Journal of
`Pharmaceutical
`Sciences
`
`JANUARY 1977
`VOLUME 66 NUMBER 1
`
`REVIEW ARTICLE
`
`Pharmaceutical Salts
`
`STEPHEN M. BERGE **, LYLE D. BIGHLEY *, and
`DONALD C. MONKHOUSE
`
`Keyphrases 0 Pharmaceutical salts-general
`pharmacy, physico-
`chemical properties, bioavailability, pharmaceutical properties, toxi-
`Salts, pharmaceutical-general
`pharmacy, physico-
`cology, review
`chemical properties, bioavailability, pharmaceutical properties, toxi-
`Physicochernical properties-dissolution,
`solubility,
`cology, review
`stability, and organoleptic properties of pharmaceutical salts, review
`Bioavailability-formulation
`effects, absorption alteration and phar-
`macokinetics of pharmaceutical salts, review 0 Toxicology-pharma-
`ceutical salts, review
`
`2
`4
`5
`7
`
`CONTENTS
`Potentially Useful Salts ..................................
`Physicochemical Studies ........................
`..........................
`Dissolution Rate ....
`Solubility ..........
`..........................
`Organoleptic Properties .........................
`...............................
`s Properties ....................
`.....................
`.....................
`Absorption Alteration . .
`.........................
`General Pharmacy .......................................
`14
`..........
`.......... 14
`Pharmacological Effect
`Dialysis ...............................................
`14
`Miscellaneous .........................................
`14
`Toxicological Considerations ..............................
`15
`Toxicity of Salt Ion .....................................
`15
`Toxicity of Salt Form ...................................
`15
`Conclusions .............................................
`16
`References ..............................................
`16
`
`10
`11
`11
`
`The chemical, biological, physical, and economic char-
`acteristics of medicinal agents can be manipulated and,
`hence, often optimized by conversion to a salt form.
`Choosing the appropriate salt, however, can be a very
`difficult task, since each salt imparts unique properties to
`the parent compound.
`
`Salt-forming agents are often chosen empirically. Of the
`many salts synthesized, the preferred form is selected by
`pharmaceutical chemists primarily on a practical basis:
`cost of raw materials, ease of crystallization, and percent
`yield. Other basic considerations include stability, hy-
`groscopicity, and flowability of the resulting bulk drug.
`Unfortunately, there is no reliable way of predicting the
`influence of a particular salt species on the behavior of the
`parent compound. Furthermore, even after many salts of
`the same basic agent have been prepared, no efficient
`screening techniques exist to facilitate selection of the salt
`most likely to exhibit the desired pharmacokinetic, solu-
`bility, and formulation profiles.
`Some decision-making models have, however, been de-
`veloped to help predict salt performance. For example,
`Walkling and Appino (1) described two techniques, “de-
`cision analysis” and “potential problem analysis,” and
`applied them to the selection of the most suitable deriva-
`tive of an organic acid for development as a tablet. The
`derivatives considered were the free acid and the potassi-
`um, sodium, and calcium salts. Both techniques are based
`on the chemical, physical, and biological properties of these
`specific derivatives and offer a promising avenue for de-
`veloping optimal salt forms.
`Information on salts is widely dispersed throughout the
`pharmaceutical literature, much of which addresses the
`use of salt formation to prolong the release of the active
`component, thereby eliminating various undesirable drug
`properties (24). This review surveys literature of the last
`25 years, emphasizing comparisons between the properties
`of different salt forms of the same compound. Included also
`is a discussion of potentially useful salt forms. Our purpose
`is twofold: to present an overview of the many different
`salts from which new drug candidates can be chosen and
`
`Vol. 66, No. 1 , January 19771 1
`
`Liquidia's Exhibit 1034
`IPR2020-00770
`Page 1
`
`

`

`Table I-FDA-Approved Commercially Marketed Salts
`Anion
`Percent”
`
`Acetate
`Benzenesulfonate
`Benzoate
`Bicarbonate
`Bitartrate
`Bromide
`Calcium edetate
`CamsyIateh
`Carbonate
`Chloride
`Citrate
`Dihydrochloride
`Edetate
`EdisylateC
`Estolate
`Esylate“
`Fumarate
`Gluceptatef
`Gluconate
`Glutamate
`Glycollylarsanilateg
`Hexylresorcina te
`Hydrabamine”
`Hydrobromide
`Hydrochloride
`H ydroxynaphthoate
`
`Cation
`
`1.26
`0.25
`0.51
`0.13
`0.63
`4.68
`0.25
`0.25
`0.38
`4.17
`3.03
`0.51
`0.25
`0.38
`0.13
`0.13
`0.25
`0.18
`0.51
`0.25
`0.13
`0.13
`0.25
`1.90
`42.98
`0.25
`
`Anion
`
`lodide
`Isethionate‘
`Lactate
`Lactobionate
`Malate
`Maleate
`Mandelate
`Mesylate
`Methylbromide
`Methylnitrate
`Methylsulfate
`Mucate
`Napsylate
`Nitrate
`Pamoate (Embonate)
`Pantothenate
`Phosphateldiphosphate
`Polygalacturonate
`Salicylate
`Stearate
`Subacetate
`Succinate
`Sulfate
`Tannate
`Tartrate
`Teoclate’
`Triethiodide
`
`Percenta
`
`2.02
`0.88
`0.76
`0.13
`0.13
`3.03
`0.38
`2.02
`0.76
`0.38
`0.88
`0.13
`0.25
`0.64
`1.01
`0.25
`3.16
`0.13
`0.88
`0.25
`0.38
`0.38
`7.46
`0.88
`3.54
`0.13
`0.13
`
`Percent“
`
`Cation
`
`Percent“
`
`Organic:
`B e n z a t h i d
`Chloroprocaine
`Choline
`Diethanolamine
`Ethylenediamine
`Meglumine‘
`Procaine
`Camphorsulfonate. 1,2-Ethanedisulfonate.
`Lauryl sulfate.
`0 Percent is based on total number of anionic or cationic salts in use through 1974.
`c Ethanesulfonate. f Glucoheptonate. g p-Glycollamidophenylarsonate. N,N’-Di(dehydroabiety1)ethylenediamine. 2-Hydroxyethanesulfonate.
`1 8-C hlorotheophyllinate. N,N‘-Dibenzylethylenediamine. N- Met,hylglucamine.
`
`0.66
`0.33
`0.33
`0.98
`0.66
`2.29
`0.66
`
`Metallic:
`Aluminum
`Calcium
`Lithium
`Magnesium
`Potassium
`Sodium
`Zinc
`
`0.66
`10.49
`1.64
`1.31
`10.82
`61.97
`2.95
`
`to assemble data that will provide, for the student and
`practitioner alike, a rational basis for selecting a suitable
`salt form.
`
`POTENTIALLY USEFUL SALTS
`Salt formation is an acid-base reaction involving either
`a proton-transfer or neutralization reaction and is there-
`fore controlled by factors influencing such reactions.
`Theoretically, every compound that exhibits acid or base
`characteristics can participate in salt formation. Particu-
`larly important is the relative strength of the acid or
`base-the acidity and basicity constants of the chemical
`species involved. These factors determine whether or not
`formation occurs and are a measure of the stability of the
`resulting salt.
`The number of salt forms available to a chemist is large;
`surveys of patent literature show numerous new salts being
`synthesized annually. Various salts of the same compound
`often behave quite differently because of the physical,
`chemical, and thermodynamic properties they impart to
`the parent compound. For example, a salt’s hydrophobicity
`and high crystal lattice energy can affect dissolution rate
`and, hence, bioavailability. Ideally, it would be desirable
`if one could predict how a pharmaceutical agent’s prop-
`erties would be affected by salt formation.
`Tables I and I1 list all salts that were commercially
`marketed through 1974. The list was compiled from all
`agents listed in “Martindale The Extra Pharmacopoeia,”
`
`2 1 Journal of Pharmaceutical Sciences
`
`26th ed. (7). Table I categorizes all salt forms approved by
`the Food and Drug Administration (FDA), while Table I1
`lists those not approved by the FDA but in use in other
`countries. (Only salts of organic compounds are considered
`because most drugs are organic substances.) The relative
`frequency with which each salt type has been used is cal-
`culated as a percentage, based on the total number of an-
`ionic or cationic salts in use through 1974. Because of
`simple availability and physiological reasons, the mono-
`protic hydrochlorides have been by far the most frequent
`choice of the available anionic salt-forming radicals, out-
`numbering the sulfates nearly six to one. For similar rea-
`sons, sodium has been the most predominant cation.
`Knowledge that one salt form imparts greater water
`solubility, is less toxic, or slows dissolution rate would
`greatly benefit chemists and formulators. In some cases,
`such generalizations can be made. Miller and Heller (8)
`discussed some properties associated with specific classes
`of salt forms. They stated that, in general, salt combina-
`tions with monocarboxylic acids are insoluble in water and
`lend themselves to repository preparations, while those of
`dicarboxylic acids confer water solubility if one carboxylic
`group is left free. Pamoic acid, an aromatic dicarboxylic
`acid, is an exception since it is used as a means of obtaining
`prolonged action by forming slightly soluble salts with
`certain basic drugs. Saias et al. (9) reviewed the use of this
`salt form in preparing sustained-release preparations.
`More recently, latentiation of dihydrostreptomycin (10)
`
`Liquidia's Exhibit 1034
`IPR2020-00770
`Page 2
`
`

`

`Table II-Non-FDA- Approved Commercially Marketed
`Salts
`
`Anion
`
`Adipate
`Alginate
`Aminosalicylate
`An hydromethylenecitrate
`Arecoline
`Aspartate
`Bisulfate
`Butylbromide
`Camphorate
`Digluconate
`Dihydrobromide
`Disuccinate
`Glycerophosphate
`Hemisulfate
`Hydrofluoride
`Hydroiodide
`Methylenebis(salicy1ate)
`Napadisylate
`Oxalate
`Pectinate
`Persulfate
`Phenylethylbarbiturate
`Picrate
`Propionate
`Thiocyanate
`Tosylate
`Undecanoate
`
`PPrrPnta
`
`0.13
`0.13
`0.25
`0.13
`o.i3
`0.25
`0.25
`0.13
`0.13
`0.13
`0.13
`0.13
`0.88 o.i3
`0.13
`0.25
`0.13
`0.13
`0.26
`_.
`0.13
`0.13
`0.13
`0.13
`0.13
`0.13
`0.13
`0.13
`
`~
`
`Percent a
`
`Cation
`Organic:
`Benethamine
`Ckmizoled
`Diethylamine
`Piperazine
`Tromethamine‘
`Metallic:
`Barium
`0.33
`0.98
`Bismuth
`a Percent is based on total number of anionic and cationic salts in use
`through 1974. 1,5-Naphthalenedisulfonate. N-Benzylphenethylamine.
`l-p-Chlorobenzyl-2-pyrrolidin-l’-ylmethylbenzimidazole. Tris(hy-
`droxymethy1)aminomethane.
`
`0.33
`0.33
`0.33
`0.98
`0.33
`
`using pamoic acid resulted in the formation of a delayed-
`action preparation. Numerous studies using pamoate salts
`are dispersed throughout the literature (11-15).
`Alginic acid also has been used to prepare long-acting
`pharmaceuticals. Streptomycin alginate was prepared (16)
`and shown to be effective in sustained-release prepara-
`tions. A striking example of a long-acting alginate salt is
`that of pilocarpine. When dispersed in sterile water and
`dried to a solid gel, this compound was found useful in the
`preparation of long-acting ophthalmic dosage forms (17).
`While liquid preparations of the alginate and hydrochlo-
`ride salts possess similar miotic activity, studies showed
`that solid pilocarpine alginate flakes constricted pupil size
`more effectively and increased the duration of miosis sig-
`nificantly when compared with the liquid preparations.
`Solid dose pilocarpine may be more uniformly available,
`because it diffuses more slowly through the gel matrix
`which holds the drug in reserve. In contrast, drops of the
`commonly employed solution dosage form release the dose
`immediately to the conjunctival fluid.
`MQlek et al. (18) devised a unique way of prolonging
`action through salt formation; they showed that the dis-
`tribution of several antibiotics could be markedly altered
`by merely preparing macromolecular salts. Since macro-
`molecules and colloidal particles have an affinity for the
`lymphatic system, streptomycin, neomycin, viomycin, and
`
`streptothrycin were combined with high molecular weight
`compounds such as polyacrylic acids, sulfonic or phos-
`phorylated polysaccharides, and polyuronic derivatives.
`Parenteral administration of these compounds produced
`low blood levels of the antibiotic for long periods, while
`lymph levels were high. (In comparison, streptomycin
`sulfate gave high blood levels but low lymph levels.) This
`alteration in distribution caused the streptomycin to
`prolong its passage through the body, since lymphatic
`circulation is quite slow.
`The appropriate choice of a salt form has been found to
`reduce toxicity. It can be rationalized that any compound
`associated with the normal metabolism of food and drink
`must be essentially nontoxic. The approach of choosing
`organic radicals that are readily excreted or metabolized
`opened up a new class of substances from which to select
`a salt form. For example, certain salts of the strong base
`choline have proven to be considerably less toxic than their
`parent compound. The preparation and properties of
`choline salts of a series of theophylline derivatives were
`reported (19), and it was shown that choline theophyllinate
`possessed a greater LD50 than theophylline or its other
`salts (20). It was postulated that this agent would be less
`irritating to the GI tract than aminophylline, because “its
`basic constituent, choline, is an almost completely non-
`toxic substance of actual importance to the physiologic
`economy.” This evidence led to the preparation of choline
`salicylate (21) as an attempt to reduce the GI disturbances
`associated with salicylate administration. Clinical studies
`indicated that choline salicylate elicited a lower incidence
`of GI distress, was tolerated in higher doses, and was of
`greater benefit to the patient than was acetylsalicylic acid
`(aspirin).
`Amino acids and acid vitamins also have been used as
`salt-forming agents. Based on the evidence that coad-
`ministration of amino acids with aminoglycoside antibi-
`otics reduced their toxicity, a series of amino acid salts of
`dihydrostreptomycin was prepared (22). In all but one
`case, the acute toxicities of these salts were lower than the
`toxicity of the sulfate. The ascorbate and pantothenate
`also were synthesized and shown to be less toxic than the
`sulfate. Of the salts prepared, the ascorbate had the highest
`LD50.
`The vitamins most commonly used for forming salts
`exhibiting reduced toxicity are ascorbic and pantothenic
`acids. Keller et al. (23) were the first to use pantothenic
`acid as a means of “detoxifying” the basic streptomyces
`antibiotics. Parenteral administration of the pantothen-
`ates of streptomycin and dihydrostreptomycin had a sig-
`nificantly reduced incidence of acute neurotoxicity in cats
`as compared with the sulfates. Subsequent studies (24-28)
`supported this finding and showed that the pantothenates
`of neomycin and viomycin also are less toxic. The ascorbate
`of oleandomycin was synthesized and its pharmacological
`properties were reported (29). Upon intramuscular injec-
`tion in rats, it produced less irritation than the phos-
`phate.
`p -Acetamidobenzoic acid, an innocuous metabolite of
`folk acid present in normal blood and urine, has been used
`in preparing salts. In particular, it yields stable salts with
`amines that otherwise tend to form hygroscopic products
`with conventional acid components (30).
`Often the salt form is chosen by determining a salt
`
`Val. 66, No. 1, January 1977 1 3
`
`Liquidia's Exhibit 1034
`IPR2020-00770
`Page 3
`
`

`

`component that will pharmacologically antagonize an
`unfavorable property or properties exhibited by the basic
`agent. Salts of N-cyclohexylsulfamic acid are an example
`of the practical application of this approach. N-Cyclo-
`hexylsulfamic acid salts, better known as cyclamates, have
`a characteristic sweet, pleasing taste. Although presently
`under investigation by the FDA for potentially carcino-
`genic properties, salts incorporating this compoynd can
`render unpleasant or bitter-tasting drugs acceptable. For
`example, the cyclamates of dextromethorphan and
`chlorpheniramine exhibit greatly improved bitterness
`thresholds compared to commonly occurring salts (31).
`Furthermore, their stability in aqueous solution was de-
`scribed as good when mtiintained at a pH not greater than
`4.
`N-Cyclohexylsulfamic acid salts of thiamine hydro-
`chloride and lincomycin also have been synthesized. Thi-
`amine N-cyclohexylsulfamate hydrochloride was reported
`to have a more pleasant taste than other thiamine salts
`while having an equal or greater stability (32). Lincomycin
`cyclamate, shown to possess an enhanced thermal stability
`over its hydrochloride, was prepared (33) to test the hy-
`pothesis that reduced lincomycin absorption in the pres-
`ence of small quantities of cyclamates was due to a simple
`metathetic reaction. However, this assumption was found
`not to be true. An extensive study of the preparation and
`characterization of cyclamic acid salts of several widely
`used classes of drugs including antihistamines, antibiotics,
`antitussives, myospasmolytics, and local anesthetics was
`reported (34,35).
`Various salts of penicillin and basic amine compounds
`have been formulated in an effort to produce a long-acting,
`nonallergenic form of penicillin. Since antihistamines
`appear to mitigate the symptomatology of penicillin re-
`actions in some patients, coadministration of the two has
`been advocated. The preparation of the benzhydralamine
`salt of penicillin was an attempt to produce a repository
`form of penicillin with antiallergic properties (36). Blood
`levels achieved with this salt were comparable to those of
`penicillin G potassium; however, its antiallergic properties
`were not evaluated. In fact, the investigators noted that
`antihistamines can actually cause sensitization at times
`and stated that “despite their occasionally favorable in-
`fluence on the symptoms of penicillin sensitivity, they
`contribute directly to the potential of drug sensitivity when
`co-administered with penicillin.’’
`Silver salts of sulfanilamide, penicillin, and other anti-
`biotics have been prepared and represent cases where the
`species (ions) are complementary. When aqueous solutions
`of the salts were applied topically to burned tissue, they
`yielded the combined benefits of the oligodynamic action
`of silver and the advantages of the antibacterial agents
`(37).
`The use of 8-substituted xanthines, particularly the
`8-substituted theophyllines, as salt-forming agents was
`first reported in the preparation of a series of antihistamine
`salts (38-41). Synthesis of these xanthine salts was an at-
`tempt to find a drug to counteract the drowsiness caused
`by the antihistamines with the stimulant properties of the
`xanthines. When an electronegative group is introduced
`into the xanthine molecule at the 8-position, the elec-
`tron-drawing capacity of the substituent results in the
`creation of an acidic hydrogen at position 7. Thus, these
`
`4 /Journal of Pharmaceutical Sciences
`
`moderately strong acidic compounds can undergo salt
`formation with various organic bases.
`The 8-halotheophyllines were the first group of xan-
`thines studied as potential salt-forming agents. Since the
`report on the preparation of the 8-chlorotheophylline salt
`of diphenhydramine (42), synthesis of the 8-halotheo-
`phyllinates of a number of organic bases has been at-
`tempted. The 8-chlorotheophylline salts of quinine,
`ephedrine, and strychnine were prepared and character-
`ized (43). These salts were less water soluble than the
`corresponding free alkaloidal bases. In a similar report, the
`8-chlorotheophyllinates of three synthetic narcotics,
`meperidine, levorphanol, and metopon, were prepared
`(44).
`Pharmacological and clinical studies involving the 8-
`bromotheophylline pyrilamine salt revealed the unusual
`diuretic properties associated with the 8-halotheophylline
`portion of the compound (45,46). This finding initiated
`an investigation into the preparation of a soluble 8-bro-
`motheophylline salt of high diuretic activity. With readily
`available amines, over 30 salts. were synthesized and
`screened for diuretic activity (47). When tested against
`theophylline salts of the same amines, the 8-bromotheo-
`phyllinates showed greater activity in every case.
`With the successful formation of 8-halotheophyllinates
`of organic bases, Morozowich and Bope (48) proposed that,
`if the halogen moiety was replaced with a more electro-
`negative substituent such as a nitro group, a more acidic
`compound would be formed. Presumably, more stable salts
`would result and precipitation of the free xanthine deriv-
`ative in the stomach would be less likely to occur. On this
`premise, they successfully prepared pharmacologically
`effective 8-nitrotheophyllinates of several pharmaceuti-
`cally useful bases.
`Duesel et al. (19), in their study of choline theophylli-
`nate, prepared the 8-chloro-, 8-bromo-, and 8-nitrotheo-
`phylline salts of choline. Oral toxicity studies in mice
`showed that the LD.50 of the 8-nitrotheophyllinate was
`much greater than that of either 8-halotheophylline. In
`fact, it remained nonlethal at doses as high as 5 g.
`Polygalacturonic acid, a derivative of pectin, has been
`used to prepare quinidine salts exhibiting reduced toxicity
`(49, 50). The compound possesses special demulcent
`properties and inhibits mucosal irritation. The rationale
`for use of this agent is to reduce the ionic shock to the GI
`mucosa resulting from the flood of irritating ions liberated
`by rapid dissociation of the conventional inorganic quin-
`idine salts. Studies have shown that it is four times less
`toxic orally than the sulfate. This difference was attributed
`to the slower release of quinidine from the polygalactu-
`ronate.
`Other compounds reported to be potentially useful as
`pharmaceutical salt forms are listed in Table 111.
`
`PHYSICOCHEMICAL STUDIES
`
`Biological activity of a drug molecule is influenced by
`two factors: its chemical structure and effect at a specific
`site and its ability to reach-and
`then be removed from-
`the site of action. Thus, a knowledge of the physicochem-
`ical properties of a compound that influence its absorption,
`distribution, metabolism, and excretion is essential for a
`complete understanding of the onset and duration of ac-
`
`Liquidia's Exhibit 1034
`IPR2020-00770
`Page 4
`
`

`

`Table 111-Potentially Useful Salt Forms of Pharmaceutical Agents
`
`Salt-Forming Agent
`
`Compound Modified
`
`Modification
`
`Reference
`
`Acetylaminoacetic acid
`N - Acetyl-L-asparagine
`N - Acetylcystine
`Adamantoic acid
`Adipic acid
`N - Alkylsulfamates
`Anthraquinone-l,5-disulfonic acid
`Arabogalactan sulfate (arabino)
`Arginine
`
`Aspartate
`Betaine
`Bis(2-carboxychromon-5-yloxy)alkanes
`.
`-
`
`Carnitine
`4-Chloro-rn-toluenesulfonic acid
`Decanoate
`Diacetyl sulfate
`Dibenzylethylenediamine
`Diethylamine
`Diguaiacyl phosphate
`Dioctyl sulfosuccinate
`Embonic (pamoic) acid
`Fructose 1,6-diphosphoric acid
`
`Glucose 1-phosphoric acid, glucose
`6-phosphpric acid
`I,-Glutamine
`H ydroxynaphthoate
`2-( 4-Imidazoly1)ethylamine
`Isobutanolamine
`Lauryl sulfate
`Lysine
`
`Methanesulfonic acid
`N - Methylglucamine
`N-Methylpiperazine
`Morpholine
`2-Naphthalenesulfonic acid
`Octanoate
`Probenecid
`Tannic acid
`Theobromine acetic acid
`3,4,5-Trimethoxybenzoate
`
`Tromethamine
`
`Doxycycline
`Erythromycin
`Doxycycline
`A1 kylbiguanides
`Piperazine
`Ampicillin
`Lincomycin
`Cephalexin
`Various alkaloids
`Cephalosporins
`a-Sulfobenzylpenicillin
`Erythromycin
`Tetracycline
`7-(Aminoalkyl) theophyllines
`Metformin
`Propoxyphene
`Heptaminol
`Thiamine
`Ampicillin
`Cephalosporins
`Tetracycline
`Vincamine
`Kanamycin
`2-Phenyl-3-methylmorpholine
`Tetracycline
`Erythromycin
`Tetracycline
`Erythromycin
`Erythromycin
`Bephenium
`Prostaglandin
`Theophylline
`Vincamine
`a-Sulfobenzylpenicillin
`Cephalosporins
`Pralidoxime (2-PAM)
`a-Sulfobenzylpenicillin
`Cephalosporins
`Phenylbutazone
`Cephalosporins
`Propoxyphene
`Heptaminol
`Pivampicillin
`Various amines
`Propoxyphene
`Tetracycline
`Heptaminol
`Aspirin
`DinoDrost (mostadandin F d
`
`Solubility
`Solubility, activity, stability
`Combined effect useful in pneumonia
`Prolonged action
`Stability, toxicity, organoleptic properties
`Absorption (oral)
`Solubility
`Stability, absorption
`Prolonged action
`Toxicity
`Stability, hygroscopicity, toxicity
`Solubility
`Gastric absorption
`Activity, prolonged prophylactic effect
`Toxicity
`Organoleptic properties
`Prolonged action
`Stability, hygroscopicity
`Prolonged action
`Reduced pain on injection
`Activity
`Organoleptic properties
`Toxicity
`Toxicity
`Solubility
`Solubility
`Solubility
`Solubility
`Solubility, activity, stability
`Toxicity
`Prolonged action
`Stability
`Organoleptic properties
`Toxicity, stability, hygroscopicity
`
`Solubility
`Toxicity, stability, hygroscopicity
`Reduced Dain on iniection
`Toxicity, ‘faster o n k t of action
`Reduced pain on injection
`Organoleptic properties
`Prolonged action
`Organoleptic properties
`Prolonged action
`Activity
`Organoleptic properties
`Prolonged action
`Absorption (oral)
`Phvsical state
`
`51
`52
`53
`54
`55
`56
`57
`58
`59,60
`61
`62
`63
`64
`65
`66
`67
`68
`69
`70,71
`72
`73
`14
`75
`76
`77
`
`77
`
`52
`78
`79
`80
`81
`62
`61
`82
`62
`72
`83
`72
`84
`68
`85
`86,87
`88
`89
`68
`90
`91
`
`tion, the relative toxicity, and the possible routes of ad-
`ministration (2).
`In a review in 1960, Miller and Holland (92) stated that
`“different salts of the same drug rarely differ pharmaco-
`logically; the differeaces are usually based on the physical
`properties.” In a subsequent review (931, Wagner ex-
`panded upon this statement, asserting that, although the
`nature of the biological responses elicited by a series of
`salts of the same parent compound may not differ appre-
`ciably, the intensities of response may differ markedly.
`The salt form is known to influence a number of physi-
`cochemical properties of the parent compound including
`dissolution rate, solubility, stability, and hygroscopicity.
`These properties, in turn, affect the availability and for-
`mulation characteristics of the drug. Consequently, the
`pharmaceutical industry has systematically engaged in
`extensive preformulation studies of the physicochemical
`properties of each new drug entity to determine the most
`suitable form for drug formulation. Published information
`concerning such studies, however, is sparse. Preformula-
`tion studies have been outlined, and the influence of the
`salt form on the volatility and hygroscopicity of an agent
`under investigation was discussed (94).
`
`In one such study, methylpyridinium-2-aldoxime
`(pralidoxime) salts were investigated (95). This study set
`out to prepare a salt with water solubility adequate to allow
`intramuscular injection of a low volume (2-3 ml) thera-
`peutic dose. The original compound, the methiodide, had
`the disadvantages of limited aqueous solubility and high
`potential toxicity, since its high iodide content could result
`in iodism. On the basis of physiological compatibility,
`better water solubility, favorable stability, and relatively
`high percentage of oxime, the chloride salt of pralidoxime
`was selected for therapeutic administration; it was claimed
`that “the anion used to form the salt can confer physical
`properties of importance and significance for the formu-
`lation and administration of the compound” (95).
`Some physicochemical properties of a series of mineral
`acid salts of lidocaine also were determined (96). While the
`hydrochloride and hydrobromide were more hygroscopic,
`they were more soluble in a number of solvents than the
`nitrate; perchlorate, phosphate, or sulfate salts.
`Dissolution Rate-The dissolution rate of a pharma-
`ceutical agent is of major importance to the formulator. In
`many cases, particularly with poorly soluble drugs, this
`characteristic best reflects the bioavailability of the com-
`
`Vol. 66, No. 1, January 1977 1 5
`
`Liquidia's Exhibit 1034
`IPR2020-00770
`Page 5
`
`

`

`pound. As a rule, a pharmaceutical salt exhibits a higher
`dissolution rate than the corresponding conjugate acid or
`base at an equal pH, even though they may have the same
`equilibrium solubility. The explanation for this result lies
`in the processes that control dissolution.
`Dissolution can be described by a diffusion layer modell
`in terms of an equation developed by Nernst and Brunner
`(97):
`
`..
`
`(Eq. 1)
`where W is the mass of the solute dissolved at time t,
`d W l d t is the rate of mass transfer per unit time, D is the
`solute molecule diffusion coefficient, S is the surface area
`of the dissolving solid, h is the diffusion layer thickness,
`C is the concentration of the drug in the bulk solution at
`time t , and C,s is the saturation solubility of the solute in
`the diffusion layer.
`The driving force for dissolution in Eq. 1 is the difference
`between the saturation solubility of the drug and the
`concentration of the drug in the bulk fluid. If the drug is
`not rapidly absorbed after it dissolves, then C, the con-
`centration in the bulk solution, approaches C, and the
`dissolution rate is retarded. When this occurs, absorption
`is “absorption rate” limited (or “membrane transport”
`limited). If the absorption rate is rapid (or if the absorption
`mass transfer coefficient is much larger than DSlh of Eq.
`1)’ however, C becomes negligible compared to C, and
`dissolution occurs under “sink” conditions. Absorption is
`then said to be dissolution rate limited, which is what oc-
`curs with most poorly soluble drugs. In either case, an in-
`crease in C,, as in salt formation, increases dissolution.
`Salts often speed dissolution by effectively acting as
`their own buffers to alter the pH of the diffusion layer, thus
`increasing the solubility of the parent compound, C,, in
`that layer over its inherent solubility at the pH of the
`dissolution medium. Hence, dissolution is controlled by
`solubility in the diffusion layer which, in turn, is deter-
`mined by the pH of that layer. The influence of Ksp on the
`solubility term, Cs, and dissolution rate, should an accu-
`mulation of ions be allowed to occur, will be treated
`later.
`Nelson (98), in a study of theophylline salts, was the first
`to show the correlation between diffusion layer pH and
`dissolution rate. The major impact that this study had on
`the pharmaceutical sciences was its conclusion that, if
`other factors remained constant, the dissolution rate of a
`compound determined the rate of buildup of blood levels
`with time and the maximum levels obtained. Those salts
`of the acidic theophylline with high diffusion layer pH’s
`had greater in uitro dissolution rates than those exhibiting
`a lower diffusion layer pH. And, indeed, the rank order of
`dissolution rates correlated well with clinically determined
`blood levels. Presumably, the higher pH in the diffusion
`layer retards hydrolysis of the salt, thereby maintaining
`the anionic charge of the theophyllinate ion. This report
`led to many additional studies which illustrate the influ-
`ence of the salt form on dissolution and the beneficial ef-
`fects of changing nonionized drugs into salts.
`Juncher and Raaschou (99) demonstrated that the rank
`order of peak blood levels of penicillin V, obtained upon
`
`’ ‘ h e autbws recognize the existence of other models; this one was chosen simply
`for illustwt ive purposes.
`
`6 1 Journal of Pharmaceutical Sciences
`
`administration of three different salts and the free acid,
`was the same as the rank order of their rates of dissolution
`in uitro. While the investigators ascribed these differences
`to the solubility properties of the salts, their experiments
`actually compared dissolution rates, not solubilities. The
`relative order of dissolution rates and mean maximal blood
`levels was: potassium salt > calcium salt > free acid >
`benzathine salt.
`Nelson (100) determined dissolution rates for several
`weak acids and their sodium salts in media whose pH’s
`represented GI fluids. In all cases, the sodium salt dis-
`solved more rapidly than the free acid. This finding re-
`solved the misconception that absorption of drugs is re-
`lated only to solubility in the appropriate medium; rather,
`solubility affects absorption only to the extent that it af-
`fects dissolution rate. Absorption of drugs is a dynamic
`process, and the ultimate solubility of a drug in fluid at
`absorption sites is of limited consequence since absorption
`prevents the attainment of saturated solutions. Therefore,
`dissolution rate, more than solubility, influences absorp-
`tion since it is a preceding process.
`In two subsequent studies, Nelson and coworkers fur-
`ther illustrated the effects of changing nonionized drugs
`into salts. A report concerning tolbutamide (101), a weak
`acid, showed that the initial dissolution rate of tolbutamide
`sodium was approximately 5000 times more rapid than the
`free acid in acidic media and 300 times more rapid in
`neutral media. This difference, measured in uitro, reflected
`the differences observed between the free a