7

Dermatological Formulation and Transdermal Systems

I.

INTRODUCTION

Over the past few decades there have been many advances in our understanding of the physicochemical properties of both formulation systems and their ingredients. These have led to the ability to develop physically, chemically, and biologically stable products. There has also been a signicant increase in our knowledge of the properties of skin and the processes that control skin permeation. The ground rules for skin permeation were laid down by Scheuplein and Blank in the late 1960s and early 1970s (1), and these have been updated on a reasonably regular basis (28). We have learned, for example, that the permeation of compounds across intact skin is controlled fundamentally by the stratum corneum, and it is the chemical composition and morphology of this layer that usually determines the rate and extent of absorption (9,10). Similarly, we have discovered how to modify this barrier, by chemical or physical means and, thereby, alter the rate of diffusion of many permeating molecules (11,12). A basic deciency, however, in the application of our understanding of the barrier properties of the skin to dermatological and transdermal therapy is that this knowledge has largely been generated by investigations on normal, rather than pathological, skin. The relevance of such information to diseased skin, for which permeation characteristics are probably signicantly altered, has yet to be fully established. There is some data on transport across skin that has been articially damaged (1318), and limited information on permeation through diseased skin has been obtained in the clinic (1922). In modern-day pharmaceutical practice, therapeutic compounds are applied to the skin for dermatological (within the skin), local (regional), and for transdermal

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

(systemic) delivery. Whatever the target site or organ, it is usually a prerequisite that the drug crosses the outermost layer of the skin, the stratum corneum. However, a major function of the stratum corneum is to provide a protective barrier to the ingress of xenobiotics and to control the rate of water loss from the body: Evolution has generated a robust and durable barrier that fullls its biological function throughout an individuals lifetime. A basic, yet thorough, understanding of the structure and transport properties of this membrane is essential to the rational development of topical dosage forms, and this has been provided elsewhere in this volume. II. DRUG CANDIDATE SELECTION

Although it may appear to be a simple task to select lead compounds for pharmaceutical product development, based on therapeutic rationale and compound safety and efcacy, the practicalities of this procedure are somewhat more complex. For the most part, therapeutic efcacy is dependent on the ability of a compound to cross biological barriers, travel to the target site, and interact with specic receptors. However, as pointed out and excellently reviewed (3), it is often more appropriate in dermatological therapy to select compounds based on their inability to breach relevant biological barriers. Because the site of action may be the skin surface, the stratum corneum, the viable epidermis, the appendages, the dermis, or the local subcutaneous tissues, the rules of candidate selection will vary. For the purposes of this discussion it will be assumed that the therapeutic rationale for dermal drug delivery has been established and that a series of compounds with appropriate pharmacological activity have been identied. It will also be assumed that each compound within the series possesses equivalent chemical and physical stability. In other words, drug candidate selection need only be based on the ability to deliver the compound to its site of action. Earlier chapters in this volume have taught that the primary requirement for a compound to penetrate the skin is the ability to leave the delivery system and enter the stratum corneum. Furthermore, this characteristic is dependent on the stratum corneumvehicle partition coefcient of the compound, for which the octanolwater partition coefcient is often used as a surrogate. Whereas it is immediately apparent that a high value for this parameter will favor delivery into the stratum corneum, it will not favor movement into the more hydrophilic regions of the viable epidermis. Furthermore, the rate of diffusion through the stratum corneum and lower layers of the skin is linked to the molecular volume of the permeant. It is evident that a compound with a high octanolwater partition coefcient and a relatively high molecular volume will possess a high afnity for the stratum corneum (i.e., be substantive to the stratum corneum). This principle is used extensively in the design of sunscreen agents, for which it is not uncommon to add a medium-length or branchedchain alkyl to the UV-absorbing molecule to increase residence time in the skin and reduce systemic uptake. The use of the skin as a route of delivery into the systemic circulation was neither commercially nor scientically exploited until the 1950s, when ointments containing agents such as nitroglycerin and salicylates were developed. Angina could be controlled for several hours by applying an ointment containing 2% nitroglycerin (23). Similarly, topical salicylates could be absorbed through the skin into arthritic joints. More recently, nonsteroidal anti-inammatory agents, such as ibuprofen and

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

ketoprofen, and hormonal steroids, such as estradiol and testosterone, have been developed and marketed in semisolid preparations. A major problem with transdermal semisolid preparations, however, is that of control. Drug concentrations in plasma or duration of action, are not reliably predictable for several reasons, including the amount and area of application and dosage frequency. The specic advantages of transdermal therapy have been fully discussed elsewhere (24). Briey, transdermal devices are easy to apply, can remain in place for up to 7 days (depending on the system), and are easily removed following, or during, therapy. The reduced-dosing frequency, and the production of controllable and sustained plasma levels, tend to minimize the risk of undesirable side effects sometimes observed after oral delivery. The avoidance of extensive hepatic rst-pass metabolism is a further advantage. The major limitation to transdermal drug delivery is the intrinsic barrier property of the skin. Although marketed patch-type transdermal delivery systems are available for only a limited number of drugs (e.g., scopolamine, nitroglycerin, clonidine, estradiol, fentanyl, testosterone, and nicotine), several other candidates are at various stages of development. Many of the drugs under investigation do not intrinsically possess any signicant ability to cross the skin; therefore, ways must be found to improve their transdermal delivery. This could be achieved by the use of prodrugs designed such that they are more rapidly absorbed than the parent compound, yet are metabolized to the active species before receptor site occupancy (25,26). Physical methods, such as iontophoresis (27), electroporation (28), and sonophoresis (29) have proved experimentally useful for increasing the skin permeation of several compounds. Alternatively, the barrier may be modulated using thermodynamic strategies, or chemically modied to reduce diffusive resistance by the use of penetration enhancers, both of which are discussed in Chapter 6. Such developmental strategies will increase the number of candidate drugs for transdermal delivery in the future. III. PREFORMULATION AND FORMULATION

Preformulation encompasses those studies that should be carried out before the commencement of formulation development. The primary goal of the preformulation process is to permit the rational development of stable, safe, efcacious dosage forms, and it is concerned mainly with the characterization of the physicochemical properties of the drug substance. At the preformulation stage, the nal route of drug administration is usually undecided; therefore, any protocols must be able to cover all required aspects. The preformulation study has several distinct phases (Table 1). A detailed description of all of the studies that form part of the preformulation stage are given elsewhere (30) and will not be considered here. The only important aspect of preformulation that is specic to dermatological and transdermal formulation concerns drug delivery characteristics. These have been fully discussed elsewhere in this volume (see Chapters 4 and 5). A. Formulation of Dermatological Products

The selection of formulation type for dermatological products is usually inuenced by the nature of the skin lesion and the opinion of the medical practitioner. Kitson and Maddin (31) elegantly stated: It is idle to pretend that the therapy for skin

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

Table 1

Preformulation Tasks Listed in Approximate Chronologic Order

Preformulation Task 1. 2. 3. 4. 5. 6. 7. 8. General description of the compound Calorimetry Polymorphism Hygroscopicity Analytical development Intrinsic stability Solubility and partitioning characteristics Drug delivery characteristics

diseases, as currently practiced, has its origins in science. To this day a practicing dermatologist would prefer to apply a wet formulation (ranging from simple tapwater to complex emulsion formulations, with or without drug) to a wet lesion and a dry formulation (e.g., petrolatum) to a dry lesion. The preparation of such formulations as poultices and pastes is extemporaneous, and it is unlikely that the industrial pharmaceutical formulator will be required to develop products of this type. Solutions and powders lack staying power (retention time) on the skin and can afford only transient relief. In modern-day pharmaceutical practice, semisolid formulations are the preferred vehicles for dermatological therapy because they remain in situ and deliver the drug over extended time periods. In most cases, therefore, the developed formulation will be an ointment, emulsion, or gel. Typical constituents for these types of formulations are shown in Table 2. 1. Ointments

An ointment is classied as any semisolid containing fatty material and intended for external application (United States Pharmacopeia; USP). There are four types of ointment base, and these are listed in the USP as hydrocarbon base, absorption base, water-removable base, and water-soluble base. Only the hydrocarbon bases are completely anhydrous. The anhydrous hydrocarbon bases, which contain straight or branched hydrocarbons with chain lengths ranging from C16 to C30, which may also contain cyclic alkanes, are used principally in nonmedicated form. A typical formulation contains uid hydrocarbons (mineral oils and liquid parafns) mixed with a longer alkyl chain, higher-melting point, hydrocarbons (white and yellow soft parafn and petroleum jelly). The difference between white and yellow soft parafn is simply that the white version has been bleached. Hard parafn and microcrystalline waxes are similar to the soft parafns, except that they contain no liquid components. These anhydrous mixtures tend to produce formulations that are greasy and unpleasant to use, but the addition of solid components, such as microcrystalline cellulose, can reduce the greasiness. Improved skin feel can also be attained by the incorporation of silicone materials, such as polydimethylsiloxane oil or dimethicones. Silicones are often used in barrier formulations that are designed to protect the skin against water-soluble irritants. Although the nonmedicated anhydrous ointments are extremely useful as emollients, their value as topical drug delivery systems is limited by the relative insolu-

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Table
Function Polymeric thickeners Sample ingredients Gums Acacia Alginates Carageenan Chitosan Collagen Tragacanth Xanthan Celluloses Sodium carboxymethyl Hydroxyethyl Hydroxypropyl Hydroxypropylmethyl Mineral oil White soft parafn Yellow soft parafn Beeswax Stearyl alcohol Cetyl alcohol Cetostearyl alcohol Stearic acid Oleic acid Nonionic Sorbitan esters Polysorbates Polyoxyethylene alkyl ethers Polyoxyethylene alkyl esters Polyoxyethylene aryl ethers Glycerol esters Cholesterol Polar Water Propylene glycol Glycerol Sorbitol Ethanol Industrial methylated spirit Antimicrobial Benzalkonium chloride Benzoic acid Benzyl alcohol Bronopol Chlorhexidine Chlorocresol Imidazolidinyl urea Paraben esters Phenol Phenoxyethanol Potassium sorbate Sorbic acid Diethanolamine Lactic acid Monoethanolamine Acrylic acids Carbomers Polycarbophil Colloidal solids Silica Clays Microcrystalline cellulose Hydrogels Polyvinyl alcohol Polyvinylpyrrolidone Thermoreversible polymers Poloxamers Isopropyl myristate Isopropyl palmitate Castor oil Canola oil Cottonseed oil Jojoba oil Arachis (Peanut) oil Lanolin (and derivatives) Silicone oils Anionic Sodium dodecyl sulfate Cationic Cetrimide Benzalkonium chloride

Oil phase

Surfactants

Solvents

Polyethylene glycols Propylene carbonate Triacetin Nonpolar Isopropyl alcohol Medium-chain triglycerides Antioxidants -Tocopherol Ascorbic acid Ascorbyl palmitate Butylated hydroxyanisole Butylated hydroxytoluene Sodium ascorbate Sodium metabisulte Chelating agents Citric acid Edetic acid

Preservatives

pH adjusters

Sodium hydroxide Sodium phosphate Triethanolamine

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

bility of many drugs in hydrocarbons and silicone oils. It is possible to increase drug solubility within a formulation by incorporation of hydrocarbon-miscible solvents, such as isopropylmyristate or propylene glycol, into the ointment. Although increasing the solubility of a drug within a formulation may often decrease the release rate, it does not necessarily decrease the therapeutic effect. It is well accepted that simple determination of release rates from formulations may not be predictive of drug bioavailability. For example, when formulated in a simple white petrolatummineral oil ointment, the release rate of betamethasone dipropionate was considerably higher than when the drug was formulated at the same concentration (0.05%) in an augmented, and more clinically effective, ointment that contained propylene glycol (Fig. 1) (32). It is also important to appreciate that various grades of petrolatum are commercially available, and that the physical properties of these materials will vary depending on the source and rening process. Even slight variations in physical properties of the constituents of an ointment may have substantial effects on drug release behavior (33). The preparation of ointment formulations may appear to be a simple matter of heating all of the constituents to a temperature higher than the melting point of all

Figure 1

Release rates of betamethasone dipropionate from ointments into various receptor uids. (1) 5% hexane in acetonitrile; (2) octanol; (3) acetonitrile; (4) 60% acetonitrile in water; (5) 95% ethanol. (From Ref. 32.)

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

of the excipients and then cooling with constant mixing. The reality, however, is that the process is somewhat more complex and requires careful control over various parameters, particularly the cooling rate. Rapid cooling, for example, creates stiffer formulations in which there are numerous small crystallites, whereas a slow-cooling rate results in the formation of fewer, but larger, crystallites and a more uid product. Further information on temperature effects and ointment phase behavior are available (3436). 2. Gels

The common characteristic of all gels is that they contain continuous structures that provide solid-like properties (4). Depending on their constituents, gels may be clear or opaque, and be polar, hydroalcoholic, or nonpolar. The simplest gels comprise water thickened with either natural gums (e.g., tragacanth, guar, or xanthan), semisynthetic materials (e.g., methylcellulose, carboxymethylcellulose, or hydroxyethylcellulose), synthetic materials (e.g., carbomercarboxyvinyl polymer), or clays (e.g., silicates or hectorite). Gel viscosity is generally a function of the amount and molecular weight of the added thickener. There are a variety of semisynthetic celluloses in use as thickeners in gel formulations. These include methylcellulose (MC), carboxymethylcellulose (CMC), hydroxyethylcellulose (HEC), hydroxypropylcellulose (HPC), and hydroxypropylmethylcellulose (HPMC). These celluloses can be obtained in diverse molecular weight grades, and the higher molecular weight compounds are used at 15% (w/ w). In the development of prototype gel formulations, it is useful to evaluate a variety of different types of cellulose. For example, if clarity of the gel is a major requirement, HPMC is preferable to MC. It is also important to appreciate that some celluloses may exhibit specic incompatibilities with other potential formulation ingredients. For example, HEC is incompatible with several salts, and MC and HPC are incompatible with parabens. This latter incompatibility limits the choice of preservative for gel formulations that are based on MC and HPC. Finally, the presence of oxidative materials (e.g., peroxides, or other ingredients containing peroxide residues) in formulations gelled with celluloses should be avoided because oxidative degradation of the polymer chains may cause a rapid decrease in formulation viscosity (37). As the branched-chain polysaccharide gums, such as tragacanth, pectin, carrageenan, and guar, are of naturally occurring plant origin, they can have widely varying physical properties, depending on their source. They are usually incorporated into formulations at concentrations between 0.5 and 10%, contingent on the required viscosity. Viscosity may be enhanced synergistically by the addition of inorganic suspending agents, such as magnesium aluminum silicate. Tragacanth, a mixture of water-insoluble and water-soluble polysaccharides, is negatively charged in aqueous solution and, therefore, incompatible with many preservatives when formulated at a pH of 7 or higher. Similarly, xanthan gum, which is produced by bacterial fermentation, is incompatible with some preservatives. Alginic acid is a hydrophilic colloidal carbohydrate obtained from seaweed and the sodium salt, sodium alginate, is used at 510% as a gelling agent. Film gels may be obtained by incorporation of small amounts of soluble calcium salts (e.g., tartrate on citrate). Many gums are ineffective in hydroalcoholic gels containing more than 5% alcohol.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

The natural clay thickeners (e.g., bentonite and magnesium aluminium silicate) are useful for thickening aqueous gels containing cosolvents, such as ethanol, isopropanol, glycerin, and propylene glycol. These materials possess a lamellar structure that can be extensively hydrated. The at surfaces of bentonite are negatively charged, whereas the edges are positively charged. These clays swell in the presence of water because of hydration of the cations and electrostatic repulsion between the negatively charged faces. Thixotropic gels form at high concentrations at which the clay particles combine in a occulated structure in which the edge of one particle is attracted to the face of another. The rheological properties of these clay dispersions, therefore, are particularly sensitive to the presence of salts. Bentonite, a native colloidal hydrated aluminium silicate (mainly montmorillonite), can precipitate under acidic conditions, and formulations must be at pH 6 or higher. A synthetic clay (colloidal silicon dioxide) is also useful for thickening both aqueous and nonpolar gels. The concentration of clay usually required to thicken formulations is 210%. By far the most extensively employed gelling agents in the pharmaceutical and cosmetic industries are the carboxyvinyl polymers known as carbomers. These are synthetic high molecular weight polymers of acrylic acid, cross-linked with either allylsucrose or allyl ethers of pentaerythritol. Pharmaceutical grades of these carbomers are available (e.g., Carbopol 981NF; B. F. Goodrich Performance Materials). In the dry state, a carbomer molecule is tightly coiled, but when dispersed in water the molecule begins to hydrate and partially uncoil, exposing free acidic moieties. To attain maximum thickening effect the carbomer molecule must be fully uncoiled, and this can be achieved by one of two mechanisms (Fig. 2). The most common method is to convert the acidic molecule to a salt, by the addition of an appropriate neutralizing agent. For formulations containing aqueous or polar solvent, carbomer gellation can be induced by the addition of simple inorganic bases, such as sodium or potassium hydroxide. Less polar or nonpolar solvent systems may be neutralized with amines, such as triethanolamine or diethanolamine, or a number of alternative amine bases (e.g., diisopropanolamine, aminomethyl propanol, tetrahydroxypropyl ethylenediamine, and tromethamine) may be employed. Neutralization ionizes the carbomer molecule, generating negative charges along the polymer backbone, and the resultant electrostatic repulsion creates an extended three-dimensional structure. Care must be taken not to under- or overneutralize the formulation, as this will result in viscosity or thixotropic changes (38). Overneutralization will reduce viscosity, as the excess base cations screen the carboxy groups and reduce electrostatic repulsion. Hydrated molecules of carbomer may also be uncoiled in aqueous systems by the addition of 1020% of hydroxyl donors, such as a nonionic surfactant or a polyol, that are able to hydrogen-bond with the polymer. Maximum thickening will not be as instantaneous using this mechanism, as it is with base neutralization, and may take several hours. Heating accelerates the process, but the system should not be heated to more than 70C. Because they are synthetic, carbomer bases vary little from lot-to-lot, although batch-to-batch differences in mean molecular weight may result in variations in the rheological characteristics of aqueous dispersions (39). 3. Emulsions

The most common emulsions used in dermatological therapy are creams. These are two-phase preparations in which one phase (the dispersed or internal phase) is nely dispersed in the other (the continuous or external phase). The dispersed phase can

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

Figure 2

A tightly coiled carbomer molecule will (a) hydrate and swell when dispersed in water. (b) The molecule will completely uncoil to achieve maximum thickening when it is converted from the acid form to the salt form on neutralization.

have either a hydrophobic-based (oil-in-water creams; O/W), or be aqueous based (water-in-oil creams; W/O). Whether a cream is O/W or W/O depends on the properties of the system used to stabilize the interface between the phases. Because there are two incompatible phases in close conjunction, the physical stability of creams is always tenuous, although it may be maximized by the judicious selection of an appropriate emulsion-stabilizing system. In most pharmaceutical emulsions, the stabilizing systems comprise either surfactants (ionic or nonionic), polymers (nonionic polymers, polyelectrolytes, or biopolymers), or mixtures of these. The most commonly used surfactant systems are sodium alkyl sulfates (anionic), alkylammonium halides (cationic), and polyoxyethylene alkyl ethers or polysorbates (nonionic). These are often used alone, or in conjunction with nonionic polymerics, such as polyvinyl alcohol or poloxamer block copolymers, or polyelectrolytes, such as polyacrylic polymethacrylic acids. The physicochemical principles underlying emulsion formulation and stabilization are extremely complex and will not be covered in depth here. The interested

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

reader is referred to the volume edited by Sjo blom (40). Briey, an emulsion is formed when two immiscible liquids (usually, oil and water) are mechanically agitated. When this occurs in the absence of any form of interfacial stabilization during agitation, both liquids will form droplets that rapidly occulate and coalesce into two phases on standing. Flocculation is the term used to describe the close accumulation of two or more droplets of dispersed phase without loss of the interfacial lm, and it is largely the result of van der Waals attraction. The occulated droplets may then coalesce into one large droplet, with the loss of the interfacial lm. In practice, there is a brief period when one of the phases becomes the continuous phase because the droplets of this liquid coalesce more rapidly than the droplets of the other. Physical stability of an emulsion is determined by the ability of an additive to counteract the van der Waals attractions, thereby reducing occulation and coalescence of the dispersed phase. This may be achieved in two ways: an increase in the viscosity of the continuous phase, which will reduce the rate of droplet movement, or the establishment of an energy barrier between the droplets, or both. Although increasing the viscosity of the continuous phase will reduce the rate at which droplets occulate, in pharmaceutical shelf-life terms, a stable system can be generated in this way only if the continuous phase is gelled and the droplet diameter is smaller than 0.1 m. In pharmaceutical emulsions it is more common to develop stability using the energy barrier technique and to complement this stabilization, if necessary, by increasing the viscosity of the continuous phase. The basis of the energy barrier is that droplets experience repulsion when they approach each other. Repulsion can be generated either electrostatically, by the establishment of an electric double layer on the droplet surface, or sterically, by adsorbed nonionic surfactant or polymeric material. Electrostatic repulsion is provided by ionic surfactants that, when adsorbed at the oilwater interface, orient such that the polar ionic group enters the water. Some of the surfactant counterion (e.g., the sodium ion of sodium dodecyl sulfate) will separate from the surface and form a diffuse cloud surrounding the droplet. This diffuse cloud, together with the surface charge from the surfactant, forms the electric double layer, and electrostatic repulsion occurs when two similarly charged droplets approach each other. For obvious reasons, this method of emulsion stabilization is appropriate only for O/W formulations. In addition, it is important to appreciate that emulsions stabilized by electrostatic repulsion are extremely sensitive to the presence of additional electrolytes, which will disrupt the electrical double layer. Steric repulsion may be produced using nonionic surfactants or polymers, such as polyvinyl alcohol or poloxamers. The specic distribution of the polyethoxylated nonionic surfactants and block copolymers (Fig. 3a and b) results in the formation of a thick hydrophilic shell of polyoxyethylene chains around the droplet. Repulsion is then afforded by both mixing interaction (osmotic repulsion) and entropic interaction (volume restriction), the latter as a result of a loss of congurational entropy of the polyoxyethylene chains when there is signicant overlap. For polymeric materials without denitive hydrophobic and hydrophilic regions, the adsorption energy is critical to generation of steric repulsion. The adsorption energy must not be so low that there is no polymer adsorption, nor so high that there is complete polymer adsorption to the droplet. In either of these cases, there will be none of the loops, or tails, (see Fig. 3c) that are essential to steric repulsion. Polymeric steric repulsion is usually achieved using block copolymers, such as poloxamers, which consist of

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

Figure 3 Oil-in-water emulsions may be stabilized by (a) nonionic surfactants, (b) poloxamer block copolymers, or (c) polymeric materials. The hydrophilic chains produce repulsion by mixing interaction (osmotic) or volume restriction (entropic).

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

linked polyoxyethylene and polyoxypropylene chains. More recently, polyacrylic acid polymers linked to hydrophobic chains (Pemulen; B. F. Goodrich) have been used as primary emulsication systems in O/W formulations and are listed in the USP as carbomer 1342. These materials form very stable emulsions because the polyacrylic acid chain, anchored to the oil droplet by the alkyl methacrylate moieties, considerably increases the surface charge on the oil droplet, forming a strong electrical barrier at the interface. Furthermore, emulsion stability is enhanced by an increase in the viscosity of the continuous phase. Usually, for O/W emulsions, and always, for W/O emulsions, it is necessary to select an emulsication system based on surfactants. Although ionic surfactants can be used only for O/W emulsions, nonionic surfactants may be used for both O/W and W/O formulations. Although, at rst glance, the choice of surfactant system appears limitless (there are hundreds to select from), there are some basic guidelines to aid the formulator. As the rst priority, use of pharmaceutically approved surfactants (or those having a Drug Master File in place with the FDA) will save a considerable amount of regulatory justication. Raw material suppliers will frequently provide guidelines, although these will obviously be biased toward the use of their products. However, it should be appreciated that suppliers have considerable experience in the applications and uses of their products, and they are a very useful resource. An approximate guide to emulsion formulation is provided by the hydrophilic lipophilic balance (HLB) system that generates an arbitrary number (usually between 0 and 20) that is assigned to a particular surfactant. The HLB value of a polyoxyethylene-based nonionic surfactant may be derived from: HLB = mol% hydrophobic group 5

and the HLB of a polyhydric alcohol fatty acid ester (e.g., glyceryl monostearate) may be derived from HLB = 20

1S A

where S is the saponication number of the ester and A the acid number of the fatty acid. When it is not possible to obtain a saponication number (e.g., lanolin derivatives), the HLB can be calculated from HLB = (E P) 5

where E is the weight percent (wt%) of the polyoxyethylene chain and P the wt% of the polyhydric alcohol group in the molecule. From the foregoing equations it is apparent that hydrophilic surfactants have high HLB values, and lipophilic surfactants have low HLB values. It is generally recognized that surfactants with HLB values between 4 and 6 are W/O emulsiers and those with HLB values between 8 and 18 are O/W emulsiers. It is also generally recognized, although poorly understood, that mixtures of surfactants create more stable emulsions than the individual surfactants. The overall HLB of a surfactant mixture (HLBM) can be calculated from

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

HLBM = f HLBA (1 f )HLBB where f is the weight fraction of surfactant A. The required emulsier HLB values for several oils and waxes are given in Table 3. Importantly, the HLB system can be used only as an approximation in emulsion design, and stability of an emulsion cannot be guaranteed by the use of an emulsier mix with an appropriate HLB value. For example, creaming of an emulsion, a typical physical stability problem, is much more dependent on the viscosity of the continuous phase than the characteristics of the interfacial lm. Mixtures of surfactants create more stable emulsions than the individual surfactants. A reasonable and coherent explanation for this is given by the gel network theory (41,42). Briey, this theory relates the consistencies and stabilities of O/W creams to the presence or absence of viscoelastic gel networks in the continuous phase. These networks form when there is an amount of mixed emulsier, in excess of that required to stabilize the interfacial lm, that can interact with the aqueous continuous phase. In its simplest form, a cream consists of oil, water, and mixed emulsier. Ofcial emulsifying waxes may be cationic (cationic emulsifying wax BPC; a mixture of cetostearyl alcohol and cetyltrimethylammonium bromide, 9:1), anionic (emulsifying wax BP; a mixture of cetostearyl alcohol and sodium dodecyl sulfate, 9:1) or nonionic (nonionic emulsifying wax BPC; a mixture of cetostearyl alcohol and cetomacrogol, 4:1). The theory dictates that when the cream is formulated it is composed of at least four phases (Fig. 4): 1. 2. 3. 4. Bulk water A dispersed oil phase A crystalline hydrate A crystalline gel phase composed of bilayers of surfactant and fatty alcohol separated by layers of interlamellar-xed water

An examination of ternary systems (containing emulsifying wax and water, but no oil phase) by X-ray diffraction indicated that, for all emulsifying systems, addition of water caused swelling of the interlamellar spaces. Ionic emulsifying systems possess a greater capacity to swell than nonionic systems. Swelling in ionic systems is an electrostatic phenomenon, whereas in nonionic systems, it is due to hydration of the polyoxyethylene chains and is limited by the length of this chain. The gel network

Table 3
and Waxes

HLB Values for Several Oils Emulsion type

Constituent Liquid parafn Hard parafn Stearic acid Beeswax Castor oil Cottonseed oil

O/W 12 10 16 12 14 9

W/O 4 4 5

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

Figure 4 The gel network theory suggests that when a cream is formulated it is composed of four phases: bulk water, a dispersed oil phase, a crystalline hydrate, and a crystalline gel phase composed of bilayers of surfactant and fatty alcohol separated by layers of interlamellarxed water. (Courtesy of Dr. G. M. Eccleston.)

theory also offers an explanation for the observation that nonionic O/W creams thicken on storage. This change is related to additional gelation in the continuous phase as a result of slow hydration of the polyoxyethylene chains of the surfactant, which reduces the amount of free water in the formulation. A relatively stable emulsion formulation may be prepared from a simple fourcomponent mixture: oil, water, surfactant, and fatty amphiphile. In practice, however, things are never that straightforward. In addition to the four principle components, a pharmaceutical emulsion formulation will also contain a drug, and is likely to contain a cosolvent for the drug, a viscosity enhancer, a microbiological preservative system, a pH adjustingstabilizing buffer, and an antioxidant system. All of these additional components are required so that the formulation is capable of delivering the correct amount of drug to the therapeutic application site from a formulation that is free from microbial contamination and is essentially physically unchanged from the day of manufacture.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

B.

Preservation of Semisolid Formulations

All pharmaceutical semisolid formulations that are not sterilized unit-dose products can support the growth of microorganisms. Preservatives are ingredients that prevent or retard microbial growth and protect formulations from spoilage. The use of preservatives is required to prevent product damage caused by microorganisms during manufacture, storage, and inadvertent contamination by the patient during use. Similarly, preservatives serve to protect consumers from possible infection from contaminated products. The characteristics of an ideal preservative system are shown in Table 4. No single preservative meets all of these characteristics for all formulations (43), and it is often necessary to use a system containing a combination of individual preservatives. It is also important to appreciate that preservatives are intrinsically toxic materials so that a balance must be achieved between antimicrobial efcacy and dermal toxicity. The most commonly used preservatives in pharmaceutical products are the parabens (alkyl esters of p-hydroxybenzoic acid, such as methyl and propyl paraben). These compounds are highly effective against both gram-positive bacteria and fungi at low concentrations (e.g., 0.10.3% parabens combinations provide effective preservation of most emulsions). Because of their widespread use, the toxicological proles of the parabens have been extensively researched, and the safety in use of the lower esters (methyl, ethyl, propyl, and butyl) has been established (45). Other preservatives widely used in topical pharmaceutical formulations include benzoic acid, sorbic acid, benzyl alcohol, phenoxyethanol, chlorocresol, benzalkonium chloride, and cetrimide. Each has particular advantages and disadvantages, which makes combination preservatives particularly effective. For example, although methyl paraben is highly active against gram-positive bacteria and moderately active against yeasts and molds, it is only weakly active against gram-negative bacteria. However, a combination of methyl paraben with phenoxyethanol provides a preservative system that is also highly active against gram-negative species. The acid preservatives (benzoic and sorbic acids) are active only as free acids and formulations containing these preservatives must be buffered to acid pH values (pH < 5). A list of pharmaceutical preservatives useful in topical formulations is given in Table 5, together with their microbiological and physicochemical properties. More information on preservatives and preservative systems may be found in the British Pharmaceutical Codex (The Pharmaceutical Press) and in the excellent text by Orth (43).

Table 4
1. 2. 3. 4. 5. 6. 7.

Characteristics for an Ideal Preservative

Effective at low concentrations against a wide spectrum of microbes Soluble in the formulation at the required concentration Nontoxic and nonsensitizing to the consumer at in-use concentrations Compatible with other formulation components No physical effect on formulation characteristics Stable over a wide range of pH and temperature Inexpensive

Source: Ref. 44.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

Table 5

Microbiological and Physicochemical Properties of Selected Preservatives Antimicrobial activitya In-use Conc (%) 0.1 0.2 1.0 0.40.8 0.40.8 0.40.8 0.1 0.010.25 0.010.1 pH rangeb 25 <6.5 Wide 3.09.5 3.09.5 3.09.5 <8.5 410 410

Preservative Benzoic acid Sorbic acid Phenoxyethanol Methyl paraben Propyl paraben Butyl paraben Chlorocresol Benzalkonium Cl Cetrimide
a b

Gram 1 2 2 1 1 1 1 1 1

Gram 2 2 1 3 3 3 2 2 2

Molds 3 2 3 2 2 2 3 3 3

Yeasts 3 1 3 2 2 2 3 2 2

O/Wc 36 3.5 7.5 80 280 117190 <1 <1

1, highly active; 2, moderately active; 3, weakly active. Optimal pH range for activity. c Oilwater partition coefcient.

Interestingly, despite their widespread use and excellent safety prole, there is increasing interest in the potential systemic exposure to preservatives following application in pharmaceutical and cosmetic products. Skin penetration data are not available in the literature for many preservatives, and much of the publicly available data has been obtained under conditions inappropriate for risk assessment. By far the most available data concern the parabens. Although most studies on skin permeation of parabens have evaluated one or two of the homologous series, recently Dal Pozzo and Pastori [46] reported on the in vitro human skin permeation of six parabens (methyl, ethyl, propyl, butyl, hexyl, and octyl esters). The permeants were applied to abdominal epidermal membranes, mounted in diffusion chambers, either as solid compounds (deposited in acetone) or as saturated solutions in various vehicles, including three typical emulsion formulations (two O/W and one W/O). Following application of the unformulated pure substance, maximum ux decreased with increasing lipophilicity, from 65.0 g/cm2 h1 for methyl paraben to 13.7 g/cm2 h1 for hexyl paraben. Similarly, when applied as saturated aqueous solutions, the maximum ux decreased with increasing lipophilicity. In the latter case, however, when ux was normalized to the vehicle concentration of the permeant, the permeability coefcient increased with increasing lipophilicity. Addition of 50% propylene glycol or 20% polyethylene glycol 400 to the aqueous solutions did not alter the prole of permeation, although the permeability coefcients were somewhat reduced. On the other hand, when the parabens were dissolved in liquid parafn, the relation between permeability coefcient and lipophilicity appeared parabolic, maximum ux occurred for the butyl ester, and the highest permeability coefcient occurred for the propyl ester. These results are consistent with theory and demonstrate that the application vehicle can signicantly affect skin permeability characteristics of compounds. In emulsion systems the existence of two distinct phases (oil and water) results in distribution of the parabens according to their physicochemical characteristics, and this can be inuenced by the presence of other ingredients, such as cosolvents and

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

surfactants. Furthermore, these excipients may also affect skin barrier properties. It was observed that permeation of the parabens from two O/W emulsions was higher than expected, based on the data from simple vehicles, and that permeation from the O/W emulsions was higher than that from the W/O emulsion (46). These results were rationalized on the basis of permeant release from the formulation, and it was assumed that the external lipid phase of the W/O emulsion retained parabens. It is difcult, however, to relate this data to conditions of actual consumer exposure because, in the DalPozzo and Pastori study (46), the formulations were applied at innite dose (which effectively generated occlusive conditions) and at an articially high permeant concentration (0.7% w/w). When low nite doses are applied to the skin surface, the vehicle is continually changing. Whether the vehicle is O/W or W/O, the water content will be released by the shear forces generated by application to the skin, most of the water will evaporate, and this should tend to reduce the skin penetration of these preservatives. It is theoretically possible to reduce skin absorption of the parabens by formulation manipulation. This may be achieved by altering the distribution pattern of the preservative within the formulation or by complexation. Care must be taken, however, to ensure that any formulation modication does not interfere with the antimicrobial activity of the preservative system (47). One such modication involved the complexation of methyl paraben with 2-hydroxypropyl--cyclodextrin (48). Although the aqueous solubility of methyl paraben was increased considerably in the presence of the cyclodextrin, the percutaneous penetration of the preservative through hairless mouse skin in vivo over 24 h was reduced by 66%. Even if this system clearly had benet in terms of potential reduction in systemic absorption of the preservative, there was no indication of any possible alteration in preservative efcacy. Incorporation of butyl paraben into liposomes prepared from phosphatidylcholinecholesteroldicetylphosphate had no effect on preservative efcacy, although the antimicrobial effect was proportional to the free, and not the total, concentration of the preservative (49). A further study by these authors demonstrated that incorporation of butyl paraben in some liposomal systems had little effect on the permeation of the preservative across guinea pig skin in vivo or in vitro (50), although incorporation of increasing amounts of lipid into the liposome tended to decrease the percutaneous absorption. Although there is considerable evidence that parabens can penetrate the skin, permeation and systemic availability of intact compounds are likely to be considerably reduced by transcutaneous and systemic metabolism. Furthermore, because these preservatives are present at concentrations of 0.10.2% w/w in topical pharmaceutical formulations, in use, dermal exposure to these compounds is relatively low. In the cosmetic industry there is a trend toward preservative-free and self-preserving formulations (51). However, before taking this route, the pharmaceutical formulator must consider the potential implications on efcacy and safety of the product. C. Drug Release from Semisolid Formulations and SUPAC-SS

Determination of the ability of a semisolid formulation to release a drug, and the pattern and rate of this release, are important aspects of formulation development and optimization. However, it is also important to appreciate that the data generated must not be over interpreted. Release studies normally involve measurement of drug

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

diffusion out of a mass of formulation into a receiving medium that is separated from the formulation by a synthetic membrane (52,53). A detailed study of the data from this type of system can generate invaluable data concerning the physical state of the drug in the formulation. For example, examination of the early experimental models and their rened updates that were derived to describe drug release from semisolids, reveals that release patterns are dependent on whether the drug is present as a solution or suspension within the formulation (5456). These subtle differences, together with variations in the rate of release, may be used to determine such parameters as drug diffusivity within the formulation matrix, the particle size of suspended drug, and the absolute solubility of a drug within a complex formulation (Fig. 5) (57). However, it is generally agreed that drug release-rate data cannot predict skin permeation or bioavailability. Application of a formulation to the skin is an ephemeral situation. The formulation will undergo considerable shearing forces, solvents will evaporate, and excipients may interact with the skin and modulate bioavailability. Nonetheless, release-rate determinations are important for purposes other than formulation development and characterization. The FDA has issued a guidance document (SUPAC-SS Nonsterile Semisolid Dosage Forms, U.S. Department of Health and Human Services, Food and Drug

Figure 5 Illustration of the use of release rates from semisolid preparations to determine drug solubility within a formulation. Data shows the release rates of benzocaine from propylene glycol/water gels as a function of drug concentration in the formulation. (From Ref. 85.)

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

Administration, Center for Drug Evaluation and Research, May 1997) that recommends the use of in vitro drug-release testing in the scale-up and postapproval changes for semisolids (SUPAC-SS). The FDA intends to promote the use of this test as a quality assurance tool to monitor minor differences in formulation composition or changes in manufacturing sites, but not yet as a routine batch-to-batch quality control test. Thus, although the agency is suggesting in vitro release rate data for level 2 and level 3 changes in formulation components and composition, such data are not required for a level 1 change. In the former case, the in vitro release rate of the new or modied formulation should be compared with a recent batch of the original formulation, and the 90% condence limit should fall within the limits of 75133%. Similarly, in vitro release testing is suggested for level 2 changes in manufacturing equipment, processes, and scale-up, and level 3 changes in manufacturing site. Recently, the use of in vitro testing as a quality assurance tool has been questioned, especially for a hydrophilic formulation containing the highly watersoluble drug, ammonium lactate (58), for which the method was insufciently specic to differentiate between small differences in drug loading or minor compositional and processing changes. D. Formulation of Transdermal Products

This section concentrates on those aspects of development specic to transdermal delivery systems. 1. Pharmacokinetic Modeling

To assess the feasibility of the transdermal route for a specic therapeutic agent, several pharmacokinetic factors, including rate of absorption and elimination, must be considered. Guy and Hadgraft (59) developed an early kinetic model for topical drug delivery. Various rate constants were described including release rate from the device (rst- or zero-order); back diffusion from the stratum corneum to the patch (usually insignicant); diffusion rates of the compound across the stratum corneum and viable epidermis; reverse rate constant (enabling prediction of the stratum corneumviable epidermis partition coefcient) and the plasma clearance rate constant. This model has been used to predict the plasma concentration of several transdermally administered drugs and has shown remarkable similarities between predicted and actual plasma concentration proles, suggesting that the model is useful for evaluating the feasibility of potential transdermal drug delivery candidates. Although this model provides predictive quantitative data based mainly on physicochemical and pharmacokinetic parameters, other aspects of the percutaneous absorption process are more difcult to predict. Of most concern is the potential for metabolic degradation of the drug, but only limited data are available on quantitative aspects of cutaneous metabolism. Modeling is possible, by estimating metabolic rate constants and varying the possible residence time of the drug in the skin. More lipophilic drugs, which have a relatively long residence time in the skin, will be more exposed to cutaneous metabolic enzymes and be less bioavailable than their hydrophilic counterparts. More recent models for prediction of skin permeation and pharmacokinetics have become increasingly elaborate and take into account such parameters as the hydrogen-bonding capability of the permeant and potential vehicle effects (60).

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

2.

Biopharmaceutical Considerations

A fundamental consideration in the development of transdermal therapeutic systems is whether the dermal delivery route can provide the requisite bioavailability for drug effectiveness. This is ultimately determined by the skin-penetration rate of the drug, the potential for metabolism during permeation across the skin, and the biological half-life of the drug. Penetration rates may be modied, if necessary, by the use of penetration enhancers, but drug metabolism and plasma clearance cannot be inuenced by any simple means. Although prediction of skin penetration and bioavailability of drugs from transdermal therapeutic systems has been reasonably accurate, there is no doubt that testing of formulated patches in vitro and in vivo will continue to be the most accurate means of evaluating their usefulness. A wide variety of experimental approaches have been developed for in vitro drug permeability determination through skin, and guidelines have been established to rationalize this aspect of pharmaceutical development (see Chapter 5). In the early stages of product development, skin penetration rates from prototype vehicles and patches are usually determined in vitro using simple diffusion cells and skin from a variety of animals. Although, the use of in vitro systems provides little quantitative information on the transcutaneous metabolism of candidate drugs, a major advantage is that experimental conditions can be controlled precisely so that the only variables are in the prototype formulations. In the latter stages of product development, when quantitative skin permeation data is required, human skin should be the membrane of choice for use in in vitro systems. Although methods are available to improve the sensitivity of in vitro skin penetration measurements (61), it is essential, at this stage, to ensure that account is taken of the inherent variability in human skin permeation. Factorial design and articial neural networks have been used in the optimization of transdermal drug delivery formulations using in vitro skin permeation techniques (6264). For example, Kandimalla et al. (63) optimized a vehicle for the transdermal delivery of melatonin using the response surface method and articial neural networks. Briey, three solvents (water, ethanol, and propylene glycol) were examined either as single solvents or binary and ternary mixtures. Measurements of skin ux, lag time, and solubility were made for ten vehicles and compared with values predicted from both a response surface generated from a quartic model and an articial neural network employing a two-layered back-propagation network with all ten design points in the hidden layer. Predictability of ux using both statistical techniques was good (Table 6), suggesting that such models may be useful in preliminary formulation optimization. A major drawback of transdermal delivery systems is the potential for localized irritant and allergic cutaneous reactions. At the earlier stages of formulation development, it is, therefore, important to evaluate both drugs and excipients for their potential to cause irritation and sensitization (see Chapters 10 and 11). This is true for all transdermal systems, but especially for those that may stay in place for prolonged periods. The degree of primary and chronic irritation, and the potential to cause contact allergy, photoirritation, and photoallergy should be determined. Normally, the drug and excipients are initially separately evaluated for contact irritation and sensitization in animal models before evaluation in human subjects. It must, however, be emphasized that animal data are often not predictive of the human situation. Evaluation of skin irritation and delayed contact hypersensitivity should

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

Table 6

Experimental Versus Predicted Fluxa of Melatonin Experimental (g/cm2 h1)c 11.32 0.86 10.89 1.36 7.54 1.39 ANN Prediction (g/cm2 h1) 12.17 11.83 6.75 RSM Prediction (g/cm2 h1) 12.73 11.76 7.50

Application vehicleb W:E:P (20:60:20) W:E (40:60) W:P (40:60)

Flux was predicted on the basis of articial neural networks (ANN) or response surface methods (RSM). b W, water; E, ethanol; P, propylene glycol. c Data are means standard deviation (n = 3) and represent ux through rat dorsal skin. Source: Ref. 63.

always be carried out using the nal and complete formulation in human volunteers. Fortunately, most of the observed skin reactions to transdermal systems are transient and mild and disappear within hours of patch removal. 3. Design Considerations

All patch-type transdermal delivery systems developed to date can be described by three basic design principles: drug in adhesive, drug in matrix (usually polymeric), and drug in reservoir (Fig. 6). In the latter the reservoir is separated from the skin by a rate-controlling membrane. Although there are many differences in the design of transdermal delivery systems, several features are common to all systems including the release liner, the pressure-sensitive adhesive and the backing layer, all of which must be compatible for a successful product. For example, if a system is designed in such a way that the drug is intimately mixed with adhesive, or diffuses from a reservoir through the adhesive, the potential for interaction between drug and adhesive, which can lead to either a reduction of adhesive effectiveness, or the formation of a new chemical species, must be fully assessed. Similarly, residual monomers, catalysts, plasticizers, and resins may react to give new chemical species. Additionally, the excipients, including enhancers, or their reaction products, may interfere with adhesive systems. Incompatibilities between the adhesive system and other formulation excipients, although undesirable, may not necessarily be impeding and designs in which the adhesive is remote from the drug delivery area of the system may be developed (see Fig. 6d). There are three critical considerations in the selection of a particular system: adhesion to skin, compatibility with skin, and physical or chemical stability of total formulation and components. All devices are secured to the skin by a skin-compatible pressure-sensitive adhesive. These adhesives, usually based on silicones, acrylates, or polyisobutylene, can be evaluated by shear-testing and assessment of rheological parameters. Standard rheological tests include creep compliance (which measures the ability of the adhesive to ow into surface irregularities), elastic index (which determines the extent of stretch or deformation as a function of load and time) and recovery following deformation. Skin-adhesion performance is based on several properties, such as initial and long-term adhesion, lift, and residue. The adhesive must be soft enough to ensure initial adhesion, yet have sufcient cohesive strength to remove cleanly, leaving no residue. Because premature lift will interfere with drug delivery, the cohesive and

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

Figure 6

Typical transdermal drug delivery system designs.

adhesive properties must be carefully balanced and maintained over the period of intended application. This can be evaluated only by wear-testing, in which a placebo patch is applied to skin. Skin adhesion is affected by shape, conformability, and occlusivity and round patches tend to be more secure than those of sharply angled geometry. If the patch is able to conform to the skin contours it resists lifting and buckling with movement. The presence of water may affect adhesive properties; therefore, the occlusivity of the system must be taken into consideration. Occlusion for prolonged periods can

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

lead to excessive hydration and problems associated with microbial growth that may increase the possibility of irritation or sensitization to the various system components. The backing material and release liner can be fabricated from a variety of materials, including polyvinylchloride, polyethylene, polypropylene, ethylene vinyl acetate, and aluminium foil. The principal requirement is that they are impervious to the drug and other formulation excipients. The most useful backing materials are those that conform with the skin and provide sufcient resistance to transepidermal water loss to allow some hydration of the stratum corneum, yet maintain a healthy subpatch environment. The release liner must be easily separated from the adhesive layer without lifting off any of the adhesive to which it is bound. Liners are usually lms or coated papers, and silicone release coatings are used with acrylate- and rubber-based adhesive systems, whereas uorocarbon coatings are used with silicone adhesives (65). 4. Drug and Enhancer Incorporation

The three principal methods for incorporating the active species into a transdermal system have led to the loose classication of patches as membrane, matrix, or drugin-adhesive types. It is, however, quite possible to combine the main types of patch; for example, by placing a membrane over a matrix, or using a drug-in-adhesive in combination with a membranematrix device to deliver an initial bolus dose. Membrane patches contain a delivery rate-controlling membrane between the drug reservoir and the skin. Microporous membranes, which control drug ux by the size and tortuosity of pores in the membrane, or dense polymeric membranes, through which the drug permeates by dissolution and diffusion, may be used. Several materials can be used as rate-controlling membranes (e.g., ethylenevinyl acetate copolymers, silicones, high-density polyethylene, polyester elastomers, and polyacrylonitrile). Ideally, the membrane should be permeable only to the drug and enhancer (if present) and should retain other formulation excipients. Membranes have also been designed such that they allow differential permeation of an enhancer and drug (6668). This type of membrane, sometimes designated as a one-way membrane, is useful when the drug is present in the adhesive and the enhancer is formulated in a reservoir. Asymmetric polymeric membranes have also been evaluated for use in transdermal delivery systems (69). Asymmetric poly(4-methyl-1-pentene) membranes, fabricated using a drywet inversion method, were used to control the delivery of nitroglycerin. The release rates of nitroglycerin were strongly inuenced by the nature and amount of the nonsolvent (butanol) used, together with the solvent (cyclohexane), in the casting process. This is, perhaps, not surprising, as increasing the amount of nonsolvent increases the porosity of the cast membrane. The concept of ne-tuning delivery of a drug through a given membrane by subtle adjustment of the porosity creates some exciting new possibilities in transdermal technology (70). In all marketed membrane-controlled transdermal systems, the rate-controlling membrane is fabricated from synthetic polymeric materials. Thacharodi and Rao (71) evaluated the potential of two biopolymers (fetal calf skin collagen and chitosan) in membrane systems for delivery of nifedipine. Chitosan (deacetylated chitin) is a widely distributed major constituent of the shells of marine shellsh. It was concluded that the permeability of both biopolymers could be readily adjusted by altering the fabrication method or cross-linking and, because these polymers were biocom-

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

patible, they were more suitable for use as rate-controlling membranes in transdermal systems. A variety of materials can be used in the drug reservoir, ranging from simple formulations (such as mineral oil), to complex formulations (such as aqueousalcoholic solutions and gels, with or without various cosolvents, and polymeric materials). A denite requirement for a reservoir system is that it can permit zero-order release of the drug over the delivery period. Essentially, this requires that the reservoir material is saturated with the drug over the period of product application, which can be achieved by formulating the drug as a suspension. The second type of transdermal system is the matrix design, in which the drug is uniformly dispersed in a polymeric matrix, through which it diffuses to the skin surface. Here, the polymeric matrix, which may comprise silicone elastomers, polyurethanes, polyvinyl alcohol, polyvinylpyrrolidones, and such, may be considered the drug reservoir. Several steps are involved in the drug delivery process: principally dissociation of drug molecules from the crystal lattice, solubilization of the drug in the polymer matrix, and diffusion of drug molecules through the matrix to the surface of the skin. Many variables may affect the dissolution and diffusion rates, making it particularly difcult, but not impossible, to predict release rates from experimental or prototype formulations (72). For a drug to be released from a polymeric matrix under zero-order kinetics, the drug must be maintained at saturation in the uid phase of the matrix, and the diffusion rate of the drug within the matrix must be much greater than the diffusion rate in the skin. Several methods can be used to alter the release rate of a drug or an enhancer from a polymeric matrix, and some of these are illustrated by a study on release of several drugs from silicone matrices (73). Silicone medical-grade elastomers (polydimethylsiloxanes) are exible, lipophilic polymers, with excellent compatibility with biological tissues, that can be coformulated with hydrophilic excipients, such as glycerol, and inert llers, such as titanium dioxide, to alter release kinetics. Increasing the amount of glycerol in the matrix increased the release rate of indomethacin, propranolol, testosterone, and progesterone, whereas incorporation of inert llers (titanium dioxide or barium sulfate) tended to reduce the release rate. Hydrophilic drug-release rates from polydimethylsiloxane matrices were also increased by up to three orders of magnitude using polydimethylsiloxanepolyethylene oxide graft copolymers (74). These data demonstrate that release rates can be modulated to achieve a desired prole by simple formulation modication. Perhaps the simplest form of transdermal drug delivery device, which is now most commonly employed, is the drug-in-adhesive system. This involves formulating the drug, and enhancer if present, in an adhesive mixture that is subsequently coated onto a backing membrane, such as a polyester lm, to produce an adhesive tape. This simplicity is, however, deceptive and several factors, involving potential interaction between drug or enhancer and the adhesive, need to be considered. These can involve chemical interactions resulting in interference with adhesive performance, breakdown of the active species, or formation of new chemical entities. Additionally, the physicochemical characteristics of the drug and adhesive system may provide very different release rates for hydrophilic and hydrophobic drugs: for example, silicone adhesives are typically lipophilic, which limits solubility of hydrophilic drugs within the adhesive matrix.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

Incorporation of other excipients, such as skin permeation enhancers, into a drug-in-adhesive system may alter drug release rates and adhesive properties. For example, addition of 1% urea to a polyacrylate pressure-sensitive adhesive resulted in loss of adhesion and skin contact could not be maintained over the required period (75). A strategy to reduce the inuence of drug and enhancer on adhesive properties was to design a system in which there was no contact between these constituents and the adhesive, by limiting the adhesive to a boundary laminate that surrounded a drugenhancer-releasing layer. One disadvantage of this type of system, however, is that the drugenhancer-releasing layer may not remain in intimate contact with the skin. If high levels of liquid skin penetration enhancers are incorporated into drug-in-adhesive transdermal patches, there is likely to be a loss in cohesiveness, resulting in patch slipping and skin residues following patch removal. Cohesive strength, can be increased by high levels of cross-linking in acrylate adhesives, but this may alter both long-term bonding and drug release rates. These problems may be overcome by use of grafted copolymer adhesives, such as ARcare ETA Adhesive Systems (76), for which reinforcement is achieved mainly through phase separation of the side chain within the continuous polymer network. A variety of side chains are available and enhancer concentrations up to 30% can be incorporated without seriously affecting the adhesive properties. This work has been limited to fatty acid ester type enhancers and application to other enhancer types remains to be established. It may also be possible to maintain adhesive properties in the presence of skin penetration enhancers by using different molecular weight blends of acrylic copolymers (77,78). When enhancers are incorporated into transdermal systems it is important to appreciate that it is a fundamental requirement that the enhancer, as well as the drug, is released by the adhesive. Furthermore, it is probable that the presence of the enhancer may increase ski permeation of other formulation excipients and that this may have an inuence on local toxicity. Much remains to be done in the eld of enhancer incorporation into transdermal drug delivery systems and it is encouraging to observe the increasing efforts of adhesive manufacturers in this sphere. 5. System Manufacture and Testing

The manufacturing processes for reservoir, matrix, and drug-in-adhesive transdermal systems are, to a large extent, similar. All involve the following stages: preparing the drug; mixing the drug (with other excipients and penetration enhancers, if required) with the reservoir, matrix, or adhesive; casting into lms and drying (or molding and curing); laminating with other structural components (e.g., backing layer, rate-controlling membrane, and release liner); die-cutting; and nally, packaging. Casting and lamination are the most critical steps in the manufacturing process: tensions and pressures must be carefully controlled to provide a wrinkle-free laminate that ensures reproducible adhesive-coating thickness and uniform drug content (79). As with all controlled-release delivery systems, nal product checks include content uniformity, release-rate determination, and physical testing. Content-uniformity evaluation involves removing a random sample of patches from a batch and assaying the amount of drug present. Of the several methods available for determining drug release rates from controlled-release formulations, the U.S. Pharmaceutical Manufacturers Association (PMA) Committee (80) has recommended three: the

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

paddle over disk (which is identical with the USP paddle dissolution apparatus, except that the transdermal system is attached to a disk or cell resting at the bottom of the vessel that contains medium at 32C); the cylinder-modied USP basket (which is similar to the USP basket method, except that the system is attached to the surface of a hollow cylinder immersed in medium at 32C), and the reciprocating disk (in which patches attached to holders are oscillated in small volumes of medium, allowing the apparatus to be useful for systems delivering low concentrations of drug). Researchers at the FDA have developed a modied paddle procedure (essentially the paddle over disk method) for determining drug release from transdermal systems (81). One problem with the original method was the mode of maintaining the patch in position in the dissolution beaker, and a device to improve and maintain placement of the patch was subsequently suggested (82). Although the industry is moving rapidly in method development, much remains to be established in the eld of pharmacopeal standards for transdermal drug delivery systems. IV. REVIEW OF CLINICAL USE OF TRANSDERMAL SYSTEMS

Despite the barriers to systemic delivery of drugs through intact human skin, there have been several successful transdermal products. The remainder of this chapter describes recent clinical studies on several of these products, including those delivering nicotine, glyceryl trinitrate, estrogens, androgens, fentanyl, clonidine, and nonsteroidal anti-inammatory agents (NSAIDs). A. 1. Nicotine Smoking Cessation

Data on efcacy, safety, and pharmacoeconomics of transdermal (TD) nicotine therapy for smoking cessation up to 1994 were reviewed (86). TD nicotine more than doubled success rates of smoking cessation in motivated subjects smoking 1015, or more, cigarettes a day. Application site reactions (erythema or burning 16%, transient itch 50%) caused discontinuation in 10%, or fewer of subjects. Sleep disturbance due to nocturnal nicotine absorption occurred in less than 13% of subjects. Attempts to develop a nicotine transdermal system (TDS), with reduced adverse skin reactions, used in vitro human cadaver skin permeation to demonstrate a permeation prole comparable with those from Habitrol and Nicoderm, and a clinical study compared systemic bioavailability and pharmacokinetic proles (87). Immediate effects of TD nicotine on sleep architecture, snoring, and sleep-ordered breathing in 20 nonsmoking subjects with a history of habitual snoring receiving placebo or a nicotine TD system delivering 11 mg/24 h were reported (88). Patches were applied at 6 PM and removed after 12 h. Mean nicotine level was nondetectable with placebo and 7.8 2.3 ng/mL with active therapy. Nicotine signicantly decreased total sleep time by 33 min, sleep efciency from 89.7 to 83.5%, and rapid eye movement (REM) sleep from 18.8 to 15.1%. Although sleep index was unchanged, mean snoring intensity decreased by 1.1 dB (p = 0.01) with nicotine. Nausea (50%) and vomiting (20%) were predominant side effects. Concurrent administration of the nicotine antagonist mecamylamine and nicotine TD was evaluated (89) in a trial with 48 healthy smokers. Nicotine TD (68 weeks) plus oral mecamylamine (2.55 mg twice a day for 5 weeks) was compared

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

with nicotine TD plus placebo. Mecamylamine treatment began 2 weeks before smoking cessation. At 7 weeks with mecamylamine, the continuous abstinence rate tripled (50%) and point abstinence doubled (58%) compared with that for placebo. Continuous abstinence was threefold higher for mecamylamine at 6 months and ninefold higher at 12 months. It was concluded that agonistantagonist therapy may substantially improve smoking cessation. The pharmacokinetic and pharmacodynamic interactions between TD mecamylamine and intravenous nicotine were investigated (90). Cigarette smokers received TD mecamylamine (6 mg/24 h) and placebo patches for 7 days each. On day 5, subjects received a combined infusion of deuterium-labeled nicotine and cotinine; the disposition kinetics of nicotine and cotinine, and cardiovascular and plasma catecholamine responses to nicotine were measured. Fifty percent of subjects were studied under alkaline urine conditions and 50% under acidic urine conditions. Steady-state plasma mecamylamine concentrations were double (12.2 vs. 6.3 ng/mL), consistent with lower renal clearance (2.1 vs. 5.8 mL/min kg1) during alkaline compared with acidic urine conditions. Mecamylamine did not signicantly affect clearance of nicotine or cotinine, but did signicantly reduce the volume of distribution and inhibited the epinephrine-releasing effects of nicotine. It was concluded that mecamylamine had little effect on nicotine clearance and was not expected to affect steady-state levels during TD nicotine dosing. Reduction of the volume of distribution of nicotine by mecamylamine suggested that part of the antagonism of nicotinic CNS effects by mecamylamine may be due to a pharmacokinetic interaction, probably decreased binding to nicotine receptors, or transport of nicotine into the brain, and may decrease potential adverse cardiovascular effects of coadministered nicotine. Stable isotope-labeled nicotine was infused simultaneously with TD nicotine to determine absolute bioavailability and absorption kinetics from TD systems of different designs (91). Rapid intravenous infusion of nicotine slowed absorption of TD nicotine, probably by constricting dermal blood vessels, thereby limiting percutaneous absorption. A study evaluated safety and tolerability of a 44-mg/day dose for smoking cessation in subjects who smoked 20 or more, cigarettes per day (92). Smokers received 44 mg/day of TD nicotine for 4 weeks, followed by 4 weeks at 22 mg/day. Thirty-eight of forty patients (95%) completed the initial 4 weeks, and 36 (90%) completed the entire study. Conrmed smoking cessation rates were 65% (4 weeks) and 55% (8 weeks), and self-reported smoking cessation at 3 months was 50%. Sleep complaints were reported by 33% of subjects during the 44-mg phase. It was concluded that 44 mg/24 h nicotine patch therapy in heavy smokers was safe, tolerable, and without signicant adverse events. Efcacy and safety of 22- and 44mg doses of TD nicotine paired with minimal, individual, or group counseling to improve smoking cessation rates were compared in an 8-week clinical trial using daily cigarette smokers (15 or fewer cigarettes per day for 1 year) and random assignment to dose and counseling conditions (93). Four weeks of 22- or 44-mg transdermal nicotine followed by 4 weeks of dosage reduction (2 weeks of 22 mg followed by 2 weeks of 11 mg). Smoking cessation rates for TD doses and levels of counseling did not differ signicantly 8 or 26 weeks postquitting date. In those receiving minimal contact, 44 mg produced greater abstinence (68%) at 4 weeks than 22 mg (45%). Participants with minimal-contact adjuvant treatment were less abstinent at 4 weeks than those receiving individual or group counseling (56 vs. 67%). Although a 44-mg dose decreased desire to smoke more than a 22-mg dose,

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

this effect was unrelated to success in quitting smoking. The 44-mg dose produced signicantly higher frequencies of nausea (28%), vomiting (10%), and erythema, with edema at the patch site (30%) than did the 22-mg dose (10, 2, and 13%, respectively). It was concluded that there was no general, sustained, benet of initiating TD nicotine therapy with a 44-mg dose or providing intense adjuvant cessation treatment. The effects of using nicotine gum with nicotine TD were evaluated in healthy subjects (374) at their work-setting in a 1-year trial (94): 149 subjects received nicotine TD plus nicotine gum; 150 nicotine TD plus placebo gum; and 75 placebo TD plus placebo gum. Treatment duration was 12 weeks with a 16-h 15-mg TDS, then 10 mg (6 weeks) and 5 mg (8 weeks). Gum use was unrestricted during the rst 6 months, with recommendation to use at least four pieces a day. It was concluded that adding nicotine gum use to nicotine TD use in subjects smoking 10, or more, cigarettes a day increased abstinence rates. Efcacy of TD nicotine as an adjunct to advice and support in patients attending hospital was investigated (95) in 234 in- and outpatients with smoking-related respiratory or cardiovascular disease (ages 1875 years) advised to stop smoking. Advice was reinforced by a Smoking Cessation Counselor initially and at 2, 4, 8, and 12 weeks, supplemented by a 24-h TDS in adjusted doses over the study period. Patients, who no longer smoked at 12 weeks, were followed up at 26 and 52 weeks and then self-reported abstinence was validated. Of patients receiving TD nicotine 24/115 (21%) were veried as nonsmokers at 12, 26, and 52 weeks, compared with 17 of 119 (14%) of the placebo group. Cessation was related to increasing age and lower Fagerstrom score. Minor skin reactions and nausea were more frequent in the nicotine group (47 vs. 34% and 12 vs. 3%,, respectively), but severe skin reactions were rare (about 5%). Nicotine concentrations in gastric juice, saliva, and plasma were monitored after Nicorette TD systems (15 mg/16 h) were applied to seven healthy volunteers (96). Nicotine concentration was highest in gastric juice, followed by saliva then plasma which suggested ion-trapping of nicotine base in acidic gastric juice. The incremental cost-effectiveness (based on physician time and retail cost of nicotine TD and benets, based on quality-adjusted life years; QALYs), of addition of nicotine TD to smoking cessation counseling was investigated in men and women smokers (2569 years of age) receiving primary care (97). TD use produced one additional lifetime quitter at a cost of 7,332 dollars. Incremental cost-effectiveness of nicotine TD by age group ranged from 4,390 to 10,943 dollars per QALY for men and from 4,955 to 6,983 dollars per QALY for women. A clinical strategy involving limiting prescription renewals to patients successfully abstaining for the rst 2 weeks improved cost-effectiveness by 25%. These data provided support for routine use of nicotine TD as an adjunct to smoking cessation counseling and for health insurance coverage of TD nicotine therapy. Metanalysis estimated cost-effectiveness of nicotine TD as an adjunct to brief physician counseling during routine ofce visits (98). Depending on age, average costs per year of life saved ranged from 965 to 1,585 dollars for men and from 1,634 to 2,360 dollars for women. Incremental costs per year of life saved ranged from 1,796 to 2,949 dollars for men and from 3,040 to 4,391 dollars for women. It was concluded that the nicotine TD is cost-effective and less costly per year of life saved than other widely accepted medical practices and that physicians and third-party payers should recommend the nicotine patch to patients who wish to stop smoking.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

Assessment of the use and effectiveness of free nicotine TD among Medicaid and uninsured smokers concluded that barriers of access to effective treatment for smoking cessation need to be eliminated (99). The majority (90%) of participants found the TD system useful, 14% were abstinent for 6 months or more, and there was no support for inappropriate use. The cost-effectiveness, for the National Health Service (NHS) in the United Kingdom, of allowing general practitioners to prescribe TD nicotine patches for up to 12 weeks was evaluated (100) using data from a randomized, placebo-controlled efcacy trial of nicotine TD and a survey of associated resource use in 30 general practitioners (GP) surgeries in England. The health benet of TD nicotine treatment was calculated in the number of life years that would be saved by stopping smoking at various ages, and used an abstinence contingent treatment model to calculate incremental cost per life year saved by GP counseling with nicotine-patch treatment over GP counseling alone. If the NHS allowed prescription of nicotine TD for up to 12 weeks, the incremental cost per life year saved would be 398 pounds per person younger than 35 years; 345 pounds for ages 3544 years; 432 pounds for ages 4554 years; and 785 pounds for ages 55 65 years. Cardiovascular effects of smoking, effects of nicotine without tobacco smoke, and available data on cardiovascular risk during nicotine replacement therapy were reviewed (101). Although nicotine gum and TD are approved for over-the-counter sale, and smokers with cardiovascular disease are advised to seek physician counseling before using nicotine products, information on product safety in such patients is not readily available. Nicotine can contribute to cardiovascular disease, by hemodynamic consequences of sympathetic neural stimulation and systemic catecholamine release, but there are many other potential cardiovascular toxins in cigarette smoke. Doses of nicotine obtained by cigarette smoking usually exceed those delivered TD, and cardiovascular effects of nicotine are generally more intense when delivered rapidly by cigarette smoking, rather than more slowly by TD nicotine or nicotine gum. As the dosecardiovascular response for nicotine is at, the effects of cigarette smoking in conjunction with replacement therapy are similar to those of cigarette smoking alone. Although cigarette smoking increases blood coagulability, a major cardiovascular risk factor, TD nicotine does not appear to. Wide variations in levels of nicotine (and cotinine) have been observed after nasal and transdermal delivery. Sources of individual variability in nicotine and cotinine plasma levels after use of replacement systems or cigarette smoking were evaluated (102). Cigarette smokers received four treatments of 5-days duration each, including (a) cigarette smoking (16 cigarettes a day); (b) TD nicotine (15 mg a day); (c) nicotine nasal spray, (241-mg doses a day); (d) placebo nicotine nasal spray (24 doses a day). Disposition kinetics were determined by infusion of deuteriumlabeled nicotine and cotinine. There was considerable individual variation in daily dose of nicotine absorbed (nasal spray 5, TD 23) and in plasma nicotine and cotinine levels. Plasma nicotine levels were determined predominantly by nicotine clearance, whereas cotinine levels were determined most strongly by nicotine dose and, to a lesser extent, by clearance of cotinine and fractional conversion of nicotine to cotinine. To compensate for individual differences in clearance, individualization of dosing based on therapeutic monitoring and comparison with nicotine or cotinine levels during cigarette smoking before treatment may be required to optimize nicotine therapy.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

Determinants of variability and the utility of baseline plasma concentrations as predictors of concentrations during TD treatment was evaluated using data from smoking cessation (n = 466) and pharmacokinetic studies (n = 12) (103). Plasma concentrations of nicotine and cotinine were highly variable. Indirect estimates of plasma clearance (baseline plasma concentration per number of cigarettes per day) together with other factors accounted for 33% or less, variability during TD treatment in the smoking cessation study. In contrast, 7599% was accounted for by direct measurements of plasma clearances and systemic doses of nicotine in the pharmacokinetic study. It was concluded that plasma concentrations of nicotine and cotinine during TD nicotine treatment are poorly predicted by clinical history or baseline plasma concentrations. The frequency of adverse effects associated with use of TD nicotine was estimated by metanalysis of data from 47 reports of 35 clinical trials (104). Few adverse cardiovascular outcomes were reported, and no excess of these outcomes was detected among patients assigned to TD use. Incidence of minor adverse effects (especially sleep disturbances, nausea or vomiting, localized skin irritation, and respiratory symptoms) was clearly elevated among TD groups. Incidence of nausea or vomiting appeared lowest when the patch dose was tapered. Application of urine and serum nicotine and cotinine excretion rates for assessment of nicotine replacement in light, moderate, and heavy smokers undergoing TD therapy was investigated (105,106). Subjects were stratied as light (1015 cigarettes per day), moderate (1630 cigarettes per day), or heavy (more than 30 cigarettes per day) smokers and assigned to a daily 24-hTD system delivering a dose of 0, 11, 22, or 44 mg. Steady-state urinary excretion rates of nicotine and cotinine were attained in 2 and 3 days, respectively, at all doses, independent of smoking rate. Signicant underreplacement occurred with the 11-mg/day dose, particularly in moderate and heavy smokers (<50%), and at 22 mg/day, nicotine replacement was still less than 100% in most subjects. Only a dose of 44 mg/day provided mean replacement exceeding 100%, regardless of baseline smoking rate. Steady-state plasma concentrations of nicotine and cotinine were attained in 1 and 3 days, respectively, at all doses, independent of baseline smoking rate. Individualization of dosage is desirable and plasma cotinine levels at steady state (>3 days of therapy) can be used to calculate percentage replacement using baseline levels. Attempts were made to identify variables associated with long-term smoking cessation following hospitalization (107). Patients were assigned to (a) minimal care (MC; brief physician-delivered motivational message); (b) counseling plus active nicotine patch (CAP; motivational message, 6-week supply of nicotine TD, and extended bedside and telephone counseling); and (c) counseling plus placebo patch (CPP). At 6 months, abstinence rates were 4.9, 6.5, and 9.7% for MC, CPP, and CAP treatments, respectively, but not signicantly different. In another trial (108) 308 smokers were randomly allocated to (a) 3-g dextrose tablets and 15-mg nicotine TD; (b) dextrose and placebo patch; (c) placebo tablets and nicotine TD; (d) placebo tablets and placebo patch. The proportion of smokers abstinent in each group was 49% (dextrose plus active patch); 44% (dextrose plus placebo patch); 36% (placebo tablet plus active patch); and 30% (placebo tablet plus placebo patch). The difference between dextrose and placebo tablets (13%) was signicant, but that between active and placebo patches (6%) was not.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

Efcacy of using nicotine TD for ve months in combination with a nicotine nasal spray for 1 year was evaluated (109) in 237 smokers (22 to 66-years old) receiving either nicotine TD (5 months) with nicotine nasal spray for 1 year (n = 118) or nicotine TD with placebo spray (n = 119). TD treatment comprised 15-mg nicotine (months 13), 10-mg (month 4), and 5-mg (month 5). The combination of use of nicotine TD for 5 months with a nicotine nasal spray for 1 year was a more effective method of smoking cessation than use of a patch only. Sustained abstinence rates for the patch and nasal-spray group and the patch-only group were 51 versus 35% at 6 weeks, 37 versus 25% at 3 months, 31 versus 16% at 6 months, 27 versus 11% at 12 months, and 16 versus 9% at 6 years. Long-term smoking-cessation efcacy of varying doses of TD nicotine was evaluated 45 years after quitting among patients enrolled in the Transdermal Nicotine Study Group investigation (110). Self-reported continuous quit rate for patients assigned 21 mg (20.2%) was signicantly higher than that for patients assigned 14 mg (10.4%), 7 mg (11.8%), or placebo (7.4%). Relapse rates among treatment conditions were similar to those 1-year after cessation. Plasma cotinine replacement levels of 56 outpatient smokers using a 21-mg/ day TDS (Nicoderm CQ) were reported (111). Cotinine replacement was 35232% (mean 107%; median 90.5%). Baseline cotinine level, previous quitting attempts, gender, and Fagerstrom Tolerance Questionnaire scores were signicantly correlated with the percentage of cotinine replacement. Baseline cotinine level plus gender was the most powerful predictor combination. Predictors and timing of adverse experiences during TD nicotine therapy were investigated (112). Intervention consisted of brief behavioral counseling, a booklet containing smoking cessation advice, instructions on patch use, and a 12-week course of decreasing TD nicotine. Most adverse experiences were mild. Sleep problems occurred in 48% and usually started on the day of smoking cessation. Application site reactions occurred in 34%, most frequently after 6 days of therapy. Signicant predictors of sleep problems were female gender and successfully quitting smoking. Predictors of application site reactions were psoriasis or eczema, other skin conditions, age younger than 40 years, female gender, and trade or university education level. Substantially increased nicotine intake during therapy, compared with baseline smoking, occurred in 8% who smoked concurrently, and 4% of those who did not. It was concluded that sleep disturbance during therapy was primarily associated with tobacco withdrawal, rather than excess nicotine from TD treatment. Comparison between nicotine gum, TD, nasal spray, and inhaler was made with assessments at quit date and after 1, 4, and 12 weeks (113). Products did not differ in their effects on withdrawal discomfort, urge to smoke, or rate of abstinence. Continuous validated 12-week abstinence rates were 20% (gum), 21% (TD), 24% (spray), and 24% (inhaler). Compliance with recommended treatment use was high for TD, low for gum, and very low for spray and inhaler. A quantitative gas chromatographymass spectrometry (GCMS) method for simultaneous determination of total and free trans-3-hydroxycotinine (THOC) and cotinine (COT) in human urine during nicotine transdermal therapy was developed (114). Results from six consecutive 24-h urine collections in 71 subjects, who used daily TD nicotine doses of 11, 22, and 44 mg, showed that free THOC was 76% of total THOC, and free COT was 48% of total COT. The effect of 24-h nicotine patches in smoking cessation was evaluated among over-the-counter customers in Denmark

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

(115). A total of 522 customers who smoked ten or more cigarettes per day were randomized to either nicotine or placebo patches; 24-h patches were offered for a 3month period. Those smoking 20 or more cigarettes per day started with patches of 21 mg/day. Those smoking less started with 14-mg/day patches, and all participants were gradually reduced to 7-mg/day patches. There was a signicant increase in smoking cessation rates but, after 8 weeks of follow-up, only among smokers who used 21-mg/day patches. No signicant differences in smoking cessation rates were seen among smokers who started with low-dose nicotine or placebo patches. The Collaborative European Antismoking Evaluation (CEASE) was a multicenter, randomized, double-blind placebo-controlled smoking cessation study, the objectives of which were to determine whether higher dosage and longer duration of nicotine patch therapy increased the success rate (116). Thirty-six chest clinics enrolled 3575 smokers, who were allocated placebo or either standard- or higher-dose nicotine TD (15 and 25 mg daily) each given for 8 or 22 weeks, with adjunctive moderately intensive support. The 12-month sustained success rates were: 25-mg TD for 22 weeks, 15.4%; 25-mg TD for 8 weeks, 15.9%; 15-mg TD for 22 weeks, 13.7%; 15-mg TD for 8 weeks, 11.7%; and placebo, 9.9%. Nicotine plasma levels and safety of nicotine TD in smokers undergoing situations suspected to result in increased nicotine plasma levels were assessed (117). Effects of increasing nicotine intake through sequential administration of nicotine TD (day 2), TD followed by consumption of nicotine gum (day 3), and TD followed by gum consumption and cigarette smoking (day 4) were examined; nicotine plasma levels increased transiently after addition of each nicotine source. Mean AUCs (0 24 h) for nicotine were 453 120 ng h1 mL1 (day 2); 489 143 ng h1 mL1 (day 3); and 485 143 ng h1 mL1 (day 4). A second study evaluated effects of physical exercise on kinetics and safety of two types of nicotine transdermal device. Mean delivered dose was higher with Nicoderm than Habitrol, and the products were not bioequivalent. During a 20-minexercise period, nicotine plasma levels increased by 13 9% for Nicoderm and 30 20% for Habitrol. After exercise, subjects taking Habitrol had a higher incidence of adverse events compared with baseline values, but safety proles remained acceptable. It was concluded that both superimposed nicotine sources and physical exertion resulted in short-lived plasma nicotine elevations and temporarily increased nicotine pharmacodynamic parameters, but without increased risk. Short-term effects of TD nicotine replacement in pregnancy were examined (118). After customary smoking cessation efforts had failed, six prenatal patients (28 to 37-weeks gestation) who smoked one to two packs per day were admitted for a period of 21 h. Maternal and fetal assessments, including vital signs, biophysical prole, and electronic fetal monitoring, amniotic uid index, and umbilical artery Doppler examinations were made, and salivary levels of cotinine and nicotine levels were determined. There were no measurable differences in fetal or maternal wellbeing. During TD use salivary nicotine levels increased to 19.0 13.5 g/L at 480 min (consistent with levels in nonpregnant adults). Surprisingly, salivary cotinine concentrations were much lower (50 g/L) than those in smoking nonpregnant adults and varied little over the period that the patch was worn. Weight changes in subjects receiving variable doses of TD nicotine replacement were assessed in 70 subjects receiving placebo or to 11-, 22-, or 44-mg/ day doses of TD nicotine and 1 week inpatient treatment with outpatient follow-up

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

through 1 year (119). The study included 1 week of intensive inpatient treatment with active TD therapy for a further 7 weeks. Counseling sessions were provided weekly during patch therapy, with long-term follow-up at 3, 6, 9, and 12 months. Forty-two subjects were conrmed as nonsmokers at all weekly visits during TD therapy, and their 8-week weight change from baseline (3.0 2.0 kg) was negatively correlated with the percentage of cotinine replacement (r = 0.38, p = 0.012) and positively correlated with baseline weight and age. Men had higher 8-week weight gain (4.0 1.8 kg) than women (2.1 1.7 kg). This suggested that higher replacement levels of nicotine may delay postcessation weight gain in both men and women, but did not identify predictive factors. TD nicotine use, nicotine and cotinine levels, and fetal effects were investigated in pregnant cigarette smokers aged 18 years or older, whose fetuses were beyond 24-weeks gestational age (120). Serial measurements of mother and fetus were made at baseline while the mother was smoking, while abstaining from smoking, and while using TD nicotine therapy for 4 days in the hospital. Nonpregnant women smokers of similar age were used as comparators. No evidence of fetal compromise was seen while nicotine patch therapy was administered. Morning serum cotinine levels were signicantly higher in nonpregnant than in pregnant subjects, but afternoon levels were not signicantly different. Steady-state urinary levels of nicotine and cotinine were also not signicantly different between pregnant versus nonpregnant patients. On inpatient days 2, 3, and 4 for women not smoking, but wearing nicotine TD, the morning fetal heart rates were signicantly reduced relative to baseline when subjects were smoking. Abuse liability and dependence potential of nicotine gum, TD, spray, and inhaler were compared in 504 male and female smokers (121). No signicant differences between products in terms of satisfaction or subjective dependence, except at week 15 when no patch users rated themselves as dependent. Continued use of nicotine replacement at week 15 was related to rate of delivery of nicotine: 2% for patch, 7% for gum and inhaler, and 10% for spray. Cessation of nicotine replacement between weeks 12 and 15 was not accompanied by withdrawal discomfort or increased frequency of urges to smoke. It was concluded that abuse liability was low for all products. 2. Treatment of Ulcerative Colitis

Ulcerative colitis is largely a disease of nonsmokers and anecdotal reports have suggested that smoking and nicotine may improve symptoms (122). Patients with active ulcerative colitis were treated with either nicotine TD or placebo patches for 6 weeks. All patients had been taking mesalamine, and some were also receiving low doses of glucocorticoids. These medications were continued during the study. Incremental doses of nicotine were used and most patients tolerated 1525 mg/24 h: 17 of 35 patients in the nicotine group had complete remissions, compared with 9 of 37 patients in the placebo group. Patients in the nicotine group had greater improvement in global clinical and histological grades of colitis, lower stool frequency, less abdominal pain, and less fecal urgency. More of the nicotine group had minor side effects (23 vs. 11 in placebo group), and withdrawals owing to ineffective therapy were more common in the placebo group (3 vs. 8). The value of TD nicotine for maintenance of remission was studied in 80 patients with ulcerative colitis in remission, using either TD nicotine or placebo

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

patches for 6 months (123). Incremental doses were given for 3 weeks to achieve a maintenance dose (most tolerated 15 mg for 16 h daily). All patients were taking mesalamine at study entry, but this was stopped when maintenance nicotine doses were achieved. Twenty-two patients in the nicotine group were prematurely withdrawn from the study, 14 because of relapse and 8 for other reasons, including side effects and protocol violations. In the placebo group, 20 patients were withdrawn prematurely, 17 owing to relapse and 3 for other reasons. Among patients using 15mgnicotine patches, serum nicotine and cotinine concentrations were lower than expected, which may have reected poor compliance. Side effects were reported by 35 patients, 21 in the nicotine group and 14 in the placebo group. TD nicotine alone was no better than placebo in maintaining remission of ulcerative colitis, and early withdrawal because of side effects was more common in the nicotine group. Nicotine alone was compared with prednisolone in 61 patients with active ulcerative colitis treated with either nicotine TD or 15 mg of prednisolone for 6 weeks (124). Incremental nicotine doses were given for the rst 9 days. Of the 43 patients who completed the trial, 6 of 19 in the nicotine group achieved full sigmoidoscopic remission, compared with 14 of 24 with prednisolone. In those who completed this study, nicotine alone appeared to be of only very modest benet in acute colitis. Use of TD nicotine in mildly to moderately active ulcerative colitis was investigated (125) in 64 nonsmoking patients with mildly to moderately active ulcerative colitis despite the use of medication. These were stratied (on the basis of smoking history, extent of disease, and concomitant therapy) and assigned to daily treatment with TD nicotine (n = 31) at highest-tolerated dose (11 mg for 1 week and then 22 mg for 3 weeks) or placebo (n = 33). At 4 weeks, 39% of those who received nicotine showed clinical improvement compared with 9% who received placebo. Four patients receiving nicotine discontinued therapy because of side effects. At week 4, the nicotine group had trough serum concentrations of 12.3 8.4 ng/mL (nicotine) and 192 95 ng/mL (cotinine). It was concluded that transdermal nicotine at 22 mg/d for 4 weeks was effective in controlling the clinical manifestations of mildly to moderately active ulcerative colitis. A pilot trial of nicotine TD as an alternative to corticosteroids in ulcerative colitis was reported (126). In ten patients with mild-to-moderate clinical relapses of ulcerative colitis during mesalamine treatment and with a previous history of poorly tolerated steroids, TD nicotine (15 mg daily) was added for 4 weeks. Clinical remission was achieved in seven patients and persisted for up to 3 months after nicotine withdrawal. A second study (127) investigated long-term effects. Patients with mild-tomoderate clinical relapses of left-sided ulcerative colitis during maintenance treatment with mesalamine were allocated additional treatment with either TD nicotine or prednisone for 5 weeks. The rst consecutive 15 patients in each group with clinical and endoscopic signs of remission were followed-up for 6 months, while continuing mesalamine maintenance treatment. Relapses of active colitis were observed in 20% of patients formerly treated with nicotine and 60% of patients in the prednisone group, and relapses occurred earlier in the latter group. As patients with mild-to-moderate active colitis treated with mesalamine plus TD nicotine appeared to suffer fewer relapses than patients treated with mesalamine plus oral prednisone a long-term follow-up was carried out (128). Thirty patients with remission of distal

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

colitis were monitored for 12 months and relapsed patients retreated in a crossover manner. Recurrences were observed in 14 of 15 patients initially treated with steroids and in 7 of 15 subjects who received TD nicotine. 3. Other Indications

Individuals with major depression have a high frequency of cigarette smoking, and TD nicotine can produce short-term improvement in mood. The effects of nicotine patches (17.5 mg) on 12 nonmedicated outpatients with major depression were studied over 4 continuous days (129). Two patients dropped out of the study because of nausea and vomiting. There was signicant improvement in depression after day 2 of TD nicotine and patients relapsed 3 or 4 days after the nal nicotine dose. Although nicotine TDS produced short-term improvement of depression, with minor side effects, nicotine TD was not recommended for clinical use in depression because of the high health risks of nicotine. It was concluded that analogues might be developed that can improve depression without major risks. The therapeutic response to nicotine TD was investigated in patients with Tourettes syndrome (130). Twenty patients (17 children and adolescents, 3 adults) were studied following application of two patches (2 7 mg/24 h). There was a broad range in individual response, but each patch application produced a signicant reduction in the Yale Global Tic Severity Scale scores, for an average duration of approximately 12 weeks. This suggested that TD nicotine could be an effective adjunct to neuroleptic therapy of Tourettes syndrome. Nicotine gum and nicotine TD were used to reduce motor and vocal tics of children (age 8 years or older; weight 25 kg), adolescents, and adults (131). Reduction of tics was seen during chewing of nicotine gum, but improvement lasted no longer than 1 h after chewing. With nicotine TD, motor and vocal tics were reduced 45% over baseline in 85% of 35 subjects within 30 min to 3 h after patch application. Relief of symptoms with a single 7-mg patch, left on the skin for 24 h, persisted for variable periods up to 120 days. Application of a second patch for 24 h when symptoms returned resulted in similar reduction in tic severity and frequency, which persisted an average of 13 3 days. Short-term nicotine injections have improved attentional performance in patients with Alzheimers disease (AD), but little is known about prolonged effects of nicotine. A study evaluated clinical and neuropsychological effects of extended TD nicotine application in AD subjects over a 4-week period (132). Patients were treated with nicotine TD (Nicotrol) for 16 h/day at the following doses: 5 mg/day (week 1), 10 mg/day (weeks 2 and 3), and 5 mg/day (week 4). Nicotine signicantly improved attentional performance, with a signicant reduction in errors of omission, which continued throughout nicotine administration, and variability of reaction time for correct responses was also signicantly reduced. Nicotine did not improve performance on other tests measuring motor and memory function. 4. Poisoning with Nicotine Patches

To evaluate potential adverse effects from inadvertent exposure, three marketed TD nicotine products: Nicoderm (drug reservoir and rate-controlling membrane); Nicotinell (nicotine solution dispersed in cotton gauze between layers of adhesive); and Niconil (nicotine in gel matrix), were administered topically and orally to dogs (133). Topical nicotine doses were 12 mg/kg 24 h1 for all products, with plasma con-

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

centrations 43 ng/mL, or less. Of 12 topical exposures (with Nicotinell and Niconil) 2 were associated with clinical signs (excess salivation or emesis). Oral doses (2.8 mg/kg [one patch] to 13.4 mg/kg [two patches] over 2557 h), were two- to ninefold higher than the oral doses reported to produce severe toxicity in children, and the higher dose was within the known lethal range for dogs. Oral dosing of Nicotinell and Niconil (two patches per dog) produced vomiting in 2 of 12 exposures. No clinical signs were observed with either topical or oral dosing of Nicoderm. Characteristics and outcomes of U.S. poisoning cases, involving dermal human application of multiple nicotine TD systems, were evaluated (134). Nine cases of dermal exposure to 220 nicotine TD systems resulted from intentional misuse or suicide attempts and included concomitant exposure to other drugs in 7 of 9 cases. Mean age was 45 years, and 7 of 9 patients were female. All suffered medical complications, including seizures, other CNS changes, cardiovascular effects, and respiratory failure, but plasma nicotinecotinine levels did not correlate with the severity of illness. Eight patients were hospitalized, but all recovered. B. Glyceryl Trinitrate and Related Drugs

Measurement of plasma concentrations of glyceryl trinitrate (GTN) is very difcult owing to the unusual pharmacokinetics, with very rapid disappearance from plasma, and large intraindividual and interindividual variations (135). There is extensive rstpass hepatic extraction after oral administration and plasma levels are often undetectable. With controlled-release TD GTN systems, plasma concentrations can be maintained over 24 h, but with uctuations and important intra- and interindividual variability. After administration of GTN by any route, glyceryl dinitrates and mononitrates are found in plasma. Bioavailability and main pharmacokinetic parameters of the GTN metabolites 1,2- and 1,3-glyceryl dinitrate (1,2-GDN and 1,3-GDN) were determined following application of two types of GTN TDS in healthy volunteers (136). For 1,2-GDN, AUCs(0-Tlast) were 23.77 h ng1 mL1 (patch A) and 27.83 h ng1 mL1 (patch B). Peak plasma levels were 2.45 ng/mL (at 6.4 h) and 2.93 ng/ mL (at 8.3 h), respectively. For 1,3-GDN AUCs(0-Tlast) were 3.32 h ng1 mL1 (patch A) and 3.81 h, ng1 mL1 (patch B). Peak plasma levels were 0.35 ng/mL (at 6.4 h) and 0.41 ng/mL (at 7.9 h), respectively. Statistical comparison showed bioequivalence between these TD systems for the metabolites investigated. Typical side effects observed after nitrate therapy also occurred. The efcacy of lisinopril, TD GTN, and their combination in improving survival and ventricular function after acute myocardial infarction (AMI) was assessed (137). A total of 19,394 patients were randomized from 200 coronary care units in Italy and assigned 6 weeks of oral lisinopril (5-mg initial dose and then 10 mg daily), or open control, as well as nitrates (intravenous for the rst 24 h followed by TD GTN 10 mg/day) or open control. Lisinopril, started within 24 h after AMI symptoms began, produced signicant reductions in overall mortality and in the combined outcome measure of mortality and severe ventricular dysfunction. Administration of TD GTN did not show any independent effect on the same outcome measures. Systematic combined administration of lisinopril and GTN produced signicant reductions in overall mortality and in the combined endpoint. The effects of a GTN TDS on platelet aggregation was examined in eight normal volunteers (138). A signicant effect of TD GTN on platelet aggregation was demonstrated in the presence and

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

absence of iloprost, but the clinical signicance of the antiplatelet effect of TD GTN remained unknown. The anti-inammatory and analgesic effects of TD GTN was studied in 21 patients with mild to moderate leg varicose veins who underwent vein sclerotherapy in both legs (139). The vein in one leg was treated every 8 h with GTN and compared with a placebo ointment applied to the vein of the other leg. Inammation signs were observed in all cases 15 min after rst application. Intensity of inammation signs were 26% in GTN-treated veins and 61.5% in placebo-treated veins. One hour later only 63% of cases in the GTN group, but all cases in the placebo group, showed signs of thrombophlebitis. All veins in the GTN group were free of signs of thrombophlebitis in fewer than 48 h, whereas, of the placebo group, 45% required more than 48 h. Intermittent TD GTN therapy with a 10- to 12-hpatch-free period each day has documented clinical benets. The antianginal and anti-ischemic effects of three dose levels of TD GTN applied for 12 h daily for 30 days and the development of tolerance and rebound were assessed (140). There was a signicant increase in treadmill walking time to moderate angina in each GTN patch group, compared with placebo, at time points up to 12 h throughout the 30-day period. Secondary efcacy parameters supported the primary efcacy results,and there was no evidence of tolerance or rebound. Transdermal GTN is widely used to treat angina pectoris, but development of tolerance is a major problem (141). The effects of short (5 h) and prolonged (3 days) exposure to transdermal GTN patches on the development of tolerance in terms of hemodynamics and vascular reactivity in the conscious rabbit were, therefore, investigated. It was concluded that in the rabbit, prolonged exposure to clinical GTN patches caused hemodynamic compensation and baroreex resetting, but no evidence of vascular reactivity tolerance. The efcacy of adding transdermal GTN or oral N-acetylcysteine, or both, to conventional medical therapy was examined (142) in a trial of 200 patients with unstable angina, followed-up for 4 months. Death, myocardial infarction, or refractory angina requiring revascularization occurred in 31% of patients receiving GTN, 42% of those receiving N-acetylcysteine, 13% of those receiving GTN plus N-acetylcysteine, and 39% of those receiving placebo. There was higher probability of no treatment failure when receiving both GTN and N-acetylcysteine than with placebo, N-acetylcysteine, or GTN alone. However, combination of GTN and N-acetylcysteine was associated with a high incidence of side effects (35%), mainly intolerable headache. The relation between tolerance development, counterregulatory responses, and arterial vasodilating effects were studied in 20 patients with stable angina pectoris who were exercise tested before, after 2 h, and 24 h of nitrate patch treatment (143). Effects observed after 2 h of treatment on exercise duration, ST-segment depression, blood pressure, and heart rate were usually lost by 24 h, although effects on arterial pulse curves persisted after 24 h, with a mean change from baseline of 29%, compared with 33% at 2 h. After 24 h, a signicant decrease in hematocrit and an increase in body weight were observed. Hematocrit changes correlated with loss of clinical efcacy. It was concluded that clinical nitrate tolerance may be observed despite maintenance of arterial vasodilating effects, and that tolerance is more related to plasma volume expansion as a counterregulatory mechanism.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

A subsequent study (144) examined whether GTN TDS treatment for 24 h could induce local cutaneous changes that impaired drug delivery and clinical efcacy. Twenty angina patients were exercise-tested after 2 and 24 h of treatment and 2 h after device renewal. The TDS was either renewed at a new skin location or on the previous application site. Clinical efcacy, the effect seen on plethysmography, and GTN plasma concentrations, all tended to increase after TDS renewal, regardless of the application site indicating that cutaneous changes of clinical importance were not demonstrated. The effect of GTN in patients with chronic heart failure (CHF) treated with angiotensin-converting enzyme (ACE) inhibition was evaluated (145). High-dose (50100 mg) TD GTN or placebo were given daily for 12 h in 29 patients with CHF. Exercise time (4 h after patch application) showed progressive improvement during GTN administration. GTN decreased left ventricular end-diastolic and end-systolic dimensions and augmented LV fractional shortening. Clinical efcacy of, and patient tolerance to, sustained-release GTN TDS in the treatment of Raynauds phenomenon was reported (146). Patients had primary Raynauds disease or Raynauds phenomenon secondary to systemic sclerosis. GTN TD (0.2 mg/h) was effective in reducing both number and severity of Raynauds attacks, but headaches led to withdrawal of 8 patients and occurred in about 80% of remaining patients. A clinical study in 20 patients with shoulder pain syndrome caused by supraspinatus tendinitis was conducted to determine whether TD GTN has analgesic action in this condition (147). One 5-mg GTN (Nitroplast) patch (or placebo patch) was applied per day over 3 days in the most painful area. Follow-up showed a signicant decrease in intensity of pain at 24 and 48 h in the GTN group, but no change in the placebo group. It was concluded that GTN could be a useful approach to management of this common condition and other tendon musculoskeletal disorders. The 24-h effect of TD GTN on splanchnic hemodynamics in nine patients with biopsy-proved liver cirrhosis was evaluated (148). GTN tape, capable of releasing 15 mg of drug in 24 h, was applied to chest skin at 7 AM of the second day. After GTN application, the mean portal blood velocity and ow signicantly decreased by 18 and 22%, whereas superior mesenteric artery velocity decreased and resistance indices increased. This indicated that GTN, from a transdermal long-acting system, signicantly inuenced portal hemodynamics in liver cirrhosis, and the use of GTN was proposed for long-term clinical studies to test efcacy in preventing gastrointestinal bleeding. The role of TD GTN, as a source of exogenous nitric oxide, in the management of primary dysmenorrhea was investigated (149) in a multinational study. Eightyeight patients from six countries were evaluated during three menstrual cycles while receiving GTN (0.1 mg/h) or placebo patches. The data indicated that TD GTN, as a source of exogenous nitric oxide, was useful as a modulator of uterine contractility and represented a new and mechanistically different therapeutic alternative for management of primary dysmenorrhea. The maternal and fetal cardiovascular effects of TD GTN compared with ritodrine for acute tocolysis (150) were studied in 60 women in preterm labor. At doses required for acute tocolysis, TD GTN had minimal effects on maternal pulse, blood pressure (BP), or fetal heart rate, and signicantly fewer adverse cardiovascular effects than intravenous ritodrine. It was concluded that TD GTN may be a safer treatment for women in preterm labor.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

The efcacy of TD GTN and intravenous ritodrine as tocolytics was also evaluated in an international study (151). A total of 245 women with preterm labor and intact membranes between 24- and 36-weeks gestation were randomized to TD GTN (10 to 20 mg patch) or intravenous ritodrine. GTN and ritodrine prolonged gestation by 74% to 37 weeks. There was no signicant difference in the proportion of women receiving GTN or ritodrine who delivered within the specied days from study entry or weeks of gestation, and no serious maternal side effects were reported for either. It was concluded that there was no overall difference between GTN and ritodrine in the acute tocolysis of preterm labor, but there was a suggested advantage of GTN over ritodrine in reducing preterm delivery rate. Maternal side effect prole and treatment discontinuation rates were fewer for GTN, suggesting it was the safer alternative. The nitric oxide (NO) donor morpholinosydnonmine has been reported to inhibit insulin release in isolated pancreatic islets; accordingly, the effect of TD GTN, an alternative NO donor, on glucose-stimulated insulin release was studied in healthy, young, male volunteers (152). Oral glucose tolerance tests were performed in the presence of placebo or TD GTN (0.4 mg/h of GTN) in the same patients, with a 2-week intertest interval. Glucose-stimulated maximum increases in plasma insulin immunoreactivity were 36.3 5 and 78.8 6.1 mU/mL in the presence of active and placebo patches, respectively, although both fasting and postload blood glucose levels were equal. Active patches signicantly decreased blood pressure with a marginal increase in heart rate. It was concluded that inhibition of glucose-stimulated insulin release by TD GTN without causing hyperglycemia may be a novel component of the antianginal action mechanism of nitrates. Benzoxazinones are a potent new class of organic nitrates used in cardiovascular therapy that have a coronary vascular selectivity greater than that of GTN and isosorbide dinitrate. The ability of these new derivatives to reach therapeutic steadystate plasma concentrations after TD administration was investigated in vitro using human skin (153). Two members of this class: sinitrodil (ITF 296) and ITF 1129 were compared with GTN, isosorbide dinitrate, and nicorandil at two concentrations (0.08% w/v and saturated solutions). Sinitrodil was considered a good candidate for transdermal administration. C. Estrogens

Numerous clinical studies have compared the effects of TD and alternative delivery strategies for steroid hormones on a range of factors and, also, comparisons have been made between reservoir and matrix type TDS, as well as topical gels. The most common use of TD steroids is in hormone replacement therapy (HRT) in women. The reduction in estrogen production in menopause may cause hot ashes, sweating, mood and sleep disturbances, fatigue, and urogenital dysfunction. The effectiveness of estrogen-based HRT in ameliorating these symptoms, and in preventing long-term effects, such as osteoporosis, is well established (154). Comparative trials indicated that 625 g of oral, conjugated estrogens, 20 g oral ethinyl estradiol and 50 g TD estradiol had equivalent efcacy in relief of mild-to-moderate menopausal symptoms and prevention of bone mineral loss. Concomitant progestogen therapy is usually included, if the uterus is intact, to protect against endometrial hyperplasia and carcinoma. Addition of progestogen maintains and may enhance bone-conserving

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

effects of estrogen, and continuous regimens appear to reduce incidence of irregular menses. Adverse reactions are predominantly local skin irritation with TDS (14%) and systemic effects common to most forms of HRT, including breast tenderness, ushing, headache, and irregular bleeding, occurring in 2% or fewer of patients. Costbenet and costeffectiveness of HRT in treatment of menopausal symptoms suggested that conjugated estrogens and TD estradiol compared well with alternative therapies, such as veralipride and Chinese medicines. A combined TDS estradiolnorethisterone system was designed to deliver both estradiol and norethisterone at a constant rate for up to 4 days (155). TD norethisterone did not appear to alter the potentially benecial effects of TD estradiol on total cholesterol, low-density lipoprotein (LDL) or triglyceride levels, or metabolic parameters of bone resorption or vaginal cytology. Protection from the effects of unopposed estradiol was achieved by sequential treatment with TD estradiolnorethisterone for 2 weeks of each 28-day cycle, and most patients experienced a regular vaginal bleeding pattern with this regimen. Menopausal symptoms were improved to a similar extent during estradiol-only and combined estradiolnorethisterone phases. The system was well accepted in clinical trials and generally well tolerated, the most common adverse effect being local irritation. Replacement therapy in 12 amenorrheic adolescents with gonadal dysgenesis treated with TD estradiol 100 g (Estraderm TTS-100) twice weekly for 3 weeks, plus medroxyprogesterone acetate (MPA; Provera) for the last 11 days, following an interval of 1 week, was reported (156). No signicant changes were recorded in FSH, LH, estradiol-17, and PRL serum levels, but signicant decreases of TC values and atheromatic indices 1 (TC/ HDL) and 2 (LDL/HDL), signicant increase in apolipoproteins-A1, and benecial effect on bone mass were seen at the end of treatment. The efcacy and acceptability of a TD norethisterone device was assessed for 6 months in 18 patients of conrmed menopausal status (157). Therapy was continuous application of Estraderm TTS 50 for 28 days, with additional application of 2norethisterone acetate patches for the last 12 days, repeated for six cycles. There was signicant improvement in hot ashes and sweating, and withdrawal vaginal bleeding was established at regular intervals in 9 of 15 who completed 6 months of therapy. Histological examination of endometrial biopsies showed secretory activity in 56% of samples after 6 months, but no evidence of hyperplasia, premalignant, or malignant changes in the remaining biopsies. Efcacy and overall acceptability of 100and 200-g twice-weekly doses of Estraderm TTS in the treatment of severe PMS were compared (158). Women with severe PMS received Estraderm TTS continuously with either esterone 10 mg or MPA 5 mg, from day 17 to day 26 of each cycle. There was no difference in change in total-ESAmax between the Estraderm 100-g and 200-g groups, but there was a greater dropout rate and incidence of side effects attributed to estrogen in the higher-dosage group. Mean estradiol levels were 300 (100 g) and 573 (200 g) pmol/L, and 100 g suppressed midluteal progesterone from a mean of 35.5 to 3.4. It was concluded that the lower-dose therapy was as effective in reducing symptom levels in severe PMS and was better tolerated. Efcacy and tolerability of Menorest 50 was compared with Estraderm TTS 50 in treatment of postmenopausal symptoms in 205 women with moderate to severe vasomotor symptoms (159). After a 4-week treatment-free period, each woman received a cyclic regimen (25 days of a 4-week cycle) of Menorest 50 (matrix-type)

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

or Estraderm TTS 50 (reservoir-type) twice weekly for 12 weeks. Oral progestin was also given for 10 days each cycle. Signicant reduction in hot ashes was observed in each group compared with baseline. There were no signicant differences in mean plasma estradiol levels and mean estradiol/estrone ratio (>1.0) in both groups after 10 weeks. Menorest 50 showed better local tolerability than Estraderm TTS 50. Effects of daily intrauterine release of 20 g of levonorgestrel by an intrauterine device on climacteric symptoms, bleeding pattern, and endometrial histological features in postmenopausal women receiving transdermal estrogen replacement therapy was evaluated in 40 parous postmenopausal women over a period of 1 year (160). Twenty women receiving a continuous TD daily dose of 50 g of estradiol had a levonorgestrel-releasing intrauterine contraceptive device inserted, whereas the control group (n = 20) received a continuous oral dose of 2 mg of estradiol valerate and 1 mg of norethisterone acetate daily. Both treatment regimens effectively relieved climacteric symptoms. The effect of TD estrogen replacement therapy on lipoprotein (Lp) and other plasma lipoproteins was studied (161) in 30 women who had undergone a total abdominal hysterectomy and bilateral salpingo-oophorectomy for benign gynecological conditions treated with 1.5 mg of 17-estradiol gel applied daily for 12 consecutive months. Plasma lipoproteins were measured before treatment and at 6- and 12month intervals. There was signicant reduction in Lp levels during the rst 6 months of treatment, with median values falling from 7.87 to 6.16 mg/dL, but during the second 6 months, median concentration increased to 9.38 mg/dL. Signicant reductions in apoprotein A-I, apoprotein B, HDL-C, and HDL(3)-C were present after 6 months, but at study completion these values were no different than those at baseline. By avoiding the rst-pass effect, this method of delivery did not appear to produce the sustained changes in lipoproteins seen with oral treatment. After menopause the hemostatic balance shifts toward a latent hypercoagulable state and the effects on hemostasis of HRT with TD estradiol and oral sequential MPA were evaluated (162). The balance between procoagulant factors and inhibitors were studied in 255 women in physiological menopause for 15 years, allocated to 1 year of treatment with cyclic TD E2 (50 g/day for 21 days) plus MPA (10 mg/day from days 10 to 21), continuous TD E2 (50 g/day for 28 days) plus MPA (10 mg/day from days 14 to 25), or placebo. Continuous treatment gave signicantly lower nal values of brinogen, factor VII, antithrombin III, protein S, and heparin cofactor II than placebo. Effects of HRT on bone mineral density (BMD) and disease activity in postmenopausal women with rheumatoid arthritis (RA) were investigated in 62 patients with RA, 22 taking placebo and 40 receiving HRT (TD estradiol patches twice weekly for 48 weeks plus norithisterone tablets when clinically indicated) (163). Fifty-nine percent of placebo and 78% of HRT groups completed 48 weeks. At entry, BMD values in the lumbar spine and femoral neck were similar to those in matched controls, whereas at the distal radius, BMD was signicantly reduced to about 50% of control values. In the HRT group, spinal BMD increased signicantly by 0.94% at 48 weeks, but BMD at femoral neck and distal radius did not change in either group. In the HRT group, there was signicant improvement in well-being and articular index. Because older women often experience side effects with conventional HRT, a low-dose preparation (Estraderm 25) was compared with conventional HRT (Estraderm 50) in patients with bone loss (164). A total of 196 women were studied

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

over 1 or 2 years, with 80 reaching 3 years of treatment. In the lumbar spine, BMD increased maximally in year 1 in all groups, and the gain was maintained after 3 years. Only 3.9% of patients were nonresponders at this site after 3 years. Mean changes after 3 years were 8.1 6.8% for Estraderm 25 and 9.0 8.3% for Estraderm 50. At the femoral neck, 10.4% of patients were nonresponders after 3 years, and changes were signicant only in Estraderm 25 in women older than 67 years, and Estraderm 50 in those younger than 67 years. BMD change over 3 years at the lumbar spine and femoral neck correlated with menopausal age. Use of Estraderm 50 was not associated with a greater response of bone mass and there was no evidence of increasing BMD response as estradiol dosage per kilogram of body weight increased. It was reported that oral and transdermal 17-estradiol provided similar benets in clinical studies (165). The lowest effective doses were 0.625 mg/day for conjugated estrogens, 2 mg/day for oral 17-estradiol, 1.5 g/day for 17-estradiol gel, and 50 g/day of 17-estradiol TDS. Decrease in incidence of osteoporotic fractures was achieved only when the duration of HRT exceeded 7 years. The responses of various biochemical markers for bone turnover to TD estradiol were measured in 11 postmenopausal women over 24 weeks (166) and compared with the within-subject variability of markers in 11 untreated healthy postmenopausal women. Mean decrease in markers of bone formation ranged from 19% for procollagen type I C-terminal propeptide to 40% for procollagen type I N-terminal propeptide (PINP). The mean decrease in markers of bone resorption ranged from 10% for tartrate-resistant acid phosphatase (TRAP), to 67% for C-terminal cross-linked telopeptide. The ability to detect a response differed between markers and was not dependent on the magnitude of response to therapy. The highest number of responders were found using PINP (9 of 11) and osteocalcin (9 of 11), and free deoxypyridinoline (8 of 11) and total deoxypyridinoline (7 of 11). Lumbar spine BMD dened four patients as responders. The comparative effects on BMD in routine clinical practice use of tibolone and estrogen (unopposed or combined with cyclic progestogen) in postmenopausal women who had not previously received estrogen or other menopausal therapy were assessed (167). BMD was measured in the spine and hip at 12-month intervals over 3 years in 82 postmenopausal women referred for climacteric therapy. Thirty-ve women received tibolone, 24 TD estradiol alone, and 12 conjugated equine estrogens together with cyclic progestogen; 11 received no therapy other than calcium. Spinal BMD increased signicantly in those taking tibolone over 3 years. In those receiving conjugated equine estrogens and cyclic progestogen, spinal BMD also increased signicantly over years 1 and 2, but not year 3. Although spinal BMD rose over 3 years in women treated with TD estradiol alone, this was not signicant. No signicant change in BMD of spine or hip was observed in the control group. A signicant difference in increase of spinal BMD between treatment groups was observed at 2 years in favor of those taking tibolone or conjugated equine estrogens, compared with TD estradiol. The most common side effect and reason for discontinuation with Norplant use is bleeding disturbance and, therefore, a 6-week application of a patch releasing 100 g/day estradiol was investigated as a method of reducing this problem (168). Of 98 Norplant users, 34 had normal, and 64 abnormal, bleeding patterns. Estradiol (33) or placebo (31) TDS were randomly used to treat patients with abnormal bleeding. Although there was clinical improvement in the estradiol group, this was not signicant.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

Changes in serum lipoproteins, apoproteins, and coagulation factors, induced in postmenopausal women treated by the Gynaderm TDS (designed to deliver 50 g of 17-estradiol per day) were studied over 6 months in 53 hysterectomized, healthy, postmenopausal women (169). One patch was applied twice weekly. There were no signicant changes in levels of total cholesterol, triglycerides, HDL, or LDL, a signicant rise in apolipoprotein A-I (apo A-I) level at 3 months was not sustained after 6 months, and there was a signicant drop in apo A-II level after 6 months. Changes in apo B and Lp were not signicant. There were signicant falls in levels of antithrombin III and protein S, and a signicant rise in factor VII. Changes in levels of brinogen and protein C were not signicant. TD estradiol administration caused minimal changes in lipoprotein metabolism and the statistically signicant changes in the thrombophilia prole parallel those observed with oral HRT. A clinical study evaluated the effect of HRT on plasma lipoproteins and Lp prole in 42 menopausal women with primary hypercholesterolemia (total cholesterol >240 mg/dL) (170). Patients were randomly assigned to (a) TD estradiol, 50 g medroxyprogesterone, 10 mg/day for 12 days; (b) conjugated equine estrogens, 0.625 mg/day MPA 10 mg/day for 12 days; (c) no treatment. Total cholesterol and LDL cholesterol signicantly decreased after 6 months in both treated groups in comparison with untreated women, but HDL cholesterol and triglycerides showed minimal changes. Comparison was made between TD estradiol (0.05 mg/day) and oral norethisterone acetate (2.5 mg/day) administered for 12 days every 2 or 3 months to patients whose menopause had begun at least 4 years earlier (171). Study duration was two long cycles in each group within 710 months. Efcacy (group E/NA = 94/92%), and systemic tolerability (95/97%) were good and continuation of spaced-out treatment accepted by 88%(E) and 87%(NA). Major skin reactions occurred in 7%(E) and 4%(NA). Progestin-associated withdrawal bleedings occurred in 61% of patients; mean duration 4.3 1.9/4.8 1.6 days, and breakthrough bleeding requiring sonographic or histological work-up in 8%(E) and 13%(NA). Effects of a HRT regimen of continuous estrogen and interrupted progestogen, administered transdermally, on the endometria of 15 healthy postmenopausal women, and the pattern of bleeding and relief of menopausal symptoms were investigated in a volunteer pilot study of up to 6-months duration involving weekly application of an estrogen-only TDS releasing 50 g estradiol per day interspersed with a combined estrogen and progestogen TDS releasing 50 g estradiol and 250 g norethisterone acetate per day for 3 days (172). Transvaginal ultrasound measurements of endometrial thickness and endometrial biopsies were performed in month 3 at the end of both the estrogen-only phase of treatment and combined estrogenprogestogen phase. Treatment provided relief of hot ashes and, by month 6, 71% women who completed treatment had no vaginal bleeding. No endometrial hyperplasia or atypical changes were observed in biopsies, and ultrasound measurements demonstrated a thin endometrium. Reduced immunostaining for Ki67 was observed in endometrium from the combined phase of treatment compared with estrogen-only phase, consistent with progestogenic-antagonism of proliferation. Exposure to progestogen did not suppress steroid receptors, as similar immunostaining was observed in both treatment phases. The tolerability, adhesion, and efcacy of the matrix-type estradiol TDS, Oesclim 50, were compared with those of Estraderm TTS 50 (a reservoir-type system)

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

in a multicenter clinical trial (173). The patches were applied twice weekly for 24 days of each 28-day cycle, over 4 cycles. Oral progestogen was taken by nonhysterectomized patients for the last 12 days of estrogen therapy in each cycle. In the Oesclim 50 group, 4.2% of applications caused a local skin reaction, compared with 9.5% in the Estraderm TTS 50 group. Both were well tolerated, although 7 patients in the Oesclim 50 group, and 12 in the Estraderm TTS 50 group, discontinued owing to adverse events. There was no signicant difference between the percentages of patients with signs of hyperestrogenism (20.3% in Oesclim and 20.0% in Estraderm TTS 50 group). Adhesion was signicantly better for Oesclim 50 (6.0% detached) than for Estraderm TTS 50 (11.3% detached). Greater adhesion of Oesclim 50 was particularly apparent, with three times fewer Oesclim 50 systems becoming detached during a shower or bath. Each treatment produced signicant and comparable improvements in vasomotor symptoms, other menopausal symptoms, and gynecological assessments. A near-maximal effect on vasomotor symptoms was observed after approximately 1 month of treatment, and this was maintained for the entire treatment period. The local skin tolerability of Oesclim 50 was also compared with that of Estraderm TTS 50 (174). In the rst study, the modied DraizeShelanskiJordan method of sensitization was used to compare cutaneous tolerability of repeated applications of patches in 24 healthy postmenopausal women. This indicated no sensitizing potential or induction of allergic reactions. The second study was a multicenter clinical trial involving 283 healthy menopausal women over 4 months. In this study, 4.2% of applications in the Oesclim group provoked reactions, compared with 9.5% in the Estraderm group, and 25.9% treated with Oesclim and 39.9% receiving Estraderm experienced one or more reactions. Redness and itching were the most frequent reactions in both groups. Durations of reactions were signicantly shorter in the Oesclim group, with higher percentage of durations of less than 1 h and lower percentage of durations less than 48 h. No reactions in the Oesclim group led to premature removal, compared with 11 in the Estraderm group. One patient in the Oesclim group discontinued treatment because of an application site reaction, but seven in the Estraderm group discontinued. A comparison between efcacy and safety of two sizes of Lyrelle (matrix type) and Estraderm TTS 50 (reservoir type) TDS was made in 394 hysterectomized postmenopausal women in a multicenter trial (175). A signicant decrease in mean number of hot ashes per day was observed in all groups from the end of cycle 1, reaching 90% at the end of cycle 7. There was no signicant difference between Lyrelle 50 and Estraderm at any time point for any parameter, although betweengroup differences for Lyrelle 80 and Estraderm occurred in cycles 13 in favor of Lyrelle 80. A similar effect on blood lipid levels was observed in all groups. The efcacy, bleeding patterns, and safety of continuous TD and sequential TD progestogen therapy were compared with those of oral progestogen therapy in postmenopausal women receiving TD estrogen (176). In a 1-year (13 treatment periods, 28 days each), study, 774 postmenopausal women received 50 g/day of continuous TD estradiol with either continuous or sequential TD norethisterone acetate (NETA) in daily doses of 170 or 350 g in a single TDS or sequential oral progestogen (1 mg norethisterone [NET] or 20 mg dydrogesterone per day). The average number of hot ashes per day decreased from prestudy by over 90%, and this reduction was unaffected by different progestogen regimes. With sequential progestogen, the bleed-

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

ing incidence and number of bleeding days did not change over the course of the study, but were lower in the low-dose TD progestogen group. With continuous progestogen, the incidence of bleeding decreased in both low- and high-dose groups, from 35 and 45% in treatment period 1, to 25 and 15%, respectively, at the end of treatment. Adverse event incidence was similar in both groups, with 2336% reporting events possibly or probably related to HRT (excluding vaginal bleeding). Lipoprotein- was reduced in all but the oral progestogen group. It was concluded that continuous and sequential TD estrogenprogestogen treatments with estradiol NETA are effective and safe alternatives to continuous TD estrogen and oral sequential progestogen for treatment of menopausal symptoms. Continuous TD therapy with estradiolNETA may be more acceptable for most patients (i.e., those who wish to avoid monthly bleeds), whereas the sequential regimen may be preferable when monthly bleeding may be appropriate. The safety and efcacy of TD estrogen replacement therapy in liver-transplanted menopausal women was investigated (177). Thirty-two menopausal women who had undergone liver transplantation at least 6 months earlier, received TD estradiol replacement therapy in combination with progestin (Estracomb Ciba, 50 g/ 24 h, 250 g/24 h) if the uterus was intact, or estradiol alone (Estraderm Ciba, 50 g). Liver function and hemostatic parameters were measured at 0, 3, and 6 months and gynecological transvaginal ultrasound (TVS) performed at 0 and 6 months. Efcacy of hormonal treatment was assessed from serum concentrations of estradiol, estrone, FSH, LH, and SHBG, by measuring endometrial thickness with TVS and recording changes in subjective climacteric symptoms at 0 and 6 months. Safety was assessed by measuring liver enzyme activity, liver synthesis functions, and coagulation factors. Therapy did not impair any liver parameters measured, no thrombotic effect was detected, and hormonal effects of the regimen were veriable biochemically, clinically, and by TVS. Effects of continuous TD estradiol, with or without sequential oral MPA, on serum lipids and lipoproteins in menopausal women were investigated in 62 healthy menopausal women (178). Group A included 38 hysterectomized women treated with continuous TD estradiol only (50 g daily). Group B included 24 menopausal women, with an intact uterus, treated with TD estradiol (50 g daily) and MPA (10 mg daily for rst 12 days of each calendar month). Serum lipids and lipoproteins were reviewed after 6 months. In group A there was a small reduction in total cholesterol (5.5%) and slight lowering in LDL-cholesterol (5.7%). In group B, there were no signicant changes in total cholesterol and LDL-cholesterol. HDLcholesterol levels did not change signicantly with unopposed TD estradiol or additional sequential MPA. Serum triglyceride concentrations decreased signicantly in both groups (13.9 and 13.4%, respectively). Serum lipid changes did not differ between groups. A multicenter trial (179) compared incidence of amenorrhea in 54 postmenopausal women (mean age, 54.9 0.6 years) who underwent six 4-week cycles of continuous HRT combining progestinnomegestrol acetate 2.5 mg/day with one of three estrogens: percutaneous 17-estradiol gel (1.5 mg/day, group G), TD 17-estradiol patch (50 g/day, group P), or oral estradiol valerate (2 mg/day, group O). The rate of amenorrhea varied signicantly according to type of estrogen preparation (calculated cycle-by-cycle, rates were 6783% [group G], 2556% [group P], and 5361% [group O]). Overall rates of persistent amenorrhea were not different between groups for cycles 1 through 3, but for cycles 4 through 6, signicantly more

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

women in groups G and O (67 and 46%, respectively) experienced amenorrhea than did those in group P (12%). Amenorrhea rates for the entire six-cycle period were 78% for group G, 48% for group P, and 60% for group O, although these differences were not statistically signicant. Differences in rates could not be attributed to endometrial atrophy, because endometrial thickness did not differ signicantly among groups. Calculated as a function of the number of women included in the trial, the percentage of amenorrheic women was highest with group G, although ndings were similar for group O. Two 11-week, placebo-controlled studies (180) compared the Climara 7-day matrix patch (at two dose levels) with 625 g/day oral conjugated equine estrogen, found that both the 50- and 100-g/day estradiol patches had a positive effect on climacteric symptoms. Tolerance was good and similar for both. Studies of skin irritation and adhesion revealed that the 7-day patch was well tolerated and that, although irritation was similar to that associated with Estraderm, adhesion was superior. Absorption of estradiol was higher and more consistent from buttock than abdomen, suggesting that choice of application site may require further investigation. Efcacy, safety, and tolerability of an estradiol gel (1.0 mg of estradiol daily; Divigel/Sandrena) in HRT of postmenopausal women were compared with those of an estradiol TDS (50 g/24 h estradiol, Estraderm TTS) over 12 months with 120 postmenopausal women (181). Dydrogesterone tablets (Terolut), 10-mg daily for the rst 12 days of every month, were used as the progestogen component of therapy. Twenty-ve women without HRT served as reference group for BMD measurements. Both treatment regimens were equally effective in alleviating climacteric symptoms, preserving BMD, and were equally safe. A trend toward heavier bleeding was detected in patients treated with the estradiol TD. A nonsignicant decrease of total cholesterol and triglyceride, but no change in high-density lipoprotein cholesterol was observed in both groups. Acceptability of treatment was higher in the gel (96.4%) than patch group (90.7%). Only two (3.3%) women using the gel complained of skin irritation, whereas 28 patients (46.7%) using the patch reported this effect. Two doses of TD estradiol gel (Divigel/Sandrena) plus oral sequential MPA were compared with oral estradiol valerate plus oral sequential MPA (Divina/Dilena) in postmenopausal women with climacteric complaints or already using HRT in a 2-year comparative study (182). Groups received either (a) 1 g of gel containing 1 mg estradiol for 3 months plus 20 mg of oral MPA during last 14 days; (b) 2 g of gel containing 2 mg of estradiol for 21 days plus 10 mg oral MPA during the last 14 days; (c) 2 mg of estradiol valerate tablets for 3 weeks plus 10 mg of oral MPA during the last 10 days. With each preparation, climacteric complaints were signicantly reduced, good bleeding control was obtained, BMD was maintained, and bone turnover was reduced. Lipid parameters showed no unfavorable changes. Continuation rates were similar in all groups, with 74% of patients completing the rst year, and 94% of patients who elected to continue completing the second year. Tolerability of gel was good, with only 1.7% of patients discontinuing because of skin irritation. Estradiol and estrone concentrations and bioavailability were compared after a single dose and at a steady state during oral estradiol valerate, TD estradiol gel, and TD estradiol TDS treatments (183). In study A, 12 healthy postmenopausal women received 1.5 mg of estradiol as a TD gel or a 2-mg estradiol valerate tablet daily for 14 days. In study B, 15 postmenopausal women were treated for 18 days with 1.5-

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

mg estradiol gel or a TD system releasing estradiol 50 g/24 h (replaced every 72 h). Tablet and transdermal gel yielded similar serum estradiol proles with a peak concentration 45 h after administration. The TDS gave relatively stable estradiol levels during the mid-third of the wearing time, whereas much lower levels were observed at the beginning and end. There was no difference in uctuation of peakand-trough estradiol levels between gel (56 or 67%) and tablet (54%), but uctuation was greater with the TDS (89%). Bioavailability of estradiol from the gel was 61%, compared with the tablet, and 109%, compared with the TDS. Gel was not bioequivalent with tablet or TD and individual dose adjustments may be needed when changing administration forms. The effect of HRT on the conjunctiva in postmenopausal women was the subject of a clinical study in 11 postmenopausal women receiving TD estradiol or TD estradiol plus MPA for 4 months (184). Signicant increases in serum estradiol levels, vaginal maturation value, and mild cytological maturation changes in conjunctival epithelium were observed. The changes were signicant and the data support the view that HRT induces cytological maturation changes in conjunctival epithelium in postmenopausal women. A case of sensitization to estrogen was reported in a patient receiving Estraderm (185). A 40-year-old woman suffered from cyclic skin disorders and at each menses developed pruritus and erythematous papulovesicular lesions over the members and trunk. Prick and patch tests with alcoholic solutions of estrone alone and serum tests for antiethinylestradiol antibodies and antiprogesterone antibodies were positive. In cases of progesterone sensitization, treatment of choice is estrogen inhibition of ovulation, whereas for estrogen sensitization, antiestrogen treatment appears more effective. Bilateral ovariectomy may be required in difcult cases. The history, current clinical practice, choice of methods, and number of prescriptions and sales of HRT in the United States were reviewed (186). The percentage of women currently utilizing HRT was greater in women aged 4060 (35%), but fell with ages older than 65 (15%), and declined further in women older than 80 (7%). News media and physicians were the largest source of information on HRT, and obstetricians and gynecologists were predominant prescribers. Women who spontaneously developed menopause early and younger women undergoing castration were more likely to take TD estrogen, but 86% of U.S. prescriptions were for oral estrogens (with conjugated equine estrogens [70%] the market leaders). Of the prescriptions for women with a uterus 50% were for combined continuous estrogen and progestogen, whereas 42% contained cyclic estrogen and progestogen. Although use of HRT by postmenopausal women in the United States increased, the percentage of current users remained lower than anticipated, in spite of widespread media and educational efforts of benets. TD estrogens were used more commonly in women in the early postmenopausal period, whereas in women with a uterus, most U.S. physicians prescribed combined estrogen plus progestogen, but used oral, rather than TD, estrogen. Bioavailability, pharmacokinetics, and tolerability of two matrix transdermal delivery systems providing 50 g/24 h of estradiol were compared in 20 healthy postmenopausal women (187). Menorest (34 days suggested use) and Climara (7 days suggested use) were compared at steady-state in two 14-day treatment periods separated by a 4-week washout, with plasma estradiol monitored in the second week of each treatment. No differences between treatments relative to AUC, C(max), C(min),

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

C(average), or uctuations in plasma estradiol. T(max) was signicantly shorter for Menorest than for Climara, and C(max) and C(min) were signicantly higher for the second Menorest patch than for the rst. Three cases of erythema with Menorest and a total of 21 skin reactions in 15 subjects with Climara were reported. Systemic tolerability was similar between treatments with 8 estrogen-related adverse events in 8 subjects with Menorest and 13 events in 10 subjects with Climara. It was concluded that although the bioavailability of estradiol from these TDS was similar, the products were not bioequivalent because T(max) was signicantly shorter for Menorest than for Climara. A comparison of continuous combined TD delivery of estradiolnorethindrone acetate and estradiol alone for menopause was carried out to determine whether a continuous estradiolnorethindrone acetate TD delivery system reduced incidence of endometrial hyperplasia in postmenopausal women more than that of TD estradiol alone (188). A total of 625 postmenopausal women were assigned to one of four treatments: TD estradiol 50 g/day, or TD estradiolnorethindrone acetate, with 50g estradiol and 140, 250, or 400 g/day of norethindrone acetate. Endometrial hyperplasia was found in 37.9% in the estradiol-alone group versus 0.8%, 1%, and 1.1% in the estradiolnorethindrone acetate 50140, 50250, and 50400 groups, respectively. Uterine bleeding was less frequent in the estradiolnorethindrone acetate 50140 group. The estradiolnorethindrone acetate combination TDS showed skin tolerance comparable with that of estradiol alone. Bioavailability of two 100-g daily 17-estradiol TDS (once-a-week matrix patch and twice-a-week reservoir patch) was compared in healthy postmenopausal women (189) in a two-period, crossover study with two 8-day treatment periods separated by a minimum 7-day washout. Subjects were assigned to either (a) matrix patch applied to abdomen and worn for 7 consecutive days, or (b) reservoir patch applied to abdomen and worn for 4 days, followed immediately by a second reservoir patch worn for 3 days. Three-hours after patch application serum estradiol levels were signicantly higher than levels at time of patch application. After 12 h, mean serum estradiol level in women with matrix patches was 98.20 44.97 pg/mL, signicantly higher than in women with the reservoir patch (62.20 16.21 pg/mL). An AUC (0168 h) with the matrix patch was also higher than for reservoir patch. Left ventricular heart function and its response to long-term estrogen replacement therapy was assessed in 30 postmenopausal women, 20 of whom had modestto-severe hot ashes and 10 of whom had never had them (190). Continuous TD estradiol was given to women with surgically induced menopause, and a combination of TD estradiol and sequential MPA to those with spontaneous menopause. Although HRT signicantly improved heart function in healthy postmenopausal women, there appeared to be some minor differences in response between those with ashes and nonashers. Effects of TD estradiol on serum triglycerides in menopausal women with preexisting mild-to-moderate hypertriglyceridemia were evaluated (191). Forty-four women (posthysterectomy and maintained on 50-g unopposed estradiol for 6 months) were divided into those with normal baseline triglyceride concentrations (0.42 mmol/L) and those with raised baseline readings (>24 mmol/L). Signicant reductions in serum triglyceride concentrations occurred in both groups (9.6 and 17%, respectively). TD estradiol therapy may be a useful treatment option in menopausal women with preexisting hypertirglyceridemia.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

A summary of the tolerability and safety of Oesclim, which was developed with the objective of providing improved local skin tolerability and adhesion, while minimizing hyperestrogenic effects, has been published (192). In a comparative clinical trial, Oesclim resulted in fewer than half as many application site reactions as Estraderm TTS (4.3 vs. 9.5%), and the duration of reactions was signicantly lower in the Oesclim group. Oesclim was well tolerated in all clinical trials and reported to have a estrogen-specic tolerability comparable to Estraderm TTS. Low-dose Oesclim (25 g/day) was associated with reduction in hyperestrogenic side effects compared with higher doses. In a study of long-term Oesclim therapy, 79% of patients wished to continue therapy after 1 year, and in a follow-up study, 79.8% wished to continue at the end of 3 years. Efcacy and safety of three dosages of Oesclim, delivering 0.025, 0.050, or 0.100 mg 17-estradiol per 24 h, in treatment of moderate to severe vasomotor symptoms was evaluated in a multicenter trial (193). A total of 196 highly symptomatic menopausal women received 12 weeks of continuous unopposed treatment with one of the three dosages of Oesclim or a matching placebo patch. Reduction in frequency of moderate-to-severe vasomotor symptoms was statistically signicant compared with placebo from week 2 onward in the Oesclim 50 and 100 groups, and from week 3 onward in the Oesclim 25 group. Symptom severity was also reduced. Estrogen-related adverse events were less frequent in the Oesclim 25 group. Signicant differences in estradiol bioavailability were reported from two similarly labeled estradiol matrix TDS (Alora and Evorel) (194). The uctuation index produced by Evorel was signicantly higher than that with Alora (135 vs. 76%) and the estradiol baseline-corrected AUC was signicantly lower for Evorel than Alora (1870.6 vs. 2871.8 pg h1 mL1). Efcacy, safety and compliance with Climara 50 and Climara 100 were evaluated in 100 women (195). Reductions in weekly frequency of hot ashes were 58.6% in the Climara 50 group and 72.1% in the Climara 100 group. Of 64 patients with sleep disturbances, 51 reported some improvement, 10 had no advantage, and 3 were worse. Incidence of temporary minor side effects was dose-related (38 vs. 70%). A multicenter study assessed the efcacy, safety, and tolerability of a low-dose (0.0375 mg/day) estradiol matrix TDS for the treatment of moderate-to-severe postmenopausal hot ashes in healthy women (196). Estradiol matrix or matching placebo patches were administered over three 4-weektreatment cycles to 257 patients (130 estradiol, 127 placebo). Assessments of treatment effectiveness signicantly favored the estradiol patch over placebo. This lowest available dose estradiol TD provided signicant relief from moderate-to-severe postmenopausal hot ashes and was well tolerated. A 3-year study enrolled 277 early postmenopausal women to examine the efcacy of a matrix 17-estradiol TDS, at three dosages (25, 50, and 75 g/day) combined with sequential oral dydrogesterone 20 mg/day, in preventing bone loss (197). At 2 years, difference from placebo in percentage change from baseline of L1-4 lumbar spine BMD was 4.7 0.7% (25 g/day), 7.3 0.7% (50 g/day), and 8.7 0.7% (75 g/day). There were also signicant increases in femoral neck, trochanter, and total hip BMD with all doses of estradiol, compared with placebo. Most patients receiving estradiol also had a signicant gain (>2.08%) in lumbar spine bone mass and clinically signicant and dose-related decreases in total serum osteocalcin, serum bone alkaline phosphatase, and urinary C-telopeptide, with all three markers of bone turnover returning to premenopausal levels.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

The effects of four doses (0.025, 0.05, 0.06, and 0.1 mg/day) of a 7-day TD 17-estradiol delivery system on bone loss in postmenopausal women were evaluated in a multicenter study (198). At 24 months, doses of 0.025, 0.05, 0.06, and 0.1 mg/ day resulted in mean increases in BMD of the lumbar spine of 2.37, 4.09, 3.28, and 4.70%, respectively, and increased BMD of the total hip by 0.26, 2.85, 3.05, and 2.03%, respectively. All increases were signicantly greater than placebo. Consistent and signicant improvements in biochemical markers of bone turnover were also noted in all treatment groups. Two estradiol TDS that released 25 or 37.5 g/day were compared with a placebo patch on 156 patients in natural or surgical menopause suffering from at least ve hot ashes per day, treated continuously for 12 weeks, without progestin opposition (199). Responders (patients with fewer than three hot ashes per day at end of treatment), were 82 and 90% with 25 or 37.5 g/day, respectively, both signicantly more than placebo (44%). Efcacy and tolerability of a matrix patch delivering estradiol at doses of 0.05 and 0.10 mg/day (Estraderm MX 50, 100) in treatment of moderate to severe postmenopausal symptoms was compared (200). A total of 254 postmenopausal women received 0.10, 0.05 mg, or placebo for 12 weeks continuously. TDS were applied twice weekly to the buttocks with each patient wearing two patches simultaneously. Patches containing 0.10 and 0.05 mg estradiol were superior to placebo in reducing hot ashes per 24 h after 4, 8, and 12 weeks of treatment. For all other efcacy parameters studied, both dosage strengths were superior to placebo at all time points. It was concluded that this matrix patch offered an effective and well-tolerated dosage form and may be particularly suitable for women who experience local sensitivity to alcohol-containing systems. The effect of the administration route and cigarette smoking on plasma estrogen levels during HRT was evaluated in 14 healthy postmenopausal women (6 smokers and 8 nonsmokers) (201). All patients randomly received cyclic therapy with estradiol and norethisterone orally or TD, each for 6 months. Plasma levels of estrone, estradiol, and estrone sulfate, all were 4070% lower in smokers than nonsmokers when HRT was given orally. Oral dosing caused higher extradiol/estradiol sulfate and estrone/estradiol sulfate ratios compared with TD therapy in smokers (40.2 vs. 7.0; and 3.2 vs. 0.8, respectively). Pharmacokinetics of Fem7 (an estradiol matrix-type TDS, applied once weekly) was investigated in 36 healthy postmenopausal women at doses of 25, 50, 75, and 100 g/24 h (202). Maximum plasma estradiol and estrone concentrations occurred 1420 h after patch application, remained within the therapeutic range until removal, and returning to baseline within 12 h. Plasma estradiol concentrations increased in a dose-dependent manner for all dose levels, and plasma estrone increased for the three highest doses. Treatment was well tolerated at all dose levels and no severe adverse reactions were reported. The efcacy of two strengths of TD estradiol matrix with daily oral doses of conjugated equine estrogens in reducing the frequency of moderate-to-severe hot ashes in postmenopausal women was evaluated (203). An estradiol TDS (Alora 0.05 or 0.1 mg/day) administered twice weekly or oral doses of conjugated equine estrogens (CEE 0.625 or 1.25 mg) administered daily were given to 321 highly symptomatic postmenopausal women for 12 weeks. Results indicated no signicant differences at any time point in mean frequency or mean percentage reduction in

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

frequency of moderate-to-severe hot ashes between patients given Alora 0.1 mg/ day or CEE 1.25 mg/day, or between the Alora 0.05 mg/day and CEE 0.625 mg/day groups by week 12. There were no serious or unexpected adverse events with the TDS, and local skin tolerability was excellent. Other estrogenic effects were comparable between TD and oral administration groups, except for lower incidence of bleeding in women receiving the lower transdermal dose. The effect of estradiol TD in postmenopausal women with conrmed pollakiuria and urinary incontinence was investigated in ten women using Estraderm TTS 2 mg for 8 weeks (204). In seven cases, severity of urinary incontinence was very effective in three cases, improved in two, slightly improved in one, and no change in one. Postmenopausal women (especially those older than 60 years) prefer HRT that avoids cyclical uterine bleeding and continuous combined HRT regimens were primarily introduced to avoid bleeding and increase compliance. In a multicenter study, 136 women at least 2 years postmenopausal, with mild-to-moderate menopausal symptoms, received either Estragest TTS 0.125/25 (delivering 0.125 mg norethisterone acetate [NETA] and 25 g estradiol per day) or placebo for 6 months (205). After 4, 12, and 24 weeks the Kupperman index was signicantly lower in the Estragest group, and the severity of vaginal dryness and dyspareunia at 12 and 24 weeks was also reduced. The proportion of supercial cells increased signicantly in the Estragest, but not the placebo group. The percentage of patients reporting amenorrhea with Estragest ranged from 80 (month 2) to 87% (month 6). In a second multicenter study lasting 1 year, 441 postmenopausal women received one of three continuous combined HRT regimens: group A, Estragest TTS 0.125/25; group B, TDS delivering estradiol 50 g and NETA 0.25 mg/day, group C, oral tablets containing 2 mg estradiol and NETA 1 mg/day. During treatment cycles 46, amenorrhea was achieved in 73% (group A), 47% (group B), and 66% (group C). During treatment cycles 1012, proportions increased to 86% (A), 65% (B), and 79% (C). Bleeding patterns in groups A and C were not signicantly different, but superior to those in group B. It was concluded that Estragest TTS 0.125/25 was effective in treatment of mild-to-moderate menopausal symptoms and urogenital complaints, induced a high rate of amenorrhea and provided good endometrial protection. Effects of three commonly prescribed estrogen replacement therapies (oral conjugated equine estrogens [CEE; n = 37], oral micronized estradiol [ME; n = 25], and TD estradiol [TE; n = 24]) on the concentrations of blood sex hormone-binding globulin (SHBG), estradiol, and estrone were studied (206). Increases in SHBG concentrations were 100, 45, and 12% for subjects receiving CEE, ME, and TE regimens, respectively. Decreases in the percentage estradiol not bound to protein and increases in the percentage of estradiol bound to SHBG correlated with therapymediated changes in concentrations of this protein. Systemic bioavailability and plasma proles of 17-estradiol after application of three matrix patches: Menorest, Tradelia, and Estraderm MX, claiming to deliver 50 g/day were evaluated (207). All patches were each worn randomly by 21 postmenopausal women volunteers over 96 h, separated by an at least a 7-day washout period. Tmax (32 h) was the only pharmacokinetic parameter identical for all patches. Menorest produced the highest estradiol bioavailability, judged by the AUC(096 h) = 3967.8 1651.8 pg/mL h1, C(average) = 41.3 21.3 pg/mL, C(min) = 36.8 8.6 pg/ mL. Tradelia (AUC(096 h) = 3737.9 1637.6 pg/mL h1, C(average) = 38.9 17.0 pg/

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

mL, and C(min) = 33.8 26.7 pg/mL) was not signicantly less than Menorest. Estraderm MX showed the lowest estradiol proles (AUC(096 h) = 3192.1 1646.0 pg/mL h1, C(max) = 38.9 25.1 pg/mL, C(average) = 33.2 17.1 pg/mL). Menorest showed the smallest uctuation over the entire test period, similar to Estraderm MX, whereas Tradelia showed the highest uctuation and the highest C(max) = 48.0 20.3 pg/mL. When estradiol baseline levels, before patch application, were individually subtracted from the subsequent estradiol level, Estraderm MX was not bioequivalent to Menorest. A circadian curve pattern of estradiol plasma level was observed for all patches, and in the evening, higher plasma levels were always detected. Individual comparison of AUC(096 h) for each patch showed large interindividual variability (20008000 pg/mL h1) for all patches, but relatively small individual variability. Women with high estradiol bioavailability (high-responders) maintained high bioavailability with all patches, and women identied as low- and medium-responders remained the same, regardless of the applied patch. Side effects were approximately equal in all patches, with a maximum after 72 h. Two long-term multicenter studies compared the efcacy on climacteric symptoms of a new active matrix estradiol TDS (CAS 50-28-2) with a reference reservoir patch (both releasing 50 g/day) (208). One group received the matrix patch and the other the reservoir patch in 4-week cycles, with twice-weekly application of patches for 3 weeks, followed by 1-week washout. Progestin opposition was with MPA; 5 mg/day orally in the last 11 days of patch application in a German study and with 10 mg/day in the last 12 days of patch application in an Italian study. Each study was divided into two parts: (a) with three 4-week cycles and (b) for ten 4week cycles. In the German study both patches quickly relieved climacteric symptoms during the rst 3 weeks of application, as shown by rapid decrease of the Kupperman Index. At the end of part 1, 91% (matrix) and 96% (reservoir) group reported relief from climacteric symptoms; at the end of part 2, these were 98% and 95%, respectively. Both patches were systemically fairly well tolerated and only 4.5 (matrix) and 3.9% (reservoir) discontinued owing to adverse reactions. Relative to local skin reactions, the matrix patch was signicantly better tolerated and adhesion was better. In the Italian study both patches relieved climacteric symptoms during the rst 3 weeks of application. At the end of part 1 both patches relieved 95% of patients and at the end of part 2, 100% of patients were relieved. Patches were systemically equally fairly well tolerated with premature discontinuations for systemic adverse drug reactions in 5.0% (matrix) and 3.9% (reservoir) groups. As in the German study, matrix patches were signicantly better tolerated. The effect of short-term HRT with estradiol and norethisterone on the pharmacokinetics of phenazone was investigated in ten women at least 6 weeks after ovariectomy (209). Each patient received TD estradiol (4 mg, every 3 days) for a period of 18 days, and subsequently oral norethisterone (5 mg/day) for 10 days, with an interval of 1 day. Pharmacokinetic studies were performed before estradiol administration, on the last day of estradiol treatment and on the last day of norethisterone administration. Short-term administration of estradiol did not modify the pharmacokinetics of phenazone. Percutaneous absorption of progesterone in postmenopausal women treated by application of progesterone cream to the skin was evaluated in six postmenopausal women over a 4-week period (210). Transdermal estradiol, 0.05 mg, was applied 2 days before rst application of progesterone (30 mg/d) and continued throughout the

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

study, with patches changed twice weekly. Progesterone cream was applied once a day for 2 weeks. On days 1529 progesterone cream was applied twice daily (60 mg/d). Serum 17-estradiol and progesterone were measured over 24 h on day 1 and at weekly intervals for the study duration. Individual serum 17-estradiol concentrations ranged from 40 to 64 pg/mL, but intraindividual concentrations remained constant. Serum progesterone concentrations were 1.63.3 ng/mL. After 2 weeks of percutaneous dosing, progesterone concentrations were sustained for at least 8 h and were consistent within an individual. An increase in progesterone concentration occurred after 4 weeks, compared with 2 weeks. Individually, a signicant correlation was seen between absorption of 17-estradiol and progesterone. Serum and urinary hormone levels following short- and long-term administration of two regimens of progesterone cream in postmenopausal women were evaluated (211) in a multiple-dose study using 24 healthy postmenopausal women. Subjects were allocated to progesterone cream, 40 mg daily, or 20 mg twice daily, for 42 days. Serum progesterone was measured on days 1 and 42 before the morning dose, and at 2, 4, 6, 12, and 24 h after the morning dose. Serum FSH, estradiol, testosterone, and urinary pregnanediol-3-glucuronide were also measured on days 1 and 42. The mean progesterone concentration rose at each sampling time between days 1 and 42 and there was evidence of a rise in pregnanediol-3-glucuronide over the study course. There were no changes in FSH, estradiol, or testosterone and no differences were detected between the regimens. The use of TD progesterone cream for vasomotor symptoms and postmenopausal bone loss was also investigated (212). One hundred two healthy women, within 5 years of menopause, were assigned to TD progesterone cream or placebo. Subjects were instructed to apply 0.25 teaspoon of cream (containing 20 mg progesterone or placebo) to the skin daily. Subjects received daily multivitamins and 1200 mg of calcium; symptoms were reviewed ever 4 months. In the treatment group, 69% and 55% in the placebo group initially complained of vasomotor symptoms. Improvement or resolution of these symptoms was noted in 83% of treatment subjects and 19% of placebo subjects. However, the number of women showing a BMD gain of more than 1.2% did not differ signicantly. In addition to the widespread use of HRT TD, steroid delivery has also been investigated as a means of contraception. A once-a-week (monophasic) contraceptive TDS was designed to simultaneously deliver a low-dose combination of levonorgestrel (LNG) and 17-estradiol (E2) for fertility regulation in females (213). In vitro permeation studies using human cadaver skin indicated 6.0 0.9 g/day cm2 of LNG and 2.9 0.5 g/day cm2 of E2 could be delivered. A 7-day dermal toxicity study on six rabbits indicated minimal potential to cause skin irritation, and histopathological examination revealed only mild-to-moderate inammation. A phase 1 bioavailabilitydose proportionality clinical study, consisting of pretreatment, treatment, and posttreatment cycles, was conducted on fertile Chinese women. During the pretreatment cycle, 48 subjects were given placebo patches to study wearability (including skin irritation and adhesion tests). During the treatment cycle, each subject in the test groups received weekly application of 1 (A), 2 (B), or 3 (C) 10-cm2 patches, and group D received daily 150 g of LNG and 35 g of ethynyl estradiol, orally. The wearability study indicated patches were very well accepted. Residual assay of used TDS indicated delivery of LNG and E2 at rates of approximately 5.0 g/cm2 day1 and 4.0 g/cm2, day1, respectively, during the treatment cycle. Ra-

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

dioimmunoassay (RIA) of serum samples demonstrated therapeutically effective serum concentration of LNG and serum proles of progesterone, luteinizing hormone (LH), and follicle-stimulating hormone (FHS) also indicated that ovulation inhibition occurred in most of the subjects wearing TDS. No subject became pregnant, and posttreatment hormonal proles indicated that most returned to their normal menstrual cycle on termination of patch use. A second evaluation of TD delivery of steroids for contraception has been reported (214). A TDS changed weekly and delivering both E(2) and LNG at daily dosages of 3.8 0.8 and 2.9 0.7 g/cm2 day1, respectively, showed ovulation suppression. An alternative progestin (ST 1435) penetrated skin when formulated in acetylated lanolin or an hydroalcoholic gel and produced ovulation suppression at a dose of 2 mg/day in a small number of cycles. For contraceptive purposes TD systems must be perfectly adhesive, well tolerated locally, and achieve nearly 100% efcacy. These targets are very challenging, although the potential advantages are so high that the concept deserves further development. A pilot clinical study was carried out to evaluate the cognitive and neuroendocrine response to estrogen administration for postmenopausal women with Alzheimers disease (AD) (215). Twelve women with probable AD of mild-to-moderate severity completed the study. During an 8-week treatment period, 6 received 0.05 mg/day dosage of 17-estradiol by a patch and the remainder a placebo patch. Signicant effects of estrogen treatment were observed on attention and verbal memory. In women treated with estrogen, verbal memory enhancement was positively correlated with estradiol plasma levels and negatively correlated with plasma concentrations of insulin-like growth factor-binding protein-3 (IGFBP-3). This suggested that estrogen replacement may enhance cognition for postmenopausal women with AD. A trial assessed the effectiveness and tolerability of transdermal estrogen in men with hot ashes after hormonal therapy for prostate cancer (216). Twelve men with moderate to severe hot ashes received either low-dose (0.05 mg) or high-dose (0.10 mg) estrogen patches applied twice weekly for 4 weeks. After a 4-week washout, each patient received the alternative dose for 4 weeks. There was a signicant reduction in overall severity of hot ashes seen with both low- and high-dose estrogen patches, but a signicant reduction in daily frequency of hot ashes was seen only at the high dose. Eighty-three percent reported either mild, moderate, or major improvement in symptoms with either low- or high-dose patch. Mild, painless breast swelling or nipple tenderness was noted in 17 and 42% of the men treated with lowand high-dose estrogen, respectively. FSH levels decreased signicantly at both doses. Estradiol levels increased from 12.1 to 16.4 (low-dose) and 26.9 (high-dose) pg/mL, but there was no signicant change in serum testosterone or LH levels. D. Androgens

An important aim in treating male hypogonadism is restoration of physiological concentrations of testosterone and metabolites. New methods for testosterone delivery that have provided increased options for men requiring hormonal replacement therapy have been reviewed (217). Intramuscular administration of testosterone is associated with early, high serum levels followed by a gradual decline over the dosing interval. The TDS now available as alternatives include Testoderm (applied to the scrotum) and Androderm (applied to nonscrotal skin). Most patients achieved

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

normal serum testosterone levels with circadian variation and normal estradiol levels. Serum LH levels generally decreased, but not to suppressed levels, and Testoderm use leads to an increase in plasma dihydrotestosterone (DHT). Clinical response in mood, energy level, and sexual function were improved with both systems and were generally comparable with intramuscular injection. There were no clinically signicant changes in laboratory parameters, including prostatic specic antigen (PSA), and prostate size did not increase above normal. Skin reactions were common and may require discontinuation of therapy. Patients with inadequate scrotal size may not achieve satisfactory results with Testoderm. Although patches are more expensive than intramuscular (IM) injections, they require less frequent ofce visits and both transscrotal and transdermal systems offer a good alternative for hypogonadal men who do not desire fertility during the treatment period. Scrotal testosterone patches can produce normal serum levels mimicking diurnal variations. This was followed in hypogonadal men treated transdermally for up to 10 years (218). Eleven men (age 35.9 9.8 years) at start of study were treated with transscrotal patches (Testoderm) because of primary (n = 4) or secondary (n = 7) hypogonadism. Clinical examinations were performed every 3 months during the rst 5 years and every 6 months thereafter. On daily application of one patch, testosterone levels rose from 5.3 1.3 to 16.7 2.6 nmol/L at month 3 and remained in the normal range throughout treatment. Serum DHT rose from 1.3 0.4 to 3.9 1.4 nmol/L and estradiol from 52.3 9.3 to 71.3 9.6 pmol/L and remained stable. Patients reported no local side effects apart from occasional itching. No relevant changes occurred in clinical chemistry and hemoglobin and erythrocyte counts remained normal. Bone density increased slightly from 113.6 5.4 to 129.7 9.3 mg/cm3. In the nine patients who were younger than 50 years prostate volumes showed a small, but insignicant, increase from 16.8 1.5 to 18.8 2.1 mL during therapy. In two older patients, prostate volume remained constant or decreased slightly during therapy. Prostate-specic antigen levels were constantly low in all patients. The possibility of immediate adverse effects of short-term testosterone administration to older men with low bioavailable testosterone, especially on the symptoms of benign prostate hyperplasia, was investigated (219). A 9-week intervention with either intramuscular testosterone enanthate (200 mg every 3 weeks), TD testosterone (two 2.5-mg patches per day), or neither, was followed by a 9-week observation period. Twenty-seven men (age 74 3 years) with no medical conditions known to affect bone turnover and with total testosterone levels less than 350 ng/dL or bioavailable testosterone levels less than 128 ng/dL were included. All men receiving testosterone treatment increased levels above their own baseline, but only six of nine men receiving TD testosterone achieved bioavailable testosterone levels in the normal range for young men. No side effects were reported using intramuscular delivery, but ve of nine men using TD testosterone developed a rash. As part of a phase III multicenter study, pharmacokinetics and metabolism of a permeation-enhanced testosterone transdermal system and the inuence of the application site were investigated in 34 hypogonadal men (2165 years of age) (220). After an 8-week androgen washout period, two patches were applied to the back for 24 h. Serum concentrations of total testosterone (T), bioavailable testosterone (BT), DHT, and estradiol [E(2)] increased from hypogonadal levels into normal physiological ranges and declined to baseline levels within 24 h after system removal. Peak

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

concentrations occurred about 8 h after application for T and BT and at 13 h for DHT and E(2). Estimated half-lives were: T, 1.29 0.71 h; BT, 1.21 0.75 h; DHT, 2.83 0.97 h; and E(2), 3.53 1.93 h. The inuence of application site was evaluated by applying two patches for 24 h to the abdomen, back, chest, shin, thigh, or upper arm. Hormone proles were qualitatively similar at each site, but CSS values were signicantly different. Based on BT levels, rank ordering of sites was back > thigh > upper arm > abdomen > chest > shin. DHT/T and E(2)/T ratios showed negligible site variation. Further investigations of hormone levels, pharmacokinetics, clinical response, and safety of a permeation-enhanced testosterone transdermal system (TD) in the treatment of hypogonadal men were reported (221). This was a multicenter study with four consecutive periods: period I (3 weeks), evaluation of current androgen therapy (primarily testosterone enanthate injections (mean dose 229 mg; mean interval 26 d); period II (8 weeks), androgen washout; period III (34 weeks), singledose pharmacokinetic studies of TD systems; period IV (12 months), efcacy, safety, and steady-state pharmacokinetic evaluation of TD systems (5 mg/day nominal delivery rate of testosterone). Thirty-seven hypogonadal men 2165 years old enrolled, 34 entered periods III and IV and 29 (9 primary, 20 secondary hypogonadism) completed the study. Four patients withdrew because of adverse events. Measurements included morning serum levels of total testosterone (T); bioavailable testosterone (BT), DHT, and E(2) levels; circadian pattern of T proles and 24-h time-averaged T level; LH levels in patients with primary hypogonadism; and reduction of hypogonadal symptoms. Safety assessments included skin tolerability, prostate parameters, lipid prole, and systemic parameters. Twelve months of TD therapy normalized morning serum T levels in 93% of patients, and produced more than 80% normalization of BT, DHT, and E(2) levels. The TD system mimicked the circadian variation in T levels seen in healthy young men and normalized 24-h timeaverage T levels in 86% of patients. LH was suppressed in eight of nine men with primary hypogonadism, and normalized in ve of these. Subjective symptoms of hypogonadism, including decreased libido and fatigue, showed improvement after 24 weeks of treatment in most patients. Most adverse events were local skin reactions. Prostate assessments showed a lower prostate-specic antigen level during TD therapy compared with IM injections (0.66 vs. 1.00 g/L), but prostate size did not differ signicantly between the treatment regimens. Weight loss associated with human immunodeciency virus (HIV) infection is multifactorial in its pathogenesis, but it was speculated that hypogonadism contributed to depletion of lean tissue and muscle dysfunction (222). Effects of testosterone replacement, using Androderm, on lean body mass, body weight, muscle strength, health-related quality of life, and HIV disease markers were evaluated. Testosterone replacement in HIV-infected men with low testosterone levels was considered safe and associated with a 1.35-kg gain in lean body mass, a signicantly greater reduction in fat mass than achieved with placebo treatment, and an increased red cell count and improvement in role limitation owing to emotional problems. It was concluded that further studies were required to assess whether such supplementation can produce clinically meaningful changes in muscle function and disease outcome in HIV-infected men. Markedly decreased serum androgen levels occur in women with acquired immunodeciency syndrome (AIDS) and may be a contributing factor to the wasting

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

syndrome. A pilot study of the effects of androgen replacement therapy was conducted to determine efcacy in terms of change in serum testosterone, safety parameters, and tolerability, and to investigate testosterone effects on weight, body composition, quality of life, and functional indexes (223). Fifty-three ambulatory women with AIDS wasting syndrome, free of new opportunistic infection within 6 weeks of study initiation and with serum levels of free testosterone less than normal reference range, were enrolled. Subjects weighed 92 2% of ideal body weight, and had lost 17 1% of their maximum weight. Subjects received two placebo patches (PP), one active and one placebo patch (AP), or two active patches (AA) applied twice weekly to the abdomen for 12 weeks. Nominal delivery rates were 150 and 300 g/ day, respectively, for AP and AA groups. Serum free testosterone levels increased signicantly from 1.2 0.2 to 5.9 0.8 pg/mL (AP) and from 1.9 0.4 to 12.4 1.6 pg/mL (AA). Testosterone administration was generally well tolerated locally and systemically, with no adverse trends in hirsutism scores, lipid proles, or liver function tests. Improved social functioning and pain score were observed in APversus PP-treated patients. These data suggested that testosterone administration may improve the status of women with AIDS wasting. The literature on androgen replacement for erectile dysfunction was evaluated by metanalysis (224). Study inclusion criteria were testosterone given as the only therapy for erectile dysfunction and a clearly stated denition of response for evaluating treatment. Sixteen of 73 articles published between 1966 and 1998 were included. Overall response rate was 57%, and patients with primary versus secondary testicular failure had a response rate of 64% versus 44%. Intramuscular and oral methods of delivery were equivalent (response rates 51.3 and 53.2%, respectively) but response to TD therapy was signicantly different (80.9%). It is acknowledged that women may experience symptoms secondary to androgen deciency, and there is substantial evidence that prudent androgen replacement can be effective in relieving both physical and psychological symptoms of androgen insufciency (225). Testosterone replacement for women is now available in a variety of formulations. It appears to be safe, with the caveat that doses are restricted to the therapeutic window for androgen replacement in women, such that the benecial effects on well-being and quality of life are achieved without incurring undesirable virilizing side effects. For treatment of adult hypogonadal men, nightly 24-h application of the Androderm testosterone TDS (5 mg/day) has been demonstrated to be effective by a series of clinical pharmacokinetic studies (226). For treatment of adolescent males, physiological replacement can be approximated by modifying the dose and duration of Androderm application to mimic patterns of nocturnal testosterone secretion observed during puberty. A clinical audit was reported on the acceptability and efcacy as a treatment for hypogonadism of the rst transdermal testosterone therapy available in the United Kingdom (Andropatch), compared with existing androgen replacement options (227). Serum testosterone and questionnaire data on treatment efcacy, side effects, therapy preference, sexual dysfunction, and partners attitudes to therapy were obtained from 50 hypogonadal men prescribed long-term testosterone replacement. Eighty percent returned analyzable questionnaires and, of these, 84% experienced adverse effects with TD therapy, usually dermatological problems. Twentytwo percent elected to continue with TD therapy, 72% returned to depot, and 5% to oral therapy. The reservoir patches were judged to be too large, uncomfortable, vi-

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

sually obtrustive and noisy; thus, the pharmacokinetic advantages were largely outweighed by low patient acceptability. The pharmacokinetics of three testosteronecontaining TDS were evaluated in healthy male volunteers (228). Type 1 and 2 were nonscrotal membrane patches differing in adhesive type. Six subjects were treated with low-dose testosterone type 1, high-dose testosterone type 1, and low-dose testosterone type 2. To eliminate the inuence of endogenous serum testosterone, secretion was suppressed by the GnRH antagonist cetrorelix. Physiological testosterone levels were achieved during the 24-h application period. Maximal serum levels were achieved after 4 h with both types, and both enabled a physiological circadian prole to be achieved. Effects of TD testosterone replacement therapy using a permeation-enhanced system on plasma lipolytic enzymes (hepatic and lipoprotein lipase), LDL, and HDL subfraction concentrations were assessed (229). Ten patients with primary testicular failure were started on Testoderm therapy and evaluated before and after 3 months of treatment. Serum testosterone level increased to within the normal range in all subjects, whereas serum DHT increased to supranormal values. Plasma hepatic lipase (HL) activity increased after testosterone replacement (24.7 7.5 vs. 29.2 8.3 mol free fatty acid released per hour) and the increase in HL correlated with the increase in DHT. No signicant change was seen in the HDL2 subfraction but HDL3 decreased after treatment (0.93 0.17 vs. 0.79 0.14 mmol/L). Whether these changes adversely inuence the cardiovascular risk in the long term remains to be determined. Local skin reactions at application site are common adverse events with transdermal testosterone and an open-label, controlled pilot study evaluated whether topical pretreatment with triamcinolone acetonide, 0.1% cream, reduced the incidence or severity of chronic skin irritation in health volunteers (230). At all assessment points, more subjects had lower cumulative scores with pretreatment than without pretreatment. E. Fentanyl

Transdermal fentanyl has been used widely in the United States since it was approved in 1990 (231). The rst clinical report of TD fentanyl therapy in cancer pain involved ve patients. Pain relief was established with intravenous fentanyl and a TDS selected to deliver the same hourly dose while the intravenous infusion was tapered over 6 h. The TDS was changed every 24 h for a total of 3156 days. The initial study demonstrated steady-state plasma levels were linearly related to the fentanyl dose. A multicenter trial was conducted in 39 patients. The TD fentanyl dose was established from a conversion table based on the dose of oral immediate-release morphine required to control pain. The fentanyl patches were changed every 72 h and immediate-release morphine used on an as-required basis for incidental pain. This trial further demonstrated that patients could be converted from oral morphine to an equianalgesic dose of transdermal fentanyl and that pain relief could be maintained for a lengthy time period on an outpatient basis. The analgesic, pharmacokinetic, and clinical respiratory effects of 72-h application of two TD fentanyl patch sizes in patients undergoing abdominal hysterectomy were evaluated (232). Fentanyl TDS, releasing 50 g/h (TF-50) or 75 g/h (TF-75) fentanyl or placebo patches were applied to 120 women 2 h before abdominal hys-

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

terectomy under general anesthesia. All patients had postoperative access to supplemental morphine using patient-controlled pumps. VAS pain scores, supplementary analgesia, fentanyl plasma concentration, continuous hemoglobin saturation, respiratory pattern, and adverse effects data were collected. Visual analogue scale (VAS) pain scores and supplemental morphine use signicantly decreased in the TF-75 group in the postanesthesia care unit, and for both TF-50 and TF-75 groups for 8 48 h postoperatively. Between 5 and 36 h, the TD fentanyl groups had signicantly increased abnormal respiratory patterns, including apneic episodes and episodes of slow respiratory rate, and a signicantly increased requirement for oxygen supplementation. Nine patients in the TD fentanyl groups were withdrawn because of severe respiratory depression, but none in the placebo group. Although fentanyl plasma concentration were higher in the TF-75 than in TF-50 group, differences were not signicant. Fentanyl plasma concentration decreased signicantly 48 h after patch application. Although good analgesia was the result of this combination therapy, it was associated with a high incidence of respiratory depression requiring intensive monitoring, oxygen supplementation, patch removal in about 11% of patients, and opioid reversal with naloxone in about 8% of patients. It is generally recommended that patients should be titrated with a short-acting narcotic to control their cancer pain before they are converted to a fentanyl TDS. However, immediate TD fentanyl therapy in patients with uncontrolled cancer pain with direct titration of fentanyl TDS according to clinical necessity on a day-to-day basis, with the availability of morphine solution for rescue medication has been evaluated (233). The major objective was simplication of therapy. On average, sufcient pain control was reached within 48 h of the start of TD fentanyl, and VAS values at all follow-up times were signicantly lower than pretreatment values. Mean fentanyl doses were 70 g/h (week 1), 98 g/h (week 2), 107 g/h (week 3) and 116 g/h (week 4). Mean morphine doses as rescue medication steadily decreased from 11 mg/day in week 1 to 3 mg/day in week 4 of treatment, although differences were not signicant. It was concluded that titration with a short-acting narcotic before conversion to TD fentanyl was not necessary, provided the patients were well monitored. A comparison of TD fentanyl with placebo for postoperative analgesia in orthopedic surgery was reported (234). Forty adult patients had general anesthesia with propofol, isourane in nitrous oxide, and oxygen, and small bolus of alfentanil or sufentanil. Preoperatively, one group received TD fentanyl (75 g/h) for 72 h, and the second group a placebo patch. Morphine was given postoperatively as necessary. Eleven patients receiving fentanyl needed morphine, compared with 19 in the placebo group, and mean morphine dose was signicantly lower in the fentanyl group. One fentanyl group patient had decreased oxygen saturation and intense sedation, necessitating administration of naloxone. The mean maximum plasma fentanyl concentration was 1.63 ng/mL. An alternative fentanyl TDS was evaluated (235). Male adult surgical patients received 650 or 750 g of fentanyl intravenously as part of the induction of anesthesia, and plasma fentanyl concentrations were measured over the following 24 h. On the rst postoperative day a fentanyl TD was placed on the upper torso for 24 h and then removed. Plasma fentanyl concentrations were measured for 72 h after device application and each patients clearance and unit disposition function were determined. During the 72 h after application of the TDS, the amount of fentanyl

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

absorbed and the absorption rate were also determined. Residual fentanyl in the transdermal device was measured to determine absolute bioavailability. Of 14 subjects receiving TD fentanyl, 3 had clinically signicant fentanyl toxicity, mandating early removal of the TDS. In the remaining subjects the fentanyl concentration from 12 to 24 h varied over a 20-fold range (0.346.75 ng/mL). In subjects wearing the device for 24 h, terminal half-life of fentanyl after device removal was 16 h. Bioavailability of transdermally administered fentanyl was 63 35% and rate of fentanyl absorption from 12 to 2 h in subjects still wearing the device ranged from 10 to 230 g/h. In two subjects, the rate within the rst 6 h briey exceeded 300 g/ h and both demonstrated fentanyl toxicity, requiring early device removal. It was concluded that the Cygnus transdermal fentanyl device produced more highly variable plasma fentanyl concentrations than those reported for the Duragesic device, which is contraindicated for postoperative analgesia. The use of transdermal fentanyl was evaluated in ten patients aged 916 years with sickle cell pain crisis (236) who received TD fentanyl at 25 (n = 7) or 50 (n = 3) g/h and morphine use was >2.5 mg/h. Average TD fentanyl dose was 0.77 0.37 g/kg h1 on day 1 and 1.17 0.46 g/kg h1 on day 2. The fentanyl concentration at 24 and 48 h was 0.60 0.31 and 1.18 0.44 ng/mL, respectively. There was a signicant relation between dose and fentanyl concentration and no difference in any clinical-monitoring parameter between day 1 and day 2, although seven of ten patients reported subjective improvement in pain control. No adverse effects were noted, but it was concluded that improved understanding of doseeffect relations for TD fentanyl in children and adolescents was necessary before adequate pain control could be achieved by this route. At the start of TD fentanyl treatment, depot accumulation of the drug within skin results in a signicant delay (1748 h) before achievement of maximum plasma concentration (237). However, concomitant use of short-acting morphine maintained pain relief during the titration period, and supplementary medication decreased with duration of TD fentanyl treatment. Patient preference for TD fentanyl was indicated by the number of patient requests for continued use at the end of the study (up to 95%). Although postoperative TD fentanyl is contraindicated, supplementary patientcontrolled analgesia was signicantly reduced in patients receiving 75 g/h TD fentanyl, compared with placebo. Some patients with previously uncontrolled pain became completely pain-free. The most common adverse events during TD fentanyl therapy included vomiting, nausea, and constipation. Most serious adverse event was hypoventilation, which occurred more frequently in postoperative (4%) than in cancer patients (2%). In surgical patients, fentanyl-associated respiratory events usually occurred within 24 h of patch application, although there were isolated reports of late onset (up to 36 h). Respiratory depression was observed in a 2-year-old boy unintentionally exposed to a fentanyl TDS (238). He was found unresponsive after sleeping in bed with his grandmother and, after intubation and ventilation, a fentanyl TDS was discovered on his back. Removal of the device and treatment with naloxone resolved symptoms. This was the rst reported case of secondary exposure to a fentanyl patch causing clinically signicant respiratory depression in the pediatric population, and it emphasized a new hazard. A multicenter evaluation of TD fentanyl for cancer pain relief in 53 patients who required 45 mg or more of oral morphine daily was reported (239). After 1-

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

week stabilization on oral morphine, patients were transferred to an appropriate dose of TD fentanyl (25, 50, 75, or 100 g/h) administered by patch every 3 days. TD fentanyl was titrated to pain relief and patients were followed for up to 3 months. Mean duration of TD fentanyl use was 58 32 days. Mean daily morphine dose during the last 2 days of stabilization was 189 20 mg, and mean initial fentanyl patch dose was 58 6 g/h. The mean daily morphine dose taken as required for breakthrough pain at study completion was 35 mg. Mean nal fentanyl dosage at study completion was 169 29 g/h. Pain relief was rated as good or excellent by 82% of patients during the treatment period and 63% preferred TD fentanyl to their last analgesic therapy. Side effects related to the patch were nausea (13%), vomiting (8%), skin rash (8%), and drowsiness (4%). Thirty percent of patients reported adverse experiences related to the patch, and 17% had to be discontinued. Pain-related treatment satisfaction, patient-perceived side effects, functioning, and well-being in patients with advanced cancer receiving TD fentanyl (Duragesic) or sustained-release oral morphine (MS Contin or Oramorph SR) were compared (240). Patients were more satised with TD fentanyl than sustained-release oral morphine. Patients receiving TD fentanyl were more satised overall with their pain medication and also reported a signicantly lower frequency and effect of a pain medications side effects. Measures of pain intensity, sleep adequacy, and symptoms demonstrated no signicant differences between treatment groups. Forty-eight patients with noncancer neuropathic pain, who had participated in a randomized controlled trial with intravenous fentanyl infusions, next received prolonged TD fentanyl (241). Eighteen patients stopped prematurely owing to inadequate pain relief, side effects, or both. Pain relief among the remaining 30 patients completing the 12-weekdose titration protocol was substantial in 13 and moderate in 5, and quality of life improved by 23%. Psychological dependence or induction of depression was not observed, and tolerance emerged in only 1 patient. There was a signicant positive correlation between pain relief obtained with intravenous and prolonged TD fentanyl. It was concluded that long-term TD fentanyl may be effective in noncancer neuropathic pain without clinically signicant management problems. Testing with intravenous fentanyl may assist in selecting neuropathic pain patients who can benet from treatment by the transdermal route. Disposition of fentanyl in dogs after intravenous and TD administration was investigated (242). Each of six clinically normal beagles received intravenous fentanyl (50 g/kg body weight) and TD fentanyl (50 g/h). TD fentanyl produced average steady-state concentrations of 1.6 ng/mL. Actual delivery rate of fentanyl was 27.599.6% of the theoretical rate. Mean elimination half-life after patch removal was 1.39 h. It was concluded that TD fentanyl had clinically useful potential as an analgesic in dogs. Plasma fentanyl concentrations were obtained in six intact, mixed-breed adult dogs (two males, four females), weighing 19.9 3.4 kg, after application of three sizes of fentanyl TDS delivering 50, 75, or 100 g/h (243). Results are summarized in Table 7. TD fentanyl was evaluated in children with cancer pain (244) and measures of analgesia, side effects, and skin changes obtained for a minimum of two doses (6 treatment days). Treatment was well tolerated and 10 of 11 patients continued the treatment. Time to peak plasma concentration ranged from 18 to more than 66 h in patients using a 25-g/h patch. Compared with published data from adults, mean

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

Table 7

Comparison Between Three Transdermal Fentanyl Patches in Dogs Plasma fentanyla (ng/mL) 0.7 0.2 1.4 0.5 1.2 0.5 Total AUC (ng/h mL1) 46 12.2 101.2 41.4 80.4 38.3 Elimination half-life (app) (h) 3.6 1.2 3.4 2.7 2.5 2.0

Fentanyl delivery (g/h) 50 75 100

Levels between 24 and 72 h. Source: Ref. 243.

clearance and volume of distribution of transdermal fentanyl were equal, but variability was less. It has been claimed that fentanyl patches are less suitable for patients who are elderly or terminally ill and dying. However, a retrospective survey of 205 cancer patients who died within hospital-based home care in Norrkoping, Sweden between January 1997 and June 1998 reported that 34 patients used fentanyl TD, and in 30 it was possible to evaluate analgesic efcacy. Estimated efcacy was good (93%) or moderate (88%) in patients younger or older than 65 years of age (245). Constipation and use of laxatives were investigated in patients with chronic cancer pain treated with oral morphine and TD fentanyl (246). Of 46 patients, treated with slow-release morphine, 301000 mg/day for 6 days, 39 were switched to TD fentanyl (0.69.6 mg/day) with a conversion ratio of 100:1. Median fentanyl doses increased from 1.2 to 3.0 mg/day throughout the 30-day TD treatment period. Mean pain intensity decreased slightly after conversion, although the number of patients with breakthrough pain or who requiring immediate-release morphine as a rescue medication was higher with TD fentanyl. The number of patients with bowel movements did not change after the opioid switch, but the number of patients taking laxatives was signicantly reduced from 7887% of patients per treatment day (morphine) to 2248% (fentanyl). Constipation is an almost universal side-effect of prolonged opioid analgesia, and resulting discomfort can be more severe than the pain itself, leading to reduction of analgesic use and consequently increased pain (247). Because of higher lipophilicity, fentanyl penetrates the bloodbrain barrier more easily and lower doses are possible; thus, comparatively less opioid is available in the gastrointestinal tract to block local receptors. The cost of treating constipation among terminally ill cancer patients receiving TD fentanyl and 12-h sustained-release morphine in Ontario, Canada was examined (248). Cost of managing constipation in a patient receiving TD fentanyl and 12-h sustained-release morphine was estimated at 31.77 and 52.76 dollars, respectively, during the rst 2 weeks of treatment. When the acquisition costs of opioids are included, the two-week cost of managing a patient with TD fentanyl and 12-h sustained-release morphine was 123.24 and 119.70 dollars, respectively. It was concluded that acquisition costs alone should not dictate treatment, and care should always be tailored to the needs and preferences of individual patients. The issues involved in switching opioids to transdermal fentanyl in a clinical setting have been evaluated (249). Case records of patients treated with TD fentanyl were retrospectively examined and conversion ratios calculated. Opioid therapy was

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

switched to TD fentanyl during inpatient treatment for 53 patients and during outpatient treatment for 11 patients. Before conversion patients were treated with slowrelease morphine (48%), immediate-release morphine (17%), buprenorphine (11%), tramadol (11%), levomethadone (5%), tilidine/naloxone (5%), and piritramid (3%). Reasons for opioid rotation included inadequate pain relief (33%), patients wish to reduce oral medication (20%), gastrointestinal side effects (31%), vomiting (13%), constipation (19%), and dysphagia (27%). Reduction of side effects was reported by 10 of 19 patients. In 12 of 21 patients, when medication was switched because of inadequate pain relief, reduction in pain intensity was reported. It was concluded that conversion to TD therapy may readjust the balance between opioid analgesia and side effects. F. Anti-inammatories

A multicenter study of comparative efcacy, tolerability, and acceptability of a urbiprofen local-action transcutaneous (LAT) patch (40 mg b.d.) and piroxicam gel (3 cm, 0.5% q.d.s), was conducted in general practice in the United Kingdom in 137 men and women with soft-tissue rheumatism of the shoulder or elbow tag, epicondylitis, tendinitis, bursitis, or adhesive capsulitis (250). Patients received one therapy for 4 days before crossing over for a further 4 days, followed by 6 days of their preferred therapy. Assessment of severity of pain, tenderness, and overall clinical condition was carried out at baseline and at 4, 8, and 14 days. There was a signicant reduction in severity of pain in favor of urbiprofen LAT (42 vs. 26%). Eligible dataset analysis revealed signicant differences, in favor of urbiprofen LAT, in severity of lesion tenderness and overall change in clinical condition. Superior efcacy was also indicated at the end of the crossover phase when 69% chose to continue treatment with urbiprofen LAT. There were also signicant differences in favor of urbiprofen LAT in assessments for night pain, quality of sleep, and patients overall opinion of treatment. G. Clonidine

Pharmacokinetic and pharmacodynamic properties and safety, of a TD clonidine system (M-5041T) were evaluated after single and repeated applications (251). In the single-application study, one patch delivering 4, 6, or 8 mg was applied for 3 days to eight healthy subjects. In a repeated-application study, A (072 h), B (72 144 h), and C (144216 h) TD systems delivering 6 mg were applied in seven healthy subjects. In the single-application study, plasma clonidine level, Cmax and AUC increased in a dose-dependent manner, but not signicantly. The blood pressure (BP)-lowering effect of 8 mg was greater than that of 4 and 6 mg. Adverse effects were reported, but did not cause withdrawal. In the repeated-application study, plasma clonidine increased up to 48 h after application of patch A, and remained stable until removal of patch C. Cmax and AUC did not differ signicantly. During an active period, BP decreased signicantly during treatment, but BP at midnight did not change signicantly. Mild erythema and systemic adverse effects were reported. In a second study (252), TD clonidine (M-5041T) was compared with oral clonidine (Catapres TTS). One TDS containing 6 mg of clonidine was applied on the right chest for 3 days or one tablet of Catapres TTS (0.075 mg) was given orally every 12 h for 3 days in eight healthy subjects. Plasma clonidine concentration

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

increased gradually after application of TDS and decreased gradually after removal, whereas it increased rapidly then decreased rapidly after each dose of Catapres TTS. Elimination half-life of clonidine after patch removal was signicantly greater than after the nal dose of Catapres TTS. There was no signicant difference in Cmax, AUC, or BP-lowering between the treatments. Adverse symptoms occurred more frequently during Catapres TTS therapy and were observed when plasma clonidine concentration was relatively high. Clinical pharmacological evaluation of a clonidine TDS was undertaken in a variety of phase I volunteer studies to assess potential effects of site of application, pharmacokinetic linearity, and absolute bioavailability (253). Delivery of clonidine from Catapres TTS was inuenced by application site. The order of permeability was chest > upper arm > upper thigh. Pharmacokinetics of clonidine following TD delivery were linear with slight nonlinearity between devices of differing area. Absolute bioavailability of clonidine from Catapres TTS was approximately 60% and calculated in vivo absorption rate (4.32 1.68 g/h) was in good agreement with claimed performance (0.1 mg/day). Premedication with oral and TD clonidine in postoperative sympatholysis was investigated in 61 patients undergoing elective major noncardiac surgery (254). The treatment group were premedicated with a clonidine TDS (0.2 mg/d), applied the night before surgery and left in place for 72 h, and were given 0.3-mg oral clonidine 6090 min before surgery. Clonidine reduced enurane requirements, intraoperative tachycardia, and myocardial ischemia (1 of 28 clonidine patients vs. 5 of 24 placebo). However, postoperatively, the heart rate decreased for only the rst 5 h, and incidence of postoperative myocardial ischemia (6 of 28 clonidine vs. 5 of 26 placebo) did not differ. Clonidine signicantly reduced plasma levels of epinephrine and norepinephrine on the rst postoperative morning. It was suggested that larger doses of clonidine should be investigated for their ability to decrease postoperative tachycardia and myocardial ischemia. Surgical trauma induces diffuse sympathoadrenal activation that contributes to perioperative cardiovascular complications in high-risk patients. Regional anesthetic and analgesic techniques can attenuate this stress response and reduce rates of adverse perioperative events, but their postoperative use is logistically difcult and expensive. Hence, use of transdermal clonidine to blunt the stress response throughout the perioperative period was evaluated (255). Forty patients scheduled for major upper abdominal surgery were entered in a clinical trial. Patients received either clonidine (0.2 mg orally plus clonidine TTS-3 patch the evening before surgery plus 0.3 mg orally on call to operating room) or matched oral and transdermal placebo. Preoperative transdermal (plus oral) clonidine administration resulted in therapeutic plasma clonidine concentrations over the perioperative period (1.54 0.07 g/mL) and reduced preoperative epinephrine and norepinephrine levels by 65%. Plasma catecholamines increased in both groups after surgery, but were markedly lower over the postoperative period in patients receiving clonidine (who also had reduced frequency of postoperative hypertension). A study was performed to determine if a lower than previously reported oral transdermal clonidine regimen could reduce postoperative morphine requirements without producing systemic side effects (256). Twenty-nine healthy females undergoing elective abdominal hysterectomy received preoperative oral clonidine, 46 g/ kg plus a 7 cm2 clonidine TDS (0.2 mg/24 h), or a placebo tablet and patch. Lowdose clonidine had no potentiating effect on morphine analgesia and postoperative

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

morphine use, VAS pain scores and morphine plasma levels were similar between the groups. The clonidine group experienced a signicantly higher incidence of intraoperative and postoperative hypotension and bradycardia than the control group, but no differences were noted in incidence of nonhemodynamic side effects. Clinical and pharmacological indications suggested reduction of noradrenergic tone occurs in cluster headache, during both active and remission periods, but that sharp uctuations of the sympathetic system may trigger the attacks (257). It was postulated that continuous administration of low-dose clonidine could be benecial in the active phase by antagonizing variations in noradrenergic tone. After a run-in week, TD clonidine (57.5 mg) was administered for 1 week to 13 patients suffering from episodic or chronic cluster headache. During clonidine treatment, mean weekly frequency of attacks reduced from 17.7 7.0 to 8.7 6.6, pain intensity of attacks (VAS) from 98.0 7.2 to 41.1 36.1 mm, and duration from 59.3 21.9 to 34.3 24.6 min. This strongly suggested that TD clonidine may be effective in the preventive treatment of cluster headache. To test a previous clinical observation that approximately 25% of patients with painful diabetic neuropathy appeared responsive to clonidine, a formal clinical trial of TD clonidine was conducted (258). In stage 1, 41 patients completed a randomized, three-period crossover comparison of TD clonidine (titrated from 0.1 to 0.3 mg/day) to placebo patches. Twelve apparent responders from stage 1 were entered into the second stage, consisting of an additional four double-blind, randomized, 1week treatment periods with TD clonidine and placebo. Stage 1 showed that, for the total group, pain intensity differed little between clonidine and placebo. In stage 2, however, the 12 apparent responders from stage 1 had 20% less pain with clonidine than placebo (95% CI: 435% pain reduction), conrming that their pain was responsive to clonidine. Post hoc analysis suggested patients describing their pain as sharp and shooting were more likely to respond to clonidine. These results support the hypothesis that there is a subset of patients with painful diabetic neuropathy who benet from systemic clonidine administration and also demonstrate the value of an enriched enrollment technique in analgesic trials. Efcacy and tolerability of TD clonidine in inner-city African-American and Hispanic-American patients with essential hypertension was evaluated in multiple community-based primary care centers in a 12-week trial (259). Patients with diastolic BP higher than 90 mmHg were given TD clonidine (0.1 or 0.2 mg daily). The drug was titrated after 1 month if diastolic BP exceeded 90 mmHg. At 12 weeks, change in blood pressure, adverse effects, and patient satisfaction were assessed. Transdermal clonidine signicantly lowered BP by 15.7/12.8 18.1/9.6 mmHg, and heart rate by 3 9 beats per minute. There were no differences in BP reduction according to race, ethnicity, gender, or age. Most common adverse effects were pruritus or discomfort at the patch site, dizziness, dry mouth, and fatigue (11% discontinued owing to adverse effects). Sixty-seven percent of patients reported that TD clonidine was more convenient than oral therapy. H. Miscellaneous

Intraocular pressure (IOP)-lowering effects of a TD system containing pilocarpine that was designed to avoid common side effects in glaucoma treatment with conventional eye drops were reported (260). Two patches, each containing 30 mg of

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

pilocarpine or placebo, were applied to the supraclavicular skin of 24 patients. IOP was recorded before (22.7 5.8 mmHg) and at 12, 16, and 20 h after application. Pilocarpine TD did not signicantly reduce IOP although there were detectable plasma levels of pilocarpine at 12 (2.9 ng/mL) and 20 h (1.3 ng/mL) after administration. The therapeutic effects of selective cholinergic replacement using oral xanomeline, an M1/M4 receptor agonist, were assessed in a multicenter study of 343 patients with Alzheimers disease (261). Improvement in ADAS-Cog provided clinical evidence of involvement of the M1 muscarinic receptor in cognition, and the favorable effects of xanomeline on disturbing behavior suggested a novel approach for treatment of noncognitive symptoms. Although adverse effects (mainly gastrointestinal) associated with the oral formulation appeared to limit its use, a large-scale study investigating the safety and efcacy of transdermal xanomeline is underway. The effects of TD scopolamine on pulmonary function, symptoms, and bronchial hyperresponsiveness to methacholine were reported (262). The rst study evaluated therapeutic effects of a single scopolamine TD system (Scopoderm TTS, 2.5 cm2) on forced expiratory volume in 1 s (FEV1), reversibility, peak expiratory ow (PEF), and symptoms in ten patients with reversible airways disease. The drug was adequately taken up into the systemic circulation, but no signicant clinical effects, nor correlations between scopolamine levels and outcome parameters, were observed. Because of the possibility of subtherapeutic doses, a second study with two scopolamine TD systems was carried out in ten patients with bronchial hyperresponsiveness to methacholine. Blood and urine concentrations of free scopolamine were doubled compared with the rst study, but there were still no signicant effects on FEV1, PEF, symptoms, and bronchial hyperresponsiveness, although most of the patients now reported adverse side effects. It was concluded that TD administration of scopolamine was not clinically useful in asthma and chronic obstructive pulmonary disease. Efciency of TD scopolamine in prophylaxis of postoperative nausea and vomiting (PONV) after otoplasty was evaluated in a post hoc assessment of data (263). Of 50 otoplasty patients 25 received a scopolamine patch before general anesthesia. The placebo group received intravenous atropine (10 g/kg) during induction. Scopolamine-treated patients suffered more from moderate preoperative bradycardia (8 of 25) than atropine-treated patients (1 of 25). After unilateral otoplasty, no scopolamine-treated patients, but 50% of atropine-treated patients, suffered from PONV. After bilateral otoplasty, respective incidences were 39 and 81%. After unilateral otoplasty no patient needed droperidol, but after bilateral otoplasty, 12 of 19 atropinetreated and 4 of 18 scopolamine-treated patients needed droperidol. It was concluded that TD scopolamine offers effective prophylaxis against PONV, but does not protect from bradycardia in otoplasty. The effects of TD scopolamine (1.5 mg) and oral cyclizine (50 mg) on postural sway, optokinetic nystagmus, and circularvection in humans were investigated (264). Neither scopolamine nor cyclizine (at doses used for relief of motion sickness) had a signicant suppressive effect on these aspects of visualvestibular interaction. A pilot clinical trial of TD administration of selegiline in HIV-positive patients was performed to obtain preliminary data on safety, tolerability, and effect on HIVassociated cognitive impairment (265). Both selegiline and placebo were well tol-

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation

Transdermal

erated and improvements favoring the selegiline group were suggested on single tests of verbal memory and motorpsychomotor performance. TD patches of diltiazem hydrochloride were formulated using ethyl cellulose and polyvinylpyrrolidone. Pharmacodynamic and pharmacokinetic performance of diltiazem hydrochloride after TD administration was compared with that of oral administration in male New Zealand albino rabbits (266). Pharmacokinetic parameters were signicantly different between transdermal and oral administration. Terminal elimination half-life of TD-delivered diltiazem was similar to that of oral administration. Skin irritation studies indicated no recognizable changes after patch removal. It was concluded that relative bioavailability of diltiazem hydrochloride and therapeutic activity were vefold higher after TD than after oral administration. Transdermal histamine has been successfully used for amelioration of symptoms of both relapsingremitting and progressive multiple sclerosis (267). Of the 55 patients using TD histamine cream 67% had improvements in one or more areas (extremity strength, balance, bladder control, fatigue, activities of daily living, and cognitive functioning) sustained for up to 3 months, and 33% had improvements in three or more areas. The occlusive properties of a range of hydrocolloid patches on the penetration of triamcinolone acetonide, and hydration, were assessed in vivo using visual assessment (268). There was a signicant difference in rates of hydration of patches containing either NaCMC 39%, or pectin 39%, although changes in hydrocolloid composition did not signicantly alter the blanching response. The feasibility of use of water-activated, pH-controlled silicone reservoir devices for TD administration was investigated using timolol maleate (269). Two timolol patches were applied to the arm of 12 volunteers for 81 h and absorption compared with that from an oral tablet formulation (Blocanol). In vivo plasma levels were also compared with those predicted by kinetic simulations. Both steady-state timolol concentrations in plasma and duration could be controlled with water-activated, pH-controlled patches, although considerable, interindividual variability in TD absorption occurred owing to the high fractional skin control in timolol delivery. Timolol patches were well tolerated, skin irritation was mild, and after removal of the patches, skin changes were practically reversed in 24 h. Preclinical tests were also reported (270) including release of timolol from patches and timolol permeation across human cadaver skin. Systemic effects, local tolerance, and effectiveness on the penis of topical gel formulations containing alprostadil (prostaglandin E1) plus 5% of the penetration enhancer SEPA versus SEPA alone (placebo) were evaluated in men with erectile dysfunction (271). Application of prostaglandin E1 gel correlated positively with erectile response (6775% of patients had an erection compared with 17% of controls. A soft, stretchy adhesive patch (10 14 cm) containing 5% lidocaine (700 mg) has been used for topical treatment of pain associated with postherpetic neuralgia (272). Systemic absorption was reported as minimal in healthy subjects and patients with postherpetic neuralgia (3% of dose). In clinical trials (12 h24 days), treatment resulted in a signicant reduction in pain intensity and increased pain relief compared with vehicle patch. Transdermal drug delivery systems are increasingly popular, yet few data exist on emergency medical outcomes after exposures. A retrospective study (273) using

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

data collected through a Regional Poison Information System identied 61 cases over a 5-year period. Routes were dermal (48), oral (10), combined oral and dermal (1), parenteral use of gel residue (1), and combined oral and parenteral (1). Most exposures (72%) were managed by home telephone consultation only. Eleven of 17 patients (18%) evaluated in health care facilities were admitted, including 8 (13%) to intensive care units. Hospital admission correlated statistically with clonidine, fentanyl, oral exposure, and drug abuse. Clonidine exposure also correlated statistically with intensive care admission. One fatality was recorded; all other patients recovered uneventfully. V. CONCLUDING REMARKS

In this chapter we have described some of the considerations that are important in the design and development of pharmaceutical products intended for application to the skin. In addition, we have described some recent clinical observations of transdermal therapeutic systems. Space limitations dictated that we could not provide an exhaustive review of all factors, and the reader will appreciate that preformulation, scale-up, and safety are not covered. However, these aspects are fully covered either elsewhere in this volume or in some of the recent excellent and fully recommended texts (7,8,83). We have attempted to share our knowledge of the formulation types that are applied to the skin, together with their properties, problems, and clinical usefulness. We hope that our comments will provide the novice formulator with some insights borne of experience, and the experienced formulator with some novel insights in the eld of dermatological and transdermal products development and their use. REFERENCES
1. 2. 3. Scheuplein RJ, Blank IH. Permeability of the skin. Physiol Rev 1971; 51:702747. Idson B. Percutaneous absorption. J Pharm Sci 1975; 64:901924. Flynn GL. (1996). Cutaneous and transdermal delivery: processes and systems of delivery. In: Banker GS, Rhodes CT, eds. Modern Pharmaceutics, 3rd ed. New York: Marcel Dekker. 1996; 239298. Barry BW. Dermatological Formulatins, Percutaneous Absorption. New York: Marcel Dekker. 1983. Shah VP, Maibach HI, eds. Topical Drug Bioavailability, Bioequivalence, and Penetration. New York: Plenum Press. 1993. Schaefer H, Redelmeir TE. Skin Barrier, Principles of Percutaneous Absorption. Basel: Karger, 1996. Roberts MS, Walters KA, eds. Dermal Absorption and Toxicity Assessment. New York: Marcel Dekker. 1998. Bronaugh RL, Maibach HI, eds. Percutaneous Absorption, Drugs, Cosmetics, Mechanisms, Methodology. 3rd ed. New York: Marcel Dekker. 1999. Elias PM. Lipids and the epidermal permeability barrier. Arch Dermatol Res 1981; 270:95117. Raykar PV, Fung MC, Anderson BD. The role of protein and lipid domains in the uptake of solutes by human stratum corneum. Pharm Res 1988; 5:140150. Walters KA, Hadgraft J, eds. Pharmaceutical Skin Penetration Enhancement. New York: Marcel Dekker, 1993.

4. 5. 6. 7. 8. 9. 10. 11.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation 12. 13.

Transdermal

14.

15. 16. 17.

18. 19.

20.

21. 22. 23. 24.

25.

26.

27. 28. 29.

30. 31.

Smith EW, Maibach HI, eds. Percutaneous Penetration Enhancers. Boca Raton: CRC Press, 1995. Behl CR, Flynn GL, Barrett M, Walters KA, Linn EE, Mohamed Z, Kurihara T, Ho NFH, Higuchi WI, Pierson CL. Permeability of thermally damaged skin II: immediate inuences of branding at 60C on hairless mouse skin permeability. Burns 1981; 7: 389399. Flynn GL, Behl CR, Walters KA, Gatmaitan OG, Wittkowsky A, Kurihara T, Ho NFH, Higuchi WI, Pierson CL. Permeability of thermally damaged skin III: inuence of scalding temperature on mass transfer of water and n-alkanols across hairless mouse skin. Burns 1982; 8:4758. Bronaugh RL, Stewart RF. Methods for in vitro percutaneous absorption studies. V. Permeation through damaged skin. J Pharm Sci 1985; 74:10621066. Scott RC, Dugard PH, Doss AW. Permeability of abnormal rat skin. J Invest Dermatol 1986; 86:201207. Scott RC. In vitro absorption through damaged skin. In: Bronaugh RL, Maibach HI, eds. In Vitro Percutaneous Absorption, Principles, Fundamentals and Applications. Boca Radon: CRC Press. 1991; 129135. Bond JR, Barry BW. Damaging effect of acetone on the permeability barrier of hairless mouse skin compared with that of human skin. Int J Pharm 1988; 41:9193. van der Valk PGM, Nater JP, Bleumink E. Vulnerability of the skin to surfactants in different groups of eczema patients and controls as measured by water vapour loss. Clin Exp Dermatol 1985; 10:98103. Turpeinen M, Salo OP, Leisti S. Effect of percutaneous absorption of hydrocortisone on adrenocortical responsiveness in infants with severe skin disease. Br J Dermatol 1986; 115:475484. Turpeinen M. Absorption of hydrocortisone from the skin reservoir in atopic dermatitis. Br J Dermatol 1991; 124:358360. Fartasch M, Diepgen TL. The barrier function in atopic dry skin. Acta Derm Venereol 1992; 176(suppl):2631. Reichek N, Goldstein RE, Redwood DR. Sustained effects of nitroglycerin ointment in patients with angina pectoris. Circulation 1974; 50:348352. Cleary GW. Transdermal delivery systems: a medical rationale. In: Shah VP, Maibach HI, eds. Topical Drug Bioavailability, Bioequivalence, and Penetration. New York: Plenum Press, 1993; 1768. Anderson BD. Prodrugs and their topical use, In: Shah VP, Maibach HI, eds. Topical Drug Bioavailability, Bioequivalence, and Penetration. New York: Plenum Press. 1993; 6989. Ahmed S, Imai T, Otagiri M. Stereoselective hydrolysis and penetration of propranolol prodrugs: in vitro evaluation using hairless mouse skin. J Pharm Sci 1995; 84:877 883. Lai PM, Roberts MS. Iontophoresis. In: Roberts MS, Walters KA, eds. Dermal Absorption and Toxicity Assessment. New York: Marcel Dekker. 1998; 371414. Prausnitz MR. The effects of electric current applied to skin: a review for transdermal drug delivery. Adv Drug Deliv Rev 1996; 18:395425. McElnay JC, Benson HAE, Hadgraft J, Murphy TM. (1993). The use of ultrasound in skin penetration enhancement. In: Walters KA, Hadgraft J, eds. Pharmaceutical Skin Penetration Enhancement. New York: Marcel Dekker. 1993; 293309. Steele GW. 2001. Preformulation. In: Gibson M, ed. Pharmaceutical Preformulation and Formulation. Interpharm Press. Interpharm Book. Kitson N, Maddin S. Drugs used for skin diseases. In: Roberts MS, Walters KA, eds. Dermal Absorption and Toxicity Assessment. New York: Marcel Dekker. 1998; 313 326.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters 32. 33. 34. 35. 36. 37. 38.

Brain

39. 40. 41. 42. 43. 44.

45. 46. 47. 48.

49.

50. 51. 52. 53.

54. 55.

Zatz JL, Varsano J, Shah VP. In vitro release of betamethasone dipropionate from petrolatum-based ointments. Pharm Dev Technol 1996; 1:293298. Kneczke M, Landersjo L, Lundgren P, Fu hrer C. In vitro release of salicylic acid from two different qualities of white petrolatum. Acta Pharm Suec 1986; 23:193204. Osborne DW. Phase behavior characterization of propylene glycol, white petrolatum, surfactant ointments. Drug Dev Ind Pharm 1992; 18:18831894. Osborne DW. Phase behavior characterization of ointments containing lanolin or a lanolin substitute. Drug Dev Ind Pharm 1993; 19:12831302. Pena LE, Lee BL, Stearn JF. Structural rheology of a model ointment. Pharm Res 1994; 11:875881. Dahl T, He GX, Samuels G. Effect of hydrogen peroxide on the viscosity of a hydroxyethylcellulose-based gel. 1998; Pharm Res 15:11371140. Planas MD, Rodriguez FG, Dominguez MH. The inuence of neutralizer concentration on the rheological behaviour of a 0.1% Carbopol hydrogel. Pharmazie 1992; 47:351 355. Pe rezMarcos B, Mart nezPacheco R, Go mezAmoza, Souto C, Concheiro A, Rowe RC. Interlot variability of carbomer 934. Int J Pharm 1993; 100:207212. Sjo blom J, ed. Emulsions and Emulsion Stability. New York: Marcel Dekker. 1996. Eccleston GM. Multiple phase oil-in-water emulsions. J Soc Cosmet Chem 1990; 41: 122. Eccleston GM. Functions of mixed emulsiers and emulsifying waxes in dermatological lotions and creams. Colloids Surfaces 1997; 123:169182. Orth DS. Handbook of Cosmetic Microbiology. New York: Marcel Dekker, 1993. Takruri H, Anger CB. (1989). Preservation of dispersed systems. In: Lieberman HA, Rieger MM, Banker GS, eds. Pharmaceutical Dosage Forms: Disperse Systems. Vol. 2. New York: Marcel Dekker. 1989; 73114. Cosmetic Ingredient Review. Final report on the safety of methyl, ethyl, propyl and butyl paraben. JACT 1984; 3(5). Dal Pozzo A, Pastori N. Percutaneous absorption of parabens from cosmetic formulations. Int J Cosmet Sci 1996; 18:5766. Kurup TRR, Wan LSC, Chan LW. Interaction of preservatives with macromolecules. Part II. Cellulose derivatives. Pharm Acta Helv 1995; 70:187193. Tanaka M, Iwata Y, Kouzuki Y, Taniguchi K, Matsuda H, Arima H, Tsuchiya S. Effect of 2-hydroxypropyl--cyclodextrin on percutaneous absorption of methyl paraben. J Pharm Pharmacol 1995; 47:897900. Komatsu H, Higaki K, Okamoto H, Miyagawa K, Hashida M, Sezaki H. Preservative activity and in vivo percutaneous penetration of butyl paraben entrapped in liposomes. Chem Pharm Bull 1986; 34:34153422. Komatsu H, Okamoto H, Miyagawa K, Hashida M, Sezaki H. Percutaneous absorption of butyl paraben from liposomes in vitro. Chem Pharm Bull 1986; 34:34233430. Kabara JJ, Orth DS, eds. Preservative-Free and Self-Preserving Cosmetics and Drugs Principles and Practice. New York: Marcel Dekker. 1997. Shah VP, Elkins J, Hanus J, Noorizadeh C, Skelly JP. In vitro release of hydrocortisone from topical preparations and automated procedure. Pharm Res 1991; 8:5559. Chattaraj SC, Kanfer I. Release of acyclovir from semisolid dosage forms: a semiautomated procedure using a simple Plexiglass ow-through cell. Int J Pharm 1995; 125: 215222. Higuch T. Physical chemical analysis of percutaneous absorption process from creams and ointments. J Soc Cosmet Chem 1960; 11:8597. Higuchi T. Rate of release of medicaments from ointment bases containing drugs in suspension. J Pharm Sci 1961; 50:874875.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation 56. 57.

Transdermal

58.

59. 60.

61. 62.

63.

64.

65. 66.

67. 68.

69. 70. 71.

72.

73.

74.

Bunge AL. Release rates from topical formulations containing drugs in suspension. J Control Release 1998; 52:141148. Flynn GL, Caetano PA, Pillai RS, Abriola LM, Amidon GE. Recent perspectives in percutaneous penetration: in vitro drug release and its relationship to drug delivery. In: Brain KR, Walters KA, eds. Perspectives in Percutaneous Penetration. Vol. 6b. Cardiff, Wales: STS Publishing. 1999 (in press). Kril MB, Parab PV, Genier SE, DiNunzio JE, Alessi D. Potential problems encountered with SUPAC-SS and the in vitro release testing of ammonium lactate cream. Pharm Technol 1999; March:164174. Guy RH, Hadgraft J. The prediction of plasma levels of drugs following transdermal application. J Control Release 1985; 1:177182. Pugh WJ, Hadgraft J, Roberts MS. Physicochemical determinants of stratum corneum permeation. In: Roberts MS, Walters KA, eds. Dermal Absorption and Toxicity Assessment. New York: Marcel Dekker. 1998; 245268. Kasting GB, Filloon TG, Francis WR, Meredith MP. Improving the sensitivity of in vitro skin penetration experiments. Pharm Res 1994; 11:17471754. Giannakou SA, Dallas PP, Rekkas DM, Choulis NH. Development and in vitro evaluation of nimodipine transdermal formulations using factorial design. Pharm Dev Technol 1998; 3:517525. Kandimalla KK, Kanikkannan N, Singh M. Optimization of a vehicle mixture for the transdermal delivery of melatonin using articial neural networks and response surface method. J Control Release 1999; 61:7182. Takayama K, Takahara J, Fujikawa M, Ichikawa H, Nagai T. Formula optimization based on articial neural networks in transdermal drug delivery. J Control Release 1999; 62:161170. Marecki NM. Design considerations in transdermal drug delivery systems. Proceedings 10th Pharmaceutical Technology Conference. 1987; 311318. Yuk SH, Lee SJ, Okano T, Berner B, Kim SW. One-way membrane for transdermal drug delivery systems. I. Membrane preparation and characterization. Int J Pharm 1991; 77:221229. Yuk SH, Lee SJ, Okano T, Berner B, Kim SW. One-way membrane for transdermal drug delivery systems. II. System optimization. Int J Pharm 1991; 77:231237. Okabe H, Suzuki E, Saitoh T, Takayama K, Nagai T. Development of novel transdermal system containing d-limonene and ethanol as absorption enhancers. J Control Release 1994; 32:243247. Wang DM, Lin FC, Chen LY, Lai JY. Application of asymmetric TPX membranes to transdermal delivery of nitroglycerin. J Control Release 1998; 50:187195. Lai JY, Lin FC, Wu TT, Wang DM. On the formation of macrovoids in PMMA membranes. J Membr Sci 1999; 155:3143. Thacharodi D, Rao KP. Rate-controlling biopolymer membranes as transdermal delivery systems for nifedipine: development and in vitro evaluations. Biomaterials 1996; 17:13071311. Iordanskii AL, Feldstein MM, Markin VS, Hadgraft J, Plate NA. Modeling of the drug delivery from a hydrophilic transdermal therapeutic system across polymer membrane. Eur J Pharm Biopharm 2000; 49:287293. Pster WR, Sheeran MA, Watters DE, Sweet RP, Walters P. Methods for altering release of progesterone, testosterone, propranolol, and indomethacin from silicone matrices: effects of cosolvents and inert llers. Proc Int Symp Control Release Bioact Mater 1987; 14:223224. Ulman KL, Gornowicz GA, Larson KR, Lee CL. Drug permeability of modied silicone polymers. I. Silicone-organic block copolymers. J Control Release 1989; 10: 251260.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters 75.

Brain

76. 77. 78. 79. 80. 81. 82.

83.

84. 85.

86. 87. 88.

89.

90. 91. 92.

93. 94. 95.

96. 97.

Hille T. Technological aspects of penetration enhancers in transdermal systems. In: Walters KA, Hadgraft J, eds. Pharmaceutical Skin Penetration Enhancement. New York: Marcel Dekker. 1993; 335343. Adhesives Research, Inc., Transdermal Technical Brief, Enhancer-Tolerant Adhesives. Ko CU, Wilking SL, Birdsall J. Pressure sensitive adhesive property optimizations for the transdermal drug delivery systems. Pharm Res 1995; 12:S-143. te Hennepe HJC. Solution acrylic polymers for medical applications. National Starch & Chemical B.V. 1993. Jenkins AW. Developing the Fematrix transdermal patch. Pharm J 1995; 255:179181. PMA Committee Report. Transdermal Drug Delivery Systems. Pharmacop Forum 1986; 12:17981807. Shah VP, Tymes NW, Yamamoto LA, Skelly JP. In vitro dissolution prole of transdermal nitroglycerin patches using paddle method. Int J Pharm 1986; 32:243250. Man M, Chang C, Lee PH, Broman CT, Cleary GW. New improved paddle method for determining the in vitro drug release proles of transdermal delivery systems. J Control Release 1993; 27:5968. Roy SD. Preformulation aspects of transdermal drug delivery systems, In: Ghosh TK, Pster WR, Sum SI, eds. Transdermal and Topical Drug Delivery Systems. Buffalo Grove: Interpharm Press, 1997; 139166. Gibson M, ed. Pharmaceutical Preformulation and Formulation. Englewood, IHS Health Group, 2001. Caetano PA, Flynn GL, Farinha AR, Toscano CF, Campos RC. The in vitro release test as a means to obtain the solubility and diffusivity of drugs in semisolids. Proc Int Symp Control Release Bioact Mater 1999; 26:375376. Gourlay S. The pros and cons of transdermal nicotine therapy. Med J Aust 1994; 160: 152. Lin SS, et al. Transdermal nicotine delivery systemsmultiinstitutional cooperative bioequivalence studies. Drug Dev Ind Pharm 1993; 19:27652793. Davila DG, et al. Acute effects of transdermal nicotine on sleep architecture, snoring, and sleep-disordered breathing in nonsmokers. Am J Respir Crit Care Med 1994; 150: 469474. Rose JE, et al. Mecamylamine combined with nicotine skin patch facilitates smoking cessation beyond nicotine patch treatment alone. Clin Pharmacol Ther 1994; 56:86 99. Zevin S, Jacob P III, Benowitz NI. Nicotinemecamylamine interactions. Clin Pharmacol Ther 2000; 68:5866. Benowitz NL. Clinical-pharmacology of transdermal nicotine. Eur J Pharm Biopharm 1995; 41:168174. Fredrickson PA, et al. High-dose transdermal nicotine therapy for heavy smokers safety, tolerability and measurement of nicotine and cotinine levels. Psychopharmacology 1995; 122:215222. Jorenby DE, et al. Varying nicotine patch dose and type of smoking cessation counseling. JAMA 1995; 274:13471352. Kornitzer M, et al. Combined use of nicotine patch and gum in smoking cessation a placebo-controlled clinical-trial. Prev Med 1995; 24:4147. Campbell IA, Prescott RJ, TjederBurton SM. Transdermal nicotine plus support in patients attending hospital with smoking-related diseases: a placebo-controlled study. Respir Med 1996; 90:4751. Lindell G, Lunell E, Graffner H. Transdermally administered nicotine accumulates in gastric juice. Eur J Clin Pharmacol 1996; 51:315318. Fiscella K, Franks P. Cost-effectiveness of the transdermal nicotine patch as an adjunct to physicians smoking cessation counseling. JAMA 1996; 275:12471251.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation 98. 99. 100.

Transdermal

101. 102.

103.

104.

105.

106.

107. 108. 109. 110. 111. 112. 113. 114.

115. 116. 117.

118. 119.

Wasley MA, et al. The cost-effectiveness of the nicotine transdermal patch for smoking cessation. Prev Med 1997; 26:264270. Jaen CR, et al. Patterns of use of a free nicotine patch program for Medicaid and uninsured patients. J Natl Med Assoc 1997; 89:325328. Stapleton JA, Lowin A, Russell MAH. Prescription of transdermal nicotine patches for smoking cessation in general practice: evaluation of cost-effectiveness. Lancet 1999; 354:210215. Benowitz NL, Gourlay SG. Cardiovascular toxicity of nicotine: implications for nicotine replacement therapy. J Am Coll Cardiol 1997; 29:14221431. Benowitz NL, Zevin S, Jacob P. Sources of variability in nicotine and cotinine levels with use of nicotine nasal spray, transdermal nicotine, and cigarette smoking. Br J Clin Pharmacol 1997; 43:259267. Gourlay SG, et al. Determinants of plasma concentrations of nicotine and cotinine during cigarette smoking and transdermal nicotine treatment. Eur J Clin Pharmacol 1997; 51:407414. Greenland S, Sattereld MH, Lanes SF. A meta-analysis to assess the incidence of adverse effects associated with the transdermal nicotine patch. Drug Safety 1998; 18: 297308. Lawson GM, et al. Application of urine nicotine and cotinine excretion rates to assessment of nicotine replacement in light, moderate, and heavy smokers undergoing transdermal therapy. J Clin Pharmacol 1998; 38:510516. Lawson GM, et al. Application of serum nicotine and plasma cotinine concentrations to assessment of nicotine replacement in light, moderate, and heavy smokers undergoing transdermal therapy. J Clin Pharmacol 1998; 38:502509. Lewis SF, et al. Transdermal nicotine replacement for hospitalized patients: a randomized clinical trial. Prev Med 1998; 27:296303. West R, Willis N. Double-blind placebo controlled trial of dextrose tablets and nicotine patch in smoking cessation. Psychopharmacology 1998; 136:201204. Blondal T, et al. Nicotine nasal spray with nicotine patch for smoking cessation: randomised trial with six year follow up. Br Med J 1999; 318:285289. Daughton DM, et al. The smoking cessation efcacy of varying doses of nicotine patch delivery systems 4 to 5 years post-quit day. Prev Med 1999; 28:113118. Gariti P, et al. Cotinine replacement levels for a 21 mg/day transdermal nicotine patch in an outpatient treatment setting. Drug Alcohol Depend 1999; 54:111116. Gourlay SG, et al. Predictors and timing of adverse experiences during transdermal nicotine therapy. Drug Safety 1999; 20:545555. Hajek P, et al. Randomized comparative trial of nicotine polacrilex, a transdermal patch, nasal spray, and an inhaler. Arch Intern Med 1999; 159:20332038. Ji AJ, et al. A new gas chromatographymass spectrometry method for simultaneous determination of total and free trans-3-hydroxycotinine and cotinine in the urine of subjects receiving transdermal nicotine. Clin Chem 1999; 45:8591. Sonderskov J, et al. Nicotine patches in smoking cessation: a randomized trial among over-the-counter customers in Denmark. Ugeskr Laeger 1999; 161:593597. Tonnesen P, et al. Higher dosage nicotine patches increase one-year smoking cessation rates: results from the European CEASE trial. Eur Respir J 1999; 13:238246. Homsy W, et al. Plasma levels of nicotine and safety of smokers wearing transdermal delivery systems during multiple simultaneous intake of nicotine and during exercise. J Clin Pharmacol 1997; 37:728736. Wright LN, et al. Transdermal nicotine replacement in pregnancy: maternal pharmacokinetics and fetal effects. Am J Obstet Gynecol 1997; 176:10901094. Dale LC, et al. Weight change after smoking cessation using variable doses of transdermal nicotine replacement. J Gen Intern Med 1998; 13:915.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters 120. 121. 122. 123. 124. 125.

Brain

126. 127. 128.

129. 130. 131. 132. 133.

134. 135. 136.

137.

138. 139.

140. 141.

Ogburn PL, Jr, et al. Nicotine patch use in pregnant smokers: nicotine and cotinine levels and fetal effects. Am J Obstet Gynecol 1999; 181:736743. West R, et al. A comparison of the abuse liability and dependence potential of nicotine patch, gum, spray and inhaler. Psychopharmacology 2000; 149:198202. Pullan RD, et al. Transdermal nicotine for active ulcerative-colitis. N Engl J Med 1994; 330:811815. Thomas GAO, et al. Transdermal nicotine as maintenance therapy for ulcerative-colitis. N Engl J Med 1995; 332:988992. Thomas GAO, et al. Transdermal nicotine compared with oral prednisolone therapy for active ulcerative colitis. Eur J Gastroenterol Hepatol 1996; 8:769776. Sandborn WJ, et al. Transdermal nicotine for mildly to moderately active ulcerative colitisa randomized, double-blind, placebo-controlled trial. Ann Intern Med 1997; 126:364. Guslandi M, Tittobello A. Pilot trial of nicotine patches as an alternative to corticosteroids in ulcerative colitis. J Gastroenterol 1996; 31:627629. Guslandi M, Tittobello A. Outcome of ulcerative colitis after treatment with transdermal nicotine. Eur J Gastroenterol Hepatol 1998; 10:513515. Guslandi M. Long-term effects of a single course of nicotine treatment in acute ulcerative colitis: remission maintenance in a 12-month follow-up study. Int J Colorectal Dis 1999; 14:261262. SalinPascual RJ, et al. Antidepressant effect of transdermal nicotine patches in nonsmoking patients with major depression. J Clin Psychiatry 1996; 57:387389. Shytle RD, et al. Transdermal nicotine for Tourettes syndrome. Drug Dev Res 1996; 38:290298. Silver AA, Shytle RD, Sanberg PR. Clinical experience with transdermal nicotine patch in Tourette syndrome. CNS Spectr 1999; 4(2):6876. White HK, Levin ED. Four-week nicotine skin patch treatment effects on cognitive performance in Alzheimers disease. Psychopharmacology 1999; 143:158165. Matsushima D, Prevo ME, Gorsline J. Absorption and adverse-effects following topical and oral-administration of 3 transdermal nicotine products to dogs. J Pharm Sci 1995; 84:365369. Woolf A, et al. Self-poisoning among adults using multiple transdermal nicotine patches. J Toxicol Clin Toxicol 1996; 34:691698. Bogaert MG. Clinical pharmacokinetics of nitrates. Cardiovasc Drugs Ther 1994; 8: 693699. Hutt V, et al. Bioequivalence evaluation of the metabolites 1,2-glyceryl and 1,3-glyceryl dinitrate of 2 different glyceryl trinitrate patches after 12-h usage in healthy-volunteers. ArzneimitteForschung 1994; 44:13171321. Devita C, et al. Gissi-3effects of lisinopril and transdermal glyceryl trinitrate singly and together on 6-week mortality and ventricular function after acute myocardial infarction. Lancet 1994; 343:11151122. Andrews R, et al. Inhibition of platelet-aggregation by transdermal glyceryl trinitrate. Br Heart J 1994; 72:575579. Berrazueta JR, et al. Local transdermal glyceryl trinitrate has an antiinammatory action on thrombophlebitis induced by sclerosis of leg varicose veins. Angiology 1994; 45:347351. Parker JO, et al. Intermittent transdermal nitroglycerin therapy in angina-pectoris clinically effective without tolerance or rebound. Circulation 1995; 91:13681374. Serone AP, Angus JA, Wright CE. Baroreffex resetting but no vascular tolerance in response to transdermal glyceryl trinitrate in conscious rabbits. Br J Pharmacol 1996; 118:93104.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation 142.

Transdermal

143. 144.

145.

146. 147.

148. 149. 150. 151. 152. 153. 154.

155.

156.

157. 158.

159. 160.

161. 162.

Ardissino D, et al. Effect of transdermal nitroglycerin or N-acetylcysteine, or both, in the long-term treatment of unstable angina pectoris. J Am Coll Cardiol 1997; 29:941 947. Klemsdal TO, Mundal HH, Gjesdal K. Mechanisms of tolerance during treatment with nitroglycerin patches for 24 h. Eur J Clin Pharmacol 1996; 51:227230. Klemsdal TO, et al. Transdermal nitroglycerin: clinical and pharmacokinetic consequences of renewing the patch and the application site. Eur J Clin Pharmacol 1997; 52:379381. Elkayam U, et al. Double-blind, placebo-controlled study to evaluate the effect of organic nitrates in patients with chronic heart failure treated with angiotensin-converting enzyme inhibition. Circulation 1999; 99:26522657. Teh LS, et al. Sustained-release transdermal glyceryl trinitrate patches as a treatment for primary and secondary raynauds-phenomenon. Br J Rheumatol 1995; 34:636641. Berrazueta JR, et al. Successful treatment of shoulder pain syndrome due to supraspinatus tendinitis with transdermal nitroglycerin. A double blind study. Pain 1996; 66: 6367. Zoli M, et al. Transdermal nitroglycerin in cirrhosis. A 24-hour echo-Doppler study of splanchnic hemodynamics. J Hepatol 1996; 25:498503. Moya RA, et al. Transdermal glyceryl trinitrate in the management of primary dysmenorrhea. Int J Gynecol Obstet 2000; 69:113118. Black RS, et al. Maternal and fetal cardiovascular effects of transdermal glyceryl trinitrate and intravenous ritodrine. Obstet Gynecol 1999; 94:572576. Lees CC, et al. Glyceryl trinitrate and ritodrine in tocolysis: an international multicenter randomized study. Obstet Gynecol 1999; 94:403408. Kovacs P, et al. Effect of transdermal nitroglycerin on glucose-stimulated insulin release in healthy male volunteers. Eur J Clin Invest 2000; 30:4144. Minghetti P, et al. In vitro skin permeation of sinitrodil, a member of a new class of nitrovasodilator drugs. Eur J Pharm Sci 1999; 7:231236. Whittington R, Faulds D. Hormone replacement therapy. 1. Pharmacoeconomic appraisal of its therapeutic use in menopausal symptoms and urogenital estrogen deciency. Pharmacoeconomics 1994; 5:419445. Wiseman LR, McTavish D. Transdermal estradiol norethisteronea review of its pharmacological properties and clinical use in postmenopausal women. Drugs Aging 1994; 4:238256. Creatsas G, et al. Transdermal estradiol plus oral medroxyprogesterone acetate replacement therapy in primary amenorrheic adolescentsclinical, hormonal and metabolic aspects. Maturitas 1994; 18:105114. Normantaylor JQ, et al. Hormone replacement therapy by the transdermal administration of estradiol and norethisterone. Maturitas 1994; 18:221228. Smith RNJ, et al. A randomized comparison over 8 months of 100 mu-g and 200 mug twice weekly doses of transdermal estradiol in the treatment of severe premenstrualsyndrome. Br J Obstet Gynaecol 1995; 102:475484. Pornel B. Efcacy and safety of Menorest in two positive-controlled studies. Eur J Obstet Gynecol Reprod Biol 1996; 64(SS):S35S37. Raudaskoski TH, et al. Transdermal estrogen with a levonorgestrel-releasing intrauterine-device for climacteric complaintsclinical and endometrial responses. Am J Obstet Gynecol 1995; 172(1 pt1):114119. Haines CJ, et al. The effect of percutaneous oestrogen replacement therapy on Lp(a) and other lipoproteins. Maturitas 1995; 22:219225. Crosignani PG, Cortellaro M, Boschetti C. Effects on haemostasis of hormone replacement therapy with transdermal estradiol and oral sequential medroxyprogesterone ac-

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters

Brain

163. 164.

165. 166.

167.

168.

169.

170.

171.

172.

173. 174.

175.

176.

177.

178.

179.

etate: a 1-year, double-blind, placebo-controlled study. Thrombo Haemost 1996; 75: 476480. Macdonald AG, et al. Effects of hormone replacement therapy in rheumatoid-arthritis a double-blind placebo-controlled study. Ann Rheum Dis 1994; 53:5457. Evans SF, Davie MWJ. Low and conventional dose transdermal oestradiol are equally effective at preventing bone loss in spine and femur at all post-menopausal ages. Clin Endocrinol 1996; 44:7984. Roux C. Estrogen therapy in postmenopausal osteoporosis what we know and what we dont. Rev Rhum 1997; 64:402409. Hannon R, et al. Response of biochemical markers of bone turnover to hormone replacement therapy: impact of biological variability. J Bone Miner Res 1998; 13:1124 1133. Prelevic GM, et al. Comparative effects on bone mineral density of tibolone, transdermal estrogen and oral estrogen/progestogen therapy in postmenopausal women. Gynecol Endocrinol 1996; 10:413420. Boonkasemsanti W, et al. The effect of transdermal oestradiol on bleeding pattern, hormonal proles and sex steroid receptor distribution in the endometrium of Norplant users. Hum Reprod 1996; 11(S2):115123. Habiba M, et al. Thrombophilia and lipid prole in post-menopausal women using a new transdermal oestradiol patch. Eur J Obstet Gynecol Reprod Biol 1996; 66:165 168. Perrone G, et al. Effect of oral and transdermal hormone replacement therapy on lipid prole and Lp(a) level in menopausal women with hypercholesterolemia. Int J Fertil Menopausal Stud 1996; 41:509515. Rabe T, et al. Spacing-out of progestinefcacy, tolerability and compliance of two regimens for hormonal replacement in the late postmenopause. Gynecol Endocrinol 1997; 11:383392. Cameron ST, et al. Continuous transdermal oestrogen and interrupted progestogen as a novel bleed-free regimen of hormone replacement therapy for postmenopausal women. Br J Obstet Gynaecol 1997; 104:11841190. Rozenbaum H, et al. Comparison of two estradiol transdermal systems (Oesclim 50 and Estraderm TTS 50). 2. Local skin tolerability. Maturitas 1996; 25:175185. Rozenbaum H, et al. Comparison of two estradiol transdermal systems (Oesclim 50 and Estraderm TTS 50). 1. Tolerability, adhesion and efcacy. Maturitas 1996; 25:161 173. AlAzzawi F, et al. A randomised study to compare the efcacy and safety of a new 17 beta-oestradiol transdermal matrix patch with Estraderm TTS 50 in hysterectomised postmenopausal women. Br J Clin Prac 1997; 51:20. Rozenberg S, Ylikorkala O, Arrenbrecht S. Comparison of continuous and sequential transdermal progestogen with sequential oral progestogen in postmenopausal women using continuous transdermal estrogen: vasomotor symptoms, bleeding patterns, and serum lipids. Int J Fertil Womens Med 1997; 42(S2):376387. Appelberg J et al. Safety and efcacy of transdermal estradiol replacement therapy in postmenopausal liver transplanted womena preliminary report. Acta Obstet Gynecol Scand 1998; 77:660664. Bhathena RK, et al. The inuence of transdermal oestradiol replacement therapy and medroxyprogesterone acetate on serum lipids and lipoproteins. Br J Clin Pharmacol 1998; 45:170172. Blanc B, et al. Continuous hormone replacement therapy for menopause combining nomegestrol acetate and gel, patch, or oral estrogen: a comparison of amenorrhea rates. Clin Ther 1998; 20:901912.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation 180. 181.

Transdermal

182. 183. 184. 185. 186. 187. 188.

189.

190. 191.

192. 193.

194. 195.

196.

197. 198. 199.

200. 201.

Notelovitz M. Clinical experience with a 7-day estrogen patch: principles and practice. Gynecol Endocrinol 1998; 12:249258. Hirvonen E, et al. Effects of transdermal oestrogen therapy in postmenopausal women: a comparative study of an oestradiol gel and an oestradiol delivering patch. Br J Obstet Gynaecol 1997; 104(S16):2631. Hirvonen E, et al. Transdermal oestradiol gel in the treatment of the climacterium: a comparison with oral therapy. Br J Obstet Gynaecol 1997; 104(S16):1925. Jarvinen A, Nykanen S, Paasiniemi L. Absorption and bioavailability of oestradiol from a gel, a patch and a tablet. Maturitas 1999; 32:103113. Vavilis D, et al. The effect of transdermal estradiol on the conjunctiva in postmenopausal women. Eur J Obstet Gynecol Reprod Biol 1997; 72:9396. Coustou D, et al. Estrogen dermatitis. Ann Dermatol Venereol 1998; 125:505508. Carr BR. HRT management: the American experience. Eur J Obstet Gynecol Reprod Biol 1996; 64(SS):S17S20. Andersson TLG, et al. Bioavailability of estradiol from two matrix transdermal delivery systems: menorest and Climara. Maturitas 2000; 34:5764. Archer DF, et al. A randomized comparison of continuous combined transdermal delivery of estradiolnorethindrone acetate and estradiol alone for menopause. Obstet Gynecol 1999; 94:498503. Baracat E, et al. Comparative bioavailability study of two 100-g daily 17-beta-estradiol transdermal delivery systems: once-a-week matrix patch and twice-a-week reservoir patch in healthy postmenopausal women. Curr Ther Res Clin Exp 1999; 60:129 137. Beljic T, et al. Effect of estrogen replacement therapy on cardiac function in postmenopausal women with and without ushes. Gynecol Endocrinol 1999; 13:104112. Bhathena RK, Anklesaria BS, Ganatra AM. The treatment of hypertriglyceridaemia in menopausal women with transdermal oestradiol therapy. Br J Obstet Gynaecol 1999; 106:980982. Brackman F. Oesclim: summary of tolerability and safety. Maturitas 1999; 33(suppl 1):8388. Utian WH, et al. Efcacy and safety of low, standard, and high dosages of an estradiol transdermal system (Esclim) compared with placebo on vasomotor symptoms in highly symptomatic menopausal patients. Am J Obstet Gynecol 1999; 181:7179. Buch AB, et al. Signicant differences in estradiol bioavailability from two similarly labelled estradiol matrix transdermal systems. Climacteric 1999; 2:248253. Caserta R, et al. Efcacy and safety of a weak transdermal estradiol drug delivery system: a pharmacokinetic comparison of two different dosage of formulations. G Ital Ostet Ginecol 2000; 22:279283. Cohen L, et al. Low-dose 17-beta estradiol matrix transdermal system in the treatment of moderate-to-severe hot ushes in postmenopausal women. Curr Ther Res Clin Exp 1999; 60:534547. Cooper C, et al. Matrix delivery transdermal 17beta-estradiol for the prevention of bone loss in postmenopausal women. Osteopor Int 1999; 9:358366. Weiss SR, et al. A randomized controlled trial of four doses of transdermal estradiol for preventing postmenopausal bone loss. Obstet Gynecol 1999; 94:330336. De Aloysio D, et al. Efcacy on climacteric symptoms and safety of low dose estradiol transdermal matrix patchesa randomized, double-blind placebo-controlled study. Arzneimittelforschung 2000; 50:293300. De Vrijer B, et al. Efcacy and tolerability of a new estradiol delivering matrix patch (Estraderm MX) in postmenopausal women. Maturitas 2000; 34:4755. Geisler J, et al. Plasma oestrogen fractions in postmenopausal women receiving hormone replacement therapy: inuence of route of administration and cigarette smoking. J Endocrinol 1999; 162:265270.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters 202.

Brain

203.

204.

205. 206.

207.

208.

209. 210. 211.

212. 213. 214. 215.

216. 217. 218. 219.

220.

221.

Geyer D, et al. Pharmacokinetics of Fem7, a once-weekly, transdermal oestrogen replacement system in healthy, postmenopausal women. Gynecol Obstet Invest 1999; 48: 16. Good WR, et al. Comparison of Alora estradiol matrix transdermal delivery system with oral conjugated equine estrogen therapy in relieving menopausal symptoms. Climacteric 1999; 2:2936. Matsumoto S, et al. A study of the clinical effect of estradiol transdermal therapeutic system alone on pollakisuria and urinary incontinence in postmenopausal woman. Jpn J Urol 2000; 91:501505. Mattsson LA. Clinical experience with continuous combined transdermal hormone replacement therapy. J Menopause 1999; 6:2529. Nachtigall LE, et al. Serum estradiol-binding proles in postmenopausal women undergoing three common estrogen replacement therapies: associations with sex hormonebinding globulin, estradiol, and estrone levels. Menopause 2000; 7:243250. Rohr UD, Nauert C, Stehle B. 17beta-Estradiol delivered by three different matrix patches 50 g/day. A three way cross-over study in 21 postmenopausal women. Maturitas 1999; 33:4558. Rovati LC, et al. Efcacy on climacteric symptoms of a new estradiol transdermal patch with active matrix in comparison with a reference reservoir patch: two long-term randomized multicenter parallel-group studies. Arzneimittelforschung 1999; 49:933 943. Wojcicki J, et al. Effects of estradiol and norethisterone replacement therapy on the pharmacokinetics of phenazone. Med Sci Mon 1999; 5:754757. Burry KA, et al. Percutaneous absorption of progesterone in postmenopausal women treated with transdermal estrogen. Am J Obstet Gynecol 1999; 180:15041511. Carey BJ, et al. A study to evaluate serum and urinary hormone levels following short and long term administration of two regimens of progesterone cream in postmenopausal women. Br J Obstet Gynaecol 2000; 107:722726. Leonetti HB, Longo S, Anasti JN. Transdermal progesterone cream for vasomotor symptoms and postmenopausal bone loss. Obstet Gynecol 1999; 225228. Chien TY, et al. Transdermal contraceptive delivery systempreclinical development and clinical-assessment. Drug Dev Ind Pharm 1994; 20:633664. Sitrukware R. Transdermal application of steroid-hormones for contraception. J Steroid Biochem Mol Biol 1995; 53:247251. Asthana S, et al. Cognitive and neuroendocrine response to transdermal estrogen in postmenopausal women with Alzheimers disease: results of a placebo-controlled, double-blind, pilot study. Psychoneuroendocrinology 1999; 24:657677. Gerber GS, et al. Transdermal estrogen in the treatment of hot ushes in men with prostate cancer. Urology 2000; 55:97101. Cofrancesco J, Dobs AS. Transdermal testosterone delivery systems. Endocrinologist 1996; 6:207213. Behre HM, et al. Long-term substitution therapy of hypogonadal men with transscrotal testosterone over 710 years. Clin Endocrinol 1999; 50:629635. Kenny AM, Prestwood KM, Raisz LG. Short-term effects of intramuscular and transdermal testosterone on bone turnover, prostate symptoms, cholesterol, and hematocrit in men over age 70 with low testosterone levels. Endocr Res 2000; 26:153168. Meikle AW, et al. Pharmaockinetics and metabolism of a permeation-enhanced testosterone transdermal system in hypogonadal men: inuence of application sitea clinical research center study. J Clin Endocrinol Metab 1996; 81:18321840. Arver S, et al. Long-term efcacy and safety of a permeation-enhanced testosterone transdermal system in hypogonadal men. Clin Endocrinol 1997; 47:727737.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation 222.

Transdermal

223.

224. 225. 226. 227. 228.

229.

230.

231. 232. 233.

234.

235. 236. 237. 238. 239. 240. 241. 242. 243.

Bhasin S, et al. Effects of testosterone replacement with a nongenital, transdermal system, Androderm, in human immunodeciency virus-infected men with low testosterone levels. J Clin Endocrinol Metab 1998; 83:31553162. Miller K, et al. Transdermal testosterone administration in women with acquired immunodeciency syndrome wasting: a pilot study. J Clin Endocrinol Metab 1998; 83: 27172725. Jain P, Rademaker AW, McVary KT. Testosterone supplementation for erectile dysfunction: results of a meta-analysis. J Urol 2000; 164:371375. Davis SR. The therapeutic use of androgens in women. J Steroid Biochem Mol Biol 1999; 69:177184. Mazer NA. New clinical applications of transdermal testosterone delivery in men and women. J Control Release 2000; 65:303315. Parker S, Armitage M. Experience with transdermal testosterone replacement therapy for hypogonadal men. Clin Endocrinol 1999; 50:5762. Rolf C, et al. Pharmacokinetics of new testosterone transdermal therapeutic systems in gonadotropin-releasing hormone antagonist-suppressed normal men. Exp Clin Endocrinol Diabetes 1999; 107:6369. Tan KCB, Shiu SWM, Kung AWC. Alterations in hepatic lipase and lipoprotein subfractions with transdermal testosterone replacement therapy. Clin Endocrinol 1999; 51: 765769. Wilson DE, et al. Use of topical corticosteroid pretreatment to reduce the incidence and severity of skin reactions associated with testosterone transdermal therapy. Clin Ther 1998; 20:299306. Simmonds MA. Transdermal fentanylclinical development in the United States. Anticancer Drugs 1995; 6(S3):3538. Sandler AN. Transdermal fentanyl for pain reliefguidelines for appropriate use. CNS Drugs 1997; 7:442451. Korte W, Morant R. Transdermal fentanyl in uncontrolled cancer paintitration on a day-to-day basis as a procedure for safe and effective dose-ndinga pilot-study in 20 patients. Supp Care Cancer 1994; 2:123127. Vanbastelaere M, Rolly G, Abdullah NM. Postoperative analgesia and plasma levels after transdermal fentanyl for orthopedic surgerydouble-blind comparison with placebo. J Clin Anesth 1995; 7:2630. Fiset P, et al. Biopharmaceutics of a new transdermal fentanyl device. Anesthesiology 1995; 83:459469. Christensen ML, et al. Transdermal fentanyl administration in children and adolescents with sickle cell pain crisis. J Pediatr Hematol Oncol 1996; 18:372376. Jeal W, Beneld P. Transdermal fentanyla review of its pharmacological properties and therapeutic efcacy in pain control. Drugs 1997; 53:109138. Hardwick WE, King WD, Palmisano PA. Respiratory depression in a child unintentionally exposed to transdermal fentanyl patch. South Med J 1997; 90:962964. Sloan PA, Moulin DE, Hays H. A clinical evaluation of transdermal therapeutic system fentanyl for the treatment of cancer pain. J Pain Sympt Manage 1998; 16:102111. Payne R, et al. Quality of life and cancer pain: satisfaction and side effects with transdermal fentanyl versus oral morphone. J Clin Oncol 1998; 16:15881593. Dellemijn PLI, vanDuijn H, Vanneste JAL. Prolonged treatment with transdermal fentanyl in neuropathic pain. J Pain Sympt Manage 1998; 16:220229. Kyles AE, Papich M, Hardie EM. Disposition of transdermally administered fentanyl in dogs. Am J Vet Res 1996; 57:715719. Egger CM, et al. Comparison of plasma fentanyl concentrations by using three transdermal fentanyl patch sizes in dogs. Vet Surg 1998; 27:159166.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Walters 244. 245. 246. 247. 248.

Brain

249. 250. 251. 252.

253. 254. 255. 256. 257. 258. 259.

260. 261.

262.

263. 264. 265. 266.

Collins JJ, et al. Transdermal fentanyl in children with cancer pain: feasibility, tolerability, and pharmacokinetic correlates. J Pediatr 1999; 134:319323. Jakobsson M, Strang P. Fentanyl patches for the treatment of pain in dying cancer patients. Anticancer Res 1999; 19:44414442. Radbruch L et al. Constipation and the use of laxatives: a comparison between transdermal fentanyl and oral morphine. Palliat Med 2000; 14:111119. Haazen L, et al. The constipation-inducing potential of morphine and transdermal fentanyl. Eur J Pain 1999; 3(suppl A):915. Flynn TN, Guest JF. Cost of managing constipation among terminally ill cancer patients treated with transdermal fentanyl and 12 hour sustained-release morphine in Canada. J Med Econ 1999; 2:1532. Elsner F, et al. Switching opioids to transdermal fentanyl in a clinical setting. Schmerz 1999; 13:273278. Ritchie LD. A clinical evaluation of urbiprofen LAT and piroxicam gel: a multicentre study in general practice. Clin Rheumatol 1996; 15:243247. Fujimura A, et al. Pharmacokinetics and pharmacodynamics of a new transdermal clonidine, M-5041t, in healthy-subjects. J Clin Pharmacol 1993; 33:11921200. Fujimura A, et al. Comparison of the pharmacokinetics, pharmacodynamics, and safety of oral (Catapres) and transdermal (M-5041t) Clonidine in healthy subjects. J Clin Pharmacol 1994; 34:260265. Toon S, Phase-I pharmacokinetic assessment of the clonidine transdermal therapeutic system. Eur J Pharm Biopharm 1995; 41:184188. Ellis JE, et al. Premedication with oral and transdermal clonidine provides safe and efcacious postoperative sympatholysis. Anesth Analg 1994; 79:11331140. Dorman T, et al. Effects of clonidine on prolonged postoperative sympathetic response. Crit Care Med 1997; 25:11471152. Owen MD, et al. Postoperative analgesia using a low-dose, oraltransdermal clonidine combination: lack of clinical efcacy. J Clin Anesth 1997; 9:814. Dandrea G, et al. Efcacy of transdermal clonidine in short-term treatment of cluster headachea pilot-study. Cephalalgia 1995; 15:430433. Byassmith MG, et al. Transdermal clonidine compared to placebo in painful diabetic neuropathy using a 2-stage enriched enrollment design. Pain 1995; 60:267274. Dias VC, et al. Clinical experience with transdermal clonidine in African-American and Hispanic-American patients with hypertension: evaluation from a 12-week prospective, open-label clinical trial in community-based clinics. Am J Ther 1999; 6:19 24. Dinslage S, et al. A new transdermal delivery system for pilocarpine in glaucoma treatment. Ger J Ophthalmol 1996; 5:275280. Bodick NC, et al. The selective muscarinic agonist xanomeline improves both the cognitive decits and behavioral symptoms of Alzheimer disease. Alzheimer Dis Assoc Disord 1997; 11(S4):S16S22. Douma WR, et al. Effects of transdermal scopolamine on pulmonary function, symptoms and bronchial hyperresponsiveness to methacholine. Eur J Pharm Sci 1997; 5: 327334. Honkavaara P, Pyykko I. Effects of atropine and scopolamine on bradycardia and emetic symptoms in otoplasty. Laryngoscope 1999; 109:108112. Gowans J, et al. The effects of scopolamine and cyclizine on visualvestibular interaction in humans. J Vestib Res Equilibr Orient 2000; 10:8792. Sacktor N, et al. Transdermal selegiline in HIV-associated cognitive impairment: pilot, placebo-controlled study. Neurology 2000; 54:233235. Rama RP, Diwan PV. Comparative in-vivo evaluation of diltiazem hydrochloride following oral and transdermal administration in rabbits. Indian J Pharmacol 1999; 31: 294298.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.

Formulation 267. 268. 269. 270. 271. 272. 273.

Transdermal

Gillson G, et al. Transdermal histamine in multiple sclerosis: part oneclinical experience. Altern Med Rev 1999; 4:424428. Martin GP, et al. The inuence of hydrocolloid patch composition on the bioavailability of triamcinolone acetonide in humans. Drug Dev Ind Pharm, 2000; 26:3543. Sutinen R, et al. Water-activated, pH-controlled patch in transdermal administration of timolol II. Drug absorption and skin irritation. Eur J Pharm Sci 2000; 11:2531. Sutinen R, Paronen P, Urtti A. Water-activated, pH-controlled patch in transdermal administration of timolol I. Preclinical tests. Eur J Pharm Sci 2000; 11:1924. McVary KT, et al. Topical prostaglandin E1 SEPA gel for the treatment of erectile dysfunction. J Urol 1999; 162:726731. Comer AM, Lamb HM. Lidocaine patch 5%. Drugs 59:245249. Roberge RJ, Krenzelok EP, Mrvos R. Transdermal drug delivery system exposure outcomes. J Emerg Med 18:147151.

Copyright 2002 by Marcel Dekker, Inc. All Rights Reserved.