The Influence of Excipients on Drug Release from
`Hydroxypropyl Methylcellulose Matrices
`
`MARINA LEVINA, ALI R. RAJABI-SIAHBOOMI
`
`Colorcon Limited, Flagship House, Victory Way, Crossways, Dartford, Kent DA2 6QD, UK
`
`Received 29 October 2003; revised 26 March 2004; accepted 7 June 2004
`
`Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/jps.20181
`
`ABSTRACT: The influence of commonly used excipients, spray-dried lactose (SDL),
`microcrystalline cellulose (MCC), and partially pregelatinized maize starch (Starch
`15001) on drug release from hydroxypropyl methylcellulose (HPMC, hypromellose)
`matrix system has been investigated. A model formulation contained 30%w/w drug,
`20%w/w HPMC, 0.5%w/w fumed silica, 0.25%w/w magnesium stearate, and 49.25%w/w
`filler. Chlorpheniramine maleate and theophylline were used as freely (1 in 4) and
`slightly (1 in 120) water-soluble drugs, respectively. It was found that for both drugs,
`addition of 20 to 49.25%w/w Starch 1500 resulted in a significant reduction in drug
`release rates compared to when MCC or SDL was used. The study showed that using
`lactose or microcrystalline cellulose in the formulations resulted in faster drug release
`profiles. Partially pregelatinized maize starch contributed to retardation of both soluble
`and slightly soluble drugs. This effect may be imparted through synergistic interactions
`between Starch 1500 and HPMC and the filler actively forming an integral part within
`the HPMC gel structure. ß 2004 Wiley-Liss, Inc. and the American Pharmacists Association J
`Pharm Sci 93:2746–2754, 2004
`Keywords: hypromellose; HPMC; Starch 1500; sustained release; pregelatinized
`starch; matrix system
`
`INTRODUCTION
`
`Nonionic cellulose ethers, and most frequently
`hydroxypropyl methylcellulose (HPMC, hypro-
`mellose) have been widely studied for their applic-
`ations in oral sustained release (SR) systems.1
`When in contact with water, HPMC hydrates
`rapidly and forms a gelatinous barrier layer
`around the tablet. The rate of drug release from
`HPMC matrix is dependent on various factors
`such as type of polymer, drug, polymer/drug ratio,
`particle size of drug and polymer, and the type
`and amount of fillers used in the formulation.
`Starch is one of the most widely used excipients
`in the manufacture of solid dosage forms. Most
`native starches consist of two polymers of glucose,
`
`Correspondence to: Marina Levina (Telephone: þ44 (0)
`1322 293000; Fax: þ44 (0) 1322 627200;
`E-mail: mlevina@colorcon.co.uk)
`
`Journal of Pharmaceutical Sciences, Vol. 93, 2746–2754 (2004)
`ß 2004 Wiley-Liss, Inc. and the American Pharmacists Association
`
`that is, branched amylopectin and essentially
`linear amylose. Physically or chemically modified
`starches have been used in sustained release
`tablets because of their cold water-swelling capa-
`city and gel barrier formation. Rak et al.2 and Van
`Aerde and Remon3 studied the possibility of using
`thermally modified starches for controled drug
`release. Herman and Remon4 found that only fully
`pregelatinized starches containing a low amount
`of amylose (25% and lower) could produce a strong
`enough gel
`layer to ensure a sustained drug
`release. These findings are in agreement with
`Michailova et al.,5 who claimed that the amylose
`molecules decrease the gel cohesion and accelerate
`the erosion of the gel layer. Mulhbacher et al.6
`studied crosslinked high amylose starch deriva-
`tives as matrices for controlled release of high drug
`loadings. They found that these polymeric excipi-
`ents are able to control the release over 20 h from
`tablets loaded with 20 to 60% drug. Lenaerts et al.7
`used crosslinked high amylose starch for the
`
`2746
`
`JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 93, NO. 11, NOVEMBER 2004
`
`Par Pharm., Inc.
`Exhibit 1044
`Page 001
`
`

`

`THE INFLUENCE OF EXCIPIENTS ON DRUG RELEASE FROM HPMC MATRICES
`
`2747
`
`preparation of sustained release matrix tablets.
`They claimed the possibility for high active ingre-
`dient core loading and achieving either zero-order
`or Fickian release for most drugs. Other advan-
`tages of crosslinked high amylose starches may be
`the absence of erosion, limited swelling and the
`fact that increasing degree of crosslinking results
`in increased water uptake rate, drug release rate,
`and equilibrium swelling.7
`Partially pregelatinized maize starches are
`normally used as binder-disintegrants in immedi-
`ate release tablet formulations.8 Leach et al.9
`claimed that these materials have a very limited
`obstructive gel formation capability at the surface
`of the tablet, which makes them not particularly
`suitable for SR applications. However, the use of
`partially pregelatinized starches in combination
`with other polymers, such as hypromellose, in SR
`tablets have not been fully examined. Therefore,
`the influence of Starch 1500, in comparison to
`MCC and SDL, on drug release from HPMC 2208
`has been investigated in this study.
`
`EXPERIMENTAL
`
`Materials
`
`Chlorpheniramine maleate (CPM) was obtained
`from Avocado Research Chemicals Ltd. (Lancas.,
`UK), theophylline (TP) was obtained from Knoll
`AG (Ludwigshafen, Germany), and were used at
`30%w/w in the formulation. Aqueous solubility
`for CPM is 1 in 4 (w/w), and for theophylline is 1 in
`120 (w/w).
`To study the effect of fillers on drug release, in all
`formulations
`only
`20%w/w
`hydroxy-
`propyl methylcellulose (HPMC, hypromellose)
`(Methocel1 K4M, Dow Chemical Co., USA) was
`used. Higher HPMC levels may mask the differ-
`ences impacted by the fillers on drug release.
`Three commonly used fillers were studied:
`partially pregelatinized maize starch (PPS)
`(Starch 15001, Colorcon, Dartford, UK), spray
`dried lactose (Fast Flo1 #316, Foremost Farms,
`Wisconsin) and microcrystalline cellulose (MCC)
`(Avicel1 PH102, FMC, Brussels, Belgium). Aver-
`age particle size for Starch 1500 is 70, for MCC—
`90, and for spray dried lactose—100 microns. This
`relatively large particle size for all three materials
`can guarantee good powder flow in direct compres-
`sion applications.
`Fumed silica (Aerosil1 A-200, Degussa AG,
`Dusseldorf, Germany) was used at 0.5%w/w level
`as a flow aid and magnesium stearate (Peter
`
`Greven, Venlo, The Netherlands) was used at
`0.25%w/w level as a lubricant.
`Model formulations (Table 1) were blended in
`a Turbular mixer (Type T2A, Pleuger, Basel,
`Switzerland). All ingredients with the exception
`of magnesium stearate were blended for 10 min,
`then magnesium stearate was added and mixed for
`an additional 5 min.
`
`Bulk Properties of the Mixtures
`
`The flow and packing properties of the powder
`mixtures were determined using an automatic
`tap volumeter (STAV 2003, J. Engelsmann AG,
`Ludwigshafen am Rhein, Germany). A 250-mL
`graduated glass cylinder was used. The tapping
`frequency was 250  15 taps/min and the lift
`height 3.0  0.2 mm. One hundred grams of
`powder were carefully filled into the measuring
`cylinder ensuring a flat top surface of the powder.
`The maximum bulk volume, Vo, was recorded.
`Then tapped volume, Vf, and compressibility
`index, 100  (Vo Vf)/Vo, were determined ac-
`cording to the USP.10
`
`Tableting
`
`Tablets (333 mg, 100 mg drug load) were
`compressed on the instrumented rotary Piccola
`tablet press (Riva, Argentina) at 30 rpm using
`9-mm concave tooling, at compression forces from
`4 to 14 kN. Upper compression and ejection forces
`were recorded.
`The tablet weight and tablet weight variation
`were obtained for 20 tablets taken during each
`tableting run for each formulation. The accuracy of
`the weight determination was 1 mg.
`
`Dissolution Testing
`
`The drug release from the matrices was measured
`using a Caleva ST7 dissolution tester (G.B.
`Caleva Ltd., Dorset, UK), USP apparatus II
`
`Table 1. Model HPMC Formulations Used in This
`Study
`
`Ingredients
`
`Concentration (%w/w)
`
`Drug
`HPMC
`Filler
`Fumed silica
`Magnesium stearate
`
`30.00
`20.00
`49.25
`0.50
`0.25
`
`JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 93, NO. 11, NOVEMBER 2004
`
`Par Pharm., Inc.
`Exhibit 1044
`Page 002
`
`

`

`2748
`
`LEVINA AND RAJABI-SIAHBOOMI
`
`(paddle) at 37  18C and 100 rpm. The drug
`concentration was measured using a UV spectro-
`photometer Model CE3021 (Cecil Instruments
`Ltd., Cambridge, UK), at 271 nm for theophylline
`and at 261 nm for chlorpheniramine maleate.
`The media used were purified water and phos-
`phate buffer (pH 7.4). The buffer was prepared
`according to British Pharmacopoeia11 by adding
`250 mL of 0.2 M potassium dihydrogen orthopho-
`sphate to 393.4 mL of 0.1 M sodium hydroxide.
`For each formulation and condition, dissolution
`rates of at least three individual tablets were
`determined and means and standard deviation
`values were calculated.
`
`Contact Angle Analysis
`
`The process of water penetration into the hydro-
`philic matrix tablets was examined using FTA˚ 200
`dynamic contact angle analyser (Camtel Ltd.,
`UK) with a flexible video system allowing fast
`image acquisition (up to 60 images per second).
`Twenty-microliter droplets of purified water were
`deposited on the face surface of dry tablet samples
`by positioning the dispenser tip just above the
`surface and growing the pendant drop until its
`bottom touched the sample and the droplet
`detached. The contact angle was measured over
`the first 15 seconds as the water spread/absorbed
`and recorded as a function of time. Nonlinear
`capture timing was used with fast timing at the
`beginning of the test (15 measurements/s) and the
`slow capture (2 measurements/s) during the final
`absorption stage.
`
`RESULTS AND DISCUSSION
`
`Tableting Properties of Matrices
`
`All formulations, regardless of type of excipient,
`had good flow (Table 2) with compressibility index
`
`of no more than 20. Tablet weight variations for
`all batches prepared in this study were found to be
`less than 1%, also an indication of good flow.
`Table 2 also shows that both CPM and TP
`formulations with lactose produced the highest
`ejection forces, whereas Starch 1500 due to its
`inherent lubricity produced the lowest ejection
`forces.
`All tablets had high mechanical strength. The
`rank order for tablet breaking force was: formula-
`tions containing MCC > spray dried lactose > PPS.
`
`Influence of Different Fillers and Compression
`Force on Drug Release
`Several authors12–17 have stated that compres-
`sion force had very little (not statistically sig-
`nificant) effect on drug release from HPMC
`matrices. However, in this study it was found
`that the applied compression force influenced
`drug release rate (Table 3), the extent of which
`was dependent on the type of filler used. The time
`taken for 50% drug release from formulations
`manufactured at different compression forces
`indicates that drug release become slower with
`increasing applied force. This effect is particularly
`profound when comparing tablets manufactured
`at a very low compression force of 4 kN with
`the tablets manufactured at higher compression
`forces of 10 and 14 kN. Depending on the com-
`pressibility behavior of the fillers, the porosity of
`the matrices may be reduced with increasing
`compression force, leading to slower water uptake
`and water front movement into the matrix, which
`in turn, may lead to slower drug release.
`Figures 1 and 2 show drug release profiles from
`matrices compressed at 4 and 14 kN, for chlorphe-
`niramine maleate and theophylline, respectively.
`Drug release from tablets made with lactose as a
`tiller was the fastest. Matrices containing partially
`pregelatinized starch produced the slowest drug
`release at all compression forces for both drugs.
`
`Table 2. Powder and Tablet Characterization of HPMC Matrix Formulations Studied Here
`
`Drug
`
`Filler
`
`CPM PPS
`MCC
`lactose
`PPS
`MCC
`lactose
`
`TP
`
`Bulk Volume
`(g/cm3) n ¼ 3
`141  1
`200  1
`194  2
`84  1
`230  2
`197  0
`
`Tapped Volume
`(g/cm3) n¼ 3
`115  1
`166  0
`165  0
`71  1
`185  1
`172  0
`
`Compress.
`Index
`
`18
`17
`15
`15
`20
`13
`
`Tablet Ejection
`Force (N)
`374  22
`530  27
`1079 48
`82  3
`96  4
`238  9
`
`Tablet Weight Variation
`(%) n¼ 20
`
`0.2–0.4
`0.4–0.7
`0.1–0.6
`0.2–0.4
`0.1–0.8
`0.1–0.9
`
`JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 93, NO. 11, NOVEMBER 2004
`
`Par Pharm., Inc.
`Exhibit 1044
`Page 003
`
`

`

`THE INFLUENCE OF EXCIPIENTS ON DRUG RELEASE FROM HPMC MATRICES
`
`2749
`
`Table 3. The Influence of Compression Force on
`Drug Release (T50%) from HPMC Matrices Containing
`Different Fillers
`
`T50% (min) for Tablets Manufactured
`at Various Compression Forces
`
`Drug
`
`Filler
`
`CPM PPS
`MCC
`lactose
`PPS
`MCC
`lactose
`
`TP
`
`4 kN
`215  2
`185  2
`95  2
`290  1
`230  1
`190  2
`
`10 kN
`380  2
`280  2
`160  2
`470  1
`340  1
`200  2
`
`14 kN
`420  2
`300  2
`175  2
`470  1
`360  1
`230  2
`
`The drug release differences between tablets cont-
`aining excipients such as lactose and MCC can be
`attributed mainly to the excipients solubility.
`However, the effect of Starch 1500 on drug release
`cannot be explained only by its solubility in water.
`It is more soluble compared to MCC, and produces
`slower drug release. Use of partially pregelati-
`nized starch in HPMC matrices may bring about
`different effects resulting from interactions be-
`tween HPMC and Starch 1500 that can affect the
`properties of the gel layer around the tablet.
`To investigate the mechanism of drug release
`and to compare the performance of various matrix
`formulations, the percent drug released versus
`time profiles were used. Data corresponding to
`5–60% release show a good fit to the Power Law
`Model18 expressed in eq. 1:
`Mt=Minf ¼ ktn
`
`ð1Þ
`
`Figure 1. The influence of compression force (4 and
`14 kN) on chlorpheniramine maleate release in water
`from HPMC matrices containing different fillers. [Color
`figure can be seen in the online version of this article,
`available on the website, www.interscience.wiley.com.]
`
`Figure 2. The influence of compression force (4 and
`14 kN) on theophylline release in water from HPMC
`matrices containing different fillers. [Color figure can be
`seen in the online version of this article, available on the
`website, www.interscience.wiley.com.]
`
`where Mt is the amount of drug released at time t,
`Minf is the amount of drug released after infinite
`time, k is a kinetic constant incorporating struc-
`tural and geometric characteristics of the tablet,
`and n is the diffusional exponent indicative of the
`drug release mechanism. The values of the kinetic
`constant (k), the release exponent (n), and correla-
`tion coefficient (R2) determined from the drug
`release data are presented in Table 4. The correla-
`tion coefficients for the data were >0.99. For
`matrix tablets, an n value of near 0.5 indicates
`diffusion control, and an n value of near 1.0 indic-
`ates erosion or relaxation control.19,20 Intermedi-
`ate values suggest that diffusion and erosion
`contribute to the overall release mechanism. The
`values of n and k are inversely related. A very
`high k value may suggest a burst drug release
`from the matrix.21
`Values of n for all matrices studied here were
`between 0.54 and 0.81, indicating an anomalous
`behavior corresponding to diffusion, erosion, and
`swelling mechanisms. In all these matrices avail-
`ability of the water within the gel structure is
`also limited, and therefore a dissolution-controlled
`release is also involved. Comparing tablets
`manufactured at the same compression force,
`separately for chlorpheniramine maleate and
`theophylline, a linear trend of decreasing n values
`can be observed from PPS to MCC and to lactose.
`Matrices containing lactose exhibited a drug
`release closer to a diffusion-controlled process
`compared to MCC and Starch 1500.
`Slower drug release from matrices with prege-
`latinized starch may be due to a slower penetration
`
`JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 93, NO. 11, NOVEMBER 2004
`
`Par Pharm., Inc.
`Exhibit 1044
`Page 004
`
`

`

`2750
`
`LEVINA AND RAJABI-SIAHBOOMI
`
`Table 4. Values of the Kinetic Constant (k), Diffusional Exponent (n) Derived
`from Equation 1 and Correlation Coefficients (R2), for HPMC Matrices Containing
`Different Fillers
`
`Filler
`
`PPS
`
`MCC
`
`Lactose
`
`Compression
`Force
`
`4 kN
`10 kN
`14 kN
`4 kN
`10 kN
`14 kN
`4 kN
`10 kN
`14 kN
`
`CPM
`
`k
`
`n
`
`R2
`
`k
`
`1.1332
`0.5638
`0.3861
`1.4910
`0.7197
`0.6304
`3.5188
`1.2356
`1.2152
`
`0.6878
`0.8048
`0.8081
`0.6759
`0.7426
`0.7708
`0.5822
`0.7268
`0.7367
`
`0.9999
`0.9948
`0.9976
`0.9976
`0.9971
`0.9967
`0.9993
`0.9961
`0.9966
`
`1.2495
`0.8673
`0.7816
`2.4406
`1.1485
`1.1077
`2.6826
`2.6563
`2.6339
`
`TP
`
`n
`
`0.6517
`0.6591
`0.6755
`0.5540
`0.6371
`0.6451
`0.5497
`0.5508
`0.5614
`
`R2
`
`0.9982
`0.9997
`0.9997
`0.9994
`0.9996
`0.9998
`0.9952
`0.9915
`0.9956
`
`of the water front towards the central core of the
`matrix. Matrices with swelling restrictions, like
`those with Starch 1500, exhibit a shift towards
`drug release by erosion mechanism.22 Tablets with
`partially pregelatinized starch would result in a
`more concentrated gel and increased gel tortuos-
`ity. Thus, the diffusional path would become more
`convoluted and the diffusion rate would therefore
`decrease. The effect of increased tortuosity and a
`delayed water penetration is expressed as low
`kinetic constant k values for tablets made with
`Starch 1500.
`Although HPMC hydration and gel formation is
`not affected by changes in pH23 (at pH ranges of
`gastrointestinal tract), the pH of the dissolution
`fluid is known to affect release rates of drugs
`from HPMC matrices.24 Attempts have been
`made to quantify the influences of the solutions
`containing phosphate and chloride ions at differ-
`ent ionic strengths on dissolution rates from
`HPMC SR tablets.25 In this study the effect of
`phosphate buffer (pH 7.4) on the matrix integrity
`and drug release from HPMC compacts containing
`different fillers was investigated. No significant
`changes in drug dissolution in buffer compared to
`water medium were observed for chlorphenira-
`mine maleate (Fig. 3). Theophylline release in
`phosphate buffer compared to water was slightly
`different for lactose and MCC containing matrices
`(Fig. 4). Theophylline dissolution profiles for
`tablets made with pregelatinized starch were
`similar in water and in buffer. Drug release from
`matrices containing Starch 1500 in both water and
`phosphate buffer was slower than when lactose or
`MCC was used.
`
`Influence of Starch 1500 Concentration on
`Drug Release from HPMC Matrices
`
`Figures 5 and 6 show drug release profiles from
`HPMC matrices containing partially pregelati-
`nized starch and lactose at different ratios, for
`CPM and TP, respectively. For both drugs, as the
`level of PPS increased the dissolution of drugs
`became significantly slower. Data in the range of
`5–60% drug release were fitted into eq. 1, and the
`results are shown in Table 5. The correlation
`coefficients for most of the data were >0.99. For
`chlorpheniramine maleate matrices studied here,
`the values of n ranged from 0.7367 to 0.8081,
`and the k values ranged from 0.3861 to 1.2152.
`For theophylline tablets, the values of n ranged
`
`Figure 3. Chlorpheniramine maleate release from
`HPMC matrices containing different fillers manufac-
`tured at 14 kN in water and in phosphate buffer (pH 7.4).
`[Color figure can be seen in the online version of this
`article, available on the website, www.interscience.
`wiley.com.]
`
`JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 93, NO. 11, NOVEMBER 2004
`
`Par Pharm., Inc.
`Exhibit 1044
`Page 005
`
`

`

`THE INFLUENCE OF EXCIPIENTS ON DRUG RELEASE FROM HPMC MATRICES
`
`2751
`
`Figure 4. Theophylline release from HPMC matrices
`containing different fillers manufactured at 14 kN in
`water and phosphate buffer (pH 7.4). [Color figure can be
`seen in the online version of this article, available on the
`website, www.interscience.wiley.com.]
`
`Figure 6. Effect of Starch 1500 levels on theophylline
`release from HPMC matrices manufactured at 14 kN.
`[Color figure can be seen in the online version of this
`article, available on the website, www.interscience.
`wiley.com.]
`
`from 0.5614 to 0.6755, and the k values ranged
`from 0.7816 to 2.6339. Values of n for all matrices
`studied here were between 0.56 and 0.81, indicat-
`ing an anomalous behavior corresponding to
`diffusion, erosion, and swelling mechanisms.
`Comparing tablets with the same drug, sepa-
`rately for chlorpheniramine maleate and theo-
`phylline, a linear trend of increasing n values can
`be observed with an increase in PPS concentra-
`tion. Matrices containing more lactose exhibited a
`drug release closer to a diffusion-controlled pro-
`cess compared to tablets containing higher levels
`of Starch 1500. Thus, the effect seen with Starch
`1500 is not just a spatial effect due to the presence
`of any filler, but PPS actively contributes to the
`dissolution kinetics. This contribution is imparted
`
`through possible contribution of Starch 1500 in
`gel formation of HPMC, that is, the filler actively
`forming an integral structure within the HPMC
`gel layer at lower concentrations of HPMC in the
`formulation.
`Michailova et al.26 characterized HPMC/prege-
`latinized starch hydrogels as ‘‘filled’’ composite
`systems where starch filler functions as a support-
`ing frame, while the linear hypromellose forms the
`continuous disperse medium. In comparison with
`the cellulose derivative, the pregelatinized starch
`hydrates to a considerably lower degree due to the
`formation of intramolecular hydrogen bonds in the
`highly branched amylopectin.27 These bonds sup-
`press the polymer segments’ mobility and dimin-
`ish the degree of HPMC/pregelatinized starch
`hydration28 resulting in a reduced gel layer diffu-
`sivity and decreased drug velocity from matrices
`containing higher pregelatinized starch quantity.
`For this reason, at 20% of HPMC and low concent-
`ration of the pregelatinized starch gel structure is
`quite porous with increased diffusion capability.
`With the increase in PPS concentration (35–49%),
`the swelled starch particles form strong support-
`ing structure with comparatively strong rigidity.
`This HPMC/PPS gel structure may explain the
`slower drug release with increasing pregelatinized
`starch concentration in the formulation.
`
`Figure 5. Effect of Starch 1500 levels on chlorphenir-
`amine mleate release from HPMC matrices manufac-
`tured at 14 kN. [Color figure can be seen in the online
`version of this article, available on the website, www.
`interscience.wiley.com.]
`
`Testing of Water Absorption Rate
`
`Drug release from HPMC matrix tablets is based
`on the glassy transition of the polymer into a
`rubbery gel that occurs as a result of water
`absorption/hydration of
`the polymer
`in the
`
`JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 93, NO. 11, NOVEMBER 2004
`
`Par Pharm., Inc.
`Exhibit 1044
`Page 006
`
`

`

`2752
`
`LEVINA AND RAJABI-SIAHBOOMI
`
`Table 5. Values of the Kinetic Constant (k), Diffusional Exponent (n) Derived from Equation 1 and Correlation
`Coefficients (R2), for HPMC Matrices Containing Various Levels of Starch 1500 and Manufactured at 14 kN
`
`PPS
`Concentration
`(%w/w)
`
`Lactose
`Concentration
`(%w/w)
`
`0.00
`20.00
`35.00
`49.25
`
`49.25
`29.25
`14.25
`0.00
`
`k
`
`1.2152
`0.9771
`0.8780
`0.3861
`
`CPM
`
`n
`
`0.7367
`0.7462
`0.7516
`0.8081
`
`R2
`
`0.9966
`0.9957
`0.9992
`0.9976
`
`k
`
`2.6339
`2.4985
`0.9678
`0.7816
`
`TP
`
`n
`
`0.5614
`0.6116
`0.6639
`0.6755
`
`R2
`
`0.9956
`0.9893
`0.9999
`0.9997
`
`matrix. The drug release mechanism is deter-
`mined by the structural characteristics of the gel
`layer (swelling, uniformity of polymer hydration,
`diffusion capability, and gel strength), and by gel
`layer erosion. Therefore, rapid gel
`formation
`(rubbery phase) to prevent rapid ingress of water
`into the matrix as well as high gel strength are
`critical
`factors in drug release from HPMC
`matrices. It was found that water penetration
`into tablets containing Starch 1500 was much
`slower compared to matrices containing MCC or
`lactose (Fig. 7). This observation was confirmed by
`contact angle measurements (Fig. 8). Table 6
`shows that the initial contact angle for all the
`samples was similar (57–728) and less than 908,
`indicating good surface wettability behavior of
`these matrices, when the water drop flattens out
`and spreads on the tablet surface. However, for
`MCC and lactose containing matrices, the water
`droplet was rapidly absorbed into the matrix
`(within 2–7 s), which was much faster (6–13
`times) than for the matrices containing Starch
`1500 (>30 s). It was also found that the rate of
`contact angle change was significantly faster for
`chlorpheniramine maleate as a freely water
`soluble drug compared to theophylline.
`
`The presence of free water within the gel layer
`plays an important part in drug movement across
`this barrier. Decreased availability of free water
`may lead to decreased drug diffusion across the gel
`layer. Partially pregelatinized starch and hypro-
`mellose combinations may be producing a gelled
`interlocked frame consisting of HPMC fibers and
`amylose reinforced by the swollen starch gran-
`ules.29,30 This network restrains water penetra-
`tion into SR matrices and prevents fast drug
`release.31
`
`CONCLUSIONS
`
`All HPMC SR formulations had good powder
`flow, tablet weight uniformity, and mechanical
`strength. Formulations with lactose produced the
`highest ejection forces. On the other hand, par-
`tially pregelatinized starch due to its inherent
`lubricity produced the lowest ejection forces.
`All formulations regardless of type of filler
`resulted in a slow drug release for both candidate
`drugs. Drug release was found to be affected by
`
`Figure 7. Water droplet and its absorption into (a)
`PPS and (b) microcrystalline cellulose or lactose contain-
`ing HPMC matrices. [Color figure can be seen in the
`online version of this article, available on the website,
`www.interscience.wiley.com.]
`
`Figure 8. Contact angle measurements for water
`droplets on the surface of HPMC matrices containing
`different fillers. [Color figure can be seen in the online
`version of this article, available on the website, www.
`interscience.wiley.com.]
`
`JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 93, NO. 11, NOVEMBER 2004
`
`Par Pharm., Inc.
`Exhibit 1044
`Page 007
`
`

`

`THE INFLUENCE OF EXCIPIENTS ON DRUG RELEASE FROM HPMC MATRICES
`
`2753
`
`Table 6. Contact Angle Analysis of Purified Water on the Surface of HPMC Matrices
`Containing Different Fillers
`
`Drug
`
`CPM
`
`TP
`
`Filler
`
`PPS
`MCC
`lactose
`PPS
`MCC
`lactose
`
`Initial Contact
`Angle (Degrees)
`
`Absorption Time
`(Seconds)
`
`Rate of Contact Angle
`Change (Degree/s)
`
`71
`65
`72
`62
`60
`57
`
`>30.0
`2.4
`7.0
`>30.0
`6.7
`5.4
`
`1.1
`14.5
`6.6
`0.5
`4.6
`4.2
`
`applied compression force. At all compression
`forces and with both drugs, when Starch 1500
`was used, drug release was slower compared to
`formulations containing MCC or lactose. Similar
`results were produced in phosphate buffer. These
`results may suggest that partially pregelatinized
`starch is not an inert filler in HPMC matrices (with
`low HPMC contents), but it actively contributes to
`the mechanism of drug release.
`It was shown that for both drugs, increasing
`concentrations of Starch 1500 (20, 35 and
`49.25%w/w) in the formulations caused a decrease
`in drug release rates. Therefore, use of blends of
`Starch 1500 with other fillers (e.g. lactose) can be
`used for tailoring the desired release profile of
`HPMC matrix systems.
`It was found that water absorption into tablet
`containing partially pregelatinized starch was
`much slower compared to matrices containing
`MCC or lactose. This observation was confirmed
`by contact angle analysis. These results may
`explain the slower drug release from HPMC
`matrices containing Starch 1500 compared to
`those containing MCC or lactose.
`
`REFERENCES
`
`1. Rajabi-Siahboomi AR, Jordan MP. 2000. Slow
`release HPMC matrix systems. Eur Pharm Rev
`5:21–23.
`2. Rak J, Chalabala M, Mandak M. 1983. Modified
`starches—New auxiliary substances in the produc-
`tion of tablets. Acta Fac Pharm 37:5–27.
`3. Van Aerde P, Remon JP. 1988. In vitro evaluation
`of modified starches as matrices for sustained
`release dosage forms. Int J Pharm 45:145–152.
`4. Herman J, Remon JP. 1989. Modified starches as
`hydrophilic matrices for controlled oral delivery. II.
`In vitro drug release evaluation of thermally
`modified starches. Int J Pharm 56:65–70.
`5. Michailova V, Titeva S, Minkov E. 1998. Influence
`of pH on molsidomine release from unlimited
`
`swelling hydrogel matrices. Proc 2nd Word Meeting
`APGI/APV, pp. 321–322.
`6. Mulhbacher J,
`Ispas-Szabo P, Lenaerts V,
`Mateescu MA. 2000. Cross-linked high amylose
`starch derivatives as matrices for release control
`from high drug dosages. Proceed Int Symp Con-
`trolled Release Bioact Mater 27:8421.
`7. Lenaerts V, Moussa I, Dumoulin Y, Mebsout F,
`Chouinard F, Szabo P, Mateescu MA, Cartilier L,
`Marchessault R. 1998. Cross-linked high amylose
`starch for controlled release of drugs: Recent
`advances. J Controlled Release 53:225–234.
`8. Cunningham CR. 1999. Maize starch and super-
`disintegrants in direct compression formulation.
`Pharm Manufac Rev December:22–24.
`9. Leach HW, McCowen LD, Schoch TJ. 1959.
`Structure of the starch granule. I. Swelling and
`solubility patterns of various starches. Cereal
`Chem 36:534–544.
`10. United States Pharmacopeia XXV. 2002. Rockville,
`MD: The US Pharmacopeial Convention, Inc.
`11. British Pharmacopoeia. 2001. London: Stationery
`Office. A120.
`12. Ford JL, Rubinstein MH, Hogan JE. 1985.
`Formulation of sustained release promethazine
`hydrochloride tablets using hydroxypropylmethyl-
`cellulose matrices. Int J Pharm 24:327–338.
`13. Ford JL, Rubinstein MH, Hogan JE. 1985. Propra-
`nolol hydrochloride and aminophylline release
`from matrix tablets containing hydroxypropyl-
`methylcellulose. Int J Pharm 24:339–350.
`14. Ford JL, Rubinstein MH, Hogan JE. 1985.
`Dissolution of a poorly water soluble drug, indo-
`methacin,
`from hydroxypropylmethylcellulose
`controlled release tablets. J Pharm Pharmacol 37:
`33P.
`15. Dahl TC, Calderwood T, Bormeth A, Trimble K,
`Piepmeir E. 1990. Influence of physico-chemical
`properties of hydroxypropylmethylcellulose on
`naproxen release from sustained release matrix
`tablets. J Controlled Release 14:1–10.
`16. Liu C, Kao Y, Chen S, Sokoloski TD, Sheu MT.
`1995. In-vitro and in-vivo studies of the diclofenac
`sodium controlled-release matrix tablets. J Pharm
`Pharmacol 47:360–364.
`
`JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 93, NO. 11, NOVEMBER 2004
`
`Par Pharm., Inc.
`Exhibit 1044
`Page 008
`
`

`

`2754
`
`LEVINA AND RAJABI-SIAHBOOMI
`
`17. Velasco MV, Ford JL, Rowe P, Rajabi-Siahboomi
`AR. 1999. Influence of drug:hydroxypropylmethyl-
`cellulose ratio, drug and polymer particle size and
`compression force on the release of diclofenac
`sodium from HPMC tablets. J Controlled Release
`57:75–85.
`18. Spiepmann J, Peppas NA. 2001. Modelling of drug
`release from delivery systems based on hydroxy-
`propyl methylcellulose (HPMC). Adv Drug Deliv
`Rev 48:139–157.
`19. Ford JL, Mitchell K, Rowe P, Armstrong DJ, Elliott
`PNC, Rostron C, Hogan JE. 1991. Mathematical
`modelling of drug release from hydroxypropyl-
`methylcellulose matrices: Effect of temperature.
`Int J Pharm 71:95–104.
`20. Espinoza R, Hong E, Villafuerte L. 2000. Influence
`of admixed citric acid on the release profile of
`pelanserin hydrochloride from HPMC matrix
`tablets. Int J Pharm 201:165–173.
`21. Ebube NK, Hikal A, Wyandt CM, Beer DC,
`Miller LG, Jones AB. 1997. Sustained release of
`acetaminophen from heterogeneous matrix tablets:
`Influence of polymer ratio, polymer loading, and
`co-active on drug release. Pharm Dev Technol 2:
`161–170.
`22. Vogireaux V, Ghaly ES. 1994. Fickian and relaxa-
`tional contribution quantification of drug release in
`a swellable hydrophilic polymer matrix. Drug Dev
`Ind Pharm 20:2519–2526.
`23. Amaral MH, Lobo JMS, Ferreira DC. 2000.
`Polymer erosion and drug release study of hydro-
`philic matrices. Proc 19th Pharm Tech Conf.,
`pp. 204–211.
`
`24. Mitchell K, Ford JL, Armstrong DJ, Elliot PNC,
`Rostron C, Hogan JE. 1990. The influence of
`additives on the cloud point, disintegration and
`dissolution of hydroxypropylmethylcellulose gels
`and matrix tablets. Int J Pharm 66:233–242.
`25. Rajabi-Siahboomi AR, Adler J, Davies MC, Melia
`CD. 1994. Particle swelling and the mechanism of
`failure of HPMC matrices. Proc 3rd Assoc. 21.
`26. Michailova V, Titeva S, Kotsilkova R, Krusteva E,
`Minkov E. 2001. Influence of hydrogel structure on
`the process of water penetration and drug release
`from mixed hydroxypropylmethyl cellulose/ther-
`mally pregelanized waxy maize starch hydrophilic
`matrices. Int J Pharm 222:7–17.
`27. Hanselmann R, Burchard W, Ehrat M, Widmer
`HM. 1996. Structural properties of fractionated
`starch polymers and their dependence of the dis-
`solution process. Macromolecules 29:3277–3282.
`28. Ferry JD, Lomellini P. 1999. Melt rheology of
`randomly branched polystyrenes. J Rheol 43:
`1355–1372.
`29. Ott M, Hester EE. 1965. Gel formation as related to
`concentration of amylose and degree of starch
`swelling. Cereal Chem 42:476–484.
`30. Caarnali JO, Zhou Y. 1996. An examination of the
`composite model for starch gels. J Rheol 40:221–
`234.
`31. Wierik GHPT, Eissens AC, Bergsma J, Arends-
`Scholte AW, Lerk CF. 1997. A new generation of
`starch products as excipient in pharmaceutical
`tablets. II. High surface area retrograded pregela-
`tinized potato starch products in sustained-release
`tablets. J Controlled Release 45:25–33.
`
`JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 93, NO. 11, NOVEMBER 2004
`
`Par Pharm., Inc.
`Exhibit 1044
`Page 009
`
`