Water—Inso|ub|e
`Drug
`Formulation
`
`I I I
`
`Edited by
`Rong Liu
`
`‘9 Interpharm/CRC
`
`B<)c;1R;lLm1 London New York
`
`\X/'.1shil1gLnr1.I).C.
`
`MYLAN - EXHIBIT 10
`
`MYLAN - EXHIBIT 1025
`
`

`
`Library ol' Congress Cataloging-in-Publication l)ata
`
`
`Wztter-insoluble drug lormulzltion I Rong l.iu, editor.
`p.
`cm.
`'
`Includes l7ll\ll()gl'11[)l]lC1ll relet‘ei1ccs and index.
`ISBN I—574‘)l-l05~8
`
`()U—()33~lS(l
`
`I. Solutions (|’l1arnr.tcy) 2.
`RSZUI .S()W38
`2()()()
`
`6| 5'42~«dt:2 I
`
`|)rugs'—Soluhility,
`
`l. Liu. Jung.
`
`is quoted with
`This book contains inlorinution obtained from authentic and highly regarded sources. Reprinted ll1L1lL’I'lLtl
`permission, and sources are indicated. A wide vmiety of rctcrences are listed. l{ensonuh|e efforts have been made to publish
`reliable data and intbrmution, but the authors and the publisher cannot ztsstllne responsibility for the validity ofall Intttcriuls
`or for the consequences of their use.
`
`Y
`Y
`-
`an metms clccmmic or Illccllallliclll.
`form or h
`in an
`he re noduecd or lI'L1llSl11lllt.‘Ll
`art mat
`Neither this boo]; nor an
`illcludillg [3h0l0CODyirI:_1. Illltlvlllllllllg. and recording, or hy nny inl'orinnlion storage or retrieval system, \vithoul prior
`permission in writing from the puhlis‘lter.
`
`The consent ol'Cl{C Press LLC does not extend to copying for general distribution, for promotion, for creating new works,
`or for resale. Specific permission must be Ol7lLlll1C(l in writing from CRC I’ress LLC for such copying.
`
`Direct all inquiries to CRC l’rcss l.l.(.‘. 2000 N.W. Corpontle l3lvd., lioeu Raton, liloridu 3343 I.
`
`'l'r:t(lenIark Notice: Product or corporate IMIIIICS may he tI':1t|emnI'ks or rcgis‘tcred trntletnmks, und are used only for
`identilication and explanation. without intent to ittfliitgc.
`
`Visit the CRC Press Web site at www.crcpress.c0in
`
`© 2000 by CRC Press LLC
`lnlerphawm is an imprint of CRC Press l.LC
`
`No claim to original U.S. (iovernrnent works
`ll1lCl‘llilll0i1tll St;1Itdm'd Book Number l~574‘)|-105-8
`Lihrziry o|' (‘ongI'ess Curd Nun1l3er()()-l)334S()
`Printed in the United Slnles of America
`I
`2 3 4 5 6 7 8 9 ()
`Printed on acid-lice paper
`
`

`
`Contents
`
`ix
`
`Preformulation Studies
`
`Formulation Development
`
`Methods of Manufacture
`
`Processing Issues
`
`Product Testing and Specifications
`
`Applications
`
`Future Perspective
`
`References
`
`15. Alteration of the Solid State of the Drug Substance:
`Polymorphs, Solvates, and Amorphous Forms
`
`Michael J. Joz,wiukow.s'ki
`
`Theoretical and Practicai Considerations
`
`Special Considerations for Polymorphs
`
`Solvates and Hydrates of Drugs
`
`The Utility of/\morphous (Noncrystalline) Forms
`
`Strategy forWater-Insoluble Drug Formulation Using Metastable
`Sohds
`
`References
`
`501
`
`502
`
`507
`
`511
`
`513
`
`515
`
`516
`
`517
`
`525
`
`526
`
`546
`
`549
`
`555
`
`561
`
`562
`
`16. Solubilization Systems—The Impact of Percolation Theory and
`Fractal Geometry
`569
`
`Hcms L(’Il(’Il]7(’I'_L{(’I‘
`Silvia Komvu El-A rim’
`
`Introduction to Percolation Theory
`
`References
`
`Experiences with a Novel Fluidized Bed System Operating under
`Vacuum Conditions
`
`Applications
`
`An Alternz-1t'ive Novel Process Technology to Obtain Well—soluble
`Substances: Atmospheric Spray—free7.e Drying
`
`570
`
`589
`
`591
`
`596
`
`597
`
`

`
`_l§_
`
`Alteration of the Solid State of the
`Drug Substance: Polymorphs,
`Solvates, and Amorphous Forms
`
`
`
`Michael I. Iozwiakowski
`3M Plzarmacezmeals
`
`St. Paul, M1Tm1.esoIa
`
`The maximum solubility of a drug substance is a function of the nature of the solid
`phase in equilibrium with a specified solvent system at a given temperature and pres-
`sure. Solubility is an equilibrium constant for the dissolution ofthe solid into the sol-
`vent and thus depends on the competition of solutezsolvent
`interactions and
`solid:solid interactions. Alteration of the solid phase of the drug substance can influ-
`ence its solubility and dissolution properties by affecting the molecular interactions
`in the solid.
`
`A crystal of higher free energy will yield an apparent higher solubility than a lower
`energy stable crystal form of the same molecular structure. In the lowest energy solid
`state, the energetically favorable solidzsolid interactions reduce the escaping tendency of
`the molecules, and thus fewer molecules dissolve in a given solvent under the same set
`of environmental conditions. Crystalliiie polymorphs, solvates and hydrates, and amor-
`phous forms of drug substances have been used to change the thermodynamic driving
`force for dissolution and to increase the apparent solubility of poorly soluble drugs.
`Unlike solul)ilizati()n techniques that change the nature of the solvent environ-
`ment (cosolvent systems, emulsions, micellization) or the chemical identity of the dis-
`solved solute (salt formation, complexation, pro-drugs), manipulation ofthe solid state
`of the drug substance results in only a transient change in the system. Since the solvent
`
`525
`
`

`
`526 Water-Insoluble Drug For/rmlrifiorz
`
`and the chemical form are identical, the system will ultimately revert to the lowest-
`energy solid phase in equilibrium with the solvent, with the lowest solubility. (Irystal
`growth and dissolution have been used to assign relative physical stability for poly-
`morphs by observing the direction of the transformation under a microscope under
`controlled temperature in contact with a solvent. The rate of tt'ansfor1nation in contact
`with a solvent is normally too fast to consider solution or suspension dosage forms of
`metastable solids. Systems with unusually large energy barriers, slow reversion kinet-
`ics, or excipients to retard crystallization can be useful in litnited circumstances.
`The most practical use ofthis technique is to alter the solid phase in dry dosage
`forms where molecular mobility is greatly reduced. Metastable forms of solid drugs
`are often stable to physical transformation in the time context required for marketable
`formulations. Tablets, capsules, lyophilized powders, granules for constitution, and
`other solid dosage forms are ideal systems for incorporation of metastable solid
`phases. In most cases, the brief exposure to gastrointestinal (GI) lluids does not result
`in conversion to the lower solubility form prior to generating the desired enhanced
`effect. Solid-state transformations and transformations induced by adsorbed water
`during long—term storage can still l)e problematic. Any consideration of formulating
`metastable solid phases must balance the expected gain in efficacy with the potential
`for reversion to the less—favorable form prior to patient use. This involves both an
`understanding of the phase diagrams (which forms are physicaily stable under which
`conditions) and the physical principles governing transformation kinetics.
`in this chapter,
`the theoretical and practical considerations for the use of
`metastable solids in formulations to gain a solubility or dissolution-rate advantag ' are
`explored. lixperiments are suggested that identify the potential solid forms of the drug
`and elucidate the potential advantages and disadvantages. Specific examples of the
`degree ofenhancement that can be expected and special considerations for each type
`ofsolid are covered (polymorphs, solvates, and amorphous forms).
`
`THEORETICAL AND PRACTICAL CONSIDERATIONS
`
`Importance of the Solid State of the Drug
`
`()rigz'n of the hjfect ofsolizl State on S0lubz'lity
`
`When a medicinal chemist discovers a new chemical entity (NCIE) with a desired phar-
`macological effect, structure-activity relationships are used to optimize the series for
`activity. Aqueous solubility, partition coefficient, crystallinity, melting point, particle
`size, and hygroscopicity, all of interest to the formulator of this NCIE, will also vary
`within the series ofdrug candidates. Because the biological activity is often estimated
`by target enzyme binding studies in very dilute media, solubility may not be opti-
`mixed simultaneously. If an ioni'/.al)le drug candidate is selected, the choice of free
`acid/base form versus the salt forms again produces a myriad of possible physical
`properties. 'l‘he alteration of solubility by judicious choice of the salt was covered in a
`
`

`
`/llI'emt’iorz oft/re Solid State oftlze Drug Subslarzcc
`
`527
`
`previous chapter. ln 111any cases the chosen salt or acid/base can crystallize in a vari-
`ety of possible arrangements, each of which l1as the possibility of different‘ physical
`properties. This includes the possibility of polymorphs, solvates, or noncrystalline
`(amorphous) forms. Thus, the solid phase chosen for development is the third deci-
`sion tnade by the pharmaceutical scientists that has a major impact on the ultimate
`physical properties ofthe NCIE, including the solubility. ln a typical development pro-
`gram, the number of candidates decreases at each stage:
`
`Stage 1. Selection of best chemical structure (1()0-1000)
`
`Stage II. Selection ofacid/base/salt form (3-25)
`
`Stage lll. Selection ofsolid phase for development (1-3)
`
`Each of these stages produces molecules ofvarying solubility by virtue of a change in
`the crystalline lattice, the last stage being the only one in which the chemical identity
`or counterion identity is unchanged.
`Yalkowsky (1081) has developed equations describing solubility as a function of
`both hydrophobicity and crystal lattice forces. This has led to the observation that
`melting point or heat effusion, both a function of the strength of forces holding mole-
`cules together in the solid state, can correlate with solubility within a homologous
`series. This is because the disruption ofthe crystal forces is a necessary prerequisite to
`the release ofindividual molecules into the solvent for dissolution. Grant and I-Iiguchi
`
`(1990) sunnnari'/.ed correlations in the case of diphenylhydantoin derivatives and
`substituted pteridines in their book on organic compound solubility. Wells (1988) has
`noted that in the series of phenols with hydroxy substituents, the high-melting para-
`form (hydroquinone) has a much lower solubility than the ortho- or meta-derivatives.
`Morelock et al. (1994) have used melting points and retention times to correlate with
`aqueous solubility in a series of reverse transcriptase inhibitors. They have found this
`useful in selecting the drug candidate that can possess optimum biological parame-
`ters and in guiding the further synthetic effort. The same factors can govern the solu-
`bility differences between drug-salt forms or solid phases, since, fundamentally, the
`change is brought about by a difference in solidzsolid forces in each case. Wells (1988)
`has shown that riboflavin polymorphs follow a similar inverse correlation with melt-
`ing point. Form lll melts at 180° to 185°C and is soluble to greater than 1000 mg/ml. in
`water. Forms 1 and ll melt at 270° to 290°C and have solubilities ofless than 100n1g/n1L.
`The success of altering solubility by manipulating the solid phase of the drug
`depends on which factor dominates the aqueous solubility behavior, the hydropho-
`bicity, or the lattice forces. When the molecule is too lipophilic to have adequate
`aqueous solubility, cosolvents, pro-drugs, or emulsions are effective in increasing the
`solubility. Altering the solid state i11 this case may have little effect on it.s solubility,
`since poor aqueous solubility is due to the molecular lipophilicity. In contrast, cosol-
`vents and emulsions do not have much impact if the reason for low solubility is the
`stability of the crystal lattice. When a drug has low solubility and a high melting point
`(>250°C), it is likely that disruption of the lattice is needed to increase the effective
`
`

`
`528 Water- Insoluble Drug I<'orm.ulm.‘iorz
`
`form (polymorphs or
`solubility. This can be done by altering the crystal
`solvates/hydrates) or by producing the amorphous form and stabilizing it to sponta-
`neous crystallization. The effect of altering the salt form of the drug is discussed in a
`separate chapter and has also been reviewed by Berge, Highly, and Monkhouse
`(I977).
`
`Historical Perspective and Defirzirimzs
`
`The effect of the solid state on drug solubility has been known for decades, and pio-
`neer articles by llaleblian and McCrone (1969); Shefter and lliguchi (1963); and
`Higuchi et al. (1963) have formed the basis for furtherstudies in this area. More recent
`reviews by Shefter (1981), /\bdou (I989), Byrn (1982), Wall (1986), Brittain (1995), and
`Fiese and liagen (1986) have summarized the effects ofpolyrnorphism or solvate for-
`mation on drug solubility. The existence ofdifferent internal crystalline arrangements
`for the same chemical structure has been termed polyIIz01'plu'.s‘m. Verma and Krishna
`(1966) have observed that the great majority ofsubstances seem to be capable of1nul-
`tiple solid states. The work of Kuhnert—Brandstatter in steroids and barbiturates (as
`reviewed by llaleblian and McCrone 1969) seems to indicate that simple organic drug
`molecules with multiple functional groups can arrange into numerous crystal-
`packing structures. McCrone, McCrone, and Delly (1987): Carstensen (1993); and Byrn
`(1982) have reviewed the seven different crystal systems, which are uniquely identi-
`fied by the length ofthe axes and the angles between them in the unit cell, Many drugs
`crystallize into multiple polymorphic forms, especially monoclinic,
`triclinic, or
`orthorhombic types (Wall 1986; Borka and llaleblian I990; Giron 1995); this relatively
`common diversity in solid forms gives the formulation scientist variations in physical
`properties to exploit.
`in general, the crystalline form with the closest packing (greatest density) and
`the highest melting point is the stable form. The stable form will have the lowest sol-
`ubility and the lowest free energy of the different solid phases of the drug. All other
`phases with the same composition are termed metastable forms at this temperature
`and pressure. in practical terms, the energy barrier for conversion to the stable form
`can be high enough that the metastable forms can be examined and formulated. This
`is especially true if the free energy difference is small (leading to a small driving force)
`or ifsignificant bond breaking, molecular motion, and bond formation are required
`for transformation.
`
`if the crystals contain solvent molecules within the lattice structure in defined
`locations and stoichiometry, these are referred to as solvates (hydrates if the solvent is
`water). The term psemlopolymorph has been used historically but is not as specific
`and should be avoided if the composition is known. in most cases, the solvated form
`of the drug is the least-soluble form in that solvent (e. g., hydrates are the low-energy
`forms in water). Due to additional mixing terms, solvates are often more soluble in a
`solvent of different composition than the nonsolvated crystal (Shefter and lliguclii
`1963). The types ofsolvates that appear in pharmaceutical systems and their proper-
`ties of interest are covered later in this chapter.
`
`

`
`/lllerarimz ofthe Solid State of the Dmg Sitbslmice 529
`
`Amorphous forms can be made for most pharmacettticals by producing the solid
`form faster than the molecules can arrange into a crystalline lattice. Noncrystalline
`solids may have some short-range order but lack the long-range periodicity and reg-
`ular intertnolecular bonding ofcrystalline solids. Their synthesis and properties differ
`markedly from their crystalline forms, as will be described in detail in the section on
`amorphous drugs. In general, amorphous forms are high-energy, low-density solids
`that can yield transient dissolution rates much greater than crystalline solids.
`Liqnirl crystals are an intermediate state in which the molecules in a crystal can
`undergo a secondary phase transition to a mesophase, which gives them tnobility in
`one or two directions. They are birefringent but possess [low properties like a liquid
`phase.
`l.yotropic liquid crystals form upon uptake of water into a system, which
`increases its tnobility, and thermotropic liquid crystals can be disrupted by heating
`above a transition temperature. Cromolyn sodium (Cox, Woodard, and McCrone
`1971 ), the l lM(}-(Jo/\ reductase inhibitor SQ 33(S()() (Brittain, Ranadive, and Serajuddin
`1995), and the leukotriene 1)‘! antagonist 1. 660,711 (Vadas, ’l‘oma, and Zografi 199])
`are examples ofpharmaceuticals that can form liquid crystals.
`The crystal lmbit or external shape may differ when drugs are recrystallized frotn
`different solvent systems without changing the internal structure. The presence of
`additives, the rate ofcooling, the degree ofagitation, and the degree ofsaturation can
`all affect crystal habit (Byrn 1982). llabit can affect bulk properties such as density and
`flowability ((Iarstensen 1993) or influence the ability to filter crystals during purifica-
`tion. Chow and Grant (1988) have shown that the dissolution rate ofacetaminophen
`can be altered two or three times by modifying the length-to~wid'th ratio through
`incorporation of additives. In general, habit effects on solubility are transient and ofa
`magnitude eqttivalent to particle-size—reduction techniques.
`In summary, most drugs are developed as crystalline forms, which have the
`greatest physical and chemical stability, so that their purity can be increased during
`recrystalli7,ation. Knowledge ofthe potential polymorphic forms may allow the devel-
`opment. scientist
`to find a metastable form with the prerequisite stability and
`increased dissolution rate to make it the desirable marketed form. Anhydrate forms
`usually give faster dissoltttion rates and higher aqueous solubilities than the hydrated
`form. Other solvates are not commonly used in pharmaceutical systems due to poten-
`tial toxicity of the solvent but may provide additional solubility enhancetnent. /\mor—
`phous forms have the highest free energy with the greatest degree of solubility
`enhancement but are the most difficult to stabilr/.e against transformation to the sta-
`
`ble crystal form. The remainder of the chapter describes the advantages and disad-
`vantages ofeach type ofsolid and gives numerous examples from the pharmaceutical
`literature where alteration of the solid form resulted in increased solubility.
`
`Methods to Stmly the Solid State dmingPrefi)rm.uI(1tion Screening
`
`After initial selection of the candidate drug and its salt form (if applicable), it is rec-
`ommended that a purposeful effort be undertaken to examine the solid states avail-
`
`able for development consideration. In the absence ofsuch programs, new crystalline
`
`

`
`530 Water-Insoluble Drug F0mmlcm.'0n
`
`forms may be discovered by accident from precipitation of less-soluble phases or a
`change in appearance of the bulk drug during scale—up. lfsufficient formulation, ana-
`lytical, and toxicology work has been completed and it becomes necessary to change
`the solid state, it may significantly delay the development program. ll‘ tl1e newly (lis-
`covered form is the more stable modification, it may be difficult to reproduce the
`metastable form. Carstensen (1993) noted that the diazepam tablet development pro-
`gram was complicated by the crystallization of a more stable polymorph after clinical
`trials had begun. If both forms can still be synthesized and they prove to be bioequiv-
`alent,
`the consequences of changing the form later in development are greatly
`reduced.
`
`A recent publication by Byrn et al. (1995) suggests a minimal preforrnulation
`program for examining the drug solid state and the regulatory implications ofvarious
`scenarios. The information that may be useful at this early phase includes the
`
`- Number ofsolid phases that exist for this drug
`
`- Relative physical and chemical stability of these phases
`
`0 Solubility ofeach form in relevant media
`
`- Resistance of metastable forms to conversion during processing
`
`0 Possible means to stabilize amorphous or metastable forms, ifneeded
`
`Another major consideration at this point is to decide on the patentability of any
`new crystal forms discovered to have significant practical advantages over those in the
`original patent. Byrn and Pfeiffer (1992) have listed >350 patents on crystal forms in
`the pharmaceutical patent literature granted for showing advantages in terms ofsta-
`bility, formulation, solubility, bioavailability, purification, hygroscopicity, preparationl
`synthesis, recovery, and prevention ofprecipitation.
`‘
`Different crystal forms can be sought by recrystallization experiments varying
`the solvent‘ system, temperature, precipitation method, and level ofsupersaturation.
`Precipitation methods may include slow evaporation of the solvent, addition of anti-
`solvents, or saturation at high temperatures followed by cooling. The solvent systems
`chosen for study are one of the key aspects of these experiments. Water must be
`included because ofits physiological significance, the possibility ofcontamination of
`crystallization solvents, and the possibility of moisture uptake during storage. Cer-
`tainly the usual recrystallization solvents for the drug synthesis and purification
`scheme and any alternate systems that may be used for scale—up of this process
`should also be studied. Mixtures used in the final steps ofthe synthetic process should
`be examined, including azeotropes and solvents with small amounts of miscible
`wat.er. A suggested list for solvents that have been known to produce different crys-
`talline forms is provided in Table 15.1 (Wells 1988; Byrn and Pfeiffer 1992; Byrn et al.
`1995).
`
`Solid forms isolated from these systems should be subjected to characterization
`techniques such as hot-stage or polarized light microscopy, differential scanning
`calorimetry (DSC),
`thermogravimetric analysis (TGA), X—ray diffraction (XRD),
`
`

`
`
`
`/tltemriou ofrhe Solid State ofllie Drug Substrmce 53]
`
`Table 15. 1
`
`Partial List of Solvents for Preformulation Studies Looking
`for Different Crystal Forms
`
`Water
`
`Ethanol
`Acetone
`
`Methanol
`
`lsopropanol
`Acetonitrile
`
`lithyl acetate
`Dimethylformamide
`Diethyl other
`
`llexane
`Methylene chloride
`Glacial acetic acid
`
`Any other solvents used in the last steps ofsynthcsis
`Aqueous mixtures with the above
`
`infrared spectroscopy (IR), t7’l‘-Raman spectroscopy, and solid-state nuclear magnetic
`resonance (NMR). The solubility oleach new form isolated should be studied in water
`or the solvent/solution of pharmaceutical interest. Composition versus vapor pres-
`sure data should be obtained to understand the stability of any hydrated forms with
`respect to the anhydrate and other stoichiometric hydrates. if the transition to the
`stable form is last, intrinsic dissolution measurements can be used to estimate the rel-
`
`ative solubilities of new crystal forms.
`’l‘he method used to identify new solid phases depends in part on the expertise
`of the investigator and the availability olequipment and in part on the properties of
`the solid phases. llaleblian and McCrone (1969) have demonstrated the power of the
`polari7.ed light microscope and hot-stage microscope in studying the phase relation-
`ships between solid forms versus temperature. Byrn (1982), Wall H986), and Surya‘-
`narayanan (1989) have noted that only X-ray tlillraction and single crystal X-ray stud-
`ies uniquely identify a solid phase unambiguously, since other methods depend on
`properties that may or may not change with a change in crystal lattice structure. if the
`new phase is a solvate, additional techniques are needed to identify the solvent of
`crystalli7.ation and its stoichiometry. Differential scanning calorimetry has been used
`extensively in polymorph investigations since the melting point is often the first ll‘t(ll-
`cation of a new crystalline form (Giron 1995). Other phenomena such as solid~st'atC
`transitions and water-loss endotherms a1'e often discernible horn the thertnograni.
`l.indenbaum and Mc(.}raw (1985) have shown how solution calorimetry can be used
`to assign relative enthalpy differences between polymorphs of drugs. It is best to use
`a combination of techniques to elucidate the nature ol‘ the new crystal forms, as one
`teclmique may not be able to dil'l'erentiate the forms. Fol‘ example, the two polymor~
`phic forms ol‘ amiloride IICI dihydrate could not be distinguished by IR spectra or
`microscopic morphology (lo7.wiakowski, Williams, and llathaway 1993) but are easily
`identified by X-ray dit't‘raction.
`ll‘ a metastable form is identilied that may be needed to produce the desired or
`optimal pharmacological effect, further studies should be done to define the relative
`stability of this form and the proper storage conditions to prevent conversion.
`/\
`metastable form should not be used when the more stable crystalline form produces
`
`

`
`532 W(tt'er—1/zsoluble Drug Formulation
`
`the desired effects; in this case the potential benefits probably do not offset the risk of
`conversion between production ofthe drug product and patient use. The stresses dur-
`ing manufacture of the product need to be studied as well, since compression,
`milling, granulation, and lyophilization can all change the solid state of the drug
`between synthesis and formulation. in some cases, appropriate precautions, resistant
`packaging, or stabilizing excipients can keep the drug in the optimum form despite its
`rnetastability. Once some of these properties are known, a rational decision can be
`made on whether to proceed with development ofa metastable solid based on a ther-
`apeutic need.
`
`Properties Dependent on the Solid State
`
`Once the different solid phases have been identified, their physical properties can be
`compared to find the form that is optimum for drug product development. Verma
`(1966) described polymorphic forms ofcommon substances that illustrated how their
`physical properties can be quite different. The cubic form of carbon (di2l1n0l1d) is
`hard, dense (3.5 g/cc), brilliantly clear, and a poor conductor. in contrast, the hexago-
`nal form of the same element (graphite) is soft, less dense (2.2 g/cc), dull in appear-
`ance, and a good conductor. Optical properties of different solid forms can change
`their color; mercuric iodide is red (tetragonal form) or yellow (orthorhombic form).
`Dt'ug substances, which tend to be larger organic molecules with multiple
`hydrogen bonding sites, are especially prone to different crystalline arrangements
`that produce variable physical properties. Table 152 lists physical and chemical prop—
`erties ofpharmaceuticals that have been cited in the literature as depending on the
`solid state of the drug. Some of these may provide significant advantages in drug
`development. Form 1 of celiprol ll(2l is much less hygroscopic than the form ll poly-
`morph (Narurkar et al. 1988). Form A chlorpropamide forms tablets with greater l1ard—
`ness than those ofform C under identical compression forces (Matsumoto et al. 1991).
`Phenobarbitone also has crystalline forms that differ in compressibility profiles (Shell
`1963). The two polymorphs of methylprednisolone have (lifferent chemical stability
`profiles when exposed to identical temperatures and humidities of storage (Munshi
`and Simonelli 1970). Crystal size and shape can alter the filterability or syringeability
`ofsuspensions and affect the weight uniformity of tablets and capsules.
`
`Properties of Drug Substances Dependent on the Solid State
`
`Table 15.2
`
`Solubility
`Dissolution rate
`Bioavailability
`Chemical stability
`lilowability
`Melting point
`Compressibility
`Particle size/shape
`Density
`llygroscopicity
`Suspension viscosity
`llilterability
`(joloy
`Tablet hardness
`:%__:_,
`
`

`
`/lflerarion ofrhe S0lt'(f Stare ofthe Drug SuI)smnce 533
`
`While there are numerous examples of miscellaneous physical property differ-
`ences between different solid phases of the same drug substance, the excess free
`energy of metastable states is the most important. The higher energy state, a conse-
`quence of decreased crystal lattice energy, produces a greater molecular mobility and
`thermodynamic escaping tendency in metastable solids. This leads to faster dissolu-
`tion t'ates and greater solubilities, which can have formulation and therapeutic i1npli-
`cations for pharmaceuticals. Drugs with poor aqueous solubility are more likely to
`show enhanced bioavailability when metastable solids are used, because their oral
`absorption tends to be dissolution-limited.
`
`Advantages of Using Metastable Solids
`
`f)is.s‘0lIl.ti0Iz Rate Irrrpr'or/emertli
`
`The Noyes-Whitney equation for the dissolution ofsolids into a solvent (Noyes and
`Whitney 1897) can be used to calculate the rate ofdrug concentration Cincrease with
`time 1:
`
`(IC
`(it
`
`.
`= K/l C ~— C
`i
`4,
`
`J
`
`(l)
`
`where A is the surface area of the solid exposed to solvent, C‘ is the saturation con-
`centration or solubility, and Kis a constant including the diffusion coefficient of the
`solute, the thickness of the unstirred diffusion layer, and the volume ofsolvent. Rotat-
`ing die methods have been developed, such as the Woods die apparatus, which main-
`tain a constant surface area during the initial phase of dissolution experiments. Under
`sink conditions, where C‘ >> C, and with constant surface area /1, the intrinsic disso-
`lution rate (IDR) can be obtained. The [DR (units mg/cin“ /min) is directly propor-
`tional to solubility and depends on the intrinsic dissolution properties of the drug in
`the media, not the dissolution method:
`
`5
`IDR: K(C)
`
`(2)
`
`Changes in the solid state can influence the dissolution rate through the surface
`area term or the solubility term. Surface area differences can arise from simple particle
`size effects between different crystal fornts and also from shape factors. Different crystal
`habits and shapes can alter the exposed surface area without a change in median parti-
`cle size measurements, since these are often calculated by methods that assume spher-
`ical shapes. Abdou (1989) has reviewed the effect of crystalline state on the dissolution
`rate ofpharmaceuticals and how this contributes to bioinequivalence ofvarious forms.
`l)il’ferences in solubility between different crystal forms alter the driving force for
`dissolution, controlled by the difference between the solution concentration and the
`saturation concentration (C5 — C). Hamlin, Northam, and Wagner (1965) have shown
`that dissolution rate correlates well with solubility for a large number of pharmaceu-
`tical compounds varying in solubility from 0.0] to it) mg/ml. at 37°C. Nicklasson and
`
`

`
`0034
`
`0.032
`
`0.028
`
`
`
`S0/ubi/iiy(molar)
`
`0.024
`
`534 Warer~[rrs0lrrbZc Drug Formulatio/r
`
`
`
`Figure 15.1 Dissolution rates for anhydrous and hydrated [3-cyclodextrin in water at
`40° C. [Source.' Reprinted from C(u'b0hydr(rle Research, vol. 143, by Jozwiakowski and
`(Jonnors, Aqueous Solubility ol"l‘hree (Iyclodextrins, pp. 51-59, Copyright 1985, with
`permission lrorrr Elsevier Science]
`
`r-—-4 Anhydrous
`
`0——O Hydrate
`
`90
`
`Time (min)
`._j.______.
`
`Brodin (1984) have shown that using cosolverrt nrixtures for drugs with poor aqueous
`solubility produces a good correlation between dissolution rate and solubility.
`The dissolution rate improvement for rrrost rrretastable solids is only trarrsient;
`ultimately the excess solid in equilibrium with the solvent converts to the lowest
`energy phase. lligure 15.1 tlozwiakowski and Connors 1985) illustrates a typical disso-
`lution pr'ol“ile for a metastable solid and a stable solid. In this case, [3-cyclodextrin dis-
`solution proliles were measured in 40°C distilled water and plotted on a molar basis
`(where the molecular weight dil‘l'erence is inconsequential). The c.orrcerrtration of the
`stable form in water at room temperature (:1 dodecalrydrate) gradually increases to
`the limit of its solubility (().()298 M).
`The rnetastable form (the arrlrydrate from oven drying) shows a rapid initial dis-
`solrrtiorr rate over the lirst 10 min, peaks at less than 30 rrrirr, and their declines to the
`solubility lirrrit as the excess solid converts to the hydrated l'or'm, as verified by
`microscopy. The same behavior is exhibited by metastable and stable polymorphic
`lorrrrs, e.g., form I and form ll ol’rrreprobanrate (Clernerrts and Popli 1973) and lornr I
`and form lI olgepirorre ll(Il {Behme et al. 1985). Ultimately, the sarrre eqrrilibriunr sol-
`ubility will be reached regardless of the direction of approach, although the kinetics of
`this transition can vary considerably for dr'l‘l‘erent drugs.
`Dissolution rate irnproverrrerrt is often characterized by the intrinsic dissolution
`rates or the change in the initial rate, since in many cases the advantage decreases
`with time. Table 15.3 shows sorrre data from the plrarmaceulical literature that iridi-
`cates the typical dissolrrtion rate increases that can be obtained by altering the solid
`
`

`
`
`
`/llrermion oftlie Solid State of the Drug Su.1)sIancc 535
`
`Table 15.3
`
`Comparison of Initial Dissolution Rates of Drugs upon Alteration
`of the Solid State
`
`Drug
`
`Solid Form
`
`Relative Rate
`
`Reference
`
`Sulphathia7,ole
`
`'l‘egal"ur
`
`Dillunisal
`
`ll
`l
`III
`
`on
`[3
`y
`l
`II