Antimicrobial Preservatives Part Three: Challenges Facing Preservative Systems

Introduction

The third and final article in this series deals with the many external challenges that face formulators who are developing preservation systems for multi-use oral, topical and parenteral medicinal products. This article provides current perspectives on safety, regulatory, public relations issues. Mention is also made, where relevant of uses of antimicrobial agents in activities such as disinfection procedures and in non-pharmaceutical products. Two case studies are presented to highlight the very real issues encountered when trying to preserve multi-use pharmaceutical products.

Safety and Side Effects of Preservatives

The ideal preservative should only show biological activity against microbial organisms, with little or no adverse effects on mammalian cells. However, as these agents are broad-based cytoplasmic poisons, this is an extremely challenging remit and, in reality most preservatives have some effect against both microbial and mammalian cells. As Paracelsus, the 16th century physician and father of modern day toxicology observed, it is the “dose that makes the poison”. Therefore, most regulatory agencies will challenge the applicant to show antimicrobial efficacy at the lowest feasible concentration.

FDA states that, “All ingredients used in a licensed product…, shall meet generally accepted standards of purity and quality. Any preservative used shall be sufficiently nontoxic so that the amount present in the recommended dose of the product will not be toxic to the recipient...” (FDA, 2017).1 It is important to understand that “preservatives are not examined in isolation, rather, 21 CFR 610.15(a) specifically directs that preservatives be examined in the context of the overall product and the recommended dose”.2 Thus, “FDA evaluates whether a preservative contained in a biological product is at such levels that the finished product itself, when used at the recommended dose, is not toxic to the recipient”.

The EMA has similar guidelines for excipients (preservatives and antioxidants in particular)3 , and a similar approach to reviewing and approving products. The addition of antimicrobial preservatives should be supported by rationale as to (i) why the excipient is included, (ii) proof of efficacy, (iii) the method of control in the finished product, (iv) details of the labelling of the finished product and (v) relevant safety information. EMA originally indicated that labelling must be aligned with the relevant EU council directives (92/27/EEC and 81/851/EEC); however, if the product is in a multi-dose container, but there is no preservative present, i.e. preservative-free because (a) it is intended for single use only, e.g. cytotoxics, (b) the product is self-preserving or (c) the product is oil based, e.g. ointments; then the applicant should indicate the absence of the preservative(s). Importantly, EMA indicated that “this would not only emphasize the risk associated with the use of such products, but also aid the physician to specifically identify a product without a preservative”. Thus the original intent was to emphasize that there were risks and balances inherent in the selection or non-selection of preservatives. That is preservatives, by their nature and function are potentially toxic and pose a safety risk to vulnerable populations but, equally non-preserved formulations that are intended for single use only, but are in fact used in a multi-dose fashion pose equally significant risks of microbial contamination to all patient populations. EMA broadened this guidance to include all excipients in the marketing dossier4 and this was further updated in 2006.5

Of particular concern to the EMA is the use of preservatives in pediatric products. Those preservatives covered under labelling requirements include parabens, benzalkonium chloride, benzoic acid, benzyl alcohol, boric acid and ethanol. Antimicrobial preservatives are “normally added to prevent microbial proliferation arising under in use conditions. The properties are due to certain chemical groups which are usually harmful to living cells and might therefore be associated with certain risks when used in humans”.5 The EMA articulated that, “Any risk identified for an excipient and in particular a CMR (carcinogen, mutagen and reproductive risk) substance would be acceptable only on condition that this excipient cannot be substituted with a safer available alternative, or that toxicological effects in animal models are considered not relevant for humans (e.g. species specific, very large safety ratio) or where the overall benefit/risk balance for the product outweighs the safety concern with the product”.6 The key principle is that “inclusion of antimicrobial preservatives in a finished product need special justification. The use of these substances should be avoided in general, especially when considering the suitability of related formulations to the pediatric population. The concentration used should be at the lowest feasible level”.

There are many food preservatives that are considered to be too toxic to be used in pharmaceuticals, e.g. sulphites, nitrites, nitrates, etc.; however, equally, there are a number of pharmaceutical preservatives, e.g. phenols and organomercurials that are routinely used in parenteral and vaccine products, that would be considered to be too toxic as additives in food. This reinforces the value of a Microbial Risk Assessment in determining which factors are the most important from a risk/benefit perspective.

Alcohols are generally considered to be safe, with the exception of ethanol. Current EU labelling for alcohol is extensive and focusses on three main exposure levels; ≥ 75 mg/kg/day, 6 - < 75 mg/kg/day and 1- 6 mg/kg/day dependent on age and route of administration.7 In addition, benzyl alcohol is not recommended for use in parenteral products for pediatrics due to fatal toxic syndrome in low weight neonates and this ban was subsequently broadened to cover all products for this age-range.8 Indeed, this material should not be used in infant and children’s products up to 3 years of age due to a combination of fatal toxic syndrome and adverse allergenic reactions. Benzyl alcohol has been allocated an ADI (acceptable daily intake) of 0-5 mg/kg/day.9 Sensitization to benzyl alcohol is generally low in topical products.10 Similarly, the long chain alkyl alcohols, cetyl and stearyl alcohol, are infrequent sensitizers.10 2-Phenylethanol can be mildly irritant to skin, eye and mucous membranes.11,12

Carboxylic acids, e.g. benzoic acid can be gastro-irritant and also generally mildly irritant.13,14 Current EU labelling indicates that benzoates are mildly irritant to the skin, eyes and mucous membranes and that in parenteral products they may increase the risk of jaundice in new born infants.15 The ADI based on its use as a food additive is currently 0-5 mg/kg.16 Sensitivity is generally considered to be mild in nature.14 Similarly, sorbic acid sensitization is considered relatively uncommon.10 There are some clinical findings suggesting effects on “activity and attention in children, however their relevance has not been evaluated nor confirmed”.17 Similarly, allergic dermatitis and allergic conjunctivitis are considered to be relatively infrequent in nature. Systemic toxicity has not been reported.10

Boric acid is also commonly used as a preservative in ophthalmic products, topical products, i.e. creams and ointments as well as cosmetic products. Human and animal studies show that more than 90% of the administered dose of inorganic borate is excreted unchanged as boric acid.18 An oral permitted daily exposure (PDE) limit for boron has been assessed at 10 mg Boron/day. SCCS (Scientific Committee on Consumer Safety) set an upper intake level (UIL) in food of 10 mg B/person/day in adults and they considered that this UIL is also applicable to pregnant and lactating women. In addition, the UIL values for children were derived from the UIL for adults on an extrapolated comparative body surface area basis and values of 3, 4, 5, 7, and 9 mg boron/person/day for children aged 1–3, 4–6, 7–10, 11–14 and 15–17 years of age, respectively, were derived. These UIL values only apply to the intake of boron as boric acid and borates.19 Current EU labelling20 is slightly more conservatives and indicates that the following age specific limits (mg boron/person/day) apply:
< 2 years, 1 mg; < 12 years, 3 mg; < 18 years, 7; > 18 years, 10 mg.

The parabens are considered unsuitable for parenteral and ophthalmic use due to irritancy.21-23 There are also numerous reports of delayed hypersensitivity to their topical use.10 However these reactions are fairly uncommon.24 However, in the EU there are also concerns about oral use of parabens with respect to the potential estrogenic activity of these preservatives, especially in neonates and the more general pediatric population.25 A recent reflection paper indicated that methyl paraben is not associated with any estrogenic issues on male or female reproductive organs in developing juvenile rodents or in any embryo-fetal development studies and consequently concentrations up to 0.2% (equating to a maximum daily intake of 140 mg/day) are deemed safe, irrespective of the age of the pediatric population. In contrast, the situation with respect to propyl paraben is much less clear cut. However, based on the effects on female reproductive organs a conservative NOEL (no observed effect level) of 100 mg/ kg/day, this equates to a very conservative PDE value of 2mg/kg/day can be assigned. In addition, although there was no clear cut position articulated with respect to ethyl paraben, it is clear from the position paper that the estrogenic effects are linked with increasing size of the alkyl group.25 As such increased regulatory focus can be expected for pediatric products using these preservatives. Consequently, companies should avoid propyl paraben in pediatric formulations and assess very carefully whether combinations of methyl and ethyl paraben are appropriate for the intended use; both from a scientific and regulatory perspective. The situation with respect to the use of parabens in products destined to be only used in the adult population are less clear, as such it is difficult to provide similar clear and unambiguous advice.

Adverse reactions attributable to phenol are fairly rare, probably reflecting the low inclusion levels that are typically utilized. The tolerable daily intake (TDI) for phenol has been cut from 1.5 mg/kg/ day (i.e. 75 mg for a 50 kg adult) to 0.5 mg/kg/day (i.e. 25 mg for a 50 kg adult). The European Food Safety Authority (EFSA) contact materials, enzymes, flavorings and processing aids (CEF) was asked to re-assess the TDI based on findings from a new safety study.26 Phenol is used in a variety of US consumer products including mouthwashes, throat lozenges, and throat sprays. It is also currently used as a preservative in three FDA-approved available vaccines, Pneumovax 23, Typhim Vi and ACAM2000 and each of these vaccines contains 0.25% phenol.27 It is also present (sometimes in combination with m-cresol) in many of the more mature biopharmaceutical products, these being largely protein replacement or signaling proteins. More recent, monoclonal antibody products do not contain preservatives, being contained in single-dose vials or auto-injector-type presentations.

Chlorocresol, while less toxic than phenol can still be an irritant to skin, eyes and mucous membranes. It cannot be used in intrathecal, intra-cisternal or peridural injections.28,29 It is also used as a cosmetic preservative in skin care and suntan formulations. Chlorocresol is approved for use as a preservative by the European Union at concentrations of up to 0.2%; however, it can still be used at different concentrations when used for other functions and it is banned from use in topical products intended to contact mucous membranes. Chlorocresol is approved as an indirect food additive in the United States, i.e. not present in the final product.30 In clinical studies, 2% chlorocresol appears to be a skin irritant.

Chloroxylenol is generally considered to be less irritating than chlorocresol.10 Cross sensitivity to chloroxylenol has been reported.32,33 It is commonly used in cosmetic products as an antimicrobial agent at concentrations of up to 5.0% w/v.32 It is readily absorbed through the human skin and gastrointestinal tract. Neat chloroxylenol, i.e. 100% was a moderate irritant in rabbit’s eyes, in contrast a 0.1% chloroxylenol solution is non-irritant when applied to rabbit’s skin.32 Chloroxylenol is non-mutagenic in the Ames test (± S9 activation). No carcinogenicity or satisfactory teratogenicity studies have been reported in the literature. In clinical studies, formulations containing up to 1.0% chloroxylenol were reported to be non-sensitizing and non-irritating to the human skin. There was a 1% incidence of skin sensitization amongst the 1752 dermatitis patients exposed to 1.0% of chloroxylenol. On the basis of all of the available safety data it was concluded that chloroxylenol was safe when used as a preservative in cosmetic products.33 Hexachlorophene usage has declined because of concerns over neurotoxicity.10

Sensitivity to the quaternary ammonium compounds (QACs) e.g. benzalkonium chloride (BKC, also known as BAC in US) is infrequent, but it is a contact irritant and can exacerbate pre-existing dermatoses.10 The safety profile of BKC (or BAC) as an orally dosed agent is relatively benign; BKC is not genotoxic, carcinogenic and does not show reproductive toxicity, but it is rarely used in oral products. Where data are available there are no significant differences in adverse event profiles between adults and children. Consistent evidence of BKC-related toxicity were not evident from a review of various clinical investigations34 and no lower safety limits for BKC for the general population have been recommended.35 Most of the shorter term clinical or non-clinical safety data are ambiguous.

  • A one-year ophthalmic study in rabbits and monkeys using eye drops with BKC concentrations of 0.005% and 0.01% showed no adverse ophthalmological changes, including irritation or corneal damage.36
  • Current EU labelling indicates that for ocular products BKC may cause eye irritation. Levels in ophthalmic products available in the US range from 0.003-0.02%. Care should be taken to avoid the long-term use in ophthalmic preparations, “as this weakly allergenic but highly toxic compound exerts dose- and time-dependent effects”.38
  • BKC is suspected of causing nasal ciliary beat stasis when dosed 8-times daily in rodents.37
  • BKC is known to discolor soft contact lenses. Patients should remove contact lenses prior to application and wait at least 15 minutes before reinsertion.
  • BKC is an irritant in topical products and may cause skin reactions.
  • BKC may cause bronchospasm in respiratory products and should be limited to 10 micrograms/delivered dose.35

A maximum concentration of 0.02% of benzethonium chloride is recommended for ophthalmic and parenteral products.39

Historically, organomercurial preservatives were widely used in topical and parenteral formulations. However, concerns about the toxicity of mercurial compounds are regularly expressed40 leading to usage constraints, e.g. intra-vaginal use.41 FDA recently commented on the substitution of the organomercurial preservative thimerosal in vaccines.2 They concluded that only a small number of licensed and approved vaccines in the US still contain thimerosal preservative. Epidemiological data indicated that there is no causal relationship between thimerosal-preservative containing vaccines and autism or other pediatric developmental disorders (see public relations discussion on this subject). All available evidence supports the Agency’s conclusion that all currently licensed vaccines containing thimerosal preservative have been proven safe under the applicable regulatory requirements. Substituting thimerosal preservatives in currently licensed vaccine products has the potential to modify the safety and effectiveness, as well as the stability, of these vaccines. Tellingly, FDA indicated that a comprehensive development program would be required to establish the safety and effectiveness of non-preserved or vaccines preserved with alternate preservatives. Similarly, EMA and WHO have also provided commentary on the same issue.42 Thiomersal is the only preservative that is used in centrally authorized human vaccines (mainly Influenza pandemic vaccines) within the EU. Alternative preservatives in some human vaccines have been registered nationally or via mutual recognition practices/MRP/DCP, for example, phenol in Typherix®, Typhim Vi® and Pneumovax II® and 2-phenoxyethanol in Tetravac®, Pediacel®, Revaxis®, Repevax®. There is movement towards the elimination of preservatives in vaccines i.e. using single dose presentations. Many new vaccines are developed as preservative-free, single dose presentations. There are some multi-dose vaccines without preservative, i.e. Celvapan®/Vepacel®. After opening, Celvapan®/Vepacel® is required to be used within 3 hours. However, in comparison, Pandemrix® and Focetria® (preserved, multi-use) should be used within 24 hours. The use of thiomersal appears to be mainly based on historical experience, e.g. Influenza vaccines. However, there are few comparative data available comparing the efficacy of thiomersal and any alternative preservative(s). Indeed, there are risks inherent in any replacement strategy; (i) substituting an established, efficacious preservative with an unproven alternative may increase the risk/benefit ratio, rather than decreasing it, i.e. the new preservative may require higher concentrations and there are significant uncertainties with regard to the effect on processing/stability/efficacy of both the preservative and the active agent within the vaccine when the preservative(s) in the existing vaccine are changed.42

Thimerosal is a potent sensitizer, particularly in topical products.10 Reactions occur in soft contact lens users and up to 10% of wearers may be impacted.43 Hypersensitivity has been reported, hence it is recommended that it should not be used in either eye drops44 or vaccines.45 Indeed, EMA and FDA have recommended the general phasing-out of thimerosal in vaccines.46,47 However, despite a concerted media campaign (see Public Relations Challenges) and public loss of confidence, regulatory bodies found no link between thimerosal and incidences of autism.48

Biguanides, e.g. chlorhexidine are used primarily as topical disinfectants and oral mouthwashes.49 They still have a significant role to play in the disinfection of catheters and skin surfaces prior to surgical intervention50, but their role as broad-based preservatives in non-sterile drug products is diminishing. Levels in ophthalmic preparations must not exceed 0.05%.51

The formaldehyde donators, e.g. imidurea52, bronopol53 are widely used in cosmetic and topical products, being generally considered non-irritant (up to 0.1%) and non-toxic.54 However, there is a growing body of evidence linking both with sensitization, caused by the release/action of formaldehyde, particularly in vulnerable populations.10,55 Sensitization rates to formaldehyde and formaldehydereleasing agents are reportedly lower in Europe than in the United States due to stricter regulations regarding their use.56 A study on formaldehyde release from eight preservatives (e.g. methenamine [MA], paraformaldehyde [PF], poly(p-toluenesulfonamide-co-formaldehyde) [PTSAF], quaternium-15 [QU], imidazolidinyl urea [IU], diazolidinyl urea [DU], dimethyloldimethyl hydantoin [DMDM] and bronopol [BP]) under various conditions was recently undertaken.57 Formaldehyde release was dependent on the matrix, pH, time and temperature. Formaldehyde release was in the order of PF > DU > DMDM ≈ QU ≈ IU > MA > BP > PTSAF.57

EDTA is often used as a preservative enhancer, being included in ophthalmic and intranasal products, enabling the inclusion level of the potentially irritating BKC to be reduced. A recent report highlighted the use of EDTA in combination with N-hydroxy-methyl-glycinate (NIG), as a novel preservative system in multi-dose ophthalmic topical medications.58 Rabbit corneal cells (SIRC) were challenged against a series of ophthalmic preservatives, i.e.; benzalkonium chloride (BKC), polyquaternium-1 (PQ-1), sodium perborate (SP) and NIG±EDTA at a variety of different concentrations, i.e. 0.001% and 0.002% and different treatment times, i.e. between 30 minutes to 120 hours. Limited cell toxicity was shown by NIG and SP at either concentration, whilst low toxicity was observed for PQ-1 at the highest dose for the longest duration, i.e. 120hours. BKC showed the highest toxicity, even after only 30minutes exposure. EDTA (0.1%), either alone or in combination with NIG (0.002%), showed no toxicity after 24hours. The new preservative system NIG/EDTA, at doses known with known antimicrobial properties, has very low toxicity and therefore it can be used in multi-dose ophthalmic products.58 Potassium sorbate and EDTA were used as a preservative system in a novel ion-activated in situ gel-forming estradiol (E2) eye drop formulation59 and BKC and EDTA were used as preservatives in a metoclopramide hydrochloride nasal spray.60 In a recent study, biguanide preservatives e.g. chlorhexidine and polyaminopropylbiguanide (PAPB), showed the best anti-staphylococcal activity; whereas, EDTA showed the best anti-pseudomonal preservative. A synergistic combination of chlorhexidine and EDTA showed superior anti-microbial efficacy against P. aeruginosa, than either agent alone. In addition, a combination of chlorhexidine (30 ppm), PAPB (5 ppm), and EDTA (5,000 ppm, i.e. 0.5%) was found to have 3-7 times greater anti-pseudomonal efficacy than any currently available preservative system and this novel system could help to reduce the incidence of microbial keratitis for contact lens users.61

Dose-related broncho-constriction has been observed when EDTA is used in nebulizer solutions, and its use is now discouraged in such products. EDTA salts can cause nephrotoxicity and should not be used in patients with renal impairment. EDTA can readily chelate calcium, leading to calcium depletion.62 Despite such reports EDTA is generally considered as having low toxicity.

Recently, the relative cytotoxicity of commonly used preservative in US licensed vaccines were assessed using human neuroblastoma cells.63 The authors found phenol to be the safest of these parenteral preservatives; the rank order was phenol330-fold). Worryingly, for the compounds tested, with the exception of 2-phenoxyethanol, the efficacious anti-microbial concentrations necessary to induce significant bactericidal efficacy, i.e. aligned with pharmacopoeial requirements, were significantly higher than those routinely present in US licensed vaccine/biological preparations. This implies that it is clinical safety criteria that over-ride microbial safety considerations when developing these products. The authors further commented that none of the preservatives in US licensed vaccine/biological products can be considered “ideal”, i.e. optimal efficacy with minimal toxicity, and their ability to fully comply with the requirements of the major pharmacopeias for anti-microbial preservative efficacy (AET) is in doubt. They recommended that future vaccines/biologics formulations should be sterile preparations for single dose use, eliminating the need for preservatives and an unnecessary risk to patients.

Therefore, the potential for undesirable side effects needs to be carefully considered when selecting a preservative system. At the same time these concerns need to be balanced with the low concentrations and the low probability of adverse effects, particularly if the medication is for short term use.

Public Relations Challenges

The organomercurial preservative, thiomersal has been used in vaccines since the 1930’s following the deaths of 28 children from staphylococcal infection following immunization with an unpreserved diphtheria vaccine. During the past two decades, reports claiming a causal link between pediatric vaccines and autism have claimed that thiomersal is the “culprit”.64 Regulatory agencies in some countries have accordingly moved to eliminate it or to reducing levels in pediatric vaccines.46,47 However, it is still commonly utilized in influenza vaccines.

The claim that autism was associated with a special form of mercury poisoning was subsequently refuted by the American Academy for Pediatrics.65 An independent committee of the US Institute of Medicine and the US National Institute of Health (NIH) reported that “neither thimerosal-containing vaccines nor MMR vaccine are associated with autism. The hypothesis regarding a link between autism and MMR vaccine and thimerosal-containing vaccines lack supporting evidence and are only theoretical. Future research to find the cause of autism should be directed toward other promising lines of inquiry that are supported by current knowledge and evidence and offer more promise for finding an answer.65 A recent study in Italy backed up this proposition. No link was found between the presence of thiomersal in whooping cough vaccines and autism in 1403 children, 10 years after immunization.66

Media reports have linked parabens in under-arm deodorants with breast cancer.67,68 Cancer specialists consider that there is no plausible biological mechanism to explain such a link. Furthermore, the dermal enzyme, Esterase III in human keratinocytes is present at sufficient concentrations to hydrolyze any paraben residues on skin from applied ointments and other dermal products.70

Preservation Challenges: Case Histories

The following case studies highlight some of the issues facing pharmaceutical scientists, with respect to the development of preservation systems. They highlight those programs aimed at developing (i) multi-use intra-nasal formulations and (ii) multi-use oral products.

Case Study 1: Preservation of a Multi-use Intra-Nasal Formulation

Early preclinical studies showed that the potassium salt of compound A, was a potent and selective inhibitor of the α-4 chain of the VLA4 integrin (very late antigen) and therefore potentially useful for the treatment of seasonal rhinitis. Compound A is anionic and highly soluble in water at pH 7 (200mg/ml), but the aqueous solubility decreases markedly with decreasing pH (e.g. < 2mg/ml at pH 5). In addition, the typical administered dose volume in intra-nasal products is about 100 µl or 0.1ml. The optimal dose for compound A was defined as 15 mg/ml for nasal applications, i.e. 150mg/actuation. Attempts to preserve this aqueous solution highlighted several confounding issues. The proposed high dose, which necessitated a high solubility at a target product pH of 7, restricted the use of many of the common preservatives with optimal preservative efficacy at acidic pHs (i.e. pH 4.5 – 5.0), for example benzoic acid, benzyl alcohol, sorbic acid, etc.

Incompatibility was observed between the anionic drug substance and cationic preservatives, e.g. quartnerary ammonium compounds. Inclusion of BKC or benzethonium chloride resulted in product cloudiness, and the addition of co-solvents did not resolve the problem. Synergistic mixtures of parabens and 2-phenylethanol were evaluated and whilst meeting USP AET criteria71, failed the more stringent Ph.Eur requirements72. The parabens also imparted a metallic taste to the product.

Organomercurials were not considered, from a safety and hypersensitivity perspective as were the phenolic preservatives, phenol and chlorocresol. EDTA was compatible with the drug substance, but there were concerns that EDTA could affect the binding of the drug to α-4 side chain (of the VLA-4 integrin), at the local site of action within the nose, making its inclusion in the formulation undesirable. Thus, a combination of the pH-solubility profile, incompatibility, palatability and potential for pharmacological interactions depleted the entire anti-microbial preservative palette.

Preservatives, not previously used in intra-nasal formulations and thus likely to incur additional regulatory scrutiny together with additional safety testing were assessed. However, there were still some intractable issues. 2-Phenoxyethanol was compatible with the formulation constituents, but it has a slight odor, which was considered unacceptable by the commercial groups. In addition it has a very narrow spectrum of anti-microbial activity, so is unlikely to be efficacious on its own.73

4-Chloroxylenol, which is the least sensitizing of the common phenols and piroctone olamine74 were insoluble in the formulation constituents and therefore inappropriate. Phospholipid PTC (an alkyl polyglucoside preservative)75 was physically incompatible with the formulation constituents. Sodium Iodate76 was physically and chemically compatible, but it is only utilized in oral mouthwashes. In addition, it has been linked with retinal degeneration and was therefore considered inappropriate for administration to the nasal mucosa.

Finally, purite76, a stabilized oxychloro-complex was evaluated. It was physically and chemically compatible with the formulation constituents and effective at neutral pH’s. It has similar preservative capability to parabens for fungal organisms, but was less effective against yeasts and molds. However, it would still require combination with other preservatives to meet the requirements of the Ph.Eur, as well as comprehensive safety studies for the intranasal route of administration. Development of the compound was terminated for reasons other than preservation, but the summary highlights the barriers to dosage form development and significant resources required, even for simple formulations for use in early clinical studies.

Case Study 2: Preservation of a Multi-Use Oral Formulation

The need for pediatric/geriatric dosage forms for patients who have difficulties swallowing (i.e. dysphagia) standard solid oral medications, e.g. tablets and capsules, is often the stimulus to develop an oral liquid/ suspension multi-use product. In addition to adequate preservation, the greatest formulation challenge is often the development of a palatable dosage form that children will accept.77 Indeed, poor compliance caused by unacceptable taste is a major challenge in pediatric and geriatric medicines.

Compound B was approved for use in depression and anxiety but noncompliance, particularly in pediatric and institutionalized geriatric patients was a significant clinical issue. Due to relatively high solubility (4.9mg/ml) and extremely bitter taste, most common taste-masking approaches were inappropriate. A promising strategy was the use of ion exchange resins to mask the bitter taste.78 This approach relied on the formation of an insoluble and therefore tasteless drug-resin complex between the cationic drug and a cation exchange resin (CER) at buccal pHs (ca. pH 6), which can then rapidly dissociate at gastric pHs (ca. pH 1.2) to release the drug for rapid and complete absorption.

A literature review identified the most promising resin to be the strong CER, Amberlite IRP-69, which is a compendial material, available as a pharmaceutical grade material (polacrilin potassium). IRP-69 readily formed palatable complexes with compound B and as the intrinsic pH of the complex was about pH 4, the formulation was preserved with sodium benzoate, which has a pH of optimal anti-microbial efficacy at the same pH.14 A prototype formulation showed compliance with USP71 and Ph. Eur.72 AET criteria and on that basis this product was progressed to a comparative bioavailability study versus the existing commercial tablet. However, the resin-based formulation was shown to be bio-inequivalent. This was attributable to the very strong binding efficiency of the strong CER with the cationic drug. The complex did not adequately dissociate under fasting conditions (gastric residence time about 30-minutes) resulting in incomplete absorption.

The product was re-formulated using the weak CER, Amberlite IRP-88. The intrinsic pH of this formulation was higher (ca. 6) than the previous Amberlite IRP-69 formulation and perhaps unsurprisingly given the non-optimal pH for the preservative system, the reformulated product subsequently failed the USP71 and Ph. Eur.72 AET criteria. Attempts to readjust the pH to pH 4 (optimal for sodium benzoate efficacy) improved the preservation efficacy, but adversely impacted palatability.

The parabens21-23 have a wider effective pH range (pH 4-10) than sodium benzoate and therefore a design of experiments utilizing various ratios of these parabens was evaluated. Butyl paraben was found to be incompatible and formulations discolored rapidly. Hence mixtures of methyl and propyl paraben were evaluated. Combinations of methyl and propyl paraben utilizing lower concentrations of methyl paraben (0.060-0.015% w/v) and higher concentrations of propyl paraben (0.20- 0.050% w/v) were ineffective at adequately killing fungal organisms (in particular Aspergillus brasilensis) in AET studies; which was surprising as anti-microbial efficacy of the parabens is reported to increase with increasing alkyl chain length.21-23 However, the inverse combinations utilizing higher concentrations of methyl paraben (0.20-0.050% w/v) and lower concentrations of propyl paraben (0.060-0.015% w/v) had adequate preservative efficacy and the product complied with both USP71 and Ph. Eur.72 AET criteria.

Additional investigations demonstrated that both parabens were adsorbed onto the weak CER, but they had very different binding efficiencies. The less hydrophobic methyl paraben bound less effectively to the weak CER (16-20% binding), than the more hydrophobic propyl paraben (45-56% binding) over the same concentration range. As bound preservatives are ineffective at killing micro-organisms this explained why higher concentrations of propyl paraben were less effective than corresponding concentrations of methyl paraben.

Thus, a combination of pH-preservative effectiveness, palatability, bioinequivalence, chemical incompatibility, and binding of preservatives to a key excipient, required extensive formulation activities to develop an adequately preserved, palatable and bioequivalent formulation that was subsequently commercialized.

Conclusions

This article demonstrates the many challenges facing the formulation scientist. In addition to safety and regulatory concerns, there is a greater public debate about the perceived safety of preserved medicinal products. The lay public often questions the need for preserved formulations, perhaps forgetting the significant health issues that were caused by non-preserved or poorly preserved medicines in earlier times. The two case studies demonstrate how the requirements of the preservative system to show maximal anti-microbial efficacy at minimal concentration are often conflicted with other needs of the formulation (palatability, chemical or physical incompatibilities, efficacy or bioavailability).

It is often simplistically assumed that, “it is nearly always technically possible to re-think the product development to remove them (… preservatives) or minimize their use”.17 Unfortunately the reality is that this isn’t the case and significant effort is required to develop safe, multi-use products, particularly those that are suitable for use in special populations, e.g. pediatrics.

Hence, although preservatives may face an uncertain future, it is vital that they remain available for incorporation into multi-use pharmaceutical products, to assure quality and patient safety. The evidence used to highlight concerns and calls for constraints can be contrasted with the well-established record of preservatives with respect to performance and safety. The limited number now available should not be constrained further.

In summary, given the limited choices that are currently available to serve as anti-microbial preservatives, as well as the years of clinical safety that have been demonstrated, a careful risk-to-benefit analysis should be considered before discounting preserved multi-dose formulations.79 Multi-use oral, topical and parenteral products need to be preserved if patient safety is to be assured.

Acknowledgements

Anna Slater, Trevor Shreeves and Gary Cannon (all either from or former GSK) for their input into case study 1 and Tracey Wood, Nicola Marzolini, Liz Jarvis and Padma Patel (all either from or former GSK) for their input into case study 2.

References

  1. FDA, 2017. 21 CFR Section 610.15(a). https://www.accessdata.fda.gov/scripts/cdrh/ cfdocs/cfcfr/CFRSearch.cfm?fr=610.15. Accessed on 22 September 2017.
  2. R. Ball, 2012. Substituting Thimerosal Preservative used in Vaccines: FDA perspective. http://who.int/immunization/sage/meetings/2012/april/USFDA_perspective_ thimerosal_alternatives.pdf. Accessed on 28 September, 2017.
  3. EMA, 1997. Note for guidance on inclusion of antioxidants and antimicrobial preservatives in medicinal products. CPMP/CVMP/ QWP/115/95, 08 July 1997.
  4. EMA, 2003. Note for guidance on excipients, antioxidants and antimicrobial preservatives in the dossier for application for marketing authorisation of a medicinal product. 20 February 2003, CPMP/QWP/419/03.
  5. EMA, 2006. Guidelines on excipients in the dossier for marketing authorisation of a medicinal product. EMEA/CHMP/QWP/396951/2006.
  6. EMEA, 2007. CHMP scientific article 5(3) opinion the potential risks of carcinogens, mutagens and substances toxic to reproduction when these substances are used as excipients of medicinal products for human use (EMEA/CHMP/SWP/146166/2007).
  7. EMA, 2014. Questions and Answers on Ethanol in the context of the revision of the guideline on ‘Excipients in the label and package leaflet of medicinal products for human use’ (CPMP/463/00). 23 January 2014. EMA/CHMP/507988/2013.
  8. EMA, 2014. Questions and Answers on Benzyl alcohol in the context of the revision of the guideline on ‘Excipients in the label and package leaflet of medicinal products for human use’ (CPMP/463/00). 23 January 2014.
  9. EMA/CHMP/508188/2013. SCF (Scientific Commission on Food), 2002. Opinion of the Scientific Committee on Food on Benzyl alcohol. http://ec.europa.eu/food/fs/sc/scf/ out138_en.pdf. Accessed on 26 September 2017.
  10. A.F. Fransway, 1991. The Problem of Preservation in the 1990s: III Agents with Preservation Function Independent of Formaldehyde Release. Am. J. Cont. Derm., 2, 145-174.
  11. Y. Boer, 1981. Irritation by Eye drops Containing 2-Phenylethyl Alcohol. Pharm. Weekbl (Sci)., 3, 826-827.
  12. R.M.E. Richards, R.J. McBride, 1971. Phenylethanol Enhancement of Preservatives Used in Ophthalmic Preparations. J. Pharm. Pharmac., 23, 141S-146S.
  13. C.E. Downward, L.I. Roberts, L.IJ.D. Morrow, 1995. Topical Benzoic Acid Induces the Increased Biosynthesis of PGD2 in Human Skin In Vivo. Clin. Pharmacol. Therap., 57, 441-445.
  14. P.J.Weller, 2006. Benzoic Acid Monograph in: R.C. Rowe, P.J. Sheskey, P. J. Weller (Eds.), Handbook of Pharmaceutical Excipients, Fifth Edition, Pharmaceutical Press, pp. 66-68.
  15. EMA, 2014. Questions and Answers on Benzoic acid and Benzoates in the context of the revision of the guideline on ‘Excipients in the label and package leaflet of medicinal products for human use’. (CPMP/463/00). 23 January 2014. EMA/CHMP/508189/2013.
  16. Joint FAO/WHO expert committee on food additives (JECFA), 2016. Evaluation of certain food additives and contaminants: Eightieth report on Joint FAO/WHO expert committee on food additives, WHO technical report series 995, 132 pp.
  17. P. Kozarewicz, 2010. Preservatives: Are they safe? 31 May 2010. http://www.ema.europa. eu/docs/en_GB/document_library/Presentation/2010/09/WC500096784.pdf. Accessed on 27 October 2017.
  18. IPCS (International Programme on Chemical Safety), 1998. Environmental Health Criteria 204: Boron. World Health Organisation, Geneva, Switzerland.
  19. SCCS (Scientific Committee on Consumer Safety), 2010. Opinion on boron compounds. European Commission, Directorate-General for Health and Consumers.
  20. EMA, 2014. Questions and answers on boric acid in the context of the revision of the guideline on ‘Excipients in the label and package leaflet of medicinal products for human use’ (CPMP/463/00 Rev. 1). 23 July 2015. EMA/CHMP/619104/2013.
  21. R. Johnson, R. Steer, 2006. Methylparaben Monograph in: R.C. Rowe,P.J. Sheskey, P. J. Weller (Eds.), Handbook of Pharmaceutical Excipients, Fifth Edition, Pharmaceutical Press, pp. 466-470.
  22. R. Johnson, R. Steer, 2006. Propylparaben Monograph in: R.C. Rowe, P.J. Sheskey, P. J. Weller (Eds.), Handbook of Pharmaceutical Excipients, Fifth Edition, Pharmaceutical Press, 2006, pp. 629-632.
  23. R. Johnson, R. Steer, 2006. Butylparaben Monograph in: R.C. Rowe, P.J. Sheskey, P. J. Weller (Eds.), Handbook of Pharmaceutical Excipients, Fifth Edition, Pharmaceutical Press, pp. 83-85.
  24. M. Weiner, M.L. Bernstein, 1989. Adverse Reactions to Drug Formulation Agents: A Handbook of Excipients, New York, Marcel-Deckker, pp. 298-300.
  25. EMA, 2015. Reflection paper on the use of methyl- and propylparaben as excipients in human medicinal products for oral use. EMA/CHMP/SWP/272921/2012. 22 October 2015.
  26. J. Whitworth, 2013. EFSA lowers phenol TDI. Food navigator.com. 13 May 2013. http:// www.foodnavigator.com/Market-Trends/EFSA-lowers-phenol-TDI. Accessed on 28 September 2017.
  27. FDA, 2017. 21 CFR Section 610.15(a). https://www.accessdata.fda.gov/scripts/cdrh/ cfdocs/cfcfr/CFRSearch.cfm?fr=610.15. Accessed on 22 September 2017.
  28. D.J. Brancato, 1982. Recognising Potential Toxicity of Phenol, Vet. Hum. Toxicol., 24, 29-30.
  29. S. Nema, 2006. Chlorocresol Monograph in: R.C. Rowe, P.J. Sheskey, P. J. Weller (Eds.), Handbook of Pharmaceutical Excipients, Fifth Edition, Pharmaceutical Press, pp. 171-173.
  30. IJT, 1997. Final Report on the Safety Assessment of p-Chloro-m-Cresol. Int. J. Toxicology. 16(3), 235-268.
  31. B.W. Hancock, A. Naysmith, 1975. Hypersensitivity to Chlorocresol Preserved Heparin, Br. Med. J., 3, 746-747.
  32. L.M.E. McIndoe, 2006. Chloroxylenol Monograph in: R.C. Rowe, P.J. Sheskey, P. J. Weller (Eds.), Handbook of Pharmaceutical Excipients, Fifth Edition, Pharmaceutical Press, pp. 180-181.
  33. CIR Expert Report, 1985. Final Report on the Safety Assessment of Chloroxylenol. J. Amer. Coll. Toxicol., 4(5), 147-169.
  34. EMA, 2009. EMEA public statement on antimicrobial preservatives in ophthalmic preparations for human use. EMEA/622721/2009.
  35. EMA, 2014. Questions and Answers on Benzalkonium chloride in the context of the revision of the guideline on ‘Excipients in the label and package leaflet of medicinal products for human use’ (CPMP/463/00). 22 May 2014. EMA/495737/2013.
  36. A. Okahara, K. Kawazu, 2013. Local toxicity of benzalkonium chloride in ophthalmic solutions following repeated applications. J. Toxicol. Sci., 38(4), 531-537.
  37. Y. Kuboyama, K. Suzuki, T. Hara, 1997. Nasal lesions induced by intranasal administration of benzalkonium chloride in rats. J. Toxicol. Sci., 22, 153-160.
  38. C. Baudouin, A. Labbé, H. Liang, H. Pauly, F. Brignole-Baudouin, 2010. Preservatives in eye drops: the good, the bad and the ugly. Prog. Retin. Eye Res., 29(4), 2010. DOI: 10.1016/j. preteyeres.2010.03.001.
  39. The Expert Panel of the American College of Toxicology, 1985. Final Report on the Safety Assessment of Benzethonium Chloride and Methylbenzethonium Chloride, J. Am. Coll. Toxicol., 4, 65-106.
  40. L. Lohr, 1978. Mercury Controversy Heats Up, Am. Pharm., 18, 23.
  41. A.F. Winder, N.J. Astbury, G.A.K. Sheraidah, M. Ruben, 1980. Penetration of Mercury from Ophthalmic Preservatives into the Human Eye, Lancet, ii: 237-239.
  42. G. Cireface, 2012. Alternatives to thiomersal as preservatives for vaccines. WHO Informal Consultation to develop further guidance on vaccines for the UNEP-convened Intergovernmental. Negotiating Committee Meeting 4. http://who.int/immunization/ sage/meetings/2012/april/Alternatives_thiomersal_preservatives_vaccines.pdf. Accessed on 28 September 2017.
  43. J. R. Miller, 1984. Sensitivity to Contact Lens Solutions, West J. Med., 140, 791.
  44. J.L. Ford, M.W. Brown, P.B. Hunt, 1985. A Note on the Contamination of Eye-Drops Following Use by Hospital Out-Patients, J. Clin. Hosp. Pharm., 10, 203-209.
  45. N.H. Cox, A. Forsyth, 1988. Thiomersal Allergy and Vaccination Reactions, Contact Derm., 18, 229-233.
  46. American Academy of Pediatrics, United States Public Health Service, 1999. Thimerosal in vaccines: a joint statement of the American Academy of Pediatrics and the Public Health Service, MMWR, 49, 563-565.
  47. EMEA, 1999. EMEA public statement on thiomersal containing medicinal products, 8th July 1999, EMEA publication 20962/99.
  48. Anon, 2001. Department of Health. Public Letter from the Chief Medical Officer: Current Vaccine and Immunisation Issues, 15th October 2001, Pl/CMO/2001/5.
  49. C. Tervit, L. Paquette, C.D. Toneck, B. Basrani, S. Friedman, 2009. Proportion of Healed Teeth With Apical Periodontitis Medicated With Two Percent Chlorhexidine Gluconate Liquid: A Case-Series Study. J. Endodontics, 35(9), 1182-1185.
  50. E.J. Septimus, J. Moody, 2016. Prevention of Device-Related Healthcare-Associated Infections. F1000 Res., 5, https://f1000research.com/articles/5-65/v1. Accessed on 29 October 2017.
  51. S.C. Owen, 2006. Chlorhexidine Monograph in: R.C. Rowe, P.J. Sheskey, P. J. Weller (Eds.), Handbook of Pharmaceutical Excipients, Fifth Edition, Pharmaceutical Press, pp. 163-167.
  52. R.T. Guest, 2006. Imidurea Monograph in: R.C. Rowe, P.J. Sheskey, P. J. Weller (Eds.), Handbook of Pharmaceutical Excipients, Fifth Edition, Pharmaceutical Press, pp. 359-361.
  53. S.P. Denyer, N.A. Hodges, 2006. Bronopol Monograph in: R.C. Rowe, P.J. Sheskey, P. J. Weller (Eds.), Handbook of Pharmaceutical Excipients, Fifth Edition, Pharmaceutical Press, pp. 76-78.
  54. W.E. Rosen, P.A. Berke, 1977. Germall 115: A Safe and Effective Modern Preservative, Cosmet. Toilet. 92, 88-89.
  55. H.I. Maibach, 1977. Dermal Sensitization Potential of 2-Bromo-2-nitropropane-1,3-diol (bronopol), Contact Derm., 3, 99 (1977).
  56. G. Deza, A.M. Giménez Arnau, 2017. Allergic contact dermatitis in preservatives: Current standing and future options. Curr. Opin. Allergy Clin. Immunol., 17(4), 263-268.
  57. C. Lv, J. Hou, W. Xie, H. Cheng, 2015. Investigation on formaldehyde release from preservatives in cosmetics. Int. J. Cosm. Sci., 37(5), 474-478.
  58. M. Cristaldi, M. Oliveri, G. Lupo, C.D. Anfuso, S. Pezzino, D. Rusciano, 2017. N-hydroxymethylglycinate with EDTA is an efficient eye drop preservative with very low toxicity: an in vitro comparative study. Cut. Ocular Toxicol., 1-6. 10.1080/15569527.2017.1347942.
  59. U.K. Kotreka, V.L. Davis, M.C. Adeyeye, 2017. Development of topical ophthalmic In Situ gelforming estradiol delivery system intended for the prevention of age-related cataracts. PloS ONE, 12(2), e0172306. 10.1371/journal.pone.0172306
  60. M. Menaka, V.P. Pandey, 2014. Formulation development and evaluation of Metoclopramide hydrochloride nasal spray. Int. J. Pharmac. Pharm. Sci., 6(1), 212-216.
  61. L. Lin, J. Kim, H. Chen, R. Kowalski, V. Nizet, 2016. Component analysis of multipurpose contact lens solutions to enhance activity against Pseudomonas aeruginosa and Staphylococcus aureus. Antimicrobial Agents Chemotherapy, 60(7), 4259-4263.
  62. S.C. Owen, 2006. Edetic Acid Monograph, in: R.C. Rowe, P.J. Sheskey, P. J. Weller (Eds.), Handbook of Pharmaceutical Excipients, Fifth Edition, Pharmaceutical Press, , pp. 260-263.
  63. D.A. Geier, S.K. Jordan, M.R. Geier, 2010. The relative toxicity of compounds used as preservatives in vaccines and biologics. Med. Sci. Monit., 16(5), SR21-27.
  64. P.D. Darbre, A. Alijarrah, W.R. Miller, N.G. Coldham, M.J. Sauer, G.S. Pope, 2004. Concentrations of Parabens in Human Breast Tumours, J. Appl. Toxicol., 24, 5-13).
  65. Vaccines and Autism: An Institute of Medicine Report, 18th May 2004, www.iom.edu/ Reports/2004/Immunization, accessed on 15th October 2017.
  66. A.E. Tozzi, P. Bisiacchi, V. Tarantino, B. De Mei, L. DeElial, F. Chiarotti, S. Salmaso, 2009. Neuropsychological Performance 10 Years after Immunization in Infancy with ThimerosalContaining Vaccines, Pediatrics, 123, 475-482.
  67. Anon, 2017. Antiperspirants/Deodorants and Breast Cancer (Fact Sheet). OncologyNurseAdvisor. http://www.oncologynurseadvisor.com/breast-cancer/breastcancer-risk-and-deodorant-use/article/644826/. Accessed on 30 October 2017.
  68. Manny, 2017. The Link Between Deodorant and Breast Cancer. http://www.askdrmanny. com/link-deodorant-breast-cancer/. Accessed on 30 October 2017.
  69. C. Lobermeier, C. Tschoetschel, S. Westie, E. Heymann, 1996. Hydrolysis of Parabens by Extracts from Differing Layers of Human Skin, Biol. Chem., 377, 647-651.
  70. United States Pharmacopeia General ChapterAntimicrobial Effectiveness Testing, USP 34- NF29, US Pharmacopeia, Rockville, Maryland, USA, 2010.
  71. European Pharmacopoeia 5.1.3 Efficacy of Antimicrobial Preservation, EP 6.4, European Directorate for Quality of Medicines, Strasbourg, France, 2010.
  72. S.C.Owen, 2006. 2-Phenoxyethanol Monograph in: R.C. Rowe, P.J. Sheskey, P. J. Weller (Eds.), Handbook of Pharmaceutical Excipients, Fifth Edition, Pharmaceutical Press, pp. 517-518.
  73. H.C. Sigle, M. Schäfer-Kortning, H.C. Korting, B. Hube, M. Niewerth, 2006. In Vitro Investigations on the Mode of Action of the Hydroxypyridone Antimycotics Rilopirox and Piroctone on Candida albicans, Mycoses, 49, 159-158.
  74. E.J. Fendler, R.A. Williams, M.J. Dolan, 1998. Antimicrobial Cleansing Composition Containing Chlorhexidine, an Amphoteric Surfactant and an Alkyl Polyglucoside. United States Patent 5, 719,113, 17th February 1998.
  75. D.J. Picciano, 2001. Iodine Containing Nasal Moisturizing Saline and Mouthwash Solutions, United States Patent 6,171,611, 09th January 2001.
  76. R. Noecker, 2001. Effect of Common Occular Preservatives on Occular Health, Advances in Therapy 18, 205-215.
  77. T.B. Ernest, D.P. Elder, L.G. Martini, M. Roberts, J.L. Ford, 2006. Developing Paediatric Medicines: Identifying the Needs and Recognizing the Challenges, J. Pharm. Pharmac., 59: 1043-1055.
  78. A. Palmieri, 2006. Polacrillin Potassium Monograph in: R.C. Rowe, P.J. Sheskey, P. J. Weller (Eds.), Handbook of Pharmaceutical Excipients, Fifth Edition, Pharmaceutical Press, pp 532-534.
  79. B. Marple, P. Roland, M. Benninger, Safety Review of Benzalkonium Chloride used as a Preservative in Intranasal Solutions: An Overview of Conflicting Data and Opinions, 2004. Otol. Head and Neck Surgery, 130, 131-141.
  • <<
  • >>

Join the Discussion