Pii: s1461-5347(99)00158-3

Excipients as stabilizers
Patrick J. Crowley
Excipients are better known as promoters of degradation than as sta-
mass can possess functional groups that renderthem susceptible, to some degree, to interactions bilizers of drug substances. This is not surprising. Functional groups
that lead to quality being compromised.
or residues in excipients can have the propensity to interact with
The long shelf-lives that are usually required for medicinal products reinforce the need for labile active ingredients, with attendant loss of molecular integrity
quality retention as a key consideration during or other changes in quality. Thus, the canon of work on excipients as
formulation. In contrast to foodstuffs, soft drinks stabilizers is not extensive. Nevertheless, possibilities exist to capital-
and many other commodities, it is usual to expectthe quality of medicines to be retained for as long ize on our knowledge of how a drug substance degrades, and of the
as 3–5 years. Such long ‘use by’ periods are es- properties and composition of excipients, to convert unstable drugs
pecially important for remote and climaticallyhostile regions, because of the complexities of sup- into viable products. This article discusses such approaches to prod-
ply and distribution. As there is no such thing as a uct stabilization.
totally impervious pack, it will be apparent that theprotection afforded by the pack will be limited inan environmentally hostile environment.
▼ Drug substances can be fragile entities.
It might be necessary, therefore, to consider Patrick J. Crowley
Environmental stresses, as well as those associ- developing a formulation that compensates for ated with turning a drug into a medicine, all have any basic deficiencies in stability of the drug sub- the potential to cause changes that compromise stance. In such cases, excipients must play a key quality. Such stresses can cause molecular com- role in turning an unstable drug into an accept- position to change to some extent. This reduces the amount of active ingredient in the medi- cation and might generate novel molecular enti- General considerations
ties that could compromise safety. Physical The potential for excipients to prevent or retard changes can lead to altered dissolution or other degradation will be determined by the factors delivery properties. Sedimentation in liquid that cause the molecular transformation of drug products, consequent to altered solubility, could present safety as well as efficacy problems.
• environmental components, such as water propensity for change on the part of the drug substance and clarify the strategy for develop- • stresses during conversion to the dosage ment of the dosage form. Drug–excipient form, such as size reduction, compaction or compatibility studies might further constrain or clarify the options for formulation. Packaging • interactions between adjacent molecules of a suitable to protect the product offers additional drug, or between functional groups on the scope for providing a stable product. Such ‘avoid- ance’ tactics might suffice for many dosage formdevelopment programmes, but not in all cases.
If excipients are to act as stabilizers they must Drug substances that are elaborated by semisyn- thetic means and those emanating from biotech-nology-based programmes usually have proper- Moisture-related degradation
ties that can render them ‘intrinsically unstable’.
Water can be associated with the drug or the Even heterocyclic compounds of low molecular excipients. It can be incorporated during 1461-5347/99/$ – see front matter 1999 Elsevier Science. All rights reserved. PII: S1461-5347(99)00158-3 manufacture of the dosage form or acquired from the environ- ment during processing, packaging or storage. Its ubiquitousnature and capability to exist as a vapour means that water is virtually impossible to avoid and difficult to control, particu- larly if the drug substance is hygroscopic. Its molecular mass is low, so modest amounts can be significant in terms of molar reactivity. It is also capable of diffusing, to a greater or lesserextent, through packaging materials, pack seals, or throughcompacted solid dosage forms.
Excipients with affinity for moisture might be expected to mitigate moisture sensitivity. Thus, formulation with a sub-stance having a greater affinity for water compared with the drug could mean that moisture in the product is sequestratedby the excipient.
Perrier and Kesselring used nitrogen sorption isotherms to predict the effect of common excipients on the stability of ni-trazepam1. They showed that stability in binary mixes was not affected by the drug/excipient ratio or by the specific surface Normalized decomposition rate constant (10 area of the excipient. Instead, it correlated with the nitrogenadsorption energy of the excipients, determined using the BETequation. Materials with higher adsorption energies caused less The primary aim of their study was to determine the ‘inert- Figure 1. Normalized decomposition rate constants of nitrazepam
ness’ of excipients, rather than their utility as stabilizers. How- in relation to nitrogen adsorption energy of various materials. ever, the approach is equally germane to ranking the desiccat- Figure reproduced, with permission, from Ref. 1.
ing capability of excipients. If the binding energy of watervapour for the excipient exceeds the binding energy for thedrug, the excipient should sequestrate any available moistureand act as a stabilizer. Competition between materials with (methanol, ethanol, isopropanol) might be present from the avidity for moisture should lead to its redistribution, based on synthesis and isolation of the drug, or from the process used relative affinity, until equilibrium is reached: the material with for manufacturing the dosage form. Formaldehyde has the ca- the greatest affinity takes the lion’s share of available water. If pability to participate in most of the known reactions of alde- the active ingredient in the dosage form can co-exist with its hydes and is present in many materials, including components equilibrium moisture content, it will be stabilized.
of packaging2,3. Parts per million levels might cause significant Perrier and Kesselring assumed that the binding energy for degradation because of its low molecular weight.The capability water vapour was comparable to that for nitrogen, or was of of formaldehyde to be adsorbed by and interact with a drug the same rank order. This assumption can be questioned. The substance can be determined during preformulation4. If it can- forces of attraction between water vapour and nitrogen for an not be eliminated it might well be that it too, like moisture, can excipient substrate could be very different because of differ- ences in dipole moment. Nevertheless, the rank order seen in Materials such as amorphous silica and microcrystalline cel- Fig. 1 is impressive and the approach merits further study. It lulose are powerful sorbents5–7 and might thus be capable of might be that dynamic vapour sorption instrumentation, functioning as ‘scavengers’ of volatile residues. Care should be which, in recent years, has been greatly enhanced in terms of taken, however, to ensure that desorption does not occur dur- sensitivity and precision, could be usefully employed to deter- ing subsequent handling or storage of product.Thus the factors mine the binding energies of common excipients for water that affect desorption of the residue from the excipient sub- vapour (or to confirm the rank order relationship in Fig. 1).
Such characterization would be a boon to the formulator deal-ing with moisture-labile materials.
Degradation by oxidation
Moisture is not the only residue with a significant vapour Oxidation is probably second only to hydrolytic breakdown as pressure that can be present in dosage forms. Lower alcohols a cause of loss of quality.The reactions are usually complex and precipitated by many factors that are difficult to separate and clarify. The facile notion that stabilization is a matter of formu- lating with an antioxidant is rarely so simple in practice, es-pecially with solid-state systems. The molecular association required to attenuate a reaction might not be readily attainable However, formulation additives have been effective stabili- zers in vitamin preparations. Tocopherol, butylated hydroxy- anisole, butylated hydroxytoluene and propylgallate have all been used to stabilize vitamins A and D (Refs 8–11). Ascorbic acid solutions have been stabilized by a combination of chelat- ing agent and antioxidant12. Ascorbic acid has also been stabi- lized by magnesium, calcium or aluminium stearate13. It is not clear, however, whether these materials effected stabilization per se or whether it was produced by antioxidants that are usually present in stearates and other fatty acids.
Oxidation can be catalysed by exposure to air or light, the presence of trace residues (for example, heavy metals), or by other components in the formulation. It can also result from acombination of all of these. Different mechanisms require dif- ferent approaches to stabilization. It is only by acquiring a basicunderstanding of the degradation process that a rational ap- proach to stabilization can be developed.
It is also possible that an additive in the excipient (such as an antioxidant) can be a surreptitious stabilizer. If this additive is replaced or removed by the supplier (for perfectly valid rea-sons), the impact on product stability could be disastrous.
Figure 2. Spectral overlay for stabilization of photolabile drugs.
Knowledge of the basic mechanism of stabilization means that (a) UV absorption spectra of nifedipine (1) and the stabilizers
effective change control arrangements can be negotiated with riboflavine 5-phosphate (2) and circumin (3). (b) Solution stability of
nifedipine (1) in the presence and absence of curcumin (2).
Figure reproduced, with permission, from Ref. 15.
Photodegradation
Exposure to light can precipitate a plethora of degradation re-
actions. These can include addition reactions in unsaturated
Spilgies used the spectral overlay approach to stabilize solu- systems, substitution reactions, polymerization, isomerization tions of a photolabile ␤-lactam using acceptable food colorants and photo-oxidation14. Suitable light-resistant packaging can, having UV spectra that went some way to providing spectral in many cases, prevent or reduce degradation. However, degra- dation during product use is also a possibility and it might be A similar approach was used by Sanderson et al. to stabilize a necessary to stabilize the formulation itself.
potential anti-psoriasis agent for application as an ointment18.
The concept of spectral overlay was pioneered by Thoma and Although stable in the formulated, packaged product, the active Klimek15,16.This approach involves formulating with a material ingredient was photolabile. Isomerization and polymerization whose UV absorption spectrum overlaps (or substantially over- reactions occurred when the ointment was exposed as a thin laps) that of the compound requiring stabilization. The impact film to simulated sunlight. In-use degradation was therefore a of damaging radiation will thus be attenuated as the excipient possibility and stabilization needed to be considered.
‘competes’ with the active compound for the photons from the Table 1 shows the effect of including benzophenones (agents radiation source. Thoma showed that the photolabile calcium used in sunscreens) on photostability.The reduced degradation antagonist nifedipine can be effectively stabilized by the natural can be ascribed to the partial spectral overlay provided by the food colorant curcumin, or by riboflavine15. Neither of these additives provided complete spectral cover but stability en- Sanderson et al. also noted that stability could be enhanced by the addition of a ‘blocker’ such as titanium dioxide and by using a brand of soft paraffin with a UV spectrum that providedpartial spectral cover (Table 2 and Fig. 5). It can be seen from Table 2 that the presence of the antioxidant ␣-tocopherol did not augment the stabilizing effect of the grade of paraffin in question. Thus the enhancement seen is more likely to be aspectral overlay effect (Fig. 5). Some aromatic residue or otheradditive in the paraffin might be responsible for the spectral cover.These findings highlight the need to be aware of the basic stabilization mechanism and the need for change control sys- tems for excipients. Change of supplier of paraffin could lead to In a similar vein, the presence of the UV absorber oxybenzone in the film coat was shown to stabilize sulphisomidine tablets18.
Successful application of the spectral overlay approach re- quires excipients with the appropriate absorption spectra. Obvi- ously they must also be free from pharmacological activity andbe non-toxic. Hence, the list of potentially useful materials is limited. Nevertheless, it is an elegant approach to stabilization ifthe pack cannot be relied on to provide the requisite protection.
Other modes of degradation
Some degradation reactions do not involve species other than
the active ingredient. Isomerization, dimerization and polym- erization and other forms of molecular rearrangement are not uncommon, particularly for drugs of large molecular mass orthose of biological origin. No other species need be involved Figure 3. Effect of food colorants on the stability of the ␤-lactam
BRL42715B. (a) UV absorption spectra. In an aqueous solution of 20 ␮g
in such reactions, although they might act as catalysts or initia- mlϪ1, quinoline yellow (1); yellow-orange S (2); and BRL42715B (3). tors. Ostensibly, it might seem that molecules with an intrinsic (b) Effect of additives on solution stability. Yellow-orange S 5.0 mg mlϪ1
‘self-destruct’ capability would be the most difficult to stabi- (black triangle), 0.5 mg mlϪ1 (black circle), 0.05 mg mlϪ1 (black square)and water (control sample) (cross); quinoline yellow 5.0 mg mlϪ1 (white lize. It is paradoxical, therefore, that some of the more success- triangle), 0.05 mg mlϪ1 (white circle), and 0.05 mg mlϪ1 (white square).
ful attempts to stabilize labile drugs have concerned materials Figure reproduced with permission from H. Spilgies (unpublished).
that are derived either directly or indirectly from biologicalsources.
Table 1. Effect of UC-absorbing agents on stability of an
anti-psoriatic compound in a paraffin-based ointment
(samples exposed for one hour in a Sol-2 light cabinet)
Stabilizer
Level (% w/w)
Degradation %
Figure 4. UV spectra of anti-psoriasis drug and benzophenones.
Oxybenzone (1); Dioxybenzone (2); and Anti-psoriasis drug (3).
Figure reproduced, with permission, from Ref. 17.
readily degrade both in solution and as solids, and such behaviour Table 2. Effect of grade of paraffin and use of ‘Blocker’
can limit their utility. However, it appears that cyclodextrins are on stability
uniquely suited to the stabilization of these labile materials and Description
Degradation %
several successful attempts have been reported. Stable complexeshave been formed with PGE (prostacyclin) and PGF (Refs 22,23). It is interesting that, in the latter case, stabilization is ef- fected by both the ␣- and the ␤-forms of cyclodextrin, although the mode of molecular encapsulation is different for each type.
The ␣-form accommodates the cyclopentane portion of the mol-ecule within the cavity, whereas, in the case of ␤-cyclodextrin, Cyclodextrins are cyclic glucose polymers with the unique one of the side-chains of the prostaglandin is inserted as shown in capability to accommodate hydrophobic ‘guest’ moieties Fig. 6 (Ref. 24). Either way it would appear that the reactive moi- within the annulus. Non-covalent links ‘anchor’ guest to host.
eties are rendered less labile by ‘spatial’ effects.
Cyclodextrins have been widely reported as enhancing the sol- The molecular inclusion properties of cyclodextrins would ubility and dissolution rate of poorly soluble drugs.Their use as suggest that their suitability as stabilizers is limited to cases in stability enhancers also capitalizes on their unique molecular which degradation is the result of molecular rearrangement. It complexation capability.This can result in improved stability of is interesting, therefore, that cyclodextrin complexation inhibits compounds with a tendency to sublime, such as clofibrate and the oxidative degradation of vitamin D (Ref. 25). On reflection, isosorbide19,20. Chemical stability can be enhanced when the this is not surprising. Positioning of the labile moiety within the reactive groups are accommodated in the cyclodextrin cavity, cyclodextrin cavity might render it less vulnerable, in the steric thereby preventing inter- or intramolecular reactions.The effect sense, to attack by many kinds of degrading species.
of such molecular encapsulation is shown in Table 3 for a It should be stated that cyclodextrin complexation is no prostaglandin E (PGE ) derivative21. There was apparently a guarantee that instability will be remedied. In some cases, the double benefit with the use of this compound in that the dis- effect might be opposite to that desired. It has been reported solution rate was also significantly enhanced.
that incorporation in ␤-cyclodextrin accelerates vitamin K de- Prostaglandins, being derivatives of arachidonic acid, have a composition in solution26, and there are other accounts of basic molecular structure that is characterized by a cyclopentane complexation being of little benefit or of variable effects27.This nucleus with two hydrocarbon side-chains giving a ‘hairpin’ con- ought not to be surprising. Generally, the hydrophobic part of figuration. Like many materials of natural origin, prostaglandins the guest molecule will be accommodated within the cyclo-dextrin cavity, but the labile group(s) might reside in a differ-ent part and not be protected by encapsulation. Indeed, theycould be rendered more vulnerable. Here again, knowledge ofthe basic modes of degradation should indicate the chances ofsuccessful stabilization by molecular incorporation.
Figure 5. UV absorption spectra of anti-psoriasis drug and soft
paraffins. Anti-psoriasis drug (SB 201993) (hatched line); Oxybenzone
Figure 6. Modes of encapsulation of prostaglandin F and cyclodextrins.
(dotted line); Dioxybenzone (hatched line). Figure reproduced, with (a) Shows ␣-cyclodextrins, and (b) shows ␤-cyclodextrins. Figure
reproduced, with permission, from Ref. 24.
hydroxypropyl derivative appears to be a safer option for par- Table 3. Effect of cyclodextrin complexation on stability of
enteral administration (see the recent excellent review on cy- dimethyl prostaglandin E1 (storage at 60°C)
clodextrins by Stella and Rajewski for a more comprehensive Degradation %
treatise on many facets of cyclodextrin properties28).
Cyclodextrin
Cyclodextrins will not be a panacea for all stability problems, ‘as is’
but clearly they have a niche and are being increasingly used incommercial pharmaceutical products.
The experiences with cyclodextrins illustrate the potential of ‘steric stabilization’. If a drug can be ‘fixed’ to an excipient, by adsorption or other non-covalent means, the possibility exists that, on occasion, labile groups will be made less available to Table reproduced, with permission, from Ref. 21.
incoming moieties with which it might react. The possibilitythat dissolved drugs can be ‘structured’ by using appropriate The molecular size of cyclodextrins, relative to most drug solvent systems or soluble additives is an attractive concept for substances, means that their use is limited to highly potent, stabilization. In this context, prostacyclin is stabilized by vari- low-dose drugs. Otherwise, the large cyclodextrin/drug ratio ous albumins29. Human serum albumin also inhibits the aggre- in an equimolar complex means that a dosage form is likely to gation and oxidation of solutions of iron protoporphyrin. Ma- be too large to be practical. The process for preparation of the terials as diverse as polyvinylpyrrolidone (PVP), caffeine, niacin complex can also be a drawback. The most effective mode of and antioxidants also enhanced protoporphyrin stability in so- complex formation appears to involve dissolution of the drug lution and in lyophilized solid30. The beneficial effects of albu- and cyclodextrin in a suitable solvent, standing to allow the mins on prostacyclin stability suggest that Mother Nature complex to form, and then drying. If the drug is very unstable, might have evolved ways of stabilizing natural mediators. This there could be problems with maintaining molecular integrity could be fertile ground that merits exploration for strategies for during the solution phase, particularly if heating is required and the time for complex formation is prolonged. Dry encap- Mupirocin (pseudomonic acid A) is an antibacterial for topi- sulation (micro grinding) also appears to be feasible, but the cal infections. It is of natural origin, being derived from fer- consistency of such a preparative technique might be open to mentation of Pseudomonas fluorescens. It readily degrades in solution and in the solid state31. Degradation in the solid state is initiated Like all ‘novel’ excipients, cyclodextrins cannot be used with- by rupture of the epoxide ring followed by rearrangement to out due consideration of the safety implications. Both ␣- and two bicyclic compounds (Fig. 7). No other molecular species ␤-cyclodextrins are present in a number of commercial prod- appears to be involved and it was therefore difficult to concep- ucts, although approval seems to vary from country to country.
tualize strategies for stabilization. However, the finding that re- Their use in oral preparations seems to be gaining general ac- arrangement was reduced when the compound was dissolved ceptance. Inclusion in parenteral products is more problemati- in polyethylene glycol was capitalized on to develop a viable cal, particularly if inclusion levels are high. Renal toxicity has commercial ointment formulation32. What was surprising was been reported for both forms, possibly as a result of depressed that the drug dissolved in this vehicle was more stable than as a solubility in the environment of the proximal tubule. The solid, reversing the normal trend for materials to be more stablein the solid state than in solution (Table 4). Although the basicmeans of stabilization has not been elucidated, it is difficult to Table 4. Effect of formulation on stability of mupirocin
Condition
Degradation %
(months)
Drug substance
Drug in solution
(2% in PEG 400)
Figure 7. Modes of degradation of mupirocin.
avoid the conclusion that some steric hindrance effect prevents claims, as evidenced by the number of patent claims referenced in this article. Thus there are potential commercial advantagesfor organizations with the will to make the effort and the skills Overview
to provide inventive approaches to stabilization.
Many products contain excipients that can be categorized as sta-bilizers in a general sense. Using suspending agents to prevent References
sedimentation, adding a preservative to prevent microbial spoilage or a buffer to adjust pH for optimum stability are all 2 Desai, D.S. et al. (1994) Int. J. Pharm. 110, 257–265 examples of excipients being added to enhance product stabil- ity. However, such approaches are part of the stock-in-trade of 4 Stephenson, S.A. et al. (1998) J. Pharm. Pharmacol. 50 (Suppl.), 122 the formulator and are expounded in many articles and text- 5 Nimry, S.S. et al. (1997) Int. J. Pharm. 149, 115–121 books. Hence, they have not been discussed here. Instead, this 6 Daniels, R. et al. (1986) Drug Dev. Ind. Pharm. 12, 2127–2156 review has attempted to focus on cases in which fundamentally 7 Tobyn, M.J. et al. (1998) Int. J. Pharm. 169, 183–194 unstable drugs can be transformed into viable medicinal prod- ucts by formulating with appropriate excipients that have some direct effect on the molecular integrity of the active ingredient.
The question can be posed whether a strategy for stabilization is appropriate. In the era of combinatorial chemistry and 12 Kirkova, J. et al. (1977) Tr. Nauchno-Issled. Khim.-Farm. Zh. 11, 100 genomics-based research, the view of the pundits seems to be that unstable entities should be ‘selected out’ early in the discov- Modern Molecular Photochemistry, pp. 121–130, Benjamin ery and development process. After all, poorly stable therapeutic agents are likely to progress slowly (or not at all) to the market- place. In the environment promised by genomics, it is postu- Dtsch.Apoth. Ztg. 120, 1967–1972 lated that there will be an embarrassment of riches in terms of 17 Sanderson, F.D. et al. (1996) The Photostability of Drugs and Drug Formulations compounds to select for clinical evaluation, thus attempts to sta- (Thonnessen, H.H., ed.), pp. 141–154, Taylor & Francis bilize a poorly stable material might not seem warranted.
18 Matsuda,Y. et al. (1979) Proceedings of the 3rd Symposium on Stabilisation and However, there are other reasons why the capability to stabi- Evaluation Methodology of Pharmaceutical Preparations, 20–21 November, lize labile materials should remain an option for the formulator of medicinal products. The activity, specificity and freedom 19 Eukama, K. et al. (1983) Pharm.Acta Helv. 58, 338 from toxicity might be directly related to molecular fragility.
20 Eukama, K. et al. (1985) Int. J. Pharm. 25, 339 Materials derived from biotechnological or other natural 21 Eukama, K. et al. (1979) J. Pharm. Sci. 68, 1059–1060 sources are a case in point. Stabilization using appropriate 22 Eukama, K. et al. (1981) Chem. Pharm. Bull. 29, 213–219 excipients might be the only way in which many of these 23 Eukama, K. et al. (1984) J. Pharm. Sci. 73, 382–384 materials can be converted into viable products.
24 Hirayama, F. et al. (1980) Chem. Pharm. Bull. 28, 1975–1980 Furthermore, in the context of environmental impact, the 25 Shima, A. et al. (1977) Japanese Patent 77,130 use of excipients to stabilize an unstable active ingredient is an et al. (1982) Pharmazie 37, 725 attractive concept. A product could be developed that retains its 27 Duchene, D. et al. (1986) Drug Dev. Ind. Pharm. 12, 2193–2215 quality while the drug and other formulation components are in close association. Rapid breakdown would follow disposal (or ingestion) with the attendant separation of drug and stabi- lizer. Stable residues of potent materials would not be an issue.
30 Cannon, J.B. et al. (1995) PDA J. Pharm. Sci.Technol. 49, 77–82 It should also be noted that turning an unstable entity into a 31 Clayton, P.J. et al. (1979) J. Chem. Soc., Perkin Trans. 1, 838–846 stable product offers possibilities for intellectual property Contributions to Pharmaceutical Science & Technology Today
We welcome suggestions for short reports, opinion articles and full reviews for publication in Pharmaceutical Science & TechnologyToday. Potential authors should contact the Editorial Office in the first instance with a brief outline of the scope of the proposed contribu-tion. Article proposals should be directed to: The Editor, Pharmaceutical Science & Technology Today, Elsevier Trends Division, 68 Hills Road, Cambridge, UK CB2 1LA (tel: ϩ44 1223 315961, fax: ϩ44 1223 464430, e-mail: pstt@elsevier.co.uk).

Source: http://www.callumconsultancy.com/articles/ExcipientsAsStabilizers.pdf