Attachment 3a: Copy of Document 3 from the Rutgers University Library
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 176
`
`

`

`Attachment 3a: Copy of Document 3 from the Rutgers University Library
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 177
`
`

`

`Attachment 3a: Copy of Document 3 from the Rutgers University Library
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 178
`
`

`

`Attachment 3a: Copy of Document 3 from the Rutgers University Library
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 179
`
`

`

`Attachment 3a: Copy of Document 3 from the Rutgers University Library
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 180
`
`

`

`Attachment 3a: Copy of Document 3 from the Rutgers University Library
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 181
`
`

`

`Attachment 3a: Copy of Document 3 from the Rutgers University Library
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 182
`
`

`

`Attachment 3a: Copy of Document 3 from the Rutgers University Library
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 183
`
`

`

`Attachment 3a: Copy of Document 3 from the Rutgers University Library
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 184
`
`

`

`Attachment 3a: Copy of Document 3 from the Rutgers University Library
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 185
`
`

`

`Attachment 3a: Copy of Document 3 from the Rutgers University Library
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 186
`
`

`

`Attachment 3a: Copy of Document 3 from the Rutgers University Library
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 187
`
`

`

`Attachment 3a: Copy of Document 3 from the Rutgers University Library
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 188
`
`

`

`ATTACHMENT 3b
`ATTACHMENT 3b
`
`
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 189
`
`

`

`Attachment 3b: Rutgers University Library catalog record for Document 3
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 190
`
`

`

`Attachment 3b: Rutgers University Library catalog record for Document 3
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 191
`
`

`

`ATTACHMENT 3c
`ATTACHMENT3c
`
`
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 192
`
`

`

`Attachment 3c: Statewide Illinois Library Catalog record (1) for Document 3
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 193
`
`

`

`Attachment 3c: Statewide Illinois Library Catalog record (1) for Document 3
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 194
`
`

`

`Attachment 3c: Statewide Illinois Library Catalog record (1) for Document 3
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 195
`
`

`

`ATTACHMENT 3d
`ATTACHMENT3d
`
`
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 196
`
`

`

`Attachment 3d: Statewide Illinois Library Catalog record (2) for Document 3
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 197
`
`

`

`ATTACHMENT 3e
`ATTACHMENT3e
`
`
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 198
`
`

`

`Attachment 3e: Google Scholar list of publications citing Document 3
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 199
`
`

`

`ATTACHMENT 3f
`ATTACHMENT3f
`
`
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 200
`
`

`

`488
`
`Journal of Crystal Growth 94 (1989) 488 498
`North-Holland, Amsterdam
`
`SERENDIPITOUS PREPARATION OF CRYSTALS OF METhOTREXATE AND ATFEMPTS
`TO MODIFY ITS CRYSTAL HABIT
`
`Hak-Kim CHAN * and Igor GONDA * *
`Department of Pharmacy, University of Sydney, Sydney, NSW 2006, Amtralia
`
`Received 15 September 1988; manuscript
`
`received in final form 21 October 1988
`
`A number of techniques were tried to obtain crystals of methotrexate (MTX), but without any success. Tetragonal crystals of this
`substance were finally obtained while attempting to prepare a complex of MTX with thymidine. Crystal habit modification of MTX
`was investigated using different solvents, surfactants, “tailor-made” additives, dyes and miscellaneous other substances as well as
`physical factors (supersaturation, rate of cooling, degree of agitation, temperature, and growth on substrates and in a gel). Two other
`solid forms, in addition to the tetragonal crystals and the original powder, were found: one which has the same unit cell as the
`tetragonal crystals but presents itself as spheres which consist of aggregates of small tetragons, and an amorphous form.
`
`1. Introduction
`
`Methotrexate (MTX) (fig. 1) is a widely used
`chemotherapeutic agent. We have studied the solid
`forms of MTX [1 4] with the view to optimize its
`aerodynamic properties for the direct administra-
`tion into the human respiratory tract in the form
`of an aerosol for the treatment of lung cancer. The
`sites of deposition of the inhaled drug particles
`depend on the size, shape and density of these
`particles [51.Together with the drug solubility, the
`regional distribution dictates the concentration of
`released drug able to exert therapeutic and toxic
`effects [6]. Therefore,
`the design of suitable re-
`spirable particles of MTX and the knowledge of
`their properties are essential prerequisites of
`meaningful biological studies.
`The commercially available MTX is a powder
`with a low degree of crystallinity [3,41.Initially, we
`had investigated methods
`to crystallize MTX.
`Having succeeded in the preparation of well-dc-
`veloped tetragonal crystals of MTX [1 4], we at-
`tempted to modify the crystal habit of this sub-
`
`* Present address: College of Pharmacy, University of Mm-
`
`nesota, Health Science Unit F, 308 Harvard Street S.E.,
`Minneapolis, Minnesota 55455, USA.
`** Correspondence to this author,
`
`stance. Numerous techniques were found acciden-
`tally, developed, or proposed, in the past to change
`the external shape of crystals of the same poly-
`morph. Among the “chemical factors”, much at-
`tention has been paid to the influence of the
`solvent on the crystal habit of the solute. Exam-
`ples of
`such studies
`are given in
`table
`1.
`Solute solvent interactions have usually been used
`to interpret
`these effects [7 11]. Watson [12] re-
`ported that
`the surface morphology of a solute
`depended
`on
`solvent
`solute
`association and
`Ananikyan et a!. [13] found that the best formed
`crystals of potassium iodate and pentaborate were
`grown from the least associated solutions. There
`is, perhaps, even more extensive literature on the
`effect of “additives”, or “impurities”, on the
`crystal habit, as it
`is especially relevant
`to in-
`dustrial crystallization (e.g., refs. [20 26]). Specific
`chemical interactions have been used successfully
`to modify the crystal habits of organic molecules
`with “tailor-made” additives;
`these have molecu-
`lar structures similar to the major component of
`the crystal and their effect can be interpreted as
`
`crystal growth inhibition in specific directions
`[11,27 33]. Earlier investigators [34 36] used dyes
`
`for the same purpose. Many dye moleculs have
`functional groups capable of hydrogen bonding or
`
`0022-0248/89/$03.50 © Elsevier Science Publishers B.V.
`(North-Holland Physics Publishing Division)
`
`Attachment 3f: Copy of a publication citing Document 3
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 201
`
`

`

`H. -K C/ian, I. Gonda / Serendipitous preparation of crystals of merhotrexate
`
`coo
`
`489
`
`OCH
`
`i
`
`~olete
`
`NH
`
`/coo_
`
`H2N-<~\’~H~~OCH3
`
`Trimetrexate
`
`NH
`
`2
`
`0 ~
`
`H3C_~j~
`
`Et
`
`2
`
`~~-CH2
`
`-
`
`Methotrexate (MTX)
`
`NSC 181928
`
`H2
`
`N
`
`CH
`
`COOC2H5
`NH—CH
`
`‘CODC2H5
`
`NH2
`Et02CH~~
`
`MIX — ethyl acetate
`
`NSC 350386
`Fig. 1. The chemical structures of inethotrexate and some of the additives used in the habit modification studies.
`
`CH3
`
`interactions via the electron cloud of the ~r dcc-
`trons of the aromatic rings. They could be easily
`visualized if they adsorb on the crystal and specific
`adsorption to selected crystal faces could be read-
`ily detected. Similar interactions have been pos-
`tulated to explain the effects of surfactants on the
`habit of adipic acid [14] and gypsum [37] while
`Garti et a!.
`[38] suggested that
`the surfactants
`changing the crystal habit of stearic acid were
`causing the formation of different polymorphs.
`Physical factors have been known for a long time
`to affect the habit of crystals. Thus, the degree of
`supersaturation often plays an important part in
`the shape of crystals [14,15,17,20,39]. An empirical
`equation suggesting that elongated crystals are
`more likely to form at high supersaturation was
`reviewed by Haleblian [39]. Related to this prob-
`
`ably is the effect of high rate of cooling which
`often leads to the formation of extreme shapes of
`crystals such as long needles or dendrites (e.g.,
`refs.
`[12,17,20]).
`Temperature
`is
`an
`implicit
`parameter in factors such as supersaturation and
`rate of cooling but it was suggested that the sole
`effect of the crystallization temperature on the
`habit
`is unimportant
`if phase changes are ex-
`eluded [40]. In contrast, Khamskii [41] did report
`promotion of non-isometric forms purely as a
`result of high temperature. It should be also men-
`tioned that temperature may have a hidden effect
`when it
`is linked to the influence of impurities
`[40]. Increased agitation which promotes mass and
`heat
`transfer,
`is thought
`to enhance interfacial
`supersaturation and nucleation. This would be
`likely to encourage the formation of small, rela-
`
`Attachment 3f: Copy of a publication citing Document 3
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 202
`
`

`

`490
`
`H. K. Chan, I. Gonda / Serendipitous preparation of crystals of methotrexate
`
`Table I
`Survey of reported effects of solvents on crystal habit of various substances
`
`Compound
`
`lodoform
`
`Resorcinol
`
`Adipic acid
`
`Salol
`
`Habit
`
`Hexagonal bipyramids
`Hexagonal prisms
`
`Compact crystals
`Very fine needles
`
`Needles
`Hexagonal plates
`
`Compact crystals or plates ~
`Only plates
`
`Solvent
`
`Aniline
`Cyclohexane
`
`Water
`Benzene and carbon tetrachloride
`
`Vapour or non-polar solvents
`Water
`
`Alcohols
`benzene, carbon tetrachloride
`acetone
`
`Phthalic acid
`
`Various habits
`
`Polar solvents
`
`Oxalic acid
`
`Nitrofurantoin
`
`Succinic acid
`
`Acetanilide
`
`Prismatic crystals
`Tabular habits
`
`Tabular to needles
`Plates b)
`
`Plates
`Needles
`
`Acetone-water mixture
`Water
`
`Formic acid
`Formic acid/H
`
`Water
`Isopropanol
`
`20 or ethanol
`
`Long needles
`Less elongated crystals
`
`Benzene
`Dimethylsulphoxide, acetone, alcohols
`
`~ Also depending on supersaturation and temperature.
`b) Also depending on solvent volume ratio, stirring, etc
`
`Reference
`
`[7]
`
`[7]
`
`[14]
`
`[15]
`
`[9]
`
`[16]
`
`[17]
`
`[18]
`
`[19]
`
`as observed with
`crystals
`tively isometric
`mtrofurantoin [17] and ammonium chloride [41].
`Other physical factors reported to have caused
`changes of crystal habit are, e.g., pH [42] and the
`presence of solid substrates [43]. It should be
`emphasized that the spectrum of opportunities for
`the changes of the crystal habit is much increased
`if different polymorphic forms are taken into con-
`sideration. The distinction between different
`crystal habit due to purely external changes of
`shape as opposed to polymorphism needs to be
`appreciated [44 47] as the latter has important
`implications for the physical and chemical stabil-
`ity of the product.
`
`2. Materials and methods
`
`All chemicals used were of analytical grade, or
`better, unless stated otherwise.
`Preliminary experiments on crystal habits of MTX
`in different solvents. 2 mg of MTX (American
`
`Cyanamid Company, Pearl River, NY, USA) was
`dissolved in 1 ml of solvents and heated up if
`necessary. The volume of
`the solvent was in-
`creased up to 10 ml if dissolution was difficult.
`The solutions were allowed to cool and evaporate
`at
`room temperature for crystallization. The
`solvents used were double distilled water,
`methanol, ethanol, butanol, octanol, cyclohexane,
`carbon tetrachioride, benzene, dimethylsulphoxide
`(DMSO) and 0.1N HC1.
`Attempts to obtain MTX crystals from the vapour
`phase. MTX in a round-bottom flask was heated
`in an oil-bath. The flask was connected to a
`vertical condenser tube with circulating cold water
`to allow condensation of any vapour evolved. The
`system was under vacuum (— 50 ftTorr). Tempera-
`ture of the bath was increased gradually and the
`condenser was observed carefully for sublimation.
`No sublimates were formed. The temperature was
`then fixed at 150 160°Cfor 1 h. Afterwards, the
`temperature was further increased until the MTX
`melted with decomposition at > 200°C. Any sub-
`
`Attachment 3f: Copy of a publication citing Document 3
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 203
`
`

`

`H. -K. C/ian, I. Gonda / Serendipitous preparation of crystals of methotrexate
`
`491
`
`limates formed were collected and dissolved in
`0.1N NaOH for characterization by UV visible
`spectroscopy (Lambda 5 UV/VIS, Perkin Elmer,
`USA).
`Crystallization in thepresence of thymidine. MTX
`was crystallized from aqueous solutions contain-
`ing different amounts of thymidine (Calbiochem,
`La Jolla, CA, USA) as follows: 2 mg of MTX
`powder was weighed into a specimen tube (2 inch
`x 1 inch).
`1 ml of double distilled water and
`thymidine as required were added and heated in a
`90 100°C water bath to dissolve the MTX. The
`weight ratios of MTX/thymidine were 1 :0.5, 1: 1
`and 1 : 5. The hot solutions were then allowed to
`cool to room temperature for spontaneous nuclea-
`tion and crystallization. Afterwards (overnight),
`drops of solution containing the solid were ob-
`served under an ordinary optical microscope. A
`zoom stereomicroscope (Kyowa, Trinocular model
`SDZ-Tr-P, Japan) was subsequently used to ob-
`serve the crystals inside the specimen tubes in-situ,
`
`2.1. Systematic studies of habit modification of MTX
`
`Solvent effect. 2 mg MTX was dissolved in hot
`solvent, volume 1 10 ml as necessary, and allowed
`to cool
`to room temperature for crystallization,
`unless described otherwise. The preliminary work
`(see above) showed MTX to be insoluble in non-
`polar solvents. Therefore, only polar solvents, or
`their mixtures with water, were used.
`Surfactants. Cationic, anionic and non-ionic
`surfactants (table 2) were used. The quantities
`employed were adjusted so that the effects below
`
`Table 2
`Surfactants used in the crystal habit modification studies of
`methotrexate
`
`Surfactant
`
`Molecular Cntical
`weight
`micelle
`(Dalton)
`concentra-
`tion
`(g/dm3)
`1.06
`364
`Cetrimide
`340
`Cetylpyridinium chloride (CPC)
`0.17
`Dodecylpyridinium chloride (DPC)
`0.80 8.26
`284
`Sodium lauryl sulfate (SLS)
`2.36
`288
`Tween8O
`0.014
`1240
`_____________________________________________
`
`the critical micelle concentration could be also
`observed.
`“Tailor made” addttives. The compounds with
`structures similar to MTX were used (fig. 1). They
`were folic acid (Hopkin and Williams Ltd., Essex,
`UK), 1-deaza-7,8-dihydroptendines (NSC 181928
`and 350386)
`(Southern Research Institute,
`Alabama, USA),
`trimetrexate
`isethionic acid
`(Warner-Lambert Company, Michigan, USA) and
`methotrexate ethylacetate (synthesized by acidic
`esterification of MTX in ethanol). Because of the
`low solubility of NSC 181928, in addition to its
`saturated solution, different dilutions of this corn-
`pound were also prepared from a stock solution of
`1 mg/mi in dimethylformamide (DMF).
`Dyes. Water-soluble dyes methylorange, meth-
`ylene blue and amaranth were used.
`Other additives. These were selected for a variety
`of reasons: capability to disturb hydrogen bond-
`ing (urea), or their known effect on crystal habits
`of other crystals (Al3~,Cr3~and Fe3~)[48]. The
`optically active aminoacids were used for their
`potential
`to interact selectively with the chiral
`glutamate moiety in the MTX [27];
`the L- and
`D-glutamic acids were employed for similar rca-
`sons.
`
`2.2. Physical factors
`
`Supersaturation. 0.2, 0.3, 0.5 and 1 mg/ml hot
`aqueous solutions of MTX were allowed to cool to
`room temperature for crystallization.
`Rate of cooling. 1 ml of 2 mg/ml MTX in a 15
`ml specimen tube dissolved at 900 C was rapidly
`cooled in an ice bath. The nuclei, which formed
`immediately, were allowed to grow in the ice bath
`for several hours.
`
`.
`
`.
`
`Agitation. The nuclei were prepared as in the
`previous experiment. The suspension was stirred
`for various length of
`time (0.5 3 mm) with a
`
`magnetic bar and then allowed to stand un-
`disturbed for crystallization.
`Temperature effect. Crystallization of MTX
`aqueous solutions, S ml of 2 mg/mI dissolved at
`90°C, was allowed to proceed by slow evapora-
`tion in oil baths kept at 50 and 70°C.
`Growth on substrates. Slow cooling of hot
`saturated MTX aqueous solutions in the presence
`
`Attachment 3f: Copy of a publication citing Document 3
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 204
`
`

`

`492
`
`H. K C/ian, I. Gonda / Serendipitous preparation of crystals of methotrexate
`
`of the following surfaces was tested: cellophane
`sheet, dialysis membrane, sintered glass, porcelain
`and paraffin wax.
`1 ml of saturated aqueous
`Growth in a gel.
`MTX solution was placed on top of a layer of
`congealed gelatin in a specimen tube and allowed
`to crystallize
`
`0
`
`3 Results and discussion
`
`The origmal matenal as supplied from the
`manufacturers is a fine powder consisting of
`anhedral particles and showing a low degree of
`crystallimty [3,4]. Our first task, therefore, was to
`prepare crystals of MTX. In the preliminary ex-
`periments, it was found that MTX was practically
`insoluble in solvents of
`low polarity (butanol,
`octanol, cyclohexane, carbon tetrachloride and be-
`nzene). Water, methanol and ethanol gave anhedral
`particles when the solutions were allowed to
`evaporate at room temperature. Dimethylsulpho-
`xide (DMSO) is an extremely good solvent for
`MTX; however, it evaporates only very slowly at
`room temperature. When water was added to the
`MTX solution in DMSO, small solid aggregates
`were formed. The same solid was observed on
`cooling a supersaturated MTX solution. Attempts
`to form MTX by sublimation and condensation of
`MTX vapour failed: no observable sublimation
`took place at temperature below 160°C.On melt-
`ing, white solid product appeared on condensation
`but UV spectra of this material showed that it was
`different from MTX, presumably a decomposition
`product. Well-shaped tetragonal crystals were
`formed when MTX was crystallized in the pres-
`ence of thymidine (fig. 2); some of
`these were
`found to be twins, or aggregates, under stereo
`zoom microscope. As the thymidine concentration
`increased, the crystals were becoming to be round
`with irregular surface. Extensive investigations by
`spectroscopic and chromatographic techniques
`showed that MTX was not forming a complex
`with thymidine. The new crystalline material was
`in fact found to be a polymorph of MTX distinct
`from the original powder which also retained its
`optically active configuration [3,4]. In order
`to
`deduce why thymidine induced the formation of
`
`a
`
`Fig. 2. Micrograph of tetragonal crystals of MTX.
`
`tetragonal crystals of MTX, we tested the effects
`of (i) sugars (cf. the sugar moiety, ribose, in the
`thymidine) by adding either lactose or glucose (1
`mg/mi) to MTX solutions (ii) pH (adjusted to 4.7
`or 5.7 by HC1) and we also carried out controls by
`recrystallizing MTX from double distilled water.
`To our surprise, all the above systems including
`the controls now gave even better formed tetrago-
`nal crystals of MTX than in the original experi-
`ment with thymidine.
`Interestingly, Sutton et a!.
`[49] reported totally independently from us on the
`preparation and single crystal X-ray diffraction
`analysis of the same solid form of MTX at almosi
`the same time when we did [2].
`
`~i1Tflfl
`
`Fig. 3. Micrograph of spherically shaped crystals of MTX.
`
`Attachment 3f: Copy of a publication citing Document 3
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 205
`
`

`

`Table 3
`Effect of solvents on solid form of methotrexate
`___________________________________________
`Solvents
`Habits
`
`H.-K Chan, I. Gonda / Serendipitous preparation of crystals of methotrexate
`values (0.2—0.5 g/dm3), rounding off of the tetra-
`gons occurs and at high concentrations, crystal
`growth is inhibited and only very small particles
`are formed. Similar effects are exhibited by dyes
`(table 5) and “tailor-made” additives (table 6) as
`well as a variety of other compounds (table 7).
`The latter group contains a hydrogen bond effec-
`
`Anhedral particles
`
`493
`
`Dimetbylsulphoxide
`(DMSO)
`Dimethylformamide
`(DMF)
`
`DMF/H
`
`20
`DMSO/H20
`
`Glycerol
`
`Glycerol: H 20 (%v/v)
`50: 50
`20:80
`10:90
`
`No crystals formed
`
`tor urea, a clathrate forming agent /3-cyclodextrin,
`
`Very small particles with size
`smaller than the original powder
`
`Very fine particles, no crystals
`
`Very fine particles, plus somestepped spheres
`Well-formed smooth and stepped
`spheres
`
`Table 4
`Effect of surface active agents on solid form of methotrexate
`________________________________________________
`Surfactant
`
`Concentration3)
`Habits
`Cetnmide
`
`(g/dm2 3.5
`Spherical particles
`(no crystals)
`
`0.2
`
`Smooth spheres
`
`Glycerol 3.5 x 10 2 g/dm3 Single and twin tetragons
`Propylene glycol
`Tiny spherical particles
`_____________________________________________________
`Acetone
`Particles are almost identical to
`Acetonitrile
`the original powder as MTX is
`insoluble
`_____________________________________________________
`Ethanol
`Anhedral particles
`Methanol
`
`Methanol: H
`
`6.0:6.0
`
`20 (v/v)
`
`5.5:6.0
`5.0:6.0
`4.5 :6.0
`4.0:6.0
`
`3.5:6.0
`
`No crystals formed at first,
`but spheres appeared
`2 days later
`
`Spheres
`Rosettes of tetragons
`and spheres
`
`Tetragon twins
`
`The effects of various solvents on the crystal
`habit of MTX are shown in table 3. Water seems
`to be essential to obtain well formed crystals of
`MTX. This is somewhat surprising in view of the
`relatively loose crystal structure of the tetragonal
`form of MTX [2]. The presence of glycerol, or
`methanol,
`in water tends to promote spherical
`aggregates of MTX (fig. 3).
`The influence of surfactants is also concentra-
`tion dependent (table 4). Generally, at low con-
`3), surfactants do
`centrations (10 ~ 10 2 g/dm
`not affect
`the habit of MTX. At
`intermediate
`
`1.34x10 2
`1.34x 10 ~
`1.34 x 10 ~
`1.34x10
`
`2.5
`
`0.5
`
`0.11
`1.25 x 10 2
`1.25 X 10
`1.25x10 ‘~
`1.25 x 10
`2.0
`
`L10X10 2
`1.10x10 ~
`1.10 x 10 ~
`1.lOxlO ~
`
`1.0 2.0
`
`0.5
`
`1.21 x 10 2
`1.21 x 10
`1.21 ~<10 ~
`
`1.21 <10
`2.0
`
`Single and twin
`tetragons
`
`Spherical particles
`
`Irregularly shaped
`spheres
`
`Single and twin
`tetragons
`
`Tetragons and spheres
`
`Singleandtwin
`tetragons
`
`Very tiny particles
`
`Mostly spheres with
`some tetragons
`
`Single and twin
`tetragons
`
`Smooth spheres
`
`CPC
`
`DPC
`
`SLS
`
`Tween 80
`
`Stepped spheres
`0.2
`Single and twin
`1O~’<10 2
`tetragons
`1.0x10
`LOxlO
`1.OxlO ~
`_____________________________________________
`
`Attachment 3f: Copy of a publication citing Document 3
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 206
`
`

`

`494
`
`H. -K. Chan, I. Gonda/ Serendipitous preparation of crystals of methotrexate
`
`Table 6
`Effect of “tailor-made” additives on solid form of methotre-
`xate
`_____________________________________________
`Additives
`Concentration Habits
`3)
`(g/dm
`0.2
`
`Fohc acid
`
`Spheres and aggregates
`of spheres
`Single and
`twin tetragons
`
`Rounding increased
`with additive concentration,
`spheres with smooth
`instead of
`stepped surface
`
`Tetragons with
`stepped surface
`and rounding
`No crystals formed
`
`Rosettes
`Spheres
`Spherical aggregates
`
`Very fine particles,
`no crystals
`Spheres of various
`
`sizes with
`aggregates
`
`Single spheres
`Smooth spheres
`
`Stepped spheres
`
`Single and
`twin tetragons
`
`a) Prepared from DMF solutions.
`
`grown in gelatin formed spheres but the purity of
`this new substance was not checked. It is, in fact,
`possible,
`that gelatin reacted chemically with
`MTX.
`
`4. Conclusions
`
`Methotrexate was shown recently to exist in the
`form of at
`least three different polymorphs. The
`
`NSC-350386
`
`2 X 10 2
`2 x 10
`2x10 ~
`2.0
`0.2
`2 X 10 2
`
`2 x 10
`
`NSC-181928 ~
`
`0.5
`
`0.1
`0.01
`Saturated
`
`aqueous
`solution
`
`1 2
`
`0.4
`
`0.2
`1.0
`
`0.2
`
`0.01
`
`Trimetrexate
`
`Table 5
`Effect of dyes on solid form of methotrexate
`________________________________________________________
`Dye
`Concentration
`Habits
`(g/dm3)
`
`MTX-ethyl
`acetate
`
`aminoacids which could interfere with the incor-
`poration of the glutamate moiety into the crystal
`lattice, and gelatin and trivalent cations reported
`to have affected the crystal growth of other com-
`pounds [48]. Five-fold variation of supersaturation
`had no effect on the crystal habit of MTX. Rapid
`cooling, on the other hand, prevented the forma-
`tion of crystals of MTX. Instead, anhedral prac-
`tides as in the original powder, were obtained.
`This was possibly due to excessive nucleation
`which depleted the mother liquor of MTX. Stir-
`ring only affected the size of the crystals but not
`the habit for time <2 ruin. Further stirring gave
`very fine particles. The explanation of these ob-
`servations is again probably in the depletion of the
`mother liquor by nucleation. Elevated temperature
`(50 versus 70°C) led to less well formed tetrago-
`nal crystals but not to a major change in the habit.
`The presence of solid substrates in the crystalliza-
`tion medium did not affect the habit: tetragons
`were formed either away from the substrates, or
`on them (e.g. on the sintered glass surface). MTX
`
`Methyl orange
`
`2.0
`
`2.2 >< 10 2
`
`Coloured, amorphous
`particles similar to
`the original powder
`
`Orange-coloured,
`mainly imperfectly
`shaped spheres
`and some aggregates
`
`2.2 x 10 ~
`
`Single and
`twin tetragons
`__________________________________________________
`Green coloured,
`Methylene blue
`3.89 X 10 2
`mainly spheres some
`still retaining the
`tetragonal shape
`
`3.89 X 10 ~
`
`Amaranth
`
`2.0
`
`Single and
`twin tetragons
`Irregularly-shaped
`aggregates deep-red
`coloured
`
`6.7 x 10
`Red-coloured spheres
`2
`6.7 x 10 ~
`Single and
`twin tetragons
`___________________________________________
`
`Attachment 3f: Copy of a publication citing Document 3
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 207
`
`

`

`H. -K C/ian, I. Gonda / Serendipitous preparation of crystals of methotrexate
`
`495
`
`Table 7
`Effect of miscellaneous compounds on solid form of methotrexate
`
`Additives
`
`Urea
`
`$-cyclodextnn
`
`Concentration
`3)
`(g/dm
`1.5
`5.0 x 10 ~
`
`5.0 x 10
`
`50
`
`5.0
`
`0.5
`sxio 2
`5x10 ~
`5x10 ~
`
`Habit
`
`Stepped spheres
`
`Single and twin tetragons
`
`Single tetragons highly favoured
`
`Solubilization
`
`Spheres
`
`Single and twin tetragons
`
`dl-Histidine
`
`20(5 mg MTX used)
`
`Solubilization
`
`2.5 (5 mg MTX used)
`
`Few smooth spheres
`
`1.0
`0.2
`
`20
`
`2 0.5
`
`0.2
`
`2.0
`
`6.5
`
`2.0
`1.0
`
`0.65
`6.5 x 10 2
`6.6x10 ~
`
`10
`
`5.5
`2.5
`
`2.0
`0.67
`6.7x10 2
`
`1.2
`
`0.5
`0.2
`
`0.1
`
`0.1
`
`L-Orrnthine
`
`L Lysine
`D-Glutamic acid
`
`L-Glutamic acid
`
`Gelatin
`
`A1CI
`
`3
`CrCl3
`FeCl3
`
`Spheres and tetragons
`
`Very fine particles
`
`Stepped spheres
`
`Single tetragons with a few spheres
`
`Single and twin tetragons
`Very fine particles, no crystals
`
`Stepped spheres
`
`Single and twin tetragons
`
`Single tetragons
`
`Very fine particles
`
`Both stepped and smooth spheres
`
`Tetragons, some round and some twins
`
`Spheres and ellipses
`
`Mostly ellipses with a few spheres
`
`Spheres
`
`Fine particles only, no crystals
`(M’FX not fully dissolved as solubility in solutions is low)
`
`ix i0~
`lx iO~
`
`Single and twin tetragons in Aid3 twin tetragons and stepped spheres
`in CrCl3 stepped and smooth spheres in FeCl3
`
`Attachment 3f: Copy of a publication citing Document 3
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 208
`
`

`

`496
`
`~
`
`~
`
`,~
`
`~
`
`~
`
`O~imm
`
`Fig. 4. Micrograph of spherically shaped “rosettes” of MTX.
`
`form which exhibits the greatest tendency to exist
`as well-formed crystals, appears to be the most
`stable polymorph of MTX [1 4]. It is interesting
`that this solid was prepared in an attempt to make
`a crystalline complex of MTX with thymidine
`because it was thought at the time that well formed
`crystals of pure MTX could not be obtained,
`Following the success with preparation of
`the
`distinct tetragonal crystals of MTX, various meth-
`ods to modify its crystal habit were employed.
`Water seems to be esential for the formation of
`crystals of MTX of observable size; this is prob-
`ably the consequence of the role of water in the
`crystal
`lattice of MTX [2]. Mixtures of polar
`solvents with water and other additives all showed
`a similar concentration-dependent effect on the
`habit of MTX: at
`low concentrations, formation
`of spheres was observed; these were shown to be
`the same polymorph as the tetragonal crystals of
`MTX [3,4] and were probably mainly aggregates
`of microscopic tetragons of MTX (fig. 4). At higher
`concentration, the effect of the additives was dis-
`ruption, or prevention of formation, of the crystal-
`line structure.
`Rounding of crystals is an interesting phenome-
`non which is often desirable to improve the flow
`and compaction properties of
`solid materials
`[50 56]. Although it is not practically relevant for
`MTX which is a drug normally used in an injecta-
`ble form, nevertheless the present results suggest
`that a relatively simple manipulation may cause
`
`H. -K. Chan, I. Gonda/ Serendipitous preparation of crystals of methotrexate
`
`formation of round solid particles during crystalli-
`zation. We could suggest two explanations for this
`phenomenon. Firstly, inhibition of rate of crystal
`growth can give rise to the formation of a large
`number of small crystals which will agglomerate
`to minimize the total surface energy of the system.
`At high concentrations of additives, the inhibition
`of
`crystal growth is
`complete
`and,
`instead,
`amorphous MTX is formed Secondly we may
`consider the non specificity of the rounding effect
`to be due to intermolecular interactions at
`the
`growing crystal faces but we have to postulate that
`all
`faces of MTX are approximately equally af-
`fected by all the additives tested. This could be
`true in the general sense for MTX since all the
`faces contain both hydrophilic (e.g.
`NH2 in
`pteridine and COOH in glutamate) and hydro-
`phobic (e.g. the benzoyl and pteridine rings) groups
`[2]. Therefore, the additives, through their hydro-
`phiuic and hydrophobic parts could affect
`the
`growth of
`the nuclei, or small crystals,
`in an
`isotropic fashion. It may seem surprising that even
`“tailor-made”
`additives could behave in this
`manner. However, if one considers a typical repre-
`sentative of this class of compounds (fig. 1) folic
`acid,
`the isotropic influence can be explained.
`Folic acid has a chemical structure similar
`to
`MTX and would be therefore expected to form
`hydrogen bonds along the c-axis [2] and thus
`inhibit growth in that direction by breaking the
`chain of MTX molecules. However, since hydro-
`gen bonding is possible along other crystal direc-
`tions as well [2], the overall effect of folic acid
`could be quite isotropic. The fact that we encoun-
`tered initially a very strong resistance of MTX to
`form well formed crystals is, perhaps, an indica-
`tion that the crystal growth of this substance can
`be relatively easily inhibited, or disrupted, leading
`to the appearance of particles formed under the
`influence of non-specific, isotropic interactions.
`
`Acknowledgements
`
`We are grateful to Ms. Sandy Butler for typing
`the manuscript, to Dr. Ian Threadgold from the
`Department of Geology for encouragement and
`help with crystallography and to Dr. Andrew
`
`Attachment 3f: Copy of a publication citing Document 3
`
`Par Pharm., Inc.
`Exhibit 1004
`Page 209
`
`

`

`H. -K Chan, I. Gonda / Serendipitous preparation of crystals of methotrexate
`
`497
`
`Cheung for general advice on methotrexate. The
`following chemicals were supplied to us kindly
`free of charge: methotrexate (American Cyanamid
`Company, Pearl River, USA), 1-deaza-7,8-dihy-
`dropteridines (NSC 181928 and 350386, Phar-
`maceutical Chemistry Division, Southern Re-
`search Institute, USA), tnmetrexate isethiomc acid
`(Pharmaceutical Research Division, Warner-Lam-
`bert Company, USA) and /3-cyclodextrin (Nikon
`Shokuhin Kako Co. Ltd., Japan). H.-K.C. was
`supported by a scholarship as a part of a Umver-
`sity of Sydney Special Project Grant to I.G.
`
`References
`
`[1] H.-K. Chan, T.W. Hambley and I. Gonda, Australian J.
`Hosp. Pharm. 16 (1986) 66.
`[21 T.W. Hambley H.-K. Chan and I. Gonda, J. Am. Chem.
`Soc. 108 (1986) 2103.
`[3] H.-K. Chan, Crystal Growth and Aerodynamics of Drug
`Particles, PhD Thesis, University of Sydney (1988).
`[4] H.-K. Chan and I. Gonda, Intern. J. Pharmaceut., sub-
`mitted.
`The Science of Dosage
`in: Pharmaceutics
`[5] I. Gonda,
`Form Design, Ed. M.E. Aulton (Churchill, Livingstone,
`Edinburgh, 1988) pp. 341 358.
`[6] I. Gonda, J. Pharm. Sci. 77 (1988) 340.
`[7] A.F. Wells, Phil. Mag. 37 (1946) 184.
`[8] A.F. Wells, Disc. Faraday Soc. 5 (1949) 197.
`[9] R.J. Davey, in: Current Topics in Materials Science, Vol.
`8, Ed. F. Kaldis (North-Holland, Amsterdam, 1982) pp.
`431 479.
`[10] Z. Berkovitch-Yellin, J. Am. Chem. Soc. 107 (1985) 8239.
`[11] L. Addadi, Z. Berkovitch-Yellin, I. Weissbuch, J. Van Mu,
`L.J.W. Shimon, M. Lahav and L. Leiserowitz, Angew.
`Chem. Intern. Ed. EngI. 24 (1985) 466.
`[12] D.H. Watson,
`in: Proc. 3rd Intern. Conf. on Electron
`Microscopy, London, 1954, p. 497.
`[13] T.A. Ananikyan, A.G. Nalbandyan and H.G. Nalbandyan,
`J. Crystal Growth 73 (1985) 505.
`[14] A.S. Michaels and A.R. Colville, J. Phys. Chem 64 (1964)
`13.
`[15] W. Kleber and H. Raidt, Z. Physik, Chem. 222 (1963) 1.
`[16] J.L. Torgesen and J. Strassburger, Science 146 (1964) 53.
`[17] N. Garti and F. Tibika, Drug Dev. md. Pharmacy 6
`(1980) 379.
`[18] R.J. Davey, J.W. Mullin and M.J.l. Whiting, J. Crystal
`Growth 58 (1982) 304.
`[19] M.C. Etter, D.A. Jahn and B.S. Donahue, J. Crystal
`Growth 76 (1986) 645.
`[20] J.W. Mullin, Crystallization (Butterworths, London, 1972).
`[21] J.W. Mullin, Ed., Industrial Crystallization (Plenum, New
`York, 1976).
`
`[221R.J. Davey, in: Industrial Crystallization 78, Eds. E.J. de
`J