Dissolution profiles ofmesalazine formulations in vitro
`
`L.M.L. Stolk, R. Rietbroek, E.H. Wiltink and J.J. Tukker
`
`Introduction
`Sulfasalazine, a conjugate of mesalazine (5-
`aminosalicylic acid; 5-ASA) and sulfapyridine,
`has been used successfully for four decades in the
`treatment of inflammatory bowel disease. Re(cid:173)
`cently, it has been demonstrated that mesalazine
`is the therapeutically active moiety of sulfasal(cid:173)
`azine. In the colon the azo bond is dissociated by
`intestinal bacteria and mesalazine is set free.
`Plain mesalazine is totally absorbed in the upper
`part of the intestine. Therefore, pharmaceutical
`formulations have been designed, which can
`transport mesalazine undisturbed
`through
`stomach, duodenum and proximal jejunum and
`deliver high concentrations of mesalazine selec(cid:173)
`tively at the inflammatory sites of the distal
`small intestine and colon [1]. These formulations
`are Pentasa®, which is a 'slow-release' formu(cid:173)
`lation and Asacol® and Salofalk®, which are
`tablets with an acid-resistant coating.
`Pentasa® tablets contain 250 mg of mesalazine
`in microgranules (0.7-1 mm), which are coated
`with a semi-permeable membrane of etbylcellu(cid:173)
`lose. Mesalazine should in principle be gradually
`released from the ethylcellulose-coated granules
`at either acid or alkaline pH and release should
`take place ovex the whole length of the intestine.
`Asacol® tablets contain 400 mg mesalazine and
`the tablets are coated with an acrylic resin
`(Eudragit® S), which dissolves above pH 7.0.
`Thus, in vivo it should be transported intact until
`it reaches the ascending part of the colon where
`the intraluminal pH rises above 7 and mesal(cid:173)
`azine is liberated ..
`Salofalk® contains 250 mg mesalazine and so(cid:173)
`dium carbonate as buffering and solubilizing
`agent. The tablet has an outer coating with a
`semi-permeable membrane of ethylcellulose and
`an inner coating with an acrylic resin (Eudragit®
`L). This acrylic resin dissolves above pH 5.6. In
`this formulation mesalazine is meant to be re(cid:173)
`leased in highly dispersed form in the distal
`small bowel and the colon.
`
`In OUl' study we have investigated in vitro re-.
`lease of mesalazine from Pentasa®, Asacol® and
`Salofalk® tablets at pH 1.0, pH 6.0 and pH 7.5,
`intended to represent the acidity of the stomach,
`small and large bowel, respectively.
`The only official testing methods utilize the
`paddle and
`the rotating-basket dissolution
`apparatus as described in the United States
`Pharmacopeia XXJ
`'!'he rotating-basket
`[2].
`method has inherent weaknesses, including poor
`mechanical stability, simultaneous disinte(cid:173)
`gration and dissolution and poor homogeneity
`because of insufficient stin-ing. As an alternative
`for the pharmacopoeia! test the flow-through or
`column-type method has been proposed (3 4]. We
`used a closed-column type method, which is kin(cid:173)
`etically equivalent to a stirred-vessel system,
`but will not affect dissolution kinetics as long
`'sink conditions' prevail. However,
`this
`as
`method has the advantage of being largely auto(cid:173)
`matic. No wetting agents were added to the dis(cid:173)
`solution media, because there were no indi(cid:173)
`cations that wettability presented a problem.
`
`Methods
`
`Materials
`from Bufa
`obtained
`Mesalazine was
`(Castricum, the Netherlands, batch no. 22K88).
`All other chemicals were analytical-grade
`quality from Merck (Darmstadt, FRG).
`Pentasa® tablets (Gist-brocades, Delft, the
`Netherlands, batch no. 86BG40), Asacol® tablets
`(Cedona, Haarlem, the Netherlands, batch no.
`BN87125) and Salofalk® tablets (Tramedico,
`Weesp, the Netherlands, batch no. 88G18) were
`obtained through OPG (Utrecht, the Nether(cid:173)
`lands).
`
`Dissolution apparatus
`A Dissotest flow-through dissolution apparatus
`was used (Sotax, Basle, Switzerland; imported by
`Proton-Wilten, Etten-Lelli', the Netherlands).
`
`Keywo1·ds
`Dissolution
`Mesalazine
`Tablets, controlled release
`
`Dr. L.M.L. Stalk (correspondence)
`and E.H. Wiltink: Pharmacy
`Department, Academic Medical
`Centre, Meibergdreef 9, 1105 AZ
`Amsterdam, the Netherlands.
`R. Rietbroek and Dr. J.J. Tukker:
`Department of Biopharmaceutics,
`Faculty of Pharmacy, Utrecht
`University, Catharijnesingel 60,
`3511 GH Utrecht, the
`Netherlands.
`
`Stolk LML, Rietbroek R, Wiltink EH, Tukker JJ. Dissolution profiles of mesalazine
`formulations in vitro. Pharro Weekbl [Seil 1990;12(5):200-4
`
`Abstract
`In vitro dissolution profiles of three controlled·release mesalazine formulations were
`determined at pH 1.0, 6.0 and 7.5. A closed-column type dissolution apparatus was used. A
`reproducible gradual dissolution profile was seen for Pentasa~ at all pH values. Dissolution
`starts immediately and i.s complete after 20 h. Dissolution profiles at pH 1 and pH 7.5 ai·e
`much alike and dissolution is faster than at pH 6. The behaviour of Asacol® at different pH
`values cori·esponds with the expectations: no release at pH 6 and p.H 1, fast release at pH
`7.5. Dissolution starts after 1 hand is complete after 3 h. Mesalazine release from Salofalk.0
`tablets at pH 7.5 and pH 6.0 starts after 2 and 3 h, respectively, and is complete after 5 and
`10 h. However, after a long Jag-time (10 h) mesalazine is also released from Salofalk® tablets
`at pH 1 and dissolution is complete after 23 b.
`
`Accepted 26 July 1990.
`
`200
`
`Phai·maceutisch Weekblad Scientific edition
`
`12(5) 1990
`Mylan Exhibit 1006
`
`

`
`--
`
`ASAC0L In SIF pH 7.5
`" (• '400 mg 5-ASA)
`SALOFALK In SIF pH 7.s
`
`- PENTASA In SIF pH 7.5
`
`- I
`f
`
`Figure 1
`Dissolution pro(cid:173)
`files at pH 7.5
`(mean.± SD;
`II= 5)
`
`90
`
`80
`
`70
`
`eo
`
`50
`
`30
`
`20
`
`10
`
`0
`
`!
`
`4
`
`8
`
`0
`
`2
`
`8
`
`I() 12141818202224~
`
`Figure 3
`Dissolution profiles at pH 1.0 (mean ± SD; n :: 3)
`
`waterbath at 37.0+ 0.5 °C. The tablets were hung
`in a wire frame in the small testing chamber of
`20 ml capacity (22.6 mm diameter). The conical
`part of the testing chamber was filled with glass
`beads (diameter 1 mm) to ensure a uniform dis(cid:173)
`tribution of the flow over the entire cross section
`of the cell. The dissolution medium was pumped
`around at a flow rate of 31.0± 0.5 ml/min from
`the vessel with the dissolution medium through
`the spectrophoLometer and then through the dis(cid:173)
`solution cell back into the vessel. The solvent left
`the cell through a glass-fibre filter (Whatman
`GF/D, Proton-Wilten), that retained particles
`larger than 2.7 µm to prevent removal of un(cid:173)
`dissolved particles.
`
`Dissolution media
`The following dissolution media were used:
`- simulated intestinal fluid USP XXI (SIF) with(cid:173)
`out pancreatine, pH 7.5;
`- the same solution as mentioned above but
`brought to pH 6.0 with phosphoric acid,
`pH6.0;
`- HCl 0.1 Min water, pH 1. In order to obtain
`'sink conditions' 5 1 dissolution medium was
`used for Asacol® and 3 1 for Pentasa® and
`Salofalk®.
`
`Assay of mesalazine
`The mesalazine content of the dissolution me(cid:173)
`dium was determined by measuring UV absorp(cid:173)
`tion at 331 nm (pH 7 .5 and pH 6.0) and at 304 nm
`(pH LO), respectively. The linearity and th e pre(cid:173)
`cision of the assay were determined by measur(cid:173)
`ing the absorbance of mesalazine of known con(cid:173)
`centrations four times. The detector response
`was linear in the concentration range employed:
`0-85 mg/l. The calibration lines can be described
`
`The dissolution apparatus consisted of a dissol(cid:173)
`ution cell (standard cell, i.d. 22,6 mm), a pump, a
`3-1 or 5-1 vessel with dissolution medium and a
`spectrophotometer (Unicam SP 1800, Philips,
`Eindhoven, the Netherlands) with a 1 cm quartz
`Dow-through cell (174-QS, Hellma, The Hague,
`the Netherlands). The cumulative dissolution
`curve was drawn by a chart recorder (Unica.m AR
`25, Philips). The dissolution cell and the vessel
`were placed in a thermostatically controlled
`
`Figure 2
`Dissolution pro(cid:173)
`files at pH 6.0
`(mean ±SD;
`n = 3)
`
`IOO f
`
`90
`
`80
`
`70
`
`50
`
`12(5) 1990
`
`Pharmaceutisch Weekblad Scientific edition
`
`201
`
`

`
`asacol
`
`'ti
`Q) 100
`(I)
`C!I
`Q)
`
`80
`
`Qi ...
`::.1! 0
`
`t 60
`
`40
`
`20
`
`0
`
`0
`
`2
`
`4
`
`Figui-e 4
`Dissolution profiles (computer fit) for individual Asacol® tablets. 0 : pH 7.5;
`e: pH 7.5 after 2 h pretreatment at pH 1.0
`
`6
`- - ->
`
`8
`time
`
`10
`(hrs)
`
`salofalk
`
`'ti 120
`
`Cl)
`II)
`C!I
`~ 100
`~
`
`::.1!
`0
`
`80
`f 60
`40
`
`20
`
`0
`
`0
`
`10
`
`20
`···>
`
`30
`time (hrs)
`
`Figure 5
`Dissolution profiles (computer fit) for individual Salofalk® tablet.s.
`0 : pH 1.0; • :pH 7.5 after 2 h pretreatment at pH 1.0; • :pH 6; 0 : pH 7.5
`
`pentasa
`
`120
`
`Q)
`Cl)
`C!I
`Q) 100
`...
`::.1! 0
`
`Q)
`
`80
`
`f 60
`
`40
`
`20
`
`0
`
`0
`
`10
`
`20
`--->
`Figure 6
`Dissolution profiles (computer fit] for individual Pentasa® tablets.
`0 : pH 1.0; • :pH 6.0; D: pH 7.5
`
`30
`(hrs)
`
`time
`
`by: y = 0.021 x + 0.007 (pH 7.5); y = 0.0145 x -
`0.0004 (pH = 6.0); y = 0.024 x + 0.004 (pH 1.0)
`(y UV absorbance; x mesalazine concentration
`(mg/l); r 0.999).
`
`Dissolution profiles
`Dissolution profiles were determined at pH 7.5,
`pH 6.0 and pH 1.0 for Asacol®, Pentasa® and
`Salofalk® tablets with the Dissotest apparatus.
`Dissolution profiles of Asacol® and Salofalk®
`tablets were also determined at pH 7.5 after 2 h
`previous immersion in 0.1 M HCl in order to
`study the influence of artificial gastric fluid. The
`tablets were placed in the 'rotating basket ap(cid:173)
`paratus', described in the USP XXI filled with 11
`0.1 M HCl [2J. The stirring speed was 100 rev/
`min. After 2 h the tablets were transferred to the
`Dissotest apparatus with dissolution medium
`pH 7.5. Because mesalazine is released from
`Pentasa® at either alkaline or acid pH, Pentasa®
`tablets were not included in this experiment.
`
`Computer fit
`An emperical function was used to describe dis(cid:173)
`solution profiles: the RRSBW-distribution func(cid:173)
`tion [5]. The RRSBW-distribution functions were
`calculated with the program SYSTAT (SYSTAT,
`A vanston, Illinois, USA) on a Macintosh SE/30
`computer.
`
`Stability of mesalazine in dissolution media
`Stability of mesalazine in the dissolution me(cid:173)
`dia was determined by dissolving 250 mg mesal(cid:173)
`azine in 3 l of the dissolution media. AfteT 24 h
`in the dissolution apparatus under the same
`experimental conditions as described above, the
`mesalazine concentrations were measured. Con(cid:173)
`centrations were measured with UV absorption
`and with a reversed-phase high pressure liquid
`chromatographic (HPLC) method described in
`the literature [6J.
`
`Solubility of mesalazine at various pH values
`To 50 ml of each of the three freshly prepared
`dissolution media 1 g mesalazine was added. The
`resulting suspension was shaken during 10 min.
`Then the suspension was placed in a thermostati(cid:173)
`cally controlled water bath of 37.0 ± 0.5 °C. 1-ml
`Samples were taken after 4 and 24 h. The
`samples were filtered through a 0.8 µm mem(cid:173)
`brane
`filter
`(Sartorius, Etten-Leur,
`the
`Netherlands). Mesalazioe concentrations were
`measw·ed \Vith HPLC .
`
`Results
`
`Dissolution
`The concentration of mesalazine in the dissol(cid:173)
`ution media after 4 h was (mean ±SD; n = 3):
`11.0±0.2 g/l (pH 1); 2.3± 0.1 g/l (pH 6); 6.5±
`0.1 g/l (pH 7.5). After 24 h were 9.9± 0.l g/l
`(pH l); 2.3 ± 0.1 g/l (pH 6); 6.1±0.1 g/l (pH 7.5).
`This means that all experiments were carried
`out under 'sink conditions'.
`The dissolution profiles at pH 7.5, pH 6.0 and
`pH 1.0, respectively, ai-e demonstrated in Fig(cid:173)
`ures 1, 2 and 3. Dissolution profiles for Asacol®,
`Salofalk® and Pentasa® at different pH values
`were constructed with the help of a computer fit.
`They are demonstrated in Figw·es 4, 5 and 6 (in(cid:173)
`dividual tablets) and in figlll'es 7, 8 and 9 (mean
`clll'ves).
`Pentasa shows a reproducible gradual dissol-
`
`202
`
`Pharmaceutisch Weekblad Scientific edition
`
`12(5) 1990
`
`

`
`Asacol
`
`-~ 0
`
`(I)
`I/I
`l'O
`(I)
`
`(I) ...
`/I • • .
`
`100
`
`80
`
`60
`
`40
`
`20
`
`0
`
`0
`
`2
`
`4
`
`6
`···>
`
`8
`time
`
`1 0
`(hrs)
`
`Figure 7
`Dissolution profiles (computer fit) for AsacoL® (mean ±SD; n == 5).
`e: pH 7.5 after 2 h pretreatment at pH 1.0; 0 : pH 7.5
`
`Salofalk
`
`- 140
`
`~ 0
`
`120
`
`(I)
`I/I
`(I) 100
`l'O
`(I) ...
`. •
`
`80
`
`60
`
`/I
`I
`
`pH 7 .5 and at pH 6.0 and for the Asacol® tablets
`at pH 7 .5. The dissolution curves determined at
`pH 7.5 with or without pretreatment with 0.1 M
`HCl were almost identical. However, after im(cid:173)
`mersion in 0.1 M HCl dissolution started some(cid:173)
`what later.
`
`Stability studies
`After 24 h under experimental conditions
`mesalazine concentrations measured by HPLC
`analysis were (n = 2): 95.8% (pH 1), 94.7% (pH
`6.0), 95.6% (pH 7.5) of the original mesalazine
`level. Measured by UV absorption the figures
`are: 99.9% (pH 1), 111.9% (pH 6.0) and 100.1%
`(pH 7.5). A brown colowfog of the dissolution
`medium occurred at pH 6 and pH 7 .5 within 24 h.
`
`Discussion
`The reproducible mesalazine release from
`Pentasa® tablets is due to the principle of dif(cid:173)
`fusion-regulated release from the ethylcellulose(cid:173)
`coated microgranules. The slower dissolution
`rate at pH 6.0 in comparison ·with pH 7 .5 and
`pH 1.0 can be explained by the lower solubility of
`mesalazine at pH 6.0, being nearer the isoelec(cid:173)
`tric point of mesalazine (pH 4.3), where the com(cid:173)
`pound's solubility is lower than at the other pH
`values used. The shape of the curve indicates fast
`desintegration into granules and grndually de(cid:173)
`creasing dissolution rate [5].
`The release mechanism of Asacol® and
`Salofalk® tablets is a dissolution-regulated reser(cid:173)
`voir system. After dissolution of the coating, re(cid:173)
`lease is very fast. A possible explanation for the
`large variation of the dissolution profile of the in(cid:173)
`dividual Asacol® and Salofalk® tablets could be
`the variation of the thickness of the coating layer
`of the metacrylic metacrylate esters. The typical
`sigmoid shape of the curves indicates that desin(cid:173)
`teg:ration and dissolution are of comparable rate
`(5). The behaviour at the different pH va]ues of
`Asacol® corresponds with the expectations: fast
`release at pH 7.5 and no release at pH 6.0 and
`pH 1.0. As expected, we found fast release for
`Salofalk® at pH 7.5 and pH 6. However, after a
`long lag-time release was also measured at
`pH 1.0. Perhaps the coating is somewhat perme(cid:173)
`able at lower pH values. After fluid penetration
`the presence of sodium carbonate may be re(cid:173)
`sponsible for fast dissolution. No large changes of
`the dissolution profiles at pH 7 .5 f01: Asacol® and
`Salofalk® were seen after previous immersion in
`0.1 M HCL We have no explanation for the
`longer lag-time of Asacol® and Salofalk® after
`pretreatment with 0.1 M HCl.
`The plateau value of about 110% at pH 6.0 can
`be explained by the forming of a degradation
`product of mesalazine with high UV absorption
`at 331 nm. The same phenomenon was also found
`by Terpstra et al. [7]. They determined dissol(cid:173)
`ution profiles during 24 h with the paddle
`method (USP XX.l) at pH 2.0, 6.0 and 7.5. The re(cid:173)
`sults for Pentasa® were essentially identical to
`our results. For Asacol® and Salofalk®, the dissol(cid:173)
`ution was significantly faster than in our exper(cid:173)
`iments. This might be caused by use of the paddle
`method, thus putting more stress on the tablets
`and decreasing desintegration
`time of the
`
`40
`
`20
`
`0
`
`0
`
`1 0
`
`20
`time
`
`--·>
`
`(hrs)
`
`Figure 8
`Dissoluti.on. profiles (computer fit) (01· Salofalk® (mean ±SD; n = 3).
`0 : pH 1.0; e: pH 7.5 aft.er 2 h pretreatment o.t pH 1.0; • :pH 6.0;
`0 : pH7.5
`
`uiion profile at all pH values. Dissolution starts
`immediately and is almost complete after 20 h.
`Dissolution profiles at pH 1.0 and pH 7.5 are
`much alike and dissolution is faster than at
`pH 6.0. Dissolution of Salofalk® at pH 1.0, pH 6.0
`and pH 7.5 starts after about 10, 3 and 2 h,
`respectively, and is complete after 23, 10 and 5 h.
`Salofalk® tablets release mesalazine more rap(cid:173)
`idly when pH is rising.
`At pH 1.0 and pH 6.0 no dissolution occurred
`from Asacol® tablets. At pH 7 .5 dissolution starts
`after 1 h and is complete after 3 h.
`A large variation of the dissolution profiles
`was seen for the individual Salofalk® tablets at
`
`12(5) 1990
`
`Pharmaceutisch W eekblad Scientific edition
`
`203
`
`

`
`Pentasa
`
`120
`~
`
`0
`
`<I> 100
`Ill
`cu
`Q)
`<I> ...
`
`80
`
`/\ • • •
`
`60
`
`40
`
`20
`
`0
`
`0
`
`30
`20
`(hrs)
`ti me
`--->
`Figure 9
`Dissolution profiles (computer fit) for Pento.sa® (mean ±SD; n = 3).
`0 : pH 1.0; • :pH 6.0; 0 : pH 7.5
`
`1 0
`
`tablets. Duchateau et al., determined dissolution
`profiles with the USP rotating-basket method at
`pH 5.0; 5.5; 6.0; 6.5; 7.0 and 7.5 during 3 h (8].
`The results of this investigation are in agree(cid:173)
`ment with our results. However, caution should
`be exercised in extrapolation of results in uitro
`
`to the situation in uiuo. Recent investigations in(cid:173)
`dicate pH values that deviate from the genexally
`assumed figures: 6.6 to 7.5 in the small intestine
`and 6.4 to 7.0 in the colon. Moreover, a lai·ge
`interindividual variation was measured [9].
`
`References
`1 Jarnerot G. Newer 5-aminosalicylic acid based drugs in
`chronic inflammatory bowel disease. Drugs 1989;37:73-
`86.
`2 Anonymous. United States Pharmacopeia XX!.
`Rock-ville: United States Pharmacopeial Convention,
`1989 .
`3 Langenbucher F, Rettig H. Dissolution rate testing with
`the column method: methodology and results. Drug Dev
`Ind Pharm 1977;3:241-63.
`4 FIP Working group V Dissolution Tests. Guidelines for
`dissolution testing of solid oral products. Pharm Ind
`1981;43:334-43.
`5 Langenbucher F. Parametric representation of dissol·
`ution curves by the RRSBW distribution. PharJ!ll Ind
`1976;38:472-7.
`6 Lee EJD, Bang SB. Simple and sensitive high-perform(cid:173)
`ance liquid chromatograph assay for 5-aminosalicylic
`in serum. J
`acid and acetyl.aminosalicylic acid
`Chromatogr 1987;431:300-4.
`7 Terpstra IJ, Bavelaar JF, Klooster NTM, Groenendaal
`JW, Hespe W. In vitro dissolution of 5-aminosalicylic
`acid delivering compounds [Abstract]. 13th Inter·
`na~ional Congress of Gastroenterology. Rome, Sept
`1988.
`8 Duchateau A, Philipse R, Van der Hoek E, Conemans J.
`pH influence on in vitro release of 5-ASA (Mesalazine)
`[Abstract). Pharm Week.bl [Sci] 1989;1l<Suppl E):ElO.
`9 Evans DF, Pye G, Bramley R, Clark AG, Dyson TJ,
`Hardcastle JD. Measuxement of gastrointestinal pH
`profiles in normal ambulant human subjects. Gut 1988;
`29:1035-41.
`
`204
`
`Pharmaceutisch Weekblad Scientific edition
`
`12(5) 1990