`JOURNAL
`OF PHARMACEUTICS
`
`/ E
`
`ditor/3" in Chief
`
`P.F. D’ARCY (Belfast, N. Ireland) and W.I. HIGUCHI (Salt Lake City, UT, U.S.A.)
`(Associate Editor: I.H. RYTTING, Lawrence, KS, U.S.A.)
`
`Editorial Board
`
`‘6
`
`G. AMIDON (Ann Arbor, MI, U.S.A.)
`B.D. ANDERSON (Salt Lake City, UT, U.S.A.)
`G.S. BANKER (Minneapolis, MN, U.S.A.)
`H. BOXENBAUM (Cincinnati, OH, U.S.A.)
`H. BUNDGAARD (Copenhagen, Denmark)
`J.T. CARSTENSEN (Madison, WI, U.S.A.)
`ss. DAVIS (Nottingham, U.K.)
`A.T. FLORENCE (Glasgow, U.K.)
`J.L. FOX-(Salt Lake City, UT, U.S.A.)
`H.L. FUNG (Amherst, NY, U.S.A.)
`D. GANDERTON (London, U.K.)
`J. HADGRAFT (Cardiff, U.K.)
`M
`M. HANANO (Tokyo, Japan)
`EN. HIESTAND (Kalamazoo, MI, U.S.A.)
`T. HIGUCHI (Lawrence, KS, U.S.A.)
`N.F.H. H0 (Kalamazoo, MI, U.S.A.)
`W.B. HUGO (Nottingham, U.K.)
`T.M. JONES (Dartford, U.K.)
`K. KNUTSON (Salt Lake City, UT, U.S.A.)
`G. LEVY (Amherst, NY, U.S.A.)
`
`J.A. MOLLICA (Suffem, NY, U.S.A.)
`T. NAGAI (Tokyo, Japan)
`I.H. PITMAN (Parkville, Australia)
`G. POSTE (Philadelphia,PA, U.S.A.)
`BJ. POULSEN (Palo Alto, CA U. S.A)
`A.J. REPTA (Lawrence, KS, U.S.A.)
`J.R._ ROBINSON (Madison, WI, U.S.A.)
`TJ. ROSEMAN (Morton Grove, IL, U.S.A.)
`H. SEZAKI (Kyoto, Japan)
`.
`J. E. SHAW (Palo Alto, CA, US.A)
`E. SHEFTER (Wilmington, DE, U.SA)
`D.D SHEN (Amherst, NY, U. S.A.)
`'
`E.’ SHOTTON (Berkhampstead, U.K.)
`J. SJOGREN (Molnlycke, Sweden)
`R.S. SUMMERS (Pretoria, South Africa)
`B. TESTA (Lausanne, Switzerland)
`K. THOMA (Munich, EKG.)
`R,F. TIMONEY (Dublin, Ireland)
`E. TOMLINSON (Horsham, U.K.)
`G. ZOGRAFI (Madison, WI, U.S.A.)
`
`
`
`VOL. 32 (1986)
`
`ELSEVIER SCIENCE PUBLISHERS B.V. — AMSTERDAM
`
`ApoteX Exhibit 1010.001
`
`Apotex Exhibit 1010.001
`
`
`
`© 1986, Elsevier Science Publishers B.V. (Biomedical Division)
`
`
`
`All rights reserved. No part of this publication may be reproduced stored in a retrieval system, or transmitted, in any form' or by any
`means, electronic, mechanical, photocopying, recording or otherwise without the permission of the publisher Elsevier Science
`Publishers B.V. (Biomedical Division), P. 0. Box 1527,1000 BM Amsterdam, The Netherlands. """ -
`’
`
`"Special regulations for authors‘llpon acceptance of an article by the journal the author(s) will be asked to transfer copyright of the
`article to the publisher. This transfer will ensure the widest possible dissemination of information.
`_
`/'
`
`Submission of a paper to this journal entails the author’s irrevocable and exclusive authorization of the publisher to collect any sums
`or considerations for copying or reproduction payable by third parties (as mentioned in article 17 paragraph 2 of the Dutch
`Copyright Act of 1912 and1n the Royal Decree of June 20, 1974 (S. 351) pursuant to article 16b of the Dutch Copyright Act/of 1912)
`and/or to act in or out of Courtin connection therewith.
`
`Special regulations for readers in the U.S.A. This journal has been registered with the Copyright Clearance Center, Inc. Consent is
`given for copying of articles for personal or internal use, or for the personal use of specific clients. This consent is given on the
`condition that the copier pays through the Center the per—copy fee stated in the code on the first page of each article for copying
`beyond that permitted by Sections 107 or 108 of the US. Copyright Law. The appropriate fee should be forwarded with a copy of the
`first page of the article to the Copyright Clearance Center, Inc., 27 Congress Street, Salem, MA 01970, U.S.A. If no code appears in
`an article, the author has not given broad consent to copy and permission to copy must be obtained directly from the author. All
`articles published prior to 1980 may be copied for a per-copy fee of US$ 2.25, also payable through the Center. This consent does not
`extend to other kinds of copying, such as for general distribution, resale, advertising and promotion purposes, or for creating new
`collective works. Special written permission must be obtained from the publisher for such copying
`}
`
`Printed in The Netherlands
`
`ApoteX Exhibit 1010‘i002
`
`Apotex Exhibit 1010.002
`
`
`
`__—.4‘_..~—.1\_,
`
`
`International Journal of Pharmaceutics, 33 (1986) 201—217
`Elsevier
`
`IJP 01121
`
`1
`
`201
`
`- Salt selection for basic drugs
`
`Pharniaceutical Research and Development Department, Pfizer Central Research, Sandwich, Kent (U.K.)
`
`Philip L. Gould
`
`// .
`
`[1’
`
`‘
`
`(Received 24 March 1986)
`(Accepted 30 May 1986)
`
`Key words: Salt form selection — Pharmaceutical salts
`
`summary
`
`An attempt has been made using a Kepnerci'Tregoe decision analysis approach to previde rationale to salt selection for basic drugs.
`The selection objectives are reviewed in terms of the ‘essential’ (MUSTS) and ‘desirable’ (WANTS) issues. The desired characteristics
`
`of the salt form, given sufficient strength and toxicological suitability of the conjugate acid, are then discussed on the basis of the
`various pivotal physicochemical preperties; melting point aqueous solubility and dissolution rate, stability and hydrophobicity.
`Several trends are established which can then assist the decision of which range of salt forms to evaluate to overcome a particular
`problem with a. basic drug. Itis concluded that it is important to view the choice of salt form for development as a compromise With
`particular focus on the correctly weighted desires to obtain the best balanced choice
`
`Introduction
`
`Salt formation provides a means of altering the
`physicochemical and resultant biological char-
`acteristics of a drug Without modifying its chem-
`ical structure. The importance of choosing the
`‘correct’ salt form of a drug is well outlined in a
`published review (Berge etal., 1977) but, although
`salt form can have a dramatic influence on the
`
`overall properties of a drug, the selection of the
`salt form that exhibits the desired combination
`
`of properties remains a difficult semi-empirical
`choice.
`
`In making the selection of a range of potential
`“salts, achemicalprobessgroup con51ders1Ssues 'on
`the basis of yield, rate and quality of the crystalliQ
`sation as well as cost" and availability of the con-
`
`jugate acid. The formulation and analytical groups
`are, on the other hand, concerned with the hygro-
`scopicity, stability, solubility and processability
`profile of the salt form, while the drug metabolism
`group is cOncerned with the ‘pharmacokinetic
`aspects and the safety evaluation group on the
`toxicological effects of chronic and acute dosing
`of the drug and its conjugate acid. Thus, a clear
`compromise of properties for the salt form is
`required, but the difficulty remains of assessing
`which salt forms are best to screen for a particular
`drug candidate.
`literature has been devoted to
`Little,
`if any,
`discussing the compromise of properties for salt
`form selection. This review addresses the problem
`of salt form selection for basic drugs.
`
`' Correspondence: P.L. Gould, Pharmaceutical Group, Product
`Research and Development Laboratories, Cyanarnid of Great
`Britain Limited, Gosport, Hants, U.K.
`
`Walking and Appino (1973) have used the
`KepneraTregoe (KT)
`techniques
`(Kepner and
`
`“Approach to the salt selection process
`
`0378-5173/86/$03.50 © 1986 Elsevier Science Publishers B.V. (Biomedical Division)
`
`ApoteX Exhibit 1010.003
`
`Apotex Exhibit 1010.003
`
`
`
`202
`
`TABLE 1
`
`FDA—APPROVED COMMERCIALLY MARKETED SALTS
`
`.
`
`(
`
`.
`
`‘
`
`\‘
`
`_
`
`"
`’
`
`.
`
`~
`
`.
`
`.
`
`Tregoe, 1976) of decision analysis and potential
`problem analysis to aid the selection of a salt'
`form; Although their application is more exemg,
`
`Percent 3
`Anion
`Percent a
`Anion
`plary of the KT method rather than of the specific
`Acetate
`1.26
`Iodide
`102
`application, the rational process decision analysis
`
`‘Benzenesulfonate
`0.25
`Isothionatei
`0.88
`approach Which defines essential and desirable
`attributes as . ‘MUSTS’ and ‘WANTS’,
`respec- 3?sz
`'.
`0751
`Latter?
`5’76
`tively, provides a route to-- initially address 'the
`BicaIbmmte
`0‘13
`Lacmbmmte
`0'13
`bl
`f
`a1 f
`1
`.
`Bitartrate
`0.63
`Malate
`0.13
`pro em 0 .s
`t orm se ection.
`Bromide
`4.68
`Maleate
`303
`Calcium edetate
`0.25
`Mandelate
`0.3 8
`"
`‘
`“GO”/ “NO—G0” issues
`Camsylate b.
`-
`0.25
`Mesylate
`2.02
`Themajor f‘GO”/“NO_GQ” (MUSTS) iSSile
`Carbonate
`0.38
`Methylbromide
`0.76
`for salt selection of an ionizable drug is the con-
`Chloride
`4‘17
`Memylmmie
`0’38
`Id
`’00
`fth
`lat’v b .
`‘t Of th (1
`d
`Citrate
`_ 3.03
`Methylsulfate
`0.88
`'51 era 1 F1 0
`C re
`1 e
`331C? y
`e_ mg an
`Dihydrochloride
`0.51
`Mucate
`0.13
`the relative strength of the conjugate ac1d. Clearly
`Edetate
`025
`Napsylate
`0.25
`to form a salt the pKa of theconjugate acid has to
`Edisylate C
`0.38
`Nitrate ‘
`064
`-
`be less than or equal to the pKa1 of the basic centre
`EStOlate.
`0:13
`Pamgate
`.1-01
`he (i
`-
`(Em onate)
`Of th If t
`t'al
`f alt
`f d
`Esylate "'
`0.13
`Pantothenate
`0.25
`rugs con-
`S O
`us
`6 p0 en 1
`range 0 S
`.Fiumarate;
`0.25
`Phosphate/
`3.16-
`taining for example triazoyl bases (I; pKa~ 2) is
`_
`.
`' diphosphate
`'
`restricted to strong acids (mineral and sulphonic ‘
`Gluceptate ‘
`0.18
`7 Polygalacturonate 0.13
`but excluding the carboxylic), Whereas imidazole
`Gluconate
`0-51
`‘Sah'cylate
`”0-88
`.
`bases (II; pKa 6-7)are far less restricted and the
`'Glmamate '
`g 0-2'5'
`‘Stearate'
`0:25»
`test sco e for salt formation 000
`s for the
`Glycouylmmlate ‘ 0'13
`subacemte
`0’38
`~ gréa
`.
`P_
`.
`.
`-_
`ur
`Hexylresorcinate
`0.13
`Sucéinate
`0.38
`aliphatlc ternary amines (HIE BKa 9—10)-
`Hydrabamine h
`0.25
`Sulfate
`. 7.46
`"
`‘
`’
`Hydrobromide
`1.90
`Tannate
`0.88
`
`Hydrochloride
`42.98
`Tartrate
`3.54
`H d
`h-
`7
`y roxynap
`thoate
`
`(31-13
`N—N
`N
`'
`f \
`,
`/ \
`l
`< ) [13Ka 2l < [pKafiCfis-Il‘I
`i
`'N
`11f}
`_ E ‘
`‘
`.CH3
`I
`(
`)
`.‘
`( )
`(HI) ‘
`
`.
`
`'
`
`1
`
`[PK-a 9]
`
`.
`0.13
`TeoclateJ
`.
`.
`u
`-
`0.13
`Tnetlnodide
`
`.
`‘
`.
`'
`.
`.
`_
`'
`.
`Cation
`' Percent 3‘
`Cation
`Percent.a
`
`
`0.25
`
`'
`
`.
`.
`,
`.
`.
`.
`‘
`«
`._
`,
`The relative acid/base strength of the resultant
`salts also dictates their stability to di5proportiona-
`tion, since all salts will be acid and therefore
`otentiall
`reactive towards basic formulation ad—
`P_ _
`1
`‘
`,
`.
`.
`»
`dltlves-
`p The other essential selection issue for a salt
`
`y
`
`form is the relative toxicity of the conjugatean-fl
`ion; some salts clearly fall into a desirable cate-
`gory, some acceptable but
`less desirable (both
`“GO_”) and some undesirable (“NO GO”). A ta-
`ble of salts ‘ used in pharmaceutical products
`marketed in the US. up to-1974 is given in Table
`1. It would seem Sensible that any acid relating to
`normal metabolism, or present in food and drink
`can be regarded as a suitable candidate for prepar-
`ing salts. Clearly anions that cause irritancy to the
`
`'
`
`‘
`
`,
`
`_
`“Maura-a:
`Organic:
`0.66 ;
`Aluminium
`0.66
`Benzathine 1‘
`-10-49
`CélC§um ‘
`.0-33 .
`ChlofOPIOCaine
`1'64 ,
`Lithium.
`'
`0'33
`Gimme
`_
`1'31
`Magnesmm
`0'98
`Diethanommne
`10.82
`' Potassium
`0.66
`Ethylenediamine
`. 61.97
`Sodium
`2.29
`Megluminel
`.
`
`
`
`0.66 Zinc .Procaine 2.95
`
`‘_
`
`
`
`“ Percent is based on total number of anionic or cationic salts
`
`in use through 1974. b Camphorsulfonate. c1,2—Ethanedisul—
`fonate.
`‘1 Laurylsulfate.
`e Ethanesulfonate. fGlucoheptonate.
`gp- Glycollamidophenylarsonate.
`h N,N ’-Di(dehydroabiety1)
`ethylenediamine 1 2aHydroxyethanesulfonate.
`J 8— Chlorotheo-
`_ phyllinate.k NN’--Dibenzy1ethylenediamine. 1N—Methylgluca—
`mine.
`
`Reproduced from Berge et al. (l,9777)iwith permission of the
`copyright owner (J. Pharm. Sci).
`
`ApoteX Exhibit 1010.004
`
`Apotex Exhibit 1010.004
`
`
`
`'
`
`203
`
`
`
`0.4
`
`0.8
`
`1.2
`
`0.8
`
`1.2
`
`1.6
`
`'.04
`1.6
`LOGS mglml
`Fig. 1. Relationship between solubilityin water and salting~out
`constant at 25°C (left) and 37° C (right). Key: A =
`phenazopyridine; B = cyproheptadine; C = bromhexine; D =
`trihexyphenidyl; E = isoxsuprine; F = chlortetracycline; G =
`methacycline; H = papaverine; and I = demeclocyline.
`
`GI tract should be avoided for some types of drug,
`e.g anti—inflammatories, laxative surfactant anions
`for anti--secretory drugs and conjugateanions with
`intrinsic toxicity,e.g0. Oxalate.
`Properties desired of the salt form (WANTS)
`The desires or “WANTS’ of a salt form are
`
`dictated by the nature of the required dosage
`forms, their process "and desired biological perfor-
`mance. Thus, it is somewhat difficult to provide a
`complete overall specification of ‘WANTS’ for a
`series of salt forms, but ideally the bulk salt
`should be completely chemically stable non--hy-
`groscopic, not cause processing problems, and dis-
`solve quickly from solid dosage forms.
`Because of simple availability. and physiological
`reasons, the monoprotic hydrochlorideshave been
`by far the most frequent _‘(~40%) choice of the
`available anionic salt-forming species. Thus, there '
`is clear precedent, and an overwhelrmng argument -
`An example of a basic drug showing a strong
`on many grounds to immediately progress to the
`
`chloride-iondependencelS prazosin '
`hydfcfihlofide salt and evaluate other forms only
`if problems with the hydIOChloride emerge.
`
`Adapted from Miyazald et al. (1981). Reproduced with permis-
`sion of the copyright owner (J. Pharm. Sci).
`
`that a precipitousdrop1n drug solubility occurs as .
`the free Cl levelis increased.
`
`Prepare the hydrochloride; pros and ,cons
`Kramer and Flynn(1972) suggest that the‘solué
`bility of an amine hydrochloride generally sets the
`maximum obtainable .concentration for a 'giVen
`amine.
`Many reports (Miyazaki et al.,1980,1981)_ have
`shown that hydrochloride salt formation does not
`necessarily enhance the solubility'lof poorly solu-
`ble basic drugs and reSult in improved bioavaila-
`bility. This finding is based on the common ion
`effect of chloride on the solubility product equi-
`librium:
`‘
`
`KSp
`BH+C1(‘S)_BH+q + c1;
`
`CH3O
`
`,
`
`CHO_N\_/Nco—m
`
`.r‘
`
`.
`
`_ Ksp= 2.2><1or6 M@ 30°C
`
`Solubility/mgml'l @ 30°C
`
`Base
`Hydrochloride 1
`0.1 M-HCl water Water
`0.037
`1.40
`0.0083
`
`Hydrochloride salts therefore, have the potential
`to exhibit a reduced disSolution rate in gastric
`fluid because of—theabundanceof chloride 1511f
`(0.1—0.15 M). Indeed, the Setschenow- salting—out
`constants (k) for chloride are greatest for drugs of
`very low solubility (Fig. 1), and can decrease the
`dissolution rate of the salt to below that of the
`. free base form (Migazaki et a1., 1980), which shows
`
`.
`
`Chloride, as well as other inorganic anions have
`the potential to form insoluble complex salts with
`weak bases (Dittert' at al., 1964), which are then
`potentially less bioavailable than the free base
`form. The formation of these complex salts is
`controlled by their stability constant Kc.
`
`Drug(s) ‘——* Drugm) + IxH+E—‘ Drug- H+ (aq)
`
`Apotex Exhibit 1010.005
`
`Apotex Exhibit 1010.005
`
`
`
`204
`
`Evaluation of Kc for triamterene (Tr) yields val-
`ues of x= 0.5 for chloride, suggesting that one
`proton solubilizes two molecules of the drug, 1e.,_
`the complexIS Tr2H+Cl.
`With hydrochloride salts there is frequently an
`‘overkill’ on acid strength, which leads to a very
`low pH for-an aqueous solution (Nudelman et al.,
`1974) of the salt. This can then limit the utility of
`hydrochloride salts in certain parenteral dosage
`forms, or lead to packaging incompatibilities with
`pharmaceutical metal containers (aerosols).
`‘ Other problems frequently arise as the result of
`the polar nature of hydrochloride salts. Their high
`hydrophilic nature, favouring wettability probably
`as a result of the polar ionized groups being
`exposedon the crystal surfaces,
`leads to water
`vapour sorption (hygroscopiCity) which on occa-
`sions, ‘may be excessive. This can result
`in
`processing. difficulties (e.g. powder flow) and re-
`. duce the stability of a hydrolytically unstable drug.
`t, This latter effect is exacerbated by the [resulting
`very low pH of the loosely bound moisture.
`””— Theseiprobl'ems can'be particularly a‘cute'With
`dihydrochlorides (or disulphates). Also,
`the dif-
`ference in the strength of the basic centres in
`dihydrochloride salts can lead to a gradual loss of
`one of the hydrochloride moieties by releasehof
`hydrogen chloride gas (Lin et al., 1972) at elevated
`temperatures or under
`reduced pressure (i.e.
`freeze-drying). Also,
`their extreme polar nature
`‘ results in excessive hydroscopicity (Boatmannand
`.
`11.Johnson, 1981) eventually-1eaaing to deliques-
`cence.
`Thus, progression of a hydrochloride salt should
`be a first move, but if the problems With that salt
`form arises due to some of the reasons outlined,
`
`.1
`
`form
`then the real selection issue for a salt
`emerges—What trends are available for guidance?
`
`‘ The pivotal issues for salt selection
`
`Each drug and its associated range of dosage
`forms , will present different salt form require-
`ments, and it is perhaps best to discuss saltselec—
`tion furtherby outlining some of the trends in salt
`properties that may facilitate selection.
`
`The pivot of melting point ‘
`
`p A change in the development of 'a compound
`from the free base to a salt may be promoted by a
`need to moderate the kinetics and extent of drug
`
`‘ absorption, or to modify drug processing. Unfor-
`’ 'funately these desires may be mutually exclusive,
`as the balance between these properties is fre- _
`quently pivoted around the melting point of the
`salt form. Forlexample, an increase in melting
`point is usually accompanied by a reduction in
`salt solubility (the ideal solubility of a drug in all
`solvents decreases by an order of magnitude on an
`increase of 100°C in its melting point), whereas
`high melting crystalline salts are potentially easier
`to process.
`,
`Theincrease or decrease in melting point of a
`series of saltslS usually dependent on the control-
`ling effect of crystallinity from the conjugate an-
`ion. This is exemplified by Considering an experi-
`mental drug candidate (UK47880) which has a
`basic pK,L of 8 and therefore gives access to a
`Wide variety of salt ferms:
`'
`
`CH,
`
`/CH3
`‘
`.
`_
`s
`, UK—47880
`
`~;~:N~s~me1fingpointfi4~°-C
`
`
`.
`
`.
`
`salts prepared from planar, high melting aromatic
`.sulphonic or hydrOxyc-arboxylic acids yielded
`crystalline salts of correspondingly high melting
`point
`(see Table 2), whereas flexible aliphatic
`strong acids such as citric and dodecyl benzene
`sulphonic yielded oils. Thus,
`the comparative
`planar symmetry of the conjugate acid [appears
`important for the maintenance of high crystal
`lattice forces. This is shown by the melting point
`of the conjugate acid being highly correlated with
`the melting point of the‘resultant salt form (Fig.
`2). Therefore the highly crystalline salts are in this
`case best suited to reducing drug Solubility.
`Alternatively it ‘should also be feasible to build
`up crystal. lattice forces of drugs with good hydro—
`gen bonding potential, by considering symmetry
`and hydrogen bonding potential of the conjugate
`acid. One salt
`series of
`interest
`is
`that
`for
`
`Apotex Exhibit 1010.006
`
`Apotex Exhibit 1010.006
`
`
`
`Melting ‘point (° C) Legend
`Salt
`’ Conjugate
`Flg' 2
`acid
`
`.
`,
`UK 47880; free base
`pamoate (embonate)
`4-hydroxynaphthalene-
`1- sulphonate
`Salicylate .
`3-hydrogtynaphthalene—
`220
`2=carboxylate
`.
`‘
`2-hydroxynaphthalene-l
`B
`120
`, 145
`-ca.rboxy1ate
`'
`F
`225
`234
`anthraquinone---3-sulphonate
`—
`'
`—
`-.
`20
`dodecylb enzene ‘sulphonate
`mesylate
`‘
`1
`113*
`, 20
`J
`—
`citrate
`A
`.
`.
`20
`153
`i
`— '
`
`
`.
`
`74
`
`235
`
`280
`
`190
`158
`
`170
`156
`
`223
`
`.
`
`'
`
`_ A
`G
`
`D .
`C
`_
`E
`
`9
`j: 200‘
`2:
`m
`
`E
`2
`gr, 10°
`E 90
`5 80
`“:4
`70
`
`
`
`_
`
`50
`
`‘80
`
`300
`_ 200
`.
`10° -
`M‘ELTING‘POlNT ACIDI°C
`Fig. 2. Plot of melting point of UK47880 salts vs melting point
`of conjugate acids. Legend given in Table 2.
`‘
`
`7
`
`_,
`
`:
`
`TABLE 2
`
`300
`
`‘ 205
`
`MELTING POINT OF SALTS OF EXPERIMENTAL COM—
`POUND (UK47880) AND THE CORRESPONDING CON—
`JUGATE ACID
`.
`.
`
`
`
`
`—epi1rep‘hrine
`
`HO—CHCHZNHCH3
`
`|
`OH
`
`OH
`
`Salt form
`
`Melting
`point(° C)-
`
`epinephrine
`tartrate
`maleate
`
`157 ‘
`- 149 “-‘
`182
`
`malate
`fumarate
`
`'
`
`170 p.
`103
`
`.-
`
`‘
`
`_
`
`where small highly hydrogen bonding acids such
`as malonic andwmaleic gave higher melting salts,
`whereas the larger bitartrate and presumablysym-
`metriCally unfavoured fumaratexgave saltsof lower
`meltingpoint.
`
`,
`,
`
`Salt form
`,
`_.
`Free base
`HCl
`BIL-lactate“
`L—lactate
`2—hydroxyethane
`sulphonate
`.Mesylate
`
`‘
`
`»
`
`-,
`
`_ ,‘Sulphate
`
`mp. salt
`(°C)
`215
`331
`'172'
`192
`
`V 251
`290
`
`270 -
`
`mp. acid
`(°C)
`'
`
`.
`
`'
`
`17
`53
`
`'
`
`solubility
`(mg/m1)
`.‘7.5
`13*
`" .1850
`925
`'
`620
`300
`
`‘
`
`'
`
`20'
`
`.
`
`..
`
`The relationship between aqueousl solubility
`‘_ .“(SW)‘ and melting point is shown diagrammatically -
`.in Fig. 3, Where log SW is correlated over'a range
`. of salts with the . inverse of the melting point.
`Interestingly with this compound,.the solubility of
`the hydrochloride salt in water is only approxi—
`mately twice that of the free base, whereas the low
`"-rmelting DL-lactate provides a ZOO-fold advantage
`over. the free base in terms of solubility, which is a
`, result in part of the reduced lattice energyf
`
`,
`'
`Melting point, and aqueousrs‘olubility‘
`The trends in melting point" (m.p.) ananqueous
`solubility alluded to above are exemplified in the
`salts of a high melting antimalarial drug (Aghar-
`kar et a1. ,.1976)
`‘
`
`Melting point and chemical stability
`' The stability of organic compoundsin the solid
`stateis intimately related to the strength of the
`ficrystal lattice; Since the forces between molecules
`.in a crystal are small compared with the energy,
`
`:
`
`ApoteX Exhibit 1010.007
`
`Apotex Exhibit 1010.007
`
`
`
`206
`1D—
`
`I
`
`SOLUBILITYlgml'1
`
`P l
`
`
`
`essentially controls the formation and- extent of
`eutectic melts.
`
`. As an additional aspect to the strength of crystal
`forces, the balance of the amorphous to crystalline
`nature in solid salts can dramatically affect their
`
`‘ stability. This is exemplified by the sodium salts
`i'orfiethacrynic aC1d" (YarWood et 51., 1—983).
`
`“
`
`Sodium ethacrynate
`
`.
`
`Crystalline
`
`Amorphous
`
`m.p. (9 C) -
`% remaining after
`9 days @ 60°C
`
`200
`
`100
`
`‘
`
`—
`
`'92 .
`
`These results are consistent with the concept of an
`amorphous material being a highly viscous con-
`centrated solution and show- that the stronger
`crystalline. lattice'forces result
`in superior solid
`State stability.
`I
`
`A Mesyiate
`
`a
`Edisylate
`‘
`
`C
`L- Lactate
`
`D DL'Lactate
`
`nmL-l—l—l—_l_
`
`1.5
`
`1.5
`
`2.0
`
`2.2
`
`-
`
`[1/MELTING POlNT) x
`
`10%"1
`
`- Figo.P16t of aqueous solubility vs inverse of absolute melting
`point for a series of salts of a hydrophobic antimalarial drug.
`Data taken from Agharkar et a1. (1976)
`
`required to break chemical bonds, liquefaction of i
`the solid (and an increased frequency of molecular
`Melting point and formulation proceSsz'ng
`collisions) occurs before degradation begins. Thus
`the melting point of a compound can be an im-
`Salt formation is frequently employed to raise
`the melting point (and cryStallinity) 0f the drug
`portant factorin determining stability.
`'Degradation of solid drugs, when it is observed,
`species being processed. However, published work
`usually occurs in the surface film phase and is
`concerning this type of manipulation is somewhat
`accompanied byithe formation of a liquid phase at
`sparse
`The melting point of drug salts can dramati-
`temperatures below the normal melting point of
`cally.affectitheinphysicahslprageillmgs(o_r___salts)__
`the solid Using this so--called_ ‘liquid layer?ap_.-
`proach, Guillory and Higuchi (1962) investigated
`with low meltingpoints generally exhibit plastic
`the stability of esters of vitamin A employing the
`deformatiori (Jones, 1979) and thus during storage
`following equation to determine the relationship
`the stress exbrted by the bulk mass on the asperity
`points of interparticulate contact can lead to the
`between degradation rate andmelting point.
`formation of localized welds leading to bulkr'ag—
`AAH[1] AH
`gregation. Also,'if ‘the sublimation temperature is
`R[Tm]
`2
`.R—Td
`10w (elg. ibuprofen, m.p. 76°C), intraparticulate
`where Tm :normal melting point; Td= depressed '
`voids can be bridged by sublimed drug again
`‘ leading to aggregation. Thus on storage, the bulk
`storage temp. = storage temperature; K= degrad-
`drug salt will begin to cake and aggregat‘e''1 thereby
`ation rate constant; AH= heat of fusion.
`Thus, for aseries of related compounds subject
`altering significantly its flow, compression and
`to a storage temperature Td, the logarithm of the
`long--term dissolution properties.
`degradation rate constant is inversely related to
`Melting point also has a crucial role in drug
`processing, in particular connninution and tablet-
`the absolute melting point of
`the compounds.
`ing. Since low melting compounds tend to be
`Although this approach may be somewhat simplis—
`tic it may have utility as a method of assessing the
`plastic, rather than brittle, they comminute poorly,
`and frictional heating causes melting and deposi-
`bulk stability of non—hygroscopic salt forms.
`tion of the drug on the screens and pins of the mill _
`The melting point of a salt form also has some
`causing it
`to ‘blind’. For production 'of
`fine
`influence on its relative compatibility with drug
`pharmaceutical powders this aspect is crucial to
`combinations (Hirsch et al., 1978) or formulation
`judging the correct level of filler to allow efficient
`, excipients (Li Wan Po and Mroso, 1984) since it
`
`IogK=
`
`ApoteX Exhibit 1010.008
`
`Apotex Exhibit 1010.008
`
`
`
`manufacture using a cost-effective feed rate.
`Salt melting point can also have important
`implications for particle bonding on compression
`for tabletingl Since bonding on compression oc-
`curs by point welding at the deformed or frag-
`merited particle surfaces, then at a fixed tempera-
`ture and pressure, a lower melting species would
`be expectedto improve bonding. However,
`the
`pressure on the powder (and the eutectics formed
`with the other excipi'ents) suppress the melting
`point further. The Skotnicky equation defines the
`’fall in melting point (Tm) with the pressure on the
`solid (/PS)
`_
`
`dTfl'1 /_ —VST
`dPs
`AHf'
`
`' 207
`
`parameter in assessing the ‘Viability’ of certain
`salt forms. In general, an increase in melting point,
`usually by maximizing or encouraging crystal sym-
`metry, leads to reduced solubility in all solvents,
`but generally improved stability, particularly if
`salt formation results in a crystalline solid, and
`easier formulation processing. For a specific salt
`form for parenteral use, i.e. Where solubility and
`resultant pH is a major issue, a low melting point
`salt produced using a soluble fairly weak acid (see
`next section) probably made in situ is likely to be
`preferable.
`'
`
`The pivot of drug solubility
`
`. There are various, solubility issues that can de-
`cide the Viability of a particular salt form and i_t__is__
`perhaps worth addr'essingithese separately to iden-
`tify trends that may aidsalt selection.
`
`7
`
`Salt
`
`.
`
`.
`
`rm (°C)
`
`AHf (kJ-mol‘l)
`
`‘ Hydrochloride ' 3
`Mesylate
`~
`Tartrate
`Citrate
`Phosphate
`Acetate
`
`._
`
`.
`
`‘280
`‘
`------- 135 -~
`213
`180 .
`2'50 -
`180
`
`,
`
`where AHf = heat of fusion; VS = volume of solid;
`T = temperature; and therefore as well as those
`salts Which are intrinsically low melting, salts ‘of
`different values of AH]? would be expected to have
`Aqueous solubility per se
`different abilities to cold weld in the compression
`As indicated earlier, the solubility of a drug can
`—pr'OcessrrI'f-WWOmp‘arerf'or-examp'l'erthmelfinfi—Be—‘e—fifiafiaed dramatically by salt
`formation
`points and heats of fusion of the salts Of an
`(Agharkar et al., 1976). This enhancement may
`experimental drug candidate:
`arise from‘ a reduction in'melting point», or from
`improVed water—drug interactions. A good exam-
`ple of this is with the saltsiof-chlorhexidine (Senior;—
`1973), where increased Water solubility'was not
`' only produced by a lowering of melting point, but
`by increasing‘the hydroxylation of the conjugate
`acid.
`'
`'
`‘
`'
`‘
`
`'
`
`‘
`
`'
`
`i
`
`‘--
`
`156.5
`..: 20.5
`63.6
`27.2..
`136.5
`167.9
`
`.
`
`.
`
`the data suggest that the low melting point and
`low AHf for the meSylate salt would make it the
`most suitablecandidate, on bonding grounds, for
`a direct compression tablet. Since the melting
`points of compounds are reduced under pressure,
`the solubility of salt forms would be expected to
`increase with increasing pressure. This can poten-
`tially cause the formation of solutions of the salts
`in the film of absorbed moisture on the surface of
`the drug (and excipient) particleswhich then may
`have an effect on drug bondihglParrotTTQSZ) or g n
`cause the drug to adhere to the punches on ‘com—
`pression (Wells and Davison, 1985).
`,
`
`Conclusion
`
`-
`
`The consideration of melting point. is a key
`
`Chlorhexidine
`
`CIQNH— CNHCNH(CH2)6NH|<I:NHICI:NHOCI
`
`.
`
`H
`H
`NH NH
`
`NH NH-
`
`- Salt
`
`.
`
`, Structure
`
`above
`.
`.
`base
`dihydrochloride HC]
`di—Zhydroxy7
`
`_
`
`_
`
`_
`
`_
`
`I)“
`'cozH
`
`I
`
`naphthoate
`
`I
`
`l
`
`OH
`
`‘Melting Solubility
`point
`96 w/v
`(°C>_. @20°C
`134
`0.008
`261
`«0.06
`— ‘
`_ 0.014
`
`madame .
`dilactate
`digluconate
`
`154
`CH3C02H
`—
`, CH3CHOHC02H
`H02C(CHOH)4C02H low
`
`1.8
`1.0
`70
`
`Apotex Exhibit 1010.009
`
`Apotex Exhibit 1010.009
`
`
`
`208
`
`The above data exemplify the importance of
`considering the hydrophilic nature of the con—
`jugate anion, as well as its role in controlling
`crystallinity, when considering the potential solu—
`bility of salts.
`V
`Reduced aqueous solubility may occasionally
`be a crucial development factor for a drug, e.g. for
`an organoleptically acceptable or chemically sta-
`ble suspension. Such systems demand salts of low
`solubility, but recent experience with a series of
`purposely designed insoluble salts of an experi-
`mental drug candidate also highlighted the need to
`consider the solubility and PK? of the conjugate
`anion.
`
`-
`
`B ¥ free base of
`BH+ X75) =BHE2q) +X;q
`__ ”eacperimental drug
`,
`salt
`V
`lLKb KalL+H+_,_.
`Be) 33m)
`HX(aq) "2 HX(s)
`H+
`
`fasted state are also important. In fact in this case,
`the "bioavailability in the dog of the two salt forms
`when dosed orally from a standard capsule formu-
`lation were of the same order; 24% for the pamoate
`and 39% for the maleate
`‘
`Usually it is the - dissolution rate of a drug
`
`'which is of major importance. to the formulation
`and as a rule a salt exhibits a higher dissolution
`rate than the base at an equal pH, even though
`they have the same equilibrium solubility. This
`latter effect, which is exemplified by theophylline
`salts (Nelson, 1957), is due to the salt effectively
`acting as its own buffer to alter the pH of the
`diffusion boundary layer, thereby increasing the
`apparent solubility of the parent drug in that
`layer. Thus, administration of basic drugs as their
`salt forms (eg. tetracycline hydrochloride) ensures
`that stomach emptying rather than in vivo dissolu-
`tion will be the rate-limiting factor in their absorp-
`tion. It is al’so’pOssible that increased drug'ab'SOrp4
`tion may occur with salts due to their effect on the
`surface tension of the gastrointestinal fluids (l3erge
`et al,1977).
`"
`'
`"
`
`111:: aoove ionic equlllbl‘la Show that even sparing
`solubility of the salt means that the level of the
`conjugate anion in solution will depend markedly
`on‘the pH of the fluid. Consideration of ”the above
`for a pamoate salt, which has pKa’s of the parent
`acid of 2.5 and 3.1 and virtually insoluble un-
`i0nized form, indicates that solutions of pH 5—6
`Salt solubility and pH of salt solutions
`will drive the equilibria to the right, with full
`
`Enhancement of the aqueous solubility of a
`precipitation of the free acid HX(S) and liberation
`drug by saltf0rmation can Occur due to dif-
`of a full component in solution of the ionized base
`ferences in the pH of the saturated salt solutions
`(BH+). However, if we consider the hydroxynaph— _
`A soluble aeid salt of a weakly basic drug will
`thalene sulphonic acid (pKa = 0.11) then this sys—
`cause-the pH to drop as the salt is added to the
`tem provides ‘insolubility’ over a much wider pH
`solution. This pH drop will, in turn cause more
`’rangearid is therefore fafmoie tolerant to fluctua-
`drug to dissolve, and this process will continue
`tions in the fluid pH.
`until the pH of maximum. solubilityis reached (see
`The above aspect is important when consider-
`Fig 4). The equilibrium solubility(ies) are then
`ing the potential use of
`‘inSoluble’
`salts (e.g.
`given by:
`/
`pamoate) to control
`the absorption. of a drug
`candidate..For example,
`the in vitro dissolution
`rates of the dimaleate and pamoate salts of a drug
`candidate were compared in simulated gastric and
`* intestinal fluid. The dissolution rates were essen—
`
`‘
`
`in. the former fluid, with rapid
`tially identical
`deposition of the pamoic acid and liberation of
`the free base, whereas in the latter the pamoate
`salt exhibited a much slower dissolution rate than
`
`the maleate. Therefore ‘control’ on the drug ab-
`sorption (and toxicity) may then depend on the
`duration of gastric residence and the pH of the
`gastric contents. Thus aspects such as food vs the
`
`s = s+(1 + iopfl-PKa)
`
`,/1
`
`i.e. when the ionized form is
`for pH=pHmax,
`solubility limiting and
`
`s = so + lOPKa'“PH)
`
`1
`
`for pH= pHmax, where the unioriized form iS
`solubility limiting and pHmax is given by the solu-
`tion of the equality of pH for the above two
`
`ApoteX Exhibit 1010.010
`
`Apotex Exhibit 1010.010
`
`
`
`209
`
`Type
`
`Salt
`
`pme
`
`Hydro-
`
`chloride
`Mesylate
`Tartrate
`Citrate
`Phos—
`phate
`Acetate
`
`2.71
`2.57
`4.21
`3.30
`-
`5.31,
`5.29
`
`Conjugate acid
`
`Solu-
`bility
`(mg/
`m1)
`
`1470 ‘
`2400 .
`
`.