`
`DRUGS AND THE PHARMACEUTICAL SCIENCES
`
`VOLUME 119
`
`WUTTe)
`TUESME)
`Formulations
`
`Vehicle
`
`Stratum Corneum Lipid
`
`Solute >@ .
`
`i 6
`
`iH
`
`eq. 14
`
`
`
`
`
`ov
`
`a
`
`*
`
`molecule
`
`-
`vows COmotecule
`ieeea.
`Interaction between
`into stratum corneum
`solute and vehicle
`___(ai- ayt Ksc t
`eq. 14
`
`Si
`
`“push” effect
`
`“pull” effect
`
`ov a
`Vehicle O
`
`;
`
`Increasing affinity
`for
`stratum comeum
`
`Ss
`
`KseT
`
`edited by
`Kenneth A. Walters
`
`Noven Pharmaceuticals, Inc.
`EX2013
`Mylan Tech., Inc. v. Noven Pharma., Inc.
`T= -PLOEMOOR EZ!
`
`eel
`
`

`

`
`
`
`
`An im
`the mi
`impor
`been 1)
`Forma
`and fn
`extens
`
`velopt
`
`have i
`of tra
`in mo
`the bi
`
`strates
`volum1
`
`under,
`atives
`
`therar
`
`ISBN: 0-8247-9889.9
`
`This bookis printed on acid-free paper.
`
`Headquarters
`Marcel Dekker, Inc.
`270 Madison Avenue, New York, NY 10016
`tel: 212-696-9000; fax: 212-685-4540
`
`Eastern Hemisphere Distribution
`Marcel Dekker, AG
`Hutgasse 4, Postfach 812, CH-4001 Basel, Switzerland
`tel: 41-61-261-8482; fax: 41-61-261-8896
`
`World Wide Web
`http://www.dekker.com
`
`The publisheroffers discounts on this book when orderedin bulk quantities. For more information,
`write to Special Sales/Professional Marketing at the headquarters address above.
`
`Copyright © 2002 by Marcel Dekker, Inc. All Rights Reserved.
`
`Neither this book nor any part may be reproduced or transmitted in any form or by any means,
`electronic or mechanical, including photocopying, microfilming, and recording, or by anyinfor-
`mation storage and retrieval system, without permission in writing from the publisher.
`
`Current printing (last digit):
`19 8
`7 65 43 2
`
`|
`
`PRINTED IN THE UNITED STATES OF AMERICA
`
`
`
`

`

`
`
`Contents
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`~~] Dermatological Formulation and Transdermal Systems
`Kenneth A. Walters and Keith R. Brain
`
`Bioavailability and Bioequivalence of Dermatological
`Formulations
`Christian Surber and Adrian F. Davis
`
`Scale-up of Dermatological Dosage Forms: A Case for
`Multivariate Optimization and Product Homogeneity
`Orest Olejnik and Bruce A. Firestone
`
`Safety Considerations for Dermal and Transdermal Formulations
`Peter J. Dykes and Anthony D, Pearse
`
`Transdermal Delivery and Cutaneous Reactions
`Jagdish Singh and Howard I. Maibach
`
`Index
`
`319
`
`401
`
`499
`
`31]
`
`529
`
`549
`
`Co
`
`Keith
`
`Shere
`land,|
`
`Adria
`
`Englar
`
`Peter
`Wales
`
`Bruce
`
`Rober
`
`Jonatl
`Chath:
`
`C. Col
`Cardif
`
`Howal
`School
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`

`

`338
`
`Walters and Brain
`
`2. Biopharmaceutical Considerations
`
`A fundamental consideration in the development of transdermal therapeutic systems
`is whetherthe dermal delivery route can provide the requisite bioavailability for drug
`effectiveness. This is ultimately determined by the skin-penetrationrate of the drug,
`the potential for metabolism during permeation across the skin, and the biological
`half-life of the drug. Penetration rates may be modified, if necessary, by the use of
`penetration enhancers, but drug metabolism and plasma clearance cannot be influ-
`enced by any simple means. Although prediction of skin penetration and bioavail-
`ability of drugs from transdermal therapeutic systems has becn reasonablyaccurate,
`there is no doubt that testing of formulated patches in vitro and in vivo will continue
`to be the most accurate means ofevaluating their uscfulness.
`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 Chapter5). 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 provideslittle 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 foruse in in vitro systems. Although methodsare 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 andartificial neural networks have been used in the optimi
`zation of transdermal drug delivery formulations using in vitro skin permeation tech.
`niques (62-64). For example, Kandimalla et al.
`(63) optimized a vehicle for the
`transdermal delivery of melatonin using the response surface method and artificial
`neural networks, Briefly, three solvents (water, ethanol, and propylene glycol) were
`examined either as single solvents or binary and ternary mixtures. Measurements of
`skin flux, 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 artificial neural network employing a two-layered back-propagation network with
`all ten design points in the hidden layer. Predictability of flux using both statistical
`techniques was good (Table 6), suggesting that such models may be useful in pre-
`liminary formulation optimization.
`A major drawbackoftransdermal delivery systemsis the potential for localized
`irritant andallergic cutaneous reactions. At the earlier stages of formulation devel-
`opment, 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 maystay in place for pro-
`longed periods. The degree of primary and chronic irritation, and the potential
`to
`cause contact allergy, photoirritation, and photoallergy should be determined. Nor-
`mally, the drug and excipients are initially separately evaluated for contactirritation
`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
`
`—_—
`
`Drug Fe
`
`Table 6
`
`Applicat
`vehicle®
`
`
`
`W:E:P
`W:E (40
`W:P (40
`
`“Flux wa
`methods
`"W, wate!
`“Data are
`Source: F
`
`always
`Fortuna
`and mil
`
`3. De
`
`All pate
`three b:
`and dru
`
`by ara
`of trans
`ing the
`which i
`
`designe
`from a
`adhesiv
`mation
`mers, ¢
`Additio
`interfer
`other fc
`and de:
`
`system
`selectio
`ical or
`A
`adhesiv
`can be
`rheolog
`sive to
`stretch
`formati
`andlon
`initial <
`residue
`
`
`
`

`

`Drug Formulation and Transdermal Systems
`
`339
`
`
`
`Table 6 Experimental Versus Predicted Flux® of Melatonin
`
`Application
`Experimental
`ANN Prediction
`RSMPrediction
`vehicle”
`(ug/em* ho')
`(ug/em* ho!)
`(ug/em? h')
`
`
`12.73
`12.17
`11.32 + 0.86
`W:E:P (20:60:20)
`11.76
`11.83
`10.89 + 1.36
`W:E (40:60)
`
`
`
`7.54 + 1.39 6.75W:P (40:60) 7.50
`
`"Flux was predicted on the basis of artificial neural networks (ANN) or response surface
`methods (RSM).
`"W, water: E, ethanol; P, propylene glycol.
`‘Data are means + standard deviation (2 = 3) and represent flux through rat dorsal skin.
`Source: Ref. 63.
`
`always be carried out using the final and complete formulation in humanvolunteers.
`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 manydifferences in the design
`of transdermal delivery systems, several features are common to all systems includ-
`ing 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 for-
`mation of a new chemical] species, must be fully assessed. Similarly, residual mono-
`mers, 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 phys-
`ical 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 assessmentof rheological parameters. Standard
`rheological tests include creep compliance (which measuresthe ability of the adhe-
`sive to flowinto surface irregularities), elastic index (which determines the extent of
`stretch or deformation as a function of load and time) and recovery following de-
`formation. Skin-adhesion performance is based on several properties, such as initial
`and long-term adhesion,lift, and residue. The adhesive must be soft enoughto ensure
`initial adhesion, yet have sufficient cohesive strength to remove cleanly, leaving no
`residue. Because premature lift will
`interfere with drug delivery, the cohesive and
`
`
`
`

`

`340
`
`Walters and Brain
`
`(a) DRUG IN ADHESIVE
`
`(b) DRUG IN MATRIX
`
`(c) DRUG IN RESERVOIR
`
`(d) PERIPHERAL ADHESIVE
`
`
`——
`—
`rt
`
`Backing Layer
`
`Adhesive Layer
`Release Liner
`
`Backing Layer
`
`Polymeric Matrix
`
`Adhesive Layer
`Release Liner
`
`Backing Layer
`Reservoir
`
`Membrane
`
`Adhesive Layer
`Release Liner
`
`Backing Layer
`
`Adhesive Layer
`Reservoir
`
`Membrane
`
`ReleaseLiner
`
`Drug Fo
`
`lead toe
`increase
`Th
`material:
`acetate,
`to the dr
`those thi
`water lo
`
`subpatck
`layer wi
`films or
`rubber-b
`adhesive
`
`4, Dru
`
`The thre
`
`system F
`in-adhes
`for exan
`combina
`Me
`
`drug resi
`size anc
`
`through
`material:
`
`copolym
`acrylonil
`hancer (
`also bee
`
`drug (6¢€
`brane,
`is
`mulated
`AS
`dermal
`fabricate
`nitroglyc
`ture and
`hexane),
`amount
`fine-tuni
`
`the poro
`In
`membra
`evaluate
`membra
`
`widely «
`cluded t]
`the fabr
`
`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 contoursit 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
`
`
`
`

`

` Drug Formulation and Transdermal Systems
`
`
`341
`
`lead to excessive hydration and problemsassociated with microbial growth that may
`increasethe possibility ofirritation orsensitizationto the various system components.
`The backing material and release liner can be fabricated from a variety of
`materials, including polyvinylchloride, polyethylene, polypropylene, ethylene vinyl
`acctate, and aluminiumfoil. The principal requirementis 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 sufficient resistance to transepidermal
`waterloss 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 withoutlifting off any of the adhesive to whichit is bound, Liners are usually
`films or coated papers, andsilicone release coatings are used with acrylate- and
`rubber-based adhesive systems, whereas fluorocarbon coatings are used withsilicone
`adhesives (65),
`
`4, Drug and Enhancer Incorporation
`The three principal methods for incorporating the active species into a transdermal
`system have led to the looseclassification of patches as membrane, matrix, or drug-
`in-adhesive types. It is, however, quite possible to combine the main types ofpatch;
`for example, by placing a membrane over a matrix. or using a drug-in-adhesive in
`combination with a membrane—matrix device to deliver an initial bolus dose.
`Membrane patches contain a delivery rate-controlling membrane between the
`drug reservoir andthe skin. Microporous membranes, which control drug flux by the
`size and tortuosity of pores in the membrane. or dense polymeric membranes,
`through whichthe drug permeates by dissolution anddiffusion, may be used. Several
`materials can be used as rate-controlling membranes(e.g., ethylene—vinyl acetate
`copolymers, silicones, high-density polyethylene, polyester elastomers, and poly-
`acrylonitrile). Ideally, the membrane should be permeable only to the drug and en-
`hancer(if present) and shouldretain other formulation excipients. Membranes have
`also been designed suchthat they allow differential permeation of an enhancer and
`drug (66-68). This type of membrane, sometimes designated as a one-way mem-
`brane, is useful when the drug is present in the adhesive and the enhancer is for-
`mulated in a reservoir.
`Asymmetric polymeric membranes have also been evaluated for use in trans-
`dermal delivery systems (69). Asymmetric poly(4-methy]-1-pentene) membranes,
`fabricated using a dry—wet inversion method. were used to control the delivery of
`nitroglycerin. The release rates of nitroglycerin were strongly influenced by the na-
`ture and amount of the nonsolvent (butanol) used, together with the solvent(cyclo-
`hexane),
`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
`fine-tuning delivery of a drug through a given membrane by subtle adjustment of
`the porosity creates some exciting newpossibilities in transdermal technology (70).
`In all marketed membrane-controlled transdermal systems, the rate-controlling
`membraneis 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 shellfish. Tt was con-
`cluded that the permeability of both biopolymers could be readily adjusted byaltering
`the fabrication method or cross-linking and, because these polymers were biocom-
`
`
`
`
`
`

`

`342
`
`Walters and Brain
`
`patible, they were more suitable for use as rate-controlling membranesin 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 aqueous —al-
`coholic solutions and gels, with or without various cosolvents, and polymeric ma-
`terials). A definite 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 res-
`ervoir 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 systemis 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, poly-
`urethanes, polyvinyl alcohol, polyvinylpyrrolidones, and such, may be considered
`the drug reservoir. Several steps are involved in the drug deliveryprocess: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
`ofthe skin. Manyvariables may affect the dissolution and diffusion rates, making it
`particularly difficult, but not impossible, to predict release rates from experimental
`or prototype formulations (72). For a drug to be released from a polymeric matrix
`underzero-order kinetics, the drug must be maintained at saturation in the fluid 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 ofthese are illustrated by a study on release of
`several drugs fromsilicone matrices (73). Silicone medical-grade elastomers (poly-
`dimethylsiloxanes)are flexible, lipophilic polymers, with excellent compatibility with
`biological
`tissues,
`that can be coformulated with hydrophilic excipients, such as
`glycerol, andinert fillers, 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 ofinert fillers
`(titanium dioxide or barium sulfate) tended to reduce the release rate. Hydrophilic
`drug-release rates from polydimcthylsiloxane matrices were also increased by up to
`three orders of magnitude using polydimethylsiloxane—polyethylene oxide graft co-
`polymers (74). These data demonstrate that release rates can be modulated to achieve
`a desired profile by simple formulation modification.
`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 enhancerif present, in an adhesive mixture that is subsequently coated
`onto a backing membrane, such as a polyester film, to produce an adhesive tape.
`This simplicity is, however, deceptive and several factors, involving potential inter-
`action between drug or enhancerand the adhesive, need to be considered. These can
`involve chemical interactions resulting in interference with adhesive performance,
`breakdownof 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.
`
`Drug F
`
`Ir
`
`drug-in
`exampl
`in loss
`(75). A
`was to
`and the
`a drug-
`is that
`the ski
`
`drug-in
`resultin
`
`strengtl
`this me
`be ove
`
`System
`of the
`are ava
`
`serious
`
`ester Ly
`lished.
`
`skin pe
`copoly1
`W
`
`appreci
`is relez
`enhanc
`
`may he
`enhanc
`to obse
`
`5.
`
`Sy
`
`The mz
`system:
`the dru
`
`quired)
`moldin.
`layer, r
`aging. '
`cess: te
`laminat
`tent (76
`A
`content
`
`mity ey
`assayin
`ing dru
`Manufz
`
`
`
`

`

`
`
`Drug Formulation and Transdermal Systems
`
`343
`
`into a
`Incorporation of other excipients, such as skin permeation enhancers,
`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 andskin contact could not be maintained overthe required period
`(75). A strategy to reduce the influence of drug and enhanceron 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 drug—enhancer-releasing layer. One disadvantage of this type of system, however,
`is that the drug—enhancer-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 highlevels 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 varicty 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 estab-
`lished.
`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 systemsit is important to
`appreciate that it is a fundamental requirementthat the enhancer, as well as the drug,
`is released by the adhesive. Furthermore,
`it
`is probable that the presence ofthe
`enhancer may increase ski permeation of other formulation excipients and that this
`may have an influence onlocal toxicity. Much remains to be done in the field of
`enhancerincorporation 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 re-
`quired) with the reservoir, matrix, or adhesive; casting into films and drying (or
`molding and curing);
`laminating with other structural components (e.g., backing
`layer, rate-controlling membrane, and release liner); die-cutting: and finally, pack-
`aging. Casting and lamination are the mostcritical steps in the manufacturing pro-
`cess: tensions and pressures must be carefully controlled to provide a wrinkle-free
`laminate that ensures reproducible adhesive-coating thickness and uniform drug con-
`tent (79).
`As with all controlled-release delivery systems, final product checks include
`content uniformity, release-rate determination, and physical testing. Content-unifor-
`mity evaluation involves removing a random sample of patches from a batch and
`assaying the amount of drug present. Of the several methods available for determin-
`ing drug release rates from controlled-release formulations, the U.S. Pharmaceutical
`Manufacturers Association (PMA) Committee (80) has recommended three:
`the
`
`
`
`
`