.9–03 4.88 100.26min 133.84 234.13 38.56 8.0 4.4–03 6.64 102.47 196.47 298.944 62.66 6.0min 4.1–03 6.99 113.07 206.92 319.995 51.15 8.0 3.7–03

min 7.82maxi 110.82 259.46maxi 370.286 52.16 8.0 8.7–03 3.30 142.87 112.0 254.877 59.8 8.0 6.5–03 4.48 132.96 155.04 288.008 66.3 6.0 9.5–03

maxi 3.03 min 139.20 93.86 233.069 74.06 8.0 6.9–03 4.17 146.52 150.36 296.8810 77.15 8.0 9.3–03 3.09 150.01 100.88 250.8911 38.77 8.0 4.8–03 6.02 108.02 188.65 296.6712 67.4 6.0 5.8–03 5.01 122.68 159.19 281.8713 70.34 6.0 6.9–03 4.14 146.42 153.09 299.5114 58.55 6.0 6.2–03 4.64 139.93 167.43 307.3615 71.04 6.0 5.8–03 5.01 164.47 187.36 351.8316 84.53maxi 8.0 6.6–03 4.38 221.49maxi 258.58 480.07maxi17 53.93 8.0 5.2–03 5.59 145.04 225.43 370.4718 60.0 6.0 7.9–03 3.67 142.46 126.98 269.4419 61.1 8.0 07.2–03 4.0 100.37 76.61min 176.98min20 74.8 8.0 6.4–03 4.52 163.80 190.71 354.5121 83.36 6.0 9.4–03 3.07 142.99 85.35 228.3422 76.68 8.0 8.7–03 3.34 186.71 146.77 333.4823 41.22 8.0 4.9–03 5.87 106.18 167.67 273.8524 61.34 8.0 6.2–03 4.68 115.31 134.83 250.14

Me dian 61.22 8.0 6.3–03 4.58 139.565 157.115 292.335Mean 61.39 7.33 6.51–03 4.74 135.85 160.29 296.15SD 14.10 0.96 1.69–03 1.28 29.21 49.16 61.69

Ac cu racy 100 ± 4.31%

Pre ci sion: Within-runs C.V. 4.8%

Be tween-runs C.V. 7%

Limit of de tec tion 10 ng/ml

Low limit of quantitation

LLOQ 15 ng/ml

Range 15 – 1,000 ng/ml

Lin ear ity y = 2.091 x –43.2339*

*y = peak area, x = fluoxetine con cen tra tion (ng/ml),

–43.2339 is the in ter cept.

trol sam ples pre pared by spik ing drug-free

plasma with low, me dium, and high con cen -

tra tions of fluoxetine. The qual ity con trol

sam ples were as sayed fol low ing the as say of

ev ery 12 clin i cal sam ples. The per for mance

of the as say dur ing the anal y sis of the study

was eval u ated by the anal y sis of the qual ity

con trol (QC) sam ples. The QC sam ples were

pre pared by spik ing blank plasma with flu -

oxetine, where the lower con cen tra tion of the

QC sam ples was close to 15 ng/ml (LLOQ).

This means, that all QC sam ples were put

nearly to the end of mea sure ments of vol un -

teers’ treat ment. An a lyt i cal mea sure ments

fol low the rec om men da tions given by the

FDA Guid ance for In dus try-Bioanalytical

method val i da tion.

Results

A bioequivalence mea sure of 2 for mu la -

tions of the same drug com prises equiv a lence

with re spect to the rate and ex tent of their ab -

sorp tion. The area un der the concentration-

time curve (AUC) gen er ally serves as the

Bioequivalence of fluoxetine 7

CPH-5591 / 24.08.2006

Table 4. Bioavailability de ter mi na tion of mean Cmax of fluoxetine fol low ing the ad min is tra tion of test for mu la tion and Ref er ence for mu la tion

cap sules as stan dard ized us ing 90% con fi dence lim its. Min i mum and max i mum val ues are sig ni fied by min and maxi, re spec tively.

Sub ject Test for mu la tion Ref er ence formulation T-R T/R LogT/R EXP

1 42.91 54.63 –11.73 0.785322 –0.10495 0.900368

2 40.44min 33.89min 6.55 1.193272 0.07674 1.079761

3 61.93 38.56 23.37 1.606068 0.205764maxi 1.228463

4 60.61 62.66 –2.05 0.967284 –0.01445 0.985658

5 44.86 51.15 –6.29 0.877028 –0.05699 0.944607

6 53.84 52.16 1.68 1.032209 0.013767 1.013863

7 59.4 59.8 –0.4 0.993311 –0.00291 0.99709

8 60.16 66.3 –6.14 0.907391 –0.04221 0.958673

9 69.16 74.06 –4.93 0.933432 –0.2992 0.970526

10 71.1 77.15 –6.05 0.921581 –0.03547 0.965155

11 48.79 38.77 10.02 1.258447 0.099835 1.104989

12 43.39 67.4 –24.01min 0.643769min –0.19127 0.825909 min

13 66.26 70.34 –4.08 0.941996 –0.02595 0.974383

14 57.44 58.55 –1.11 0.981042 –0.00831 0.991722

15 63.71 71.04 –7.33 0.896819 –0.0473 0.953806

16 72.19 84.53maxi –12.34 0.854016 –0.06853 0.933762

17 54.15 53.93 0.22 1.004079 0.001768 1.00177

18 59.37 60.0 –0.63 0.9895 –0.00458 0.995426

19 56.89 61.1 –4.21 0.931097 –0.03101 0.96947

20 63.71 74.8 –11.09 0.851738 –0.06969 0.932679

21 75.83 83.36 –7.53 0.909669 –0.04112 0.959717

22 69.16 76.68 –7.52 0.90193 –0.04483 0.956163

23 94.35maxi 41.22 53.13maxi 2.288937maxi –0.359634min 1.432805maxi

24 80.09 61.34 18.75 1.305673 0.115835 1.12281

Me dian 60.39 61.22 –4.15 0.937714 0.02234 0.9724545

Mean 61.24 61.39 –0.16 1.04065 0.002244 1.008316

SD 12.69 14.10 15.04 0.328126 0.108994 0.120524

The av er age ra tio of Cmax (up per limit) = 1.059; the av er age ra tio of Cmax (lower limit) = 0.957. This com plies with the FDA lim its

(0.8 – 1.25).

Ta ble. 3. Mean (geo met ric) pharmacokinetic pa ram e ters show ing sim i lar

period/se quence ef fects de rived af ter a sin gle dose ad min is tra tion of fluoxetine

20 mg cap sule to 24 healthy male volunteers.

Pa ram e ters Test for mu la tion (20 mg) Ref er ence for mu la tion(20 mg)

Cmax (ng/ml) 61.24 ± 12.69 61.39 ± 14.1

tmax (h) 8.25 ± 1.7 7.33 ± 0.96

AUC0ॠ(ng/day/ml) 293.02 ± 52.69 296.15 ± 61.69

Rel a tive bioavailability 98.94%

Each pa ram e ter rep re sents the mean ± SD, n = 24, Cmax = max i mum plasma

con cen tra tion, tmax = time of peak plasma con cen tra tion, AUC0ॠ= area un der

the plasma con cen tra tion-time curve, rel a tive bioavailability =

AUC of fluoxicare 100

AUC of prozac0

0

® ¥

® ¥

´

char ac ter is tic of the ex tent of ab sorp tion,

while the peak con cen tra tion (Cmax) and the

time of its oc cur rence (tmax) re flect the rate of

ab sorp tion, es pe cially in fast re leas ing drug

for mu la tions [Najib et al. 2003]. Re sults from

our study can be found in Fig ures 1 and 2, and

in Ta bles 1 – 9. Fig ure 1 is a rep re sen ta tive

chromatogram of the cen tri fuged plasma sam -

ple from an in di vid ual sub ject. Fig ure 2 shows

the mean plasma con cen tra tion-time curve of

fluoxetine for both prod ucts. The peak plas -

ma con cen tra tion (Cmax) of fluoxe tine fol low -

ing the ad min is tra tion of Fluoxi care cap sules

ranged from 40.44 – 94.35 ng/ml, with a

mean value of 61.24 (± 12.69) ng/ml (Ta ble

1). The peak plasma con cen tra tion of flu -

oxetine fol low ing the ad min is tra tion of Pro -

zac cap sules ranged from 33.89 – 41.22 ng/ml

with a mean value of 61.39 (± 14.10) ng/ml

(Ta ble 2). Sta tis ti cal anal y ses re vealed no sig -

nif i cant dif fer ence be tween fluoxetine Cmax

pro duced af ter the ad min is tra tion of both Flu -

oxicare and Prozac cap sules us ing 1-way

ANOVA and Sather wait t-tests for un equal

vari ances (Ta bles 6, 7, 8).

The time-to-peak plasma con cen tra tion

(tmax) fol low ing the ad min is tra tion of test for -

mu la tion cap sules ranged from 6 – 10 hours,

with a me dian value of 8 hours (mean 8.25 ±

1.7), and the time to peak plasma con cen tra -

tion (tmax) fol low ing the ad min is tra tion of ref -

er ence for mu la tion cap sules ranged from 6 –

8 hours with a me dian value of 8 hours (mean

7.33 ± 0.96) (Ta ble 3). The ANOVA of

fluoxetine tmax fol low ing the ad min is tra tion

of test for mu la tion and ref er ence for mu la tion

cap sules, re vealed a sta tis ti cally sig nif i cant

dif fer ence be tween the 2 prep a ra tions (p =

0.026) (Ta ble 8). Since the tem po ral mea sure

(hours) was coarse, a c2-test also was per -

formed, yield ing a com pa ra ble sig nif i cant

dif fer ence be tween tmax val ues for the 2 cap -

Zaid, Bowirrat, Kort and Oscar-Berman 8

CPH-5591 / 24.08.2006

Ta ble 5. Bioavailability de ter mi na tion of mean AUC0-¥ of fluoxetine fol low ing the ad min is tra tion of test for mu la tion and ref er ence for mu la tion

cap sules as stan dard ized us ing 90% Con fi dence lim its. Min i mum and max i mum val ues are sig ni fied by min and maxi, re spec tively.

Sub ject Test for mu la tion Ref er ence formulation T-R T/R LogT/R EXP

1 258.24 286.02 –27.78 0.902874 –0.04437 0.956597

2 248.56 234.10 14.46 1.061768 0.02603 1.026372

3 396.83 298.94 97.89 1.327457 0.12302 1.130908

4 319.63 319.99 –0.36 0.998875 –0.00049 0.999511

5 399.88maxi 370.28 29.60 1.07994 0.033399 1.033963

6 225.50 254.87 –29.37 0.884765 –0.05317 0.984217

7 300.37 288.00 12.37 1.042951 0.018264 1.018432

8 202.16min 233.06 –30.90 0.867416 –0.06177 0.940097

9 322.10 296.88 25.22 1.08495 0.3541maxi 1.036044

10 286.31 250.89 35.42 1.14117 0.057353 1.05903

11 291.91 296.67 –4.76 0.983955 –0.00702 0.993

12 216.28 281.87 –65.59 0.767304 –0.11503 0.891337

13 317.00 299.51 17.49 1.058395 0.024648 1.024954

14 308.99 307.36 1.63 1.005303 0.002297 1.0023

15 329.60 351.83 –22.23 0.936816 –0.02835 0.972052

16 252.00 480.07maxi –228.07min 0.524923min –0.2799min 0.755856min

17 369.50 370.47 –0.97 0.997382 –0.00114 0.998862

18 327.16 269.44 57.72 1.214222 0.084298 1.087953

19 275.23 176.98min 98.25maxi 1.555147maxi 0.191772 1.211394maxi

20 320.82 354.51 –33.69 0.904967 –0.04337 0.95756

21 235.14 228.34 6.8 1.02978 0.012745 1.012826

22 251.69 333.48 –81.79 0.754738 –0.1222 0.884968

23 311.95 273.85 38.1 1.139127 0.056572 1.058203

24 265.64 250.14 15.5 1.061965 0.02611 1.026454

Me dian 296.14 292.34 4.22 1.0175415 0.007521 1.007563

Mean 293.02 296.15 –3.13 1.013592 –0.0027 1.00112

SD 52.69 61.69 64.19 0.199692 0.090416 0.087679

The av er age ra tio of AUCs (up per limit) = 1.038, the av er age ra tio of AUCs (lower limit) = 0.964; this com plies with the FDA lim its

(0.8 – 1.25).

sule for mu la tions (c2 = 13.59, p = 0.001).

Other in ves ti ga tors have re ported that the tmax

of fluoxetine has a wide range of vari abil ity

(from 1.5 – 12 hours) [Bergstrom et al. 1988,

Lemberger et al. 1985]. This may be due to in -

di vid ual vari a tions as well as the dif fer ences

in the man u fac tur ing pro cess.

The area un der the plasma concentration-

time curve AUC0-¥ for fluoxetine fol low ing

test for mu la tion ad min is tra tion ranged

from 202.16 – 399.88 ng/day/ml, with a mean

value of 2,930.02 (± 52.69) ng/day/ml, while

the AUC0-¥ af ter ref er ence for mu la tion ad -

min is tra tion ranged from 176.98 – 4,800.07

ng/day/ml, with a mean value of 296.15 (±

61.69) ng/day/ml (Ta bles 1, 3, 5). Sta tis ti cal

anal y ses showed no sig nif i cant dif fer ence be -

tween fluoxetine AUC0-¥ pro duced af ter the

ad min is tra tion of ei ther test for mu la tion or

ref er ence for mu la tion cap sules us ing both

1-way ANOVA and Satherwait t-tests for un -

equal vari ances (Ta ble 9). The sta tis ti cal anal -

y ses were con ducted on AUC0-4 as the ex trap -

o lated part of the AUC, and it was above 20%.

Of note, AUC0-¥ is not ad e quately re flected

be cause a trun cated area ap proach was

adapted.

The per cent age rel a tive bioavailability of

fluoxetine from test for mu la tion cap sules

com pared to ref er ence for mu la tion cap sules

was found to be 98.94% as de ter mined by the

ra tios be tween the AUC 0-¥ of test for mu la -

tion to ref er ence for mu la tion (Ta ble 3).

Summary and conclusions

24 healthy men were ad min is tered equiv -

a lent doses of test for mu la tion and ref er ence

for mu la tion in an open, cross over de sign in

which all sub jects re ceived both drugs. The

mean pharmaco kinetic pa ram e ters for both

drugs were: max i mum plasma con cen tra tion

(Cmax) for 20 mg test for mu la tion 61.24 (±

12.96) ng/ml and for ref er ence for mu la tion

61.39 (± 14.1) ng/ml. The dif fer ence be tween

these re sults was sta tis ti cally in sig nif i cant.

Also the EXP of Test/Ref er ence (T/R) was

0.964 – 1.038; us ing a 90% con fi dence limit,

which com plies with the Food and Drug Ad -

min is tra tion (FDA) limit (0.8 – 1.25). The

time to peak con cen tra tion (tmax) for test for -

mu la tion was 8.25 (± 1.7), while it was 7.33

(± 0.96) for ref er ence for mu la tion. This dif -

fer ence is sta tis ti cally sig nif i cant (0.026) (Ta -

ble 8), a dif fer ence that is within the pre vi -

ously re ported limit (13). The ar eas un der the

plasma concentration- time curve to in fin ity

for test for mu la tion and ref er ence for mu la -

tion (AUC 0-¥ (ng, day/ml)) were found to be

293.02 (± 52.69) and 296.15 (± 61.69), re -

spec tively. Also the EXP of T/R 90% con fi -

dence in ter val was 0.964 – 1.38%, and com -

Bioequivalence of fluoxetine 9

CPH-5591 / 24.08.2006

Ta ble 6. ANOVA of fluoxetine AUC0-¥ fol low ing the ad min is tra tion of test and

ref er ence cap sules.

Source of Sum of df Mean F Sig nif i cance

vari a tion squares square

Be tween groups 117.375 1.0 117.375 0.036 0.851

Within groups 151402.782 46 3291.365

To tal 151520.157 47

Ta ble. 9. Satherwait t-test for un equal vari ances of fluoxetine fol low ing the ad -

min is tra tion of test and ref er ence cap sules.

Pa ram e ter F Sig nif i cance t df

tmax 10.040 0.003 2.298 46

Cmax 0.444 0.509 0.04 46

AUC(0-¥) 0.021 0.884 0.189 46

Ta ble 7. ANOVA of fluoxetine Cmax fol low ing the ad min is tra tion of test and ref -

er ence cap sules.

Source of Sum of df Mean F Sig nif i cance

vari a tion squares square

Be tween groups 0.288 1.0 0.288 0.002 0.968

Within groups 8277.390 46 179.943

To tal 8277.679 47

Ta ble 8. ANOVA of fluoxetine tmax fol low ing the ad min is tra tion of test and ref -

er ence cap sules.

Source of Sum of df Mean F Sig nif i cance

vari a tion squares square

Be tween groups 10.083 1.0 10.083 5.281 0.026

Within groups 87.833 46 1.909

To tal 97.917 47

plies with the FDA limit (0.8 – 1.25). The

per cent age rel a tive bioavailability of the test

for mu la tion com pared to the ref er ence for -

mu la tion cap sules was found to be 98.94% as

de ter mined by the ra tios be tween the

AUC(0-¥) of the test for mu la tion to the ref er -

ence for mu la tion cap sules. We con clude that

the sta tis ti cal anal y sis of Cmax, tmax, and AUC

(0-¥) clearly sup ports the fact that Fluoxicare

cap sule (20 mg fluoxetine) pro duced by Phar -

macare Ltd. (Pal es tine) is bioequivalent to

ref er ence for mu la tion cap sule (20 mg

fluoxetine) pro duced by Eli Lilly Com pany,

Ltd. (UK).

Acknowledgment

Ac knowl edg ments to Phar ma ceu ti cal

Ser vices Cen ter, Uni ver sity of Tanta, Col -

lege of Phar macy, Tanta, Egypt, the US PHS,

Na tional In sti tute on Al co hol Abuse and Al -

co hol ism grants R37-AA07112 and K05-

AA00 219, and the Med i cal Re search Ser -

vice of the US De part ment of Vet er ans Af -

fairs. We thank the anon y mous re viewer

who made help ful com ments on the manu -

script.

References

Altamura AC, Mont gom ery SA, Wernicke JF. The ev i -

dence for 20 mg a day fluoxetine as the op tional dose

in the treat ment of de pres sion. Br J Psy chi a try. 1988;

153 (Suppl): 109-112.

Altamura AC, Mauri MC. As pects of treat ment of el derly

de pres sion: the fluoxetine ex pe ri ence. In: Free man HL

(eds). The uses of fluoxetine in clin i cal prac tice. Lon -

don: Royal So ci ety of Med i cine Press; 1991. pp.

53-59.

Altamura AC, Moro AR, Percudani M. Clin i cal pharma -

cokinetics of fluoxetine. Clin Pharmacokinet. 1994;

26: 201-214.

Aronoff GR, Bergstrom RF, Pottratz ST, Sloan RS, Wolen

RL. Fluoxetine ki net ics and pro tein bind ing in nor mal

and im paired re nal func tion. Clin Pharmacol Ther.

1984; 36: 138-144.

Beasley CM, Masica DM, Provin JH. Fluoxetine: a re -

view of re cep tor and func tional ef fects and their im pli -

ca tions. Psychopharmacology. 1992; 107: 1-10.

Benfield P, Heel PRC, Adlewis SP. Fluoxetine: a re view

of its pharmacodynamic and pharmacokinetic prop er -

ties and ther a peu tic ef fi cacy in de pres sive ill ness.

Drugs. 1986; 32: 481-508.

Bergstrom RF, Lemberger L, Farid NA, Wolen RL. Clin i -

cal phar ma col ogy and pharmacokinetics of fluoxe -

tine: a re view. Br J Psy chi a try. 1988; 153 (Suppl 3):

47-50.

DeVane CL. Pharmacokinetics of the newer an ti de pres -

sants: clin i cal rel e vance. Am J Med. 1994; 19: 13S-23S.

Fuller RW, Wong DT. Se ro to nin re-up take blocker in vi -

tro and in vivo. J Clin Psychopharmacol. 1987; 7:

365-435.

Gourion D, Perrin E, Quintin P. Fluoxetine: an up date

of its use in ma jor de pres sive dis or der in adults.

Encephale. 2004; 30: 392-399.

Lemberger L, Bergstrom RF, Wolen RL, Farid NA, Enas

GG. Fluoxetine: clin i cal phar ma col ogy and phys i o -

logic dis po si tion. J Clin Psy chi a try. 1985; 46: 14-19.

Mandel J. The statistical anal y sis of ex per i men tal data.

New York: Do ver Pub li ca tions; 1984.

Najib NM, Idkaidek N, Adel A, Mo ham med B, Al-Masri

S, Admour I, Alam SM, Dham R, Qumaruzaman S.

Com par i son of two cyclosporine for mu la tions in

healthy Mid dle East ern vol un teers: bioequivalence of

the new Sigmasporin Microoral and Sandimmun

Neoral. Eur J Pharm Biopharm. 2003; 55: 67-70.

Preskorn SH. Clin i cally rel e vant phar ma col ogy of se lec -

tive se ro to nin reuptake in hib i tors. An over view with

em pha sis on pharmacokinetics and ef fects on ox i da -

tive drug me tab o lism. Clin Pharmacokinet. 1997; 32

(Suppl 1): 1-21.

van Harten J. Clin i cal pharmacokinetics of se lec tive se ro -

to nin reuptake in hib i tors. Clin Pharmacokinet. 1993;

23: 203-220.

Zaid, Bowirrat, Kort and Oscar-Berman 10

CPH-5591 / 24.08.2006