formulation of a modified-release pregabalin tablet using hot-melt coating with glyceryl behenate

7232019 Formulation of a modified-release pregabalin tablet using hot-melt coating with glyceryl behenate

httpslidepdfcomreaderfullformulation-of-a-modified-release-pregabalin-tablet-using-hot-melt-coating 18

Formulation

of

a

modi1047297ed-release pregabalin

tablet

using

hot-meltcoating

with

glyceryl

behenate

Kyu Ho Jeongc1 Hye Seung Wooc1 Chae Jin Kimc Kyung Hwa Leec Jun Young JeonaSang Young Leea Jae-Hoon Kanga Sangkil Leeb Young Wook ChoicaResearch Laboratories ILDONG Pharmaceutical Co Ltd Hwaseong-si Gyeonggi-do 445-170 Republic of KoreabCollege of Pharmacy Keimyung University Daegu 704-701 Republic of KoreacDrug Delivery Research Laboratory College of Pharmacy Chung-Ang University Dongjak-gu Seoul 156-756 Republic of Korea

A R T I C L E I N F O

Article history

Received

22

May 2015

Received in revised form 13 August 2015

Accepted 20 August 2015

Available online 24 August 2015

Keywords

Pregabalin

Glyceryl behenate

Hot-melt coating

Drug release

Pharmacokinetic

Stability

A B S T R A C T

A modi1047297ed-release (MR) tablet of the anti-anxiety drug pregabalin (PRE) was prepared by hot-melt

coating PRE with glyceryl behenate (GB) as a release retardant and compressing to form a matrix with

microcrystalline cellulose (MCC) as a hydrophilic diluent GB-coated PRE had a size in the range of 177ndash

290mmwith good to acceptable 1047298owability Tablet hardnessdecreased slightly as GB content increased

PREreleasefromthetabletmatriceswassuccessfullymodi1047297edby altering the ratioof MCC andGB and it

was found that dissolution- or diffusion-controlled release depended on the amount of GB used Drug

release was pH-independent An accelerated stability test on themost promisingMR tablet at 40 C and

75 relative humidity for 6 months showedno signi1047297cant changes in PRE content andthe occurrenceof

total impuritiesmdashincluding PRE-lactammdashwas within acceptable limits After oral administration of the

selected MR tablet or a commercial IR capsule (Lyrica) to healthy human volunteers pharmacokinetic

parameters including T max C max

AUC0ndash24

and T 12 were compared The con1047297dence interval of AUC0ndash24waswithin the adequate range

but that of C max was inadequate This study demonstrated the potential

use of GB for PRE-containing MR formulations

atilde

2015 Elsevier BV All rights reserved

1 Introduction

Pregabalin (PRE S -(3)-amino methyl hexanoic acid) is a

structural analogue of g-aminobutyric acid which is used to treat

refractory partial seizures diabetic neuropathy post-therapeutic

neuralgia and social anxiety disorders Its main site of action

appears to be the a2-d subunit of the voltage-dependent calcium

channels that are widely distributed throughout the peripheral

and central nervous system (Gee et al 1996 Bryans and Wustrow

1999 Bian et al 2006) PRE is a highly soluble and highly

permeable drug categorized according to the biopharmaceutics

classi1047297cation system (BCS) as a class 1 compound PRE has an oralbioavailability (BA) of more than 90 with an average elimination

half life of 63 h and it is excreted unchanged in the urine (French

et al 2003) The absorption of PRE is limited to the upper

small intestine where l-amino transporters that govern PRE

absorption exclusively exist (Su et al 2005 Cundy et al 2004)In

2004 P1047297zer introduced PRE to the market under the brand name

Lyrica as a conventional immediate release (IR)-type capsule with

a recommended dosage regimen of 150ndash600 mg per day divided

into 2 or 3 doses (NDA 21446 2004 Dworkin and Kirkpatrick

2005) Therefore modi1047297ed release (MR)-type dosage forms would

be useful to reduce dosing frequency and improve patient

compliance

Orally administered controlled-release tablets are an attractive

new class of drug delivery system (Chien 1992) Controlled-release

systems are generally classi1047297ed as either monolithic matrix or

reservoir type In monolithic matrix systems the drug is distributedthroughout a polymer matrix Sustained or modi1047297ed drug release is

attained by the use of water-swellable or -erodible matrices

consisting of various polymeric excipients such as hydrophilic

polymers (hypromellose hydroxypropylcellulose sodium alginate

chitosan polyethylene oxide etc) and hydrophobic polymers

(ethylcellulose hypromellose acetate succinate methacrylic acid

co-polymers etc) Alternatively lipids can be used as a release rate-

controlling agent for solid oral dosage forms Lipidic excipients

include long-chain saturated fatty acids andor their partial

glycerides hydrogenated vegetable oils polyoxylglycerides and

Corresponding author at College of Pharmacy Chung-Ang University

221 Heuksuk-dong Dongjak-gu Seoul 156-756 Republic of Korea

Fax +82 2 826 3781

E-mail address ywchoicauackr (YW Choi)1 These authors contributed equally to this work

httpdxdoiorg101016jijpharm201508057

0378-5173atilde 2015 Elsevier BV All rights reserved

International Journal of Pharmaceutics 495 (2015) 1ndash8

Contents

lists

available

at

ScienceDirect

International

Journal

of

Pharmaceutics

j ourna l homepage wwwelsev ier comlocate i jpharm



fatty alcohols (Rosiaux et al 2014 Saraiya and Bolton 1990

Barthelemy et al 1999) The advantages of the lipid matrix system

are that it avoids drug-excipient interactions has pH-independent

drug release properties and has a high correlation between in vitro

and in vivo properties (Schroeder et al 1978) Various methods for

preparing lipid matrices have been proposed including direct

compression wet granulation the hot-melt granulationspray

method coldhot extrusion etc (Obaidat 2001 Mehta 1986

Miyagawa 1996) Sustained-release tablet formulations have been

successfully prepared using the hot-melt granulation method

(Zhang and Schwartz 2003)

Glyceryl behenate (GB) is a widely used lipid excipient which is

synthesized by an esteri1047297cation reaction between glycerol and

behenic acid (C22 fatty acid) It has a melting point of 69ndash74 C and

an HLB value of 2 GB was initially used as a tablet lubricant and

taste-masking agent but has recently found use as a controlled-

release agent Many reports have demonstrated the use of GB to

prolong or modify the release of numerous drugs including

theophylline metoprolol succinate and tramadol (Barthelemy

et al 1999 Roberts et al 2012 Obaidat 2001) Drug release from

lipophilic matrix systems is dependent on several factors such as

solubility dosing contents excipients physical dimensions of the

tablet and the level of matrix-forming agent (Roberts et al 2012)

The release of the drug from the lipid matrix is generally slow andmainly due to diffusion (Obaidat 2001) Drug release can be

controlled by the addition of hydrophilic excipients such as lactose

or microcrystalline cellulose derivatives which generate pores in

the matrix or accelerate disintegration of the matrix by causing

swelling (Obaidat 2001)

In this study in order to develop a once-a-day PRE product

various MR tablets were formulated with different amounts of GB

and microcrystalline cellulose (MCC)mdashas a release retardant and a

hydrophilic diluent respectively The in vitro drug release

characteristics of various MR tablets were investigated and the

stability of the most promising MR tablet was evaluated The

pharmacokinetic (PK) pro1047297les of the selected MR tablet and a

commercial IR product (Lyrica) were studied in healthy human

volunteers

2 Materials and methods

21 Materials

PRE and 4-isobuyl-pyrrolidin-2-one (PRE-lactam) were pur-

chased from TEVA (Bersquoer sheva Israel) GB (trade name of

Compritol1 888 ATO) was purchased from Gateffosseacute (Nanterre

France) MCC was purchased from JRS Pharma (Weissenbern

Germany) The other excipients used to prepare the tablets were of

standard pharmaceutical grade All other reagents were of

analytical grade Commercially available IR-type PRE capsules

(150 mg Lyrica P1047297zer Germany) were used as the reference for the

in

vivo

study

22 Methods

221 Preparation of MR tablets

The compositions of the various PRE-containing MR (PRE-MR)

tablets are shown in Table 1 GB was selected as a lipid coating

material to control the release of PRE and MCC was used as a

directly compressible diluent A combination of colloidal silicon

dioxide talc and magnesium stearate was used as lubricant All

constituents were weighed accurately and a hot-melt coating

technique was adopted for granulation Brie1047298y GB was melted at

80ndash90 C in a vertical granulator (FM-VG-5P Powrex Osaka

Japan) PRE was added while shearing with impeller at 300 rpm

and chopper at 2000 rpm and then the melted mass were cooled

down to 20 C and passed through no 20 mesh The coated

granules were collected with over 97 yield After homogeneous

blending with MCC and lubricant direct compression was

performed on a rotary tablet machine (Pressima IMA-KILAN

Cologne Germany) at compression force of 15kN using 112mm

oval type punches

222 Physical property observation

2221

Property

of

the

hot-melt-coated

granules The shapes of the

particles in the hot-melt-coated granules were observed using anoptical microscope system (Eclipse TE2000-U Nikon Co Tokyo

Japan) and the particle size distribution was measured using a

laser diffraction particle size analyzer (Malvern Mastersizer 2000

Malvern Instruments Ltd Malvern UK) To evaluate the 1047298ow

properties of the prepared granules apparent bulk and tapped bulk

densities were measured by the cylinder method using a powder

tester (ABD-100 Tsutsui Scienti1047297c Instruments Co Ltd Tokyo

Japan) Accurately weighed granule samples were poured into a

cylinder and the volume was measured to obtain the apparent bulk

density separately a sample was tapped 100 times to measure

tapped bulk density

2222

Property

of

PRE-MR

tablets Physical testing of PRE-MR

tablets was performed after a relaxation period of at least 24 hWeight variation tests were performed with 20 individually

weighed tablets using a balance (AB204-SFACT Analytical

Balance Mettler-Toledo Greifensee Switzerland) The thickness

and diameter of 10 tablets were measured individually using

vernier calipers (MN84 Mitutoyo Kawasaki Japan) The crushing

strength was determined using a hardness tester (C50 Holland

Nottingham UK) Tablet friability was calculated as the percentage

of weight loss (4 min 25 rpm 20 tablets) using a friabilator (FAT-

10 FineScienti1047297c Instrument Seoul Korea)

223

Drug

content

determination

by

HPLC

assay

Twenty tablets were weighed individually crushed into a

1047297ne

powder and combined to give a sample containing 300 mg of PRE

The

mobile

phase

was

poured

into

the

1047298ask

and

the

drug

wasextracted for 5 min at 60 Hz using a bath sonicator (8510-DTH

Branson Danbury CT USA) The drug contents were determined by

HPLC assay using a pump (W26905 Waters Milford MA USA)

UV detector (W2489 Waters) and a data station (Empower3

Waters) Chromatographic separation was performed using

cyano silica column (Hypersil BDS cyano 46 150 mm 5 mmThermo Fisher Scienti1047297c San Jose CA USA) at a 1047298ow rate of

10 mLmin Sample was injected and peaks were monitored at

210 nm The isocratic mobile phase was composed of acetonitrile

and buffer solution at a ratio of 1388 (vv) Buffer solution

consisted of sodium hexanesulfonate (941 g) triethylamine

(2 mL) and water (880 mL) with pH adjusted to 31 using

orthophosphoric acid

Table 1

Compositions of PRE-MR tablets

Composition F1 F2 F3 F4 F5 F6

PRE

300

300

300

300

300

300

Glyceryl behenate 0 30 60 100 200 300

Microcrystalline cellulose 2502 2202 1902 1502 502 502

Colloidal silicon dioxide 102 102 102 102 102 102

Talc 198 198 198 198 198 198

Magnesium stearate 198 198 198 198 198 198

Total weight (mg) 6000 6000 6000 6000 6000 7000

2 KH Jeong et al International Journal of Pharmaceutics 495 (2015) 1ndash8



224 In vitro release study

The release of PRE from the PRE-MR tablets was studied using

USP dissolution Apparatus II (VK7025 Varian Palo Alto CA USA)

The dissolution media used (900 mL) were pH 40 acetate buffer

(100 mM) pH 68 phosphate buffer (50 mM) or water The

temperature was maintained at 37 05 C The rotation speed

was 50 rpm After

1047297lling the vessel with every PRE-MR tablet

containing 300 mg PRE aliquots (5 mL) were withdrawn at

predetermined time intervals of 025 05 1 2 4 6 8 and 12 h

The medium was replenished with fresh buffer (5 mL) each time

Samples were analyzed using HPLC as described above

225 Photographic observation

Photographic observation of the MR tablet during dissolution

test was carried out using a digital camera (Model d2h with

35 mm-NIKKOR lens Nikon) Photographs were taken of formu-

lations F2 F3 F4 and F5 after dissolution tests in aqueous media

(initial 1 2 4 and 6 h) Tablets were removed from the medium

then dried in an oven at 40 C for 24 h and then imaged

226 Accelerated stability test

Formulation F3 was selected for stability testing according to

ICH guidelines F3 was stored at 40 2 C and 75 5 relative

humidity (RH) Drug contents and impurities were determinedover 6 months by HPLC assay Chromatographic separation was

performed using ODS column (Hypersil BDS 46 150 mm 5 mmThermo Fisher Scienti1047297c) at a

1047298ow rate of 10 mLmin Sample was

injected and peaks were monitored at 210 nm A gradient elution

was performed changing from mobile phase A to mobile phase B

Mobile phase A consisted of buffer methanol and acetonitrile

solution at a volume ratio of 801010 Buffer solution consisted of

aqueous diammonium phosphate (528 gL) with pH adjusted to

65 using orthophosphoric acid Mobile phase B was acetonitrilendash

water solution at a volume ratio of 9010 Mobile phase

composition was 100 mobile phase A for 6 min followed by a

linear increase of mobile phase B to 35 over 50 min and returning

to 100 mobile phase A for 5 min

227 In vivo PK study

2271

Human

subjects A human PK study was approved by

Ethical Committee (Protocol no ID-PRSD-1201 Seoul St Maryrsquos

Hospital Catholic University Korea) and the subjects gave written

informed consent to participate in the study Twenty-eight

subjects (age 251 32 years height 176 47 cm body weight

693 75 kg and within 20 of their ideal body weight) had no

history of cardiovascular renal or hepatic disease drug allergies

or hypersensitivity All subjects refrained from the consumption of

alcoholic beverages xanthine-containing foods smoking and

other drugs for at least 1 week prior to the study and until the

1047297nal

blood sampling

2272

Oral

administration

and

plasma

sampling A randomized

open-label 2-way crossover clinical trial was performed to

compare the PK pro1047297les of PRE in healthy male subjects after

oral administrations of either the chosen MR tablet or an IR-type

reference capsule The volunteers were fasted overnight at least

10 h prior to dosing with water and were served two low fat meals

(total 700 kcal with 20 of the calories from fat) at 45 h and 9 h

after dosing The reference (PRE 150 mg) was given twice a day

(every 12 h) and blood samples were collected at predetermined

time points (0 033 067 1 15 2 4 6 12 1233 1267 13 135 14

16 and 24 h) The MR tablet (PRE 300 mg) was given once a day and

blood samples were withdrawn at predetermined time points

(0 05 1 2 3 4 5 6 8 10 12 14 and 24 h) The collected blood

samples

were

centrifuged

immediately

at

4000

rpm

and

the

plasma was stored in light-protected container at 20 C until the

time of analysis

2273 PRE determination in plasma samples by LC ndashMSMS The

concentrations of PRE in human plasma were determined by

HPLCndashMSMS A high-performance liquid chromatography system

(UPLC I-Class system Waters) was used for analysis

Chromatographic separation was performed using a C18 column

(Unison UK-C18 20 75 mm 3 mm Imtakt Kyoto Japan) with a

mobile phase consisting of 01 formic acid containing 10 mM

ammonium acetate buffer and acetonitrile (8515 vv) at a

1047298ow

rate of 05 mLmin The LC system was coupled with a

TurboIonSpray ionization-triple quadrupole mass spectrometer

API 4000 (AB Sciex Instruments Toronto Canada) for detection

Multiple-reaction monitoring (MRM) mode with positive ion was

used for quanti1047297cation PRE 1602 gt 1421 and gabapentin

1723 gt 1372 as an internal standard (IS) The cone voltage

collision energy and dwell time were 18 V 10eV and 0025 s

respectively

A stock solution of PRE at 1 mgmL in puri1047297ed water and IS

solution (500 ngmL gabapentin in 100 methanol) were sepa-

rately prepared Plasma calibration solution was prepared at PRE

concentrations of 005 01 05 1 5 10 and 20 mgmL The

calibration curve was peak area ratio of analyteIS versusconcentration using a weighed least-square linear regression

The linear response of this method was 005ndash20 mgmL with a

coef 1047297cient of determination (r 2) value of greater than 099 Assay

method was validated for linearity accuracy precisions and

extraction ef 1047297ciency In sample preparation an aliquot of each

plasma sample (20 mL) was pipetted into an Eppendorf tube and

deproteinated with methanol by the addition of IS solution

(200 mL) The mixture was vortexed for 3 min and centrifuged at

12000 rpm for 5 min A portion of the upper layer (2 mL) was

injected into the HPLCndashMSMS system

2274 PK analysis Data analysis was performed using Phoenix

WinNonlin version 63 (Pharsight Corporation Apex NC USA)

Maximum plasma concentration (C max) and time to reach themaximum plasma concentration (T max) were determined directly

from the obtained concentrationndashtime data The areas under the

plasma drug concentrationndashtime curves from 0 to 24 h (AUC0ndash24)

and in1047297nity (AUC01) were calculated using a linear trapezoidal

rule All PK parameters were calculated based on a non-

compartmental model The bioequivalence result was calculated

as the ratio of the

C max and AUC0ndash24of the selected MR tablet (F3) to

the reference product

228 Statistical analysis

All data are expressed as mean standard deviation (SD)

Statistical signi1047297cance was checked by Studentrsquos

t -test and was

considered to be signi1047297cant at P lt 005 unless otherwise indicated

3

Results

and

discussion

31 Formulation characteristics

Particle shapes and

1047298owability are shown in Fig 1 PRE in its

unprocessed state was crystalline and particles had a geometric

mean diameter of 75678 mm As the GB content increased the GB-

coated areamdashindicated by a darkened surfacemdashincreased as did

particle size The Hausner ratio was calculated as the ratio of

tapped density versus apparent density and was between 111 and

134 which was judged according to USP guidelines as a good to

passable grade of 1047298owability

Physical properties of prepared PRE-MR tablets were evaluated

for

their

drug

content

weight

variation

hardness

friability

and

KH Jeong et al International Journal of Pharmaceutics 495 (2015) 1ndash8 3



thickness (Table 2) Drug contents were 996ndash1015 Weight

variation was within 2 of the total amount Hardness and

percentage friability were found 79ndash132kgcm2 and 016ndash025

respectively Tablet hardness was slightly decreased as GB content

increased In other words MCC content greatly in1047298uenced the

hardness variation due to its good compressibility even at lower

compaction forces Weight losses of less than 1 in the friability

test are generally acceptable (Reddy et al 2003) The advantages of

the hot-melt coating method used in this study are its simplicity

ease of use and practicality as well as the homogeneous

distribution and high loading of drug in the tablet matrix

32 Dissolution behavior

The in vitro release pro1047297les of PRE-MR tablets in water are

shown in Fig 2 Formulation F1 released 980 of PRE within 1 h

but F2 F3 and F4 took 4 6 and 8 h respectively to release almost

all PRE F5 and F6 released about 78 and 54 respectively at 12 h

In lipid matrices drug release tends to be slower at increasing

amounts of lipid excipient (Zhang and Schwartz 2000) This

behavior was consistent with earlier reports which demonstrated

that the release of theophylline phenylpropanolamine and

felodipine was ef 1047297ciently controlled by the lipid excipient (Zhang

and Schwartz 2003 Savolainen et al 2002) Meanwhile using

more MCC as a hydrophilic excipient resulted in rapid drug release

in the early stages of dissolution F2 and F3 showed faster release

than the other formulations In addition photographs of tablets

F2ndashF5 were taken during dissolution at 1 2 4 and 6 h As shown in

Fig 3 F2 and F3 were disintegrated or partially disrupted within

1 h but F4 and F5 maintained their initial shape even after 4 h As

the GB content increased tablets were less prone to disintegration

Thus we suggest that a crucial factor for disintegration is not the

tablet hardness but the MCC content in the matrix because of the

water-soluble property of MCC

Table 2

Physical properties of PRE-MR tablets

Formulation Drug content () Weight variation () Hardness (kgcm2) Friability () Thickness (mm)

F1

1013

11

08

03

132

25

017

008

586

009

F2 997 19 05 03 127 14 016 014 588 014

F3 1011 13 06 02 111 18 023 017 595 007

F4 996 08 05 03 107 15 025 012 593 013

F5 1009 17 05 03 98 17 024 015 597 011

F6 1015 20 07 04 79 15 018 013 651 012

Values represent means SD (n = 3)

Fig 2 Dissolution pro1047297les of PRE from different MR tablets in water Values

represent

means

SD

(n

=

6)

Fig

3

Photographs

of

various

MR

tablets

during

the

dissolution

test

Fig 1 Optical micrographs (40) of hot melt-coated PRE with different amounts of GB d05 and HR represent the geometric mean diameter and the Hausner ratio

respectively




The effect of GBMCC ratio on PRE release was investigated in

terms of early (1 h) and late (6 h) dissolution (Fig 4) In both stagesPRE release was proportionally decreased by increasing the GB

fraction in other words MCC increased the release of PRE During

the early stage of dissolution PRE release decreased rapidly up to

24 GB content (F3 tablet) and decreased gradually as GB was

increased further In contrast PRE release during late stage

dissolution decreased gradually up to 80 GB content (F5 tablet)

and then decreased rapidly as GB was increased further Thus PRE

release from the tablet matrices could be controlled by changing

the ratio of hydrophilic polymer to lipid excipient

Furthermore to observe the in1047298uence of dissolution media on

drug release the F3 tablet was tested in media of pH 4 and 68

and water As shown in Fig 5 the dissolution patterns were quite

similar regardless of the medium In addition drug release from

the reference product was independent of dissolution mediaeven though there was a sharp increase in earlier stage Thus

PRE release was considered as pH-independent This might

be attributed to the high solubility of PRE which is a BCS class I

drug

33

Drug

release

kinetics

In order to examine the mechanism of PRE release from MR

tablets (F2ndashF6) the results of the dissolution experiments were 1047297t

to the 1047297rst-order kinetic model (Eq (1)) the HixsonndashCrowell

equation (Eq (2)) and the PeppasndashKorsmeyer equation (Eq (3)) as

follows (Bourne 2002 Hixson and Crowell 1931 Korsmeyer et al

1983)

ln Q frac14 ln Q 0 K 1t eth1THORN

Q 1=30 Q 1=3 frac14 K HCt eth2THORN

M t M 1

frac14

K PKt n eth3THORN

where Q is the cumulative amount of drug released at time t Q 0 is

the initial amount of drug in the matrix M t M 1 is the fraction of

drug released at time

t n is the release exponent and

K 1

K HC and

K PK are the rate constants for 1047297rst-order HixsonndashCrowell and

PeppasndashKorsmeyer equations respectively Data obtained from

the mathematical equations are listed in Table 3 The F2 tablet

which contained a lower amount of GB was found to be different

from the other formulations It was quickly eroded released the

drug rapidly and correlated poorly with all three kinetic models

In

the

case

of

the

F3

tablet

coef 1047297

cient

of

determination

(r

2

)

for

the1047297rst-order HixsonndashCrowell and PeppasndashKorsmeyer models were

09958 09368 and 09356 respectively Thus the 1047297rst-order

equation provided the best 1047297t even though the other models

also showed relatively good correlations Dissolution-controlled

release can be obtained by incorporating a water-soluble drug into

a hydrophobic matrix such as wax polypropylene or ethyl

cellulose the rate-limiting step for dissolution of a drug is

diffusion across an aqueous boundary layer The rate of drug

release is controlled by the rate of penetration of the dissolution

1047298uid into the matrix in which aqueous front formation depends

on the compressed structure In the case of F4ndashF6 which

contained more GB than F3 the PeppasndashKorsmeyer model was

the best-1047297tting equation The n values of F5 and F6 were

05096 and 04912 respectively indicating that drug release

was governed by Fickian diffusion (Costa and Sousa Lobo 2001)

Unlike the dissolution-controlled systems drug release was

dependent of the rate of drug diffusion through the matrix and

not on the rate of solid dissolution

34 Selection of the optimized formulation

As the number of absorption sites for PRE is very limited in the

upper small intestine extended release over a longer period than

the gut transit time would be unnecessary for in vivo oral

administration Although the gastrointestinal transit time in

human depends on gender age body weight food intake and the

dosage form it has been estimated to be approximately 6 h

(Coupe et al 1991) The transit time in the small intestine was

around 3 h in average (Yu et al 1996) Therefore we screenedF3 and F4 tablets as candidates for the

1047297nal formulation based

Fig 4 Effect of GBMCC ratio on PRE release during early (1 h closed circle) and late

(6 h open circle) stages of dissolution

Fig 5 pH-independent dissolution pro1047297les of PRE from F3 tablet and reference

capsule

Values

represent

mean

SD

(n

=

6)

Table 3

Kinetic parameters of PRE release in water

Formulation First-order HixsonndashCrowell PeppasndashKorsmeyer

K 1 r 2 K HC r 2 K PK r 2 n

F2 076 09439 029 07072 6401 08194 01142

F3 051 09958 034 09368 5084 09356 03096

F4 032 09774 042 09559 4023 09917 03914

F5 012 09851 015 09668 2341 09954 05096

F6 006 09526 009 09364 1561 09976 04912

K 1K HC and K PK are the rate constants for the respective model r 2 coef 1047297cient of

determination

n

release

exponent




on the self-imposed requirement of more than two-thirds PRE

release by 6 h and complete dissolution within 12 h As shown

earlier the release from F2 was too fast and the release from

F5 and F6 was too slow and the formulation was incompletely

dissolved In the preliminary PK studies in dogs the plasma PRE

level after F4 tablet administration was not high enoughmdashunlike

the F3 tablet (data not shown)mdashpossibly due to differences in

early stage dissolution Ultimately the F3 tablet was selected as

the optimized formulation and subjected to further evaluation

concerning stability and PK

35 Accelerated stability assessment

PRE is known to form conjugates with lactoseby undergoing a

Maillard reaction and an Amadori rearrangement (Lovdahl et al

2002) which leads to the formation of various lactose con-

jugatesmdashincluding PRE lactammdashas degradation products The

Maillard reaction is a condensation reaction between lactose and

primary amine that produces a simple glycosylamine which

readily undergoes an Amadori rearrangement to create further

byproducts Since MCC is a polysaccharide consisting of a linear

chain of several hundred to many thousands of b(1 4) linked D-

glucose units similar degradation pathways could occur in this

formulation The major impurity of PRE-lactam has been shown

to cause seizures in animal models (Potschka et al 2000) A

simultaneous assay for PRE PRE-lactam and other unknownimpurities was successfully carried out with a good resolution

using this gradient HPLC condition (Fig 6) PRE and PRE lactam

were separated with a retention time of 61 min and 325 min

respectively and other peaks were considered to be unidenti1047297ed

impurities

An accelerated stability test was carried out according to ICH

guidelines at 40 C and 75 RH and no signi1047297cant changes in PRE

content were observed The drug content was found to be more

than 995 at the end of 6 months even though there was a slight

increase in the amount of PRE-lactam as the main degradation

product As shown in Table 4 the levels of PRE lactam and total

impurities were within acceptable limits

36 PK characteristics in human subjects

In order to 1047297nd the formulationrsquos effect on PK pro1047297les the

optimized MR tablet (F3) and an IR-type reference capsule were

administered orally to human volunteers The plasma level of PRE

was measured and plotted against time (Fig 7) The mean plasma

PRE level of F3 was signi1047297cantly higher than that of the reference

In both F3 and the reference however PRE was rapidly absorbed

Fig 6 HPLC chromatogram showing PRE (RT 612) PRE-lactam (RT 3255) and unidenti1047297ed impurity peaks Inset shows the degradation of PRE to PRE-lactam

Table 4

Changes in PRE content and impurity formation during accelerated stability test

Time period

(months)

PRE

contents

()

PRE-

lactam

()

Unknown

impurity

()

Total impurities

()

0 1011 14 003 005 008

3 997 13 003 011 014

6 996 17 005 036 041

Values represent mean SD (n = 3)

Acceptance criteria PRE contents 95ndash105 PRE-lactam lt05 unknown

impurities

lt02

each

total

impurities

lt10

Fig 7 Plasma concentrationndashtime pro1047297les of PRE after oral administration of single

dose of F3 tablet (PRE 300 mg) or two-consecutive doses of the reference capsule

(PRE 150 mg Lyrica) to healthy subjects under fasted condition Values represent

mean

SD

(n

=

28)






Saraiya D Bolton S 1990 The use of Precirol1 to prepare sustained release tabletsof theophylline and quinidine gluconate Drug Dev Ind Pharm 16 1963ndash1969

Savolainen M Khoo C Glad H Dahlqvist C Juppo AM 2002 Evaluation of controlled-release polar lipid microparticles Int J Pharm 244 151ndash161

Schroeder HG Dakkuri A DeLuca PP 1978 Sustained release from inert waxmatrixes I drugndashwax combinations J Pharm Sci 67 350ndash353

Su TZ Feng MR Weber ML 2005 Mediation of highly concentrative uptake of pregabalin by L -type amino acid transport in Chinese hamster ovary and Caco-2 cells

J

Pharmacol

Exp

Ther

313

1406ndash1415

Yu LX Crison JR Amidon GL 1996 Compartmental transit and dispersionmodel analysis of small intestinal transit 1047298ow in humans Int J Pharm 140 111ndash118

Zhang YE Schwartz JB 2000 Effect of diluents on tablet integrity and controlleddrug release Drug Dev Ind Pharm 26 761ndash765

Zhang YE Schwartz JB 2003 Melt granulation and heat treatment for waxmatrix-controlled drug release Drug Dev Ind Pharm 29 131ndash138




fatty alcohols (Rosiaux et al 2014 Saraiya and Bolton 1990

Barthelemy et al 1999) The advantages of the lipid matrix system

are that it avoids drug-excipient interactions has pH-independent

drug release properties and has a high correlation between in vitro

and in vivo properties (Schroeder et al 1978) Various methods for

preparing lipid matrices have been proposed including direct

compression wet granulation the hot-melt granulationspray

method coldhot extrusion etc (Obaidat 2001 Mehta 1986

Miyagawa 1996) Sustained-release tablet formulations have been

successfully prepared using the hot-melt granulation method

(Zhang and Schwartz 2003)

Glyceryl behenate (GB) is a widely used lipid excipient which is

synthesized by an esteri1047297cation reaction between glycerol and

behenic acid (C22 fatty acid) It has a melting point of 69ndash74 C and

an HLB value of 2 GB was initially used as a tablet lubricant and

taste-masking agent but has recently found use as a controlled-

release agent Many reports have demonstrated the use of GB to

prolong or modify the release of numerous drugs including

theophylline metoprolol succinate and tramadol (Barthelemy

et al 1999 Roberts et al 2012 Obaidat 2001) Drug release from

lipophilic matrix systems is dependent on several factors such as

solubility dosing contents excipients physical dimensions of the

tablet and the level of matrix-forming agent (Roberts et al 2012)

The release of the drug from the lipid matrix is generally slow andmainly due to diffusion (Obaidat 2001) Drug release can be

controlled by the addition of hydrophilic excipients such as lactose

or microcrystalline cellulose derivatives which generate pores in

the matrix or accelerate disintegration of the matrix by causing

swelling (Obaidat 2001)

In this study in order to develop a once-a-day PRE product

various MR tablets were formulated with different amounts of GB

and microcrystalline cellulose (MCC)mdashas a release retardant and a

hydrophilic diluent respectively The in vitro drug release

characteristics of various MR tablets were investigated and the

stability of the most promising MR tablet was evaluated The

pharmacokinetic (PK) pro1047297les of the selected MR tablet and a

commercial IR product (Lyrica) were studied in healthy human

volunteers

2 Materials and methods

21 Materials

PRE and 4-isobuyl-pyrrolidin-2-one (PRE-lactam) were pur-

chased from TEVA (Bersquoer sheva Israel) GB (trade name of

Compritol1 888 ATO) was purchased from Gateffosseacute (Nanterre

France) MCC was purchased from JRS Pharma (Weissenbern

Germany) The other excipients used to prepare the tablets were of

standard pharmaceutical grade All other reagents were of

analytical grade Commercially available IR-type PRE capsules

(150 mg Lyrica P1047297zer Germany) were used as the reference for the

in

vivo

study

22 Methods

221 Preparation of MR tablets

The compositions of the various PRE-containing MR (PRE-MR)

tablets are shown in Table 1 GB was selected as a lipid coating

material to control the release of PRE and MCC was used as a

directly compressible diluent A combination of colloidal silicon

dioxide talc and magnesium stearate was used as lubricant All

constituents were weighed accurately and a hot-melt coating

technique was adopted for granulation Brie1047298y GB was melted at

80ndash90 C in a vertical granulator (FM-VG-5P Powrex Osaka

Japan) PRE was added while shearing with impeller at 300 rpm

and chopper at 2000 rpm and then the melted mass were cooled

down to 20 C and passed through no 20 mesh The coated

granules were collected with over 97 yield After homogeneous

blending with MCC and lubricant direct compression was

performed on a rotary tablet machine (Pressima IMA-KILAN

Cologne Germany) at compression force of 15kN using 112mm

oval type punches

222 Physical property observation

2221

Property

of

the

hot-melt-coated

granules The shapes of the

particles in the hot-melt-coated granules were observed using anoptical microscope system (Eclipse TE2000-U Nikon Co Tokyo

Japan) and the particle size distribution was measured using a

laser diffraction particle size analyzer (Malvern Mastersizer 2000

Malvern Instruments Ltd Malvern UK) To evaluate the 1047298ow

properties of the prepared granules apparent bulk and tapped bulk

densities were measured by the cylinder method using a powder

tester (ABD-100 Tsutsui Scienti1047297c Instruments Co Ltd Tokyo

Japan) Accurately weighed granule samples were poured into a

cylinder and the volume was measured to obtain the apparent bulk

density separately a sample was tapped 100 times to measure

tapped bulk density

2222

Property

of

PRE-MR

tablets Physical testing of PRE-MR

tablets was performed after a relaxation period of at least 24 hWeight variation tests were performed with 20 individually

weighed tablets using a balance (AB204-SFACT Analytical

Balance Mettler-Toledo Greifensee Switzerland) The thickness

and diameter of 10 tablets were measured individually using

vernier calipers (MN84 Mitutoyo Kawasaki Japan) The crushing

strength was determined using a hardness tester (C50 Holland

Nottingham UK) Tablet friability was calculated as the percentage

of weight loss (4 min 25 rpm 20 tablets) using a friabilator (FAT-

10 FineScienti1047297c Instrument Seoul Korea)

223

Drug

content

determination

by

HPLC

assay

Twenty tablets were weighed individually crushed into a

1047297ne

powder and combined to give a sample containing 300 mg of PRE

The

mobile

phase

was

poured

into

the

1047298ask

and

the

drug

wasextracted for 5 min at 60 Hz using a bath sonicator (8510-DTH

Branson Danbury CT USA) The drug contents were determined by

HPLC assay using a pump (W26905 Waters Milford MA USA)

UV detector (W2489 Waters) and a data station (Empower3

Waters) Chromatographic separation was performed using

cyano silica column (Hypersil BDS cyano 46 150 mm 5 mmThermo Fisher Scienti1047297c San Jose CA USA) at a 1047298ow rate of

10 mLmin Sample was injected and peaks were monitored at

210 nm The isocratic mobile phase was composed of acetonitrile

and buffer solution at a ratio of 1388 (vv) Buffer solution

consisted of sodium hexanesulfonate (941 g) triethylamine

(2 mL) and water (880 mL) with pH adjusted to 31 using

orthophosphoric acid

Table 1

Compositions of PRE-MR tablets

Composition F1 F2 F3 F4 F5 F6

PRE

300

300

300

300

300

300

Glyceryl behenate 0 30 60 100 200 300

Microcrystalline cellulose 2502 2202 1902 1502 502 502

Colloidal silicon dioxide 102 102 102 102 102 102

Talc 198 198 198 198 198 198

Magnesium stearate 198 198 198 198 198 198

Total weight (mg) 6000 6000 6000 6000 6000 7000










was 50 rpm After






























2271

Human











1047297nal

blood sampling

2272

Oral

administration

and

plasma










time points (0 033 067 1 15 2 4 6 12 1233 1267 13 135 14




samples

were

centrifuged

immediately

at

4000

rpm

and

the


time of analysis









1047298ow








respectively























as the ratio of the






t -test and was


3

Results

and

discussion


Particle shapes and










for

their

drug

content

weight

variation

hardness

friability

and




































Table 2



F1

1013

11

08

03

132

25

017

008

586

009

F2 997 19 05 03 127 14 016 014 588 014

F3 1011 13 06 02 111 18 023 017 595 007

F4 996 08 05 03 107 15 025 012 593 013

F5 1009 17 05 03 98 17 024 015 597 011

F6 1015 20 07 04 79 15 018 013 651 012



represent

means

SD

(n

=

6)

Fig

3

Photographs

of

various

MR

tablets

during

the

dissolution

test


respectively





















drug

33

Drug

release

kinetics






1983)



M t M 1

frac14

K PKt n eth3THORN





K 1

K HC and







In

the

case

of

the

F3

tablet

coef 1047297

cient

of

determination

(r

2

)

for
































capsule

Values

represent

mean

SD

(n

=

6)

Table 3




F2 076 09439 029 07072 6401 08194 01142

F3 051 09958 034 09368 5084 09356 03096

F4 032 09774 042 09559 4023 09917 03914

F5 012 09851 015 09668 2341 09954 05096

F6 006 09526 009 09364 1561 09976 04912


determination

n

release

exponent































impurities
















Table 4


Time period

(months)

PRE

contents

()

PRE-

lactam

()

Unknown

impurity

()

Total impurities

()

0 1011 14 003 005 008

3 997 13 003 011 014

6 996 17 005 036 041



impurities

lt02

each

total

impurities

lt10




mean

SD

(n

=

28)










J

Pharmacol

Exp

Ther

313

1406ndash1415













was 50 rpm After






























2271

Human











1047297nal

blood sampling

2272

Oral

administration

and

plasma










time points (0 033 067 1 15 2 4 6 12 1233 1267 13 135 14




samples

were

centrifuged

immediately

at

4000

rpm

and

the


time of analysis









1047298ow








respectively























as the ratio of the






t -test and was


3

Results

and

discussion


Particle shapes and










for

their

drug

content

weight

variation

hardness

friability

and




































Table 2



F1

1013

11

08

03

132

25

017

008

586

009

F2 997 19 05 03 127 14 016 014 588 014

F3 1011 13 06 02 111 18 023 017 595 007

F4 996 08 05 03 107 15 025 012 593 013

F5 1009 17 05 03 98 17 024 015 597 011

F6 1015 20 07 04 79 15 018 013 651 012



represent

means

SD

(n

=

6)

Fig

3

Photographs

of

various

MR

tablets

during

the

dissolution

test


respectively





















drug

33

Drug

release

kinetics






1983)



M t M 1

frac14

K PKt n eth3THORN





K 1

K HC and







In

the

case

of

the

F3

tablet

coef 1047297

cient

of

determination

(r

2

)

for
































capsule

Values

represent

mean

SD

(n

=

6)

Table 3




F2 076 09439 029 07072 6401 08194 01142

F3 051 09958 034 09368 5084 09356 03096

F4 032 09774 042 09559 4023 09917 03914

F5 012 09851 015 09668 2341 09954 05096

F6 006 09526 009 09364 1561 09976 04912


determination

n

release

exponent































impurities
















Table 4


Time period

(months)

PRE

contents

()

PRE-

lactam

()

Unknown

impurity

()

Total impurities

()

0 1011 14 003 005 008

3 997 13 003 011 014

6 996 17 005 036 041



impurities

lt02

each

total

impurities

lt10




mean

SD

(n

=

28)










J

Pharmacol

Exp

Ther

313

1406ndash1415







































Table 2



F1

1013

11

08

03

132

25

017

008

586

009

F2 997 19 05 03 127 14 016 014 588 014

F3 1011 13 06 02 111 18 023 017 595 007

F4 996 08 05 03 107 15 025 012 593 013

F5 1009 17 05 03 98 17 024 015 597 011

F6 1015 20 07 04 79 15 018 013 651 012



represent

means

SD

(n

=

6)

Fig

3

Photographs

of

various

MR

tablets

during

the

dissolution

test


respectively





















drug

33

Drug

release

kinetics






1983)



M t M 1

frac14

K PKt n eth3THORN





K 1

K HC and







In

the

case

of

the

F3

tablet

coef 1047297

cient

of

determination

(r

2

)

for
































capsule

Values

represent

mean

SD

(n

=

6)

Table 3




F2 076 09439 029 07072 6401 08194 01142

F3 051 09958 034 09368 5084 09356 03096

F4 032 09774 042 09559 4023 09917 03914

F5 012 09851 015 09668 2341 09954 05096

F6 006 09526 009 09364 1561 09976 04912


determination

n

release

exponent































impurities
















Table 4


Time period

(months)

PRE

contents

()

PRE-

lactam

()

Unknown

impurity

()

Total impurities

()

0 1011 14 003 005 008

3 997 13 003 011 014

6 996 17 005 036 041



impurities

lt02

each

total

impurities

lt10




mean

SD

(n

=

28)










J

Pharmacol

Exp

Ther

313

1406ndash1415
























drug

33

Drug

release

kinetics






1983)



M t M 1

frac14

K PKt n eth3THORN





K 1

K HC and







In

the

case

of

the

F3

tablet

coef 1047297

cient

of

determination

(r

2

)

for
































capsule

Values

represent

mean

SD

(n

=

6)

Table 3




F2 076 09439 029 07072 6401 08194 01142

F3 051 09958 034 09368 5084 09356 03096

F4 032 09774 042 09559 4023 09917 03914

F5 012 09851 015 09668 2341 09954 05096

F6 006 09526 009 09364 1561 09976 04912


determination

n

release

exponent































impurities
















Table 4


Time period

(months)

PRE

contents

()

PRE-

lactam

()

Unknown

impurity

()

Total impurities

()

0 1011 14 003 005 008

3 997 13 003 011 014

6 996 17 005 036 041



impurities

lt02

each

total

impurities

lt10




mean

SD

(n

=

28)










J

Pharmacol

Exp

Ther

313

1406ndash1415


































impurities
















Table 4


Time period

(months)

PRE

contents

()

PRE-

lactam

()

Unknown

impurity

()

Total impurities

()

0 1011 14 003 005 008

3 997 13 003 011 014

6 996 17 005 036 041



impurities

lt02

each

total

impurities

lt10




mean

SD

(n

=

28)










J

Pharmacol

Exp

Ther

313

1406ndash1415













J

Pharmacol

Exp

Ther

313

1406ndash1415





formulation of a modified-release pregabalin tablet using hot-melt coating with glyceryl behenate

Documents